general results of the reflex project

8/6/2019 General Results of the REFLEX Project

http://slidepdf.com/reader/full/general-results-of-the-reflex-project 1/312



CHAPTER

Norway 52.0 Norway 49.7Estonia 51.0 Estonia 47.8Germany 45.6 Switzerland 46.3

Spain 44.9 Czech Republic 45.8Switzerland 42.4 Spain 44.5France 36.9 France 33.9Total 51.4 Total 51.6

. . Organisations at the ore ront o innovation

Te third question about the position o the company regarding its initiative ininnovation provides a complementary perspective. A ter a rapid consideration o the countries according to this dimension, the characteristics o the market, size andsector will be taken into account. A last development will compare the in ormationprovided by the REFLEX survey with the ones derived rom the European InnovationScoreboard.

able . provides answers and to the question “Is your organisation or - in caseo sel -employment – are you normally at the ore ront when it comes to adoptinginnovations, new knowledge or new methods, or is it more a ollower?” When lookingat the answers, Swiss organisations appear better ranked than in the previous tables.

At the bottom o the distribution, organisations rom the three “southern” Europeancountries o the sample, France, Spain and Italy, look more requently like ollowers.

Table 4.6Proportion o graduates working in organisations at the ore ront o innovation by country

Proportion o graduates working in organisationsat the ore ront o innovation (private sector)

Finland 60,7

Austria 56,3

Switzerland 55,0

Czech Republic 53,4Norway 53,2

United Kingdom 49,6

Germany 49,5

Estonia 47,7

Netherlands 45,6

Belgium 42,2

France 40,3

Spain 38,8

Italy 38,7

Total 48,0



The graduates in the knowledge and innovation society

Te main characteristics o the organisations at the ore ront o innovation appearclose to the ones already observed. One hal o graduates ( %) working in privatecompanies acing a very strong competition declare these organisations are at the

ore ront o innovation; this is the case or .8% o graduates in organisations wherethe strength o competition is very weak (see able . ).

Table 4.7Proportion o graduates working in organisations at the ore ront o innovation, and strength o competition

Strength o competition Proportion o graduates working inorganisations at the ore ront o innovation

Private sector Public sector1. Very weak 37.8 28.8

2. 45.2 35.43. 40.3 39.8

4. 48.6 45.0

Very strong 52.2 42.3

Total 48.1 37.2

Te same holds true or the scope o operations. Graduates working in organisations with an international scope o operations declare more requently they are at theore ront o innovation ( 8% against . % or private companies and . % against

. % or public organisations) (see able .8). When the size o the organisation isconcerned, big organisations appear more requently at the ore ront o innovation, aresult which is not particularly original: 8% o graduates working in private compa-nies with more than employees consider their company at the ore ront in inno-vation against 9% or graduates in organisations smaller than employees (see able.9). Te proportion increases regularly with the size. Te same holds true or publicorganisations, but with a more narrow range between large and small organisations.

Table 4.8Proportion o graduates working in organisations at the ore ront o innovation, and scope o oper-ations

Scope o operations Proportion o graduates working inorganisations at the ore ront o innovation

Private sector Public sectorLocal 33.2 26.4

Regional 38.4 35.7

National 42.9 40.3

International 58.0 55.1

Total 48.2 37.5



CHAPTER

Table 4.9Proportion o graduates working in organisations at the ore ront o innovation, and size o organi-sation

Size o organisation Proportion o graduates working inorganisations at the ore ront o innovation

Private sector Public sector10-49 43.1 38.6

50-99 44.8 39.7100-249 46.2 38.1

250-999 50.9 35.4

1000 or more 58.0 38.4

Total 48.5 37.7

Table 4.10

Proportion o graduates working in organisations at the ore ront o innovation, and sector o activity

Proportion o graduatesworking in organisation

at the ore ront o innovation

N Total

International Standard Industrial Classi cation

C - Mining and quarrying 55.6 133D – Manu acturing 54.7 2641E - Electricity, gas and water supply 52.1 163I - Transport, storage and communications 51.6 818K - Real estate, renting and business activities 49.3 3942J - Financial intermediation 46.9 1207A - Agriculture, hunting and orestry, shing 46.8 190G - Wholesale and retailtrade; repair o motor vehicles,motorcycles and personal and household goods 40.8 1004M – Education 40.8 3897F – Construction 39.4 569H - Hotels and restaurants 38.2 136N - Health and social work 37.4 3457

O - Other community, social and personal service activities 36.9 1035P - Activities o private households as employers andundi erentiated production activities o private households 36.4 11Q - Extraterritorial organizations and bodies 30.0 20L - Public administration and de ence; compulsory social security 29.8 1791Total 43.4 21014

Te position regarding innovation varies with the sector o activity. Whereas morethan hal o graduates working in “Mining and quarrying”, in “Manu acturing”, in“Electricity, gas and water supply”, in “ ransport, storage and communications”, orin “Real estate, renting and business activities” consider their company is at the ore-ront o innovation, less than % o graduates do the same in service activities such




as “Hotels and restaurants” or “Health and social work”, without mentioning “Publicadministration” (see able . ).

It may be interesting to compare these results with the ones provided by the “EuropeanInnovation Scoreboard / Comparative Analysis o Innovation Per ormance”prepared or the Commission, under the Lisbon strategy. Obviously, the exercise o this report is o a di erent nature, since it uses indicators, split into ve maincategories (input/innovation drivers, input/knowledge creation, input/innovationand entrepreneurship, output/applications, output/intellectual property) to evaluateand to compare the innovation per ormance o the EU member states and some othercountries. Eight indicators have been selected or the comparison, which seem to bemore directly related to innovation in organisations and to innovative employment:business R&D expenditures (% o GDP), share o medium-high-tech and high-techR&D (% o manu acturing R&D expenditures), share o enterprises receiving publicunding or innovation, SME’s using organisational innovation (% o all SMEs), saleso new-to-market products (% o total turnover), sales o new-to- rm products (%o total turnover), employment in medium-high and high-tech manu acturing (% o total work orce).

We can compare the ranking o the countries according to the selected EuropeanInnovation Scoreboard indicators and according to the proportion o graduates inorganisations to be considered at the ore ront o innovation (see able . ).

Table 4.11Ranking o countries according to the European Innovation Scoreboard indicators and to theREFLEX survey

EIS REFLEX (innovation leaders)

Country Position Country Position

Germany 1 Finland 1

Finland 2 Austria 2

Switzerland 3 Switzerland 3

Austria 4 Czech Republic 4

Belgium 5 Norway 5Italy 6 United Kingdom 6

Czech Republic 7 Germany 7

United Kingdom 8 Estonia 8

France 9 Netherlands 9

Netherlands 10 Belgium 10

Spain 11 France 11

Estonia 12 Spain 12

Norway 13 Italy 13




sector. On the other hand, French and German graduates appear to be o ten less asso-ciated to the innovative process. Special attention may be given to the case o Estoniaand Czech Republic, since graduates in these “new” European countries appear to be

requently involved in innovation activities.Table 4.12Proportion o graduates who play a role in introducing innovations in their organisation

Product or service

Private sector Public sector

Norway 60.7 Estonia 54.9

Finland 56.7 Finland 49.1

Estonia 55.2 Italy 48.0

Czech Republic 53.2 Norway 47.4

Netherlands 52.2 Czech Republic 46.6

Austria 49.7 Netherlands 45.0

Germany 48.9 United Kingdom 44.9

Switzerland 48.7 Belgium 40.0

Italy 46.7 France 39.5

United Kingdom 45.9 Austria 36.9

Belgium 43.7 Germany 33.9

France 43.0 Switzerland 31.6

Spain 41.4 Spain 30.7

Total 49.2 Total 42.7

Technology, tools or instruments


Norway 45.7 Estonia 44.2

Finland 45.5 Italy 41.6

Estonia 43.5 Czech Republic 41.3

Czech Republic 42.4 Finland 39.3

Italy 40.7 Spain 34.7

Switzerland 37.0 France 33.5

United Kingdom 36.5 Belgium 33.3

Spain 36.5 United Kingdom 31.6

France 36.3 Netherlands 28.5

Austria 34.5 Norway 27.2

Netherlands 34.0 Austria 26.5

Belgium 33.8 Switzerland 26.1

Germany 32.7 Germany 18.2




CHAPTER

Knowledge and methods


Czech Republic 69.3 Estonia 76.2

Norway 69.1 Czech Republic 72.9Finland 66.8 Finland 67.0

Estonia 66.4 Norway 62.9

Netherlands 60.4 United Kingdom 62.4

France 59.4 France 61.4

Italy 59.1 Belgium 61.1

Belgium 58.8 Germany 60.3

United Kingdom 58.8 Netherlands 59.2

Austria 56.3 Italy 58.4

Spain 54.6 Austria 57.1Switzerland 54.4 Switzerland 53.8

Germany 52.8 Spain 52.4


Another characteristic o the knowledge workers regards their networking activitiessince innovation requires ability to capture ideas outside the organisation, as Cohenand Levinthal ( 99 ) pointed out. Te Oslo manual reminds that “the presence o expert technological “gatekeepers” or receptors – individuals who, through many

means, keep abreast o new developments (including new technology and codi edknowledge in patents, the specialised press and scienti c journals), and maintainpersonal networks which acilitate ows o in ormation – can be crucial to innovation

within a organisation”.

Table 4.13Contacts o graduates with experts outside the organisation, according to the role in innovation

Introductioninnovationo product

Introductioninnovationo technology

Introductioninnovationo technology

I take the initiative in establishingpro essional contacts with experts outside theorganization ( rom 1/not at all to 5/to a very highextent)

Yes No Yes No Yes NoYes (3-5) 71.2 50.0 70,2 45.9 71.4 53.3 No (1-2) 28.8 50.0 29.8 54.1 31.6 46.6 Total 100.0 100.0 100.0 100.0 100.0 100.0

10179 11564 7777 13968 13484 8503

Graduates have been questioned to what extent they take the initiative in establishingpro essional contacts with experts outside the organization. Tose who introduceinnovation are clearly engaged in active networking, since % among them take




initiative in contacting external experts, compared with % among the other gradu-ates. Te proportions are identical in the three types o innovation (see able . ).

. . Innovation and working environment

A rst consideration regards the way the work o graduates involved in innovativeactivities may be characterised. According to the results o a regression analysis (seedetailed results in Appendix ), our most signi cant characteristics emerged: theextent o utilisation o knowledge and skills, the demand or more knowledge andskills than possessed, the de nition o the goals o the job and the decision on how todo the job. able . contrasts the answers o graduates involved in each o the threetypes o innovation against the others.

Table 4.14Characteristics o the job content o innovative graduates

Role in introductiono innovationo product

Role in introductiono innovation o technology

Role in introductiono innovation o knowledge

No Yes No Yes No YesTo what extent are knowledgeand skills utilized in your current work?

To a veryhigh extent* 31,8 40,1 32,5 41,3 27,3 41,1

To what extent does your current workdemand more knowledge and skills thanyou can actually o er? Not at all NS 13,3 9,7 NSAre you responsible or: setting goals oryour own work?


Are you responsible or: deciding how youdo your own job?


* score=5

Whatever the type o innovation, innovating graduates utilize more intensively theirknowledge and skills. More than % o them consider they utilize them to a very high extent, against % or graduates not involved in innovation o product or

service, or in innovation o technology, and even % o graduates not involved ininnovation o knowledge or method. Innovation activities don’t seem discriminating when the question regards to what extent the present work demand more knowledgeand skills than possessed. Te only slight di erence can be observed in the answers o those who declare this is not the case: % o graduates not involved in innovation o technology declare they don’t need more knowledge and skills, against % o gradu-ates involved.

Di erences are more obvious when the de nition o the contents o the job is at stake.Innovative workers are clearly more autonomous: they are more requently respon-sible to set the goals o their own work and they are also more o ten responsible orthe decisions regarding how to do their job.



CHAPTER

Around one hal o innovative graduates consider they are responsible to a very highextent or setting goals or their own work, against only % o those not involved ininnovation o knowledge and method, the di erence being slightly narrower or those

involved and not involved in innovation o technology ( 8% against %).

Te same holds true when the decisions on the tasks are considered: - % o inno-vative graduates decide to a very high extent how to do their job, compared with% o graduates not involved in innovation o knowledge, or % o graduates not

involved in innovation o product or service.

. . Innovation activities and characteristics o the organisation

As no strong di erences appeared in the previous analyses according to the type o innovation, an innovation index, which represents the means o the marks or thethree types o innovation or each graduate, will be used.

Di erent to the earlier analyses, the characteristics o the organisation will be relatedto the innovative role o the graduates, and not to the innovation behaviour o theorganisation. Tree main dimensions are considered here: the scope o operations,the sector o activity and the size. Te results are shown in able . . When thescope o operations is considered, the same logic as the one already observed appears.

Graduates are more requently involved in innovative activities when the scope o activities o their organisation is wider. Manu acturing activities but also business andnancial services promote innovation among graduates. Education and health andsocial work represent sectors where innovation is also present.

Te most striking result regards the role o the size o the organisation on innovationactivities, since graduates appear to be more requently involved in innovation activi-ties when they work in a smaller organisation.




Table 4.15Innovation and characteristics o the organisation

Dependent variable: Innovation index B*Regional scope o operations 0.15National scope o operations 0.23International scope o operations 0.56Re : local scope o operationManu acturing and other productive activities 0.35Trade, transport and other traditional services 0.13Business and nancial services and communication 0.36Education 0.27Health and social work 0.27Re : public administration

10_49 workers -0.0950_99 workers NS100_249 workers -0.11250_999 workers -0.08>1000 workers NSRe : <10 workersOther variables in the model : country

Adjusted R2 N:17159 0.10367

The reported coe cients present a level o signi cance above 1/100

In order to illustrate such a result, the proportion o graduates involved in the threetypes o innovation taken together is contrasted toward the size o the organisation.

Table 4.16Innovation activities according to the size o the organisation

Number o people workingin total organization

Proportion o graduatesinvolved in the threetypes o innovation

Total

Private sector 1-9 37.7 221410-49 26.0 217950-99 22.7 1056100-249 20.3 1224250-999 18.1 16701000 or more 19.1 3699Total 24.1 12042

Public sector 1-9 35.0 35710-49 22.2 126250-99 21.3 834100-249 21.3 1115250-999 21.0 14781000 or more 17.9 3126Total 20.7 8172



CHAPTER

able . shows clearly that graduates working in small organisations are morerequently involved in innovative activities than graduates hired by large organisa-tions, both in the public and the private sector. Less than % o graduates working

in organisations larger than employees develop the three types o innovation,compared with more than % o workers in organisations with less than employeesor with % o workers in private companies with a number o employees between and . Graduates in big organisations are part o large groups o highly quali ed andexperienced workers and have dependent positions, whereas in small organisations,they get more responsibilities in the development o the product.

. Are graduates equipped or innovation?

One o the main questions at the heart o this project regards to what extent highereducation institutions prepare graduates to ul l the tasks required by the presentknowledge and innovation societies. Since innovation is more precisely tackled in thispart, in ormation has to be provided on the links between the characteristics o thestudy programme and the ability ot per orm innovative activities.Tree main questions will be addressed here: What are the elds o study more linkedto innovation? What are the competencies more related to innovation? Are innovation

workers specialised?

. . Innovation activities and eld o study

Table 4.17Field o study and introduction o innovation

Field o education and training Introduction o innovationproduct/service

Introduction o innovationtechnology/tools/instruments

Introduction o innovation knowledge/method

% N % N % N

Education 49.2 1,954 32.0 1,943 68.6 2,012

Humanities and Arts 44.4 2,490 29.5 2,475 59.5 2,524

Social sciences, Business and Law 43.2 7,333 28.6 7,291 57.0 7,358

Science, Mathematics and Computing 44.9 2,231 47.1 2,226 65.9 2,232

Engineering, Manu acturing and Construction 54.4 3,756 53.6 3,758 64.2 3,770

Agriculture and Veterinary 54.2 507 42.3 504 68.7 514

Health and Wel are 44.6 3,228 30.1 3,210 59.4 3,248

Services 47.6 510 30.3 512 57.8 517

Total 46.5 22,009 35.7 21,919 61.1 22,175




When the eld o study is taken into account, “engineering, manu acturing andconstruction” appears to be the one which leads more requently to cope with theintroduction o innovation. % o graduates o that eld are involved in the intro-

duction o innovation o product/service and technology/tools/instruments (see able. ). Around two thirds introduce innovation o knowledge and methods. wo otherelds, “Science, mathematics and computing” and “Agriculture and veterinary”present the same pattern.

A second result regards the width o discrepancies between elds when the type o innovation is considered. Whereas the discrepancies are larger or the innovationso technologies/tools/instruments (which concern around % o graduates in theelds o education, humanities and arts, social sciences, business and law and alsohealth and wel are and services), they are very tight or the innovation o knowledgeand methods. In this last case, education and agriculture and veterinary are even thedisciplines which present the highest proportion o graduates introducing innovations( 8%).

Te Canberra Manual, which deals with human resources in science and technology notes the ollowing with respect to elds o study: “some elds, like the natural sciencesor engineering and technology, are o ten considered, at least in English-speakingareas, to be more directly relevant to S& activities than the social sciences, humani-ties or other elds”. Tis is why the manual makes a distinction within elds o study

between core, extended and completed coverage. Natural sciences, engineering andtechnology, medical sciences, agricultural sciences, social sciences represent the core, whereas humanities and other elds represent the extended. Tis classi cation appearsto be coherent with the REFLEX results, except or what has been called in our survey innovation o knowledge/method where education is predominant.

. . Characteristics o the study programme and innovation activities

What are the main characteristics o the study programme related to innovation

activities? A regression analysis has been run, rom which the most signi cant vari-ables describing the programme have been kept. Following the value o the regressioncoefcients, the most important characteristic is to participate in research projects

when in higher education (see able . 8). Other eatures o the study programmesmay induce innovative positions as well: work placements and internships, projectand problem based learning, reedom in composing the programme, multiple choiceexams, employers being amiliar with the programme. Graduates involved in innova-tion activities also strived more requently or the highest marks as a student. Tis may be related to the characteristics o the programme being more requently demandingand prestigious.



CHAPTER

Table 4.18Regression coe cients o the characteristics o the study programme

Dependent variable : Innovation index BDescription apply to study programme: programme was generally regarded as demanding .04Description apply to study programme: employers are amiliar with the content o programme .02Description apply to study programme: there was reedom in composing your own programme .03Description apply to study programme: programme was academically prestigious .04Modes o teaching and learning: participation in research projects .05Modes o teaching and learning: internships, workplacement .03Modes o teaching and learning: theories and paradigms .03Modes o teaching and learning: group assignments .02Description apply to study programme: programme had a broad ocus .02Modes o teaching and learning: project and/or problem-based learning .03Modes o teaching and learning: multiple choice exams .02Description study behaviour: I strived or the highest possible marks .02Other variables included in the model: country, eld o study, level o degree, genderN=17,942Adjusted R square=0.085

The reported coe cients present a level o signi cance above 1/100

. . What are the competences more related to innovation?

aking into consideration that the graduates are equipped with di erent types o competences, developed partially by the programs ollowed during higher education,it is important to observe what are the competences more linked to the three typeso innovation. Te probability to play a role in introducing each o the three types o innovation has been regressed separately on each o the 9 competences acquired by graduates controlling or the country. Te most signi cant results will be presented(Regression coefcient higher than . ). Six competences correspond to this criterionor the introduction o innovation o product, our in the case o the introduction o technology and nine when the introduction o knowledge is considered. Tese resultsmean that the more use ul competences are not the exactly the same, according to the

type o innovation considered.

ables . 9 to . show the means o competences or graduates who play a role inthe introduction o each type o innovation and or those who don’t.



CHAPTER

Table 4.21Means o the most di erentiated competences (Introduction o innovation o knowledge ormethod)

Abilityto presentproducts,ideas or

reports to anaudience

Abilityto come upwith newideas andsolutions

Willingnessto question

your own andothers’ ideas

Alertnessto new

opportunities

Abilityto mobilize

thecapacitieso others

Analyticalthinking

Abilityto make

yourmeaningclear toothers

Masteryo yourowneld or

discipline

Abilityto rapidly

acquire newknowledge

Yes 5.07 5.54 5.53 5.16 5.07 5.48 5.50 5.45 5.77

No 4.59 5.08 5.19 4.83 4.77 5.20 5.26 5.24 5.57

Di erence 0.48 0.46 0.36 0.33 0.30 0.28 0.24 0.21 0.20

. . Are knowledge workers specialised?

According to Peter Drucker ( 9 9), knowledge workers would be specialised. In orderto investigate such an assumption, we looked at the ollowing question: “what eld o study do you eel is most appropriate or your current work?”.

Table 4.22Field specialisation and innovation

Introduction o innovationo product or service

Introduction o innovationo technology, tools or

instruments

Introduction o innovationo knowledge or method

Yes No Yes No Yes NoExclusively own eld 30.7 33.0 31.1 32.4 32.4 31.3Own or a related eld 55.8 50.8 56.5 51.3 55.2 50.0A completely di erent eld 6.6 6.9 6.4 7.0 6.0 7.8No particular eld 6.8 9.3 6.0 9.3 6.3 10.9Total 100.0 100.0 100.0 100.0 100.0 100.0

Te most striking result o able . regards the specialisation o the graduates’ work,

whatever their situation in relation to innovation. More than 8 % consider that themost appropriate eld or their current work in exclusively their own eld or a relatedone. Nevertheless, graduates more involved in innovative activities appear to be a bitmore specialised than the others: around 88% o them consider their eld or a closeone as a prerequisite or their work.

. Innovation, occupations and rewards

A ter a rst glance at the relation between occupations and innovation, the impact o innovation activities on earnings will be studied.




. . Occupations and innovation

Te role played regarding innovation is related to the occupation o graduates. o

study this question, the occupations grouping at least two percent o the graduateshave been considered. Tese 8 occupations, which represent % o all working grad-uates, are presented in able . . Te observations show a relation with the type o innovation.

Table 4.21Participation to innovative activities o the main occupations

Occupation Proportion in thetotal o occupied

population

Proportionplaying a rolein introductiono product or

service

Proportionplaying a role inintroduction o

technology, toolsor instruments

Proportionplaying a role inintroduction o knowledge or

methodsManagers o small enterprises 2.2% 74.4% 53.7% 79.8%Production and operations managers 1.8% 71,0% 53.3% 77.2%Other specialist managers 2.9% 65.4% 46.1% 76.1%Architects, engineers and related pro essionals 9.4% 53.8% 54,0% 65.2%College, university and higher education teachingpro essionals 2.4% 48.7% 43.1% 79.1%Computing pro essionals 4.1% 50.9% 57.4% 59.1%Other teaching pro essionals 2.4% 50.2% 35,0% 72.1%

Health associate pro essionals (except nursing) 1.8% 53.8% 34.7% 67.3%Primary and pre-primary education teachingpro essionals 4,0% 49.8% 32.8% 72.4%Writers and creative or per orming artists 1.8% 56.6% 31.8% 57.7%Nursing and midwi ery pro essionals 2.3% 48,0% 29.6% 60.7%Social science and related pro essionals 5.6% 44.4% 26.8% 62.8%Secondary education teaching pro essionals 4.9% 38.2% 28.2% 66.1%Health pro essionals (except nursing) 5.3% 39.2% 34.7% 55.6%Business pro essionals 10.2% 44.2% 26.8% 56.8%Finance and sales associate pro essionals 3.4% 45.2% 26.7% 54.5%Legal pro essionals 3.7% 38,0% 20.8% 50.6%Administrative associate pro essionals 3.1% 30.2% 24.4% 46.1%Total 71.3% 48.0% 36,0% 62.8%

When it comes to the innovation o product or service, the managers o small organi-sations and the managers and operation managers are the categories at the ore ronto innovation (respectively % and % introduce innovation), ollowed by otherspecialist managers ( %), health associate pro essionals, teaching pro essionals otherthan teachers and writers/artists ( %) and architects, engineers and related pro es-sionals ( %). Tis last category appears to be one o the more innovative when theinnovation o technology, tools or instruments is at stake. Tis type o innovation



CHAPTER

is in act rather discriminatory since it concerns with intensity only some speci coccupations, computing pro essionals, managers o small enterprises, production andoperation managers, on top o the architects and engineers.

Te results underline that innovation is not only related to technical innovation. Tatappears clearly when the innovation o knowledge or methods is taken into account.In that case, university and higher education teaching pro essionals appear to be themore innovative category, together with the categories already quoted, as managerso small enterprises, production and operation managers ( 8% ). Such an intensity or higher education teachers can be linked to the research activities embodied inthe tasks o this occupation, but is also related to the teaching activities as shown by the high proportions among the primary and pre-primary education teaching pro es-sionals and among the secondary education teaching pro essionals (at least two-thirdso graduates in these categories introduce innovation o knowledge or methods).

Te importance o occupations such as managers, engineers/architects and computingpro essionals in the second type o innovation (.innovation o technology/tools/instru-ment) may allow to understand why competences as ability to come up with new ideas and solutions, to use computers, analytical thinking, willingness to questionyour own and others’ ideas emerged as strategic competences or that type o innova-tion. On the same ground, the importance o an occupation like teachers o all levelsmay explain why competences such as mastery o own eld or discipline or abilities

o presentation to an audience were important or the introduction o innovation o knowledge or method.

Tese ndings can be related to the classi cation o occupations proposed by theCanberra Manual. Te core occupations are the ollowing: physical, mathematicaland engineering science pro essionals (such as physicist, chemist, operation researchanalyst, computer system engineer, architect and mechanical engineer), li e sciencesand health pro essionals. Extended occupations comprise, amongst others, produc-tion and operations department managers, general managers, teaching pro essionals(university pro essor, school teacher), physical and engineering science associate

pro essionals, li e science and health associate pro essionals, lawyers, economists.Notwithstanding some strong similarities between the two distributions, the Canberraone and the REFLEX one, new insights need to be mentioned rom the REFLEX in ormation. Whereas managers belong to the “extended” category, they appearclearly as one o the occupations most concerned with innovation. Unsurprisingly,the same comment applies to the teaching pro essionals, who consider innovation asa main part o their duties.




. . Are innovation activities rewarded?

Regression analysis allows identi ying i innovation activities are rewarded and to

what extent they are. o play a role in the introduction o innovation is worthwhile.It represents an increase o the earnings rom % or innovation o technology/tools/instruments to % or the introduction o knowledge/methods, with some slightdi erences between sectors: innovation is more rewarded in the private sector that inthe public one (see able . ).

Table 4.24Relative earnings o innovative activities

Coe cientsModel Variables ALL PRIVATE PUBLIC

1 Introduction innovation o product or service 0.05 0.06 0,052 Introduction innovation o technology/tools/instruments 0.03 0.03 NS3 Introduction knowledge/methods 0.06 0.05 NS

Other variables: countries, sector, gender, level o degree, total working hours main work, size o the organisationDependent variable: logarithm earnings main jobAll coe cient signi cant at 1/100 level

. Conclusion

Te concept o Knowledge and In ormation Society (ies) became undoubtedly areality in Europe. Te survey con rmed that innovation represents a main tool in theday-to-day li e o most organisations, especially those more con ronted with strongcompetition and globalisation. Surveys like REFLEX may o er a di erent perspec-tive to surveys speci cally dedicated to innovation activities in organisations. Teextent o innovation extension, as perceived by graduates, is di erent rom country tocountry, according to the economic, social and political trajectory, to the culture, tothe representation o the di erent sectors o economic activity.

Graduates represent crucial actors in this innovation process: more than hal declarethey play a role in introducing innovation in their organisation. Nevertheless, innova-tion cannot be restricted to industrial processes. Its importance or services, includingthe public ones (education, health), has been noticed. Innovative graduates play theirrole o knowledge workers and expert technological gatekeepers. Teir jobs presentsome speci c characteristisc: more autonomy, more room to de ne their own goalsand the way to per orm it.

Tough innovation is more developed in big organisations, it stands out that smallones o er more opportunities or graduates to be involved in innovation activities.



CHAPTER

Some speci c competences are mobilised by the graduates who play a role in intro-ducing innovation: ability to present reports, ideas or products to an audience, tocome up with new ideas and solutions, to use computers and the internet, willing-

ness to question own and others’ ideas, alertness to new opportunities. Such compe-tences can be related to certain modes o teaching and learning. Study programmes o graduates involved in innovation have been more requently demanding. Tey o eredopportunities to develop research projects, to be part o internship activities. Projectand problem based learning seem also to represent a good basis or preparing gradu-ates to be part o the innovation society.

When the earnings are considered, innovative activities appear to be rewarded, espe-cially in the private sector. Tat represents an additional proo that these activities arerecognised as such by organisations.

Te in ormation provided by the REFLEX survey may be considered as an input orhigher education institutions which have not yet introduced re orms in their ways o teaching and learning. Competences use ul or innovative activities are produced by modes o teaching and learning which assume an active participation o the studentsto the learning process, through research projects, internships, project and problembased learning. Innovation comes rom graduates academically brilliant but alsoopen to new ideas and concepts, ready to communicate and to work with others.Conservative institutions are threatened by such an evolution.

Tough REFLEX allowed making new insights about the role o graduates in theEuropean society, complementary in ormation may be use ul. Additional knowledgeon the contents o the job o graduates may give a more complete vision o whatinnovation activities precisely represent. And data on PhD graduates would obviously complete this rst view. Tough they are less numerous than Bachelor and Mastergraduates, their role in innovative activities would have to be taken into account, inorder to get a ull view o the graduates’ participation to the innovation activities o European organisations.

Re erences

Canberra Manual ( ) Te measurement o scienti c and technological activities. Manualon the measurement o human resources devoted to S& “Canberra Manual”. OECD,Paris.

Castells M. ( ) Te Rise o the Network Society, Vol. I. Cambridge, MA; Ox ord, UK:Blackwell ( ) (second edition, )

Cohen, W., Levinthal, D. ( ). Absorptive Capacity: A new perspective on learning andinnovation. Administrative Science Quarterly, : - .Drucker P. ( ) Landmarks o tomorrow: A Report on the New Post-Modern World, Harp-

er, New York.Duru-Bellat M. ( ) L’in ation scolaire: les désillusions de la méritocratie, Seuil, Paris.




European Innovation Scoreboard ( ), Comparative Analysis o Innovation Per ormance,Pro Inno Europe, Innometrics, available on internet http://www.proinno-europe.eu/inno-metrics.html

Foray D. ( ) L’économie de la connaissance, La Découverte, Paris.Frascati manual ( ) , Proposed Standard Practice or Surveys on Research and Experimental

Development, OECD, Paris.Oslo Manual ( ): Te measurement o scienti c and technological activities. Proposed

guidelines or collecting and interpreting technological innovation data. OECD, EuropeanCommission, Eurostat, Second edition, . Available on Internet: www.oecd.org.

Reich R. Te Work o Nations, Preparing Ourselves or the st Century Capitalism, Al red Al red A.Knop , .

Schumpeter J. ( ) Te Teory o Economic Development, Harvard University Press. Wol A. ( ) Does Education Matter: Myths about Education and Economic Growth? Pen-

guin Books, London.



Chapter Mobilization o Human Resources

Jim Allen

. Clari ying concepts

In a sense, this chapter is something o an oddity in the context o this report. It isonly a slight exaggeration to say that the report as a whole is about di erent kindso human resources o higher education graduates, and most chapters pay attentionamong other things to the mobilization o the particular type o resource that is thesubject o the chapter. In this chapter, the ocus is on the mobilization o humanresources in general. Te oddity lies in the act that the ability to mobilize humanresources is itsel a human resource. In particular when the discussion turns to thecompetences which are thought to be especially relevant to mobilization o humanresources, things can get a little con using. We may even nd ourselves in a situation

where we are describing the degree to which competences important or mobilizationare themselves being mobilized.

At an abstract level, it looks deceptively simple to de ne the subject o mobilizationo human resources. In large part as a result o the learning that takes place in highereducation institutions in di erent countries, there is at any given moment a certainstock o human capital that could, at least in principle, be put to productive use inthe economy. Economic growth can be achieved not only by increasing this stock o human capital, but also by increasing the proportion that is actually being put to produc-tive use . Te main idea o this chapter is that higher education has an important roleto play in this latter area as well as the ormer, by teaching its participants how to puttheir own knowledge and skills to good use, as well as how to play a role in mobilizing

the competences o other people with whom they work.

I higher education does play such a role, this should be re ected in the ways in whichhuman resources are mobilizedin higher education. One might expect that graduates

who actively mobilize their own and others’ resources a ter leaving higher educationalready start doing so in higher education. Tis is not only a matter o putting in longhours at study, but should involve a high degree o motivation to get more out o their study than what is minimally required to pass exams. In addition, there are o tenample opportunities or students to mobilize their own capacities by taking part invarious extra-curricular activities, such as paid work, positions in student or voluntary organizations, or time spent abroad. It is important to take into account the possi-bility that that engaging in such extra-curricular activities might leave students with



CHAPTER

less time to spend on their studies. More generally, it is important to look at eatureso higher education that are related to a high degree o study e ort and motivation.

At the most basic level, the rst thing to look at when describing the mobilization o human resourcesa ter graduationis whether they are being mobilized at all. In other

words: do graduates participate in the labour orce, and i so are they in paid employ-ment? In general we can say that it is better or graduates to work than not to work,but there may be a large amount o variation among those participating in the labourorce in terms o the extent to which human resources are being mobilized. Not every

working graduate is necessarily employed or a ull working week. Tis is o coursenot always a bad thing: part time work may provide opportunities or some graduatestaking care o young children, or or participation in urther education and training.

For those who work, whether this is ulltime or part-time, a more important consider-ation is the extent to which they are able to make use o their ull potential in the timethey spend at work. Tere is an extensive literature on this subject, most o which allsunder the general heading o overeducation (see e.g. (Duncan and Ho man, 98 ;Hartog & Oosterbeek, 988; Sicherman, 99 ; Hersch, 99 ; Cohn & Khan, 99 ;Van Smoorenburg & Van der Velden, ). Although working in a job requiringone’s own level (or in some versions one’s own eld) o education is something mostgraduates would strive or, this is neither a necessary nor a sufcient condition ormobilizing one’s own capacities (see e.g. Allen & van der Velden ). In particular,

graduates can acquire skills that help them to mobilize their own capacities even whenthey are working in jobs other than they were trained or. In the context o thischapter, in which the ocus is about the ability to mobilize resources as well as theactual mobilization, this is an important point.

Tings get more complicated when we consider the act that higher education gradu-ates are not only responsible or mobilizing their own capacities, but can also be calledon to help mobilize the capacities o other workers. It should be stated at the outsetthat our data don’t allow anything like a comprehensive analysis o this aspect o mobilization. Te main limitation is that we have almost no in ormation at all about

who these others are. Are they other higher education graduates, or workers with alower level o education? What kind o work are these other workers doing? And inparticular: to what extent are these workers utilizing their capacities, and what is thecontribution o the graduates in our survey to this? Tese are questions we simply cannot answer. We can however answer other important questions, such as: to whatextent are graduates expected to work with, and particular to monitor and supervise,others? o what extent is the output o graduates interdependent with that o co-

workers? o what extent do graduates bear responsibility or setting goals or decidingstrategies or the organizations in which they work? And what kinds o competencesare they required to use in ul lling these duties?



Mobilization o Human Resources

Tis chapter is not just about describing the extent to which graduates mobilize theirown capacities, or are involved in mobilizing those o their co-workers. At least asimportant is to try to uncover the actors that contribute to this mobilization. A basic

assumption underlying this chapter is that graduates are not entirely at the mercy o the work situation they nd themselves in or the mobilization o competences, butcan actively strive to increase the level o mobilization even when the objective condi-tions are un avorable. Higher education has a role to play, by teaching its studentshow to put the available human resources – whether their own or others’ – to gooduse. How well higher education plays this role is a key element o the chapter.

I we manage to establish that higher education indeed has the capacity to in uencethe level o “mobilization” competences, the next step is to see whether this actually results in more mobilization. In doing so, it should be kept in mind that there arelimits to what higher education and higher education graduates can do. Mobilizationis likely to be in uenced as much or even more by the characteristics o the rmsand organizations in which graduates work. A key ocus o this project is to establish

whether rms and organizations do what they need to do in order to get the best outo higher education graduates.

In the next section we describe various indicators or the degree o mobilization o one’s own resources in higher education: study hours, intrinsically and extrinsically motivated study behaviour, and extracurricular activities, and use several multivar-

iate analsyses to determine which eatures o higher education are related to a highdegree o study e ort. We subsequently look at how graduates rate their own study programmes as producers o competences that are thought to be relevant to mobilizinghuman resources, and look or eatures o higher education that are related to highacquired levels o these competences. In the next section we describe various indica-tors or the degree o mobilization o one’s own resources: labour orce participationand working hours, participation in activities outside the world o work, the educa-tion-job match and (under)utilization. Following that we move on to a descriptiono several indicators o mobilization o the capacities o others, including supervision,quality control and strategic decision-making authority. Sections . and . . contain

a number o multivariate analyses aimed at exposing some important determinantso the utilization o one’s own capacities and the mobilization o others’ capacities.Section . comprises a brie conclusion.

. Mobilization o human resources during higher education

Te data contain a number o indicators o the mobilization o human resources by students during their time in higher education. Several o these re er to the amounto e ort made by students to achieve good study results: the amount o time spenteach week on studying, doing extra work above what is required to pass one’s examsand striving or higher grades. In addition, and potentially important in preparing



CHAPTER

students to mobilize their own and others’ resources a ter graduation, we have in or-mation on various kinds o other experiences gained while enrolled in higher educa-tion. A ter describing these eatures, in this section we will present the results o a

series o regression analyses aimed at uncovering eatures o higher education that arerelated to a high degree o e ort by students.

. . Study behaviour

Figure . shows three di erent indicators o the amount o e ort students put intoachieving good results. Te most seemingly straight orward o these is the totalamount o time spent on average on studying (including lectures, internships etc.). A limitation o this indicator is that it may be in uenced negatively by students’ ability and/or efciency: in order to achieve the same results, less gi ted students, or students

who are less efcient in their use o time, will need to spend more hours studying justto achieve the same results as their more talented and/or efcient peers. Particularly the possible relationship with time efciency is potentially problematic when we wishto consider this as an indicator o mobilization o one’s own human resources. For thisreason, we include two additional indicators. Te rst indicates the degree to whichstudents did extra work during higher education above what was needed to pass theirexams. Tis can be regarded as an indicator o intrinsic study motivation, since it isnot related to any obvious rewards in terms o demonstrable study achievement. In

contrast, the second indicator, the extent to which students strive or higher grades,is more an indicator o extrinsic study motivation, since such grades can improvegraduates’ CVs. Figure . shows the distribution o the three indicators across theparticipating countries.

. Both o these indicators are measured on a -point scale ranging rom “not at all” to “to a very highBoth o these indicators are measured on a -point scale ranging rom “not at all” to “to a very highextent”. Figures . and . present the percentage o answers or .




Figure 5.1Study behaviour, by country

0

15

30

45

ES IT FR CH AT DE NL BE UK NO FI EE CZ TOT

s t u d y h o u r s ( m e a n )

0

25

50

75

e x t r a w o r k / h i g h e r g r a d e s ( % )

Did extra work (%) Strived for high grades (%) Study hours (mean)

According to the three indicators, the e ort that European students put into achievinggood study results in higher education is moderate at best. Although students report

working close to a ulltime working week on their study (slightly less than hoursper week; this rises to hours or ulltime students), only % o graduates reported

doing substantial extra work above what was required to pass their exams. A higherpercentage (but still only slightly more than hal ) reported that they strived or thehighest possible grades, suggesting that the study motivation o European graduatesis more extrinsic than intrinsic.

Figure . reveals large di erences between countries in all three indicators. Study hours vary rom less than thirty in Estonia and the Czech Republic to more than in France. Te other two indicators also vary greatly between countries. With

the exception o Finland, more graduates in each country indicated that they strivedor high marks during higher education than that reported doing extra work above

what was required to pass exams. Only in Spain and the UK do more than hal o graduates report having done substantially more work than needed to pass exams,compared to a quarter or less in France, the Czech Republic and the Netherlands.Spanish and British graduates are also among the top with respect to striving orhigher grades, together with their German peers. Around two thirds o graduates inthese countries reported that they strived or higher grades, compared to only arounda third o Dutch and Flemish graduates. In general, there is little relation between themean study hours in a country and the other two indicators. A notable exception isthe Netherlands, which combines below average study hours with low levels o bothextrinsic and intrinsic study motivation. Figure . shows the same indicators by leveland type o education.



CHAPTER

Figure 5.2Study behaviour, by eld and level o education

first level

0

15

30

45

E d u c a t

i o n

H u m / A

r t s

S o c i

a l s c i e n c e

s

S c i e n c e / M

a t h . / C

o m p .

E n g i n . / M a n u

f . / C o n s t

r .

A g r i c .

/ V e t e r i

n a r y

H e a l

t h / W e l f a r e

B u s i n e s

s / L a w O t h e r

T o t a l

s . h o u r s ( m e a n )

0

25

50

75

e . w

o r k / h

. g r a d e s ( % )


second lev el

0

15

30

45

E d u c a

t i o n

H u m / A

r t s

S o c i a

l s c i e n

c e s

S c i e n c

e / M a t h

. / C o m

p .

E n g i n . / M a

n u f . / C o

n s t r .

A g r i c . / V e

t e r i n a r

y

H e a l t h / W

e l f a r e

B u s i n e s

s / L a w O t h e r

T o t a l

s . h o u r s ( m e a n )

0

25

50

75

e . w

o r k / h

. g r a d e s ( % )


Second level graduates report slightly higher study hours than rst level graduates.Te di erences by eld are more pronounced. At both levels graduates in the ‘harder’

elds such as Engineering or Health report much higher study hours than graduatesin the so ter elds like Humanities or Social Sciences. Te order is almost reversed orthe other indicators: Humanities graduates report the highest levels o both intrinsicand extrinsic motivation, while Engineering graduates report quite low levels.

. . Other experiences during higher education

Figure . shows the total number o months that students spent acquiring variouskinds o other experience while in higher education. In terms o mobilization o

. It should be noted that the number o months says nothing about the amount o time spent eachIt should be noted that the number o months says nothing about the amount o time spent eachmonth on the activity in question. It is likely that some activities, particularly experience abroad and




human resources, these indicators have rather mixed meanings. At a general levelone might argue that any kind o additional activity is a sign o an active attitudeand is there ore positive. However, some activities, in particular non study-related

work experience, are probably undertaken mainly or instrumental reasons and havelittle bearing on what students hope to do a ter graduating. In contrast internships,other study-related work experience and experience abroad may prove highly relevantto graduates’ later career development. Voluntary positions occupy an intermediateplace: while in most cases probably not directly related to graduates later career interms o substance, such experience can help students develop assertiveness and lead-erships skills that may prove invaluable.

Figure 5.3Experiences during higher education, by country (months)

0

10

20

30

40

50


m o n t h s ( m e a n )

experience abroad

internship/work placement

voluntarypositions

non study-related w.exp/

study-related w. exp.

Most striking in Figure . are the large di erences between countries. Even takinginto account the act that the number o months is a ar rom per ect indicator o actual time spent, it is surprising that the di erences are so large. Dutch graduatesspend an average o months on the included activities, compared to months orless in Spain, Italy, Flanders and the UK. Closer inspection reveals that almost hal the

experiences o Dutch students involve non-study-related work. Relevant work experi-ence is most common in Austria and Germany, and hardly occurs in Flanders and theUK. Swiss, Flemish and Dutch students spend the most time occupying voluntary positions, French and Dutch students spend most time on internships, and Austriangraduates spend the most time abroad. Figure . shows the same gures by level andtype o higher education.

internships, are more or less ulltime activities, while others, particularly voluntary positions, may involve no more than a ew hours each month.



CHAPTER

Figure 5.4Experiences during higher education, by eld and level o education (months)

rs t lev el

0

10

20

3040

experi enc e abro adint erns hip/ wo rk pl ac ementv o lun t ary p o s it io nsno n s t udy- rel at ed w.exp/s t udy-r elat ed w. exp.

m o n

t h s

( m e a n

)

E d u c a t i o n

S c i e n c e / M a t h . / C o m p .

S o c i a l s c i e n c e s

B u s i n e s s / L a w O t h e r

E n g i n . / M a n u f . / C o n s t r .

A g r i c . / V e t e r i n a r y

H e a l t h / W e l f a r e T o t a l

H u m / A r t s

s ec o nd le v el

0

10

20

30

40

E d u c a t i o n

S c i e n c e / M a t h . / C o m p .

S o c i a l s c i e n c e s

B u s i n e s s / L a w O t h e r

E n g i n . / M a n u f . / C o n s t r .

A g r i c . / V e t e r i n a r y

H e a l t h / W e l f a r e T o t a l

H u m / A r t s

m o n t h s

( m e a n

) experi enc e abro adint erns hip/ wo rk pl ac ementv o lun t ary p o s it io nsno n s t udy -r ela t ed w.exp /s t udy-r elat ed w. exp.

Second level graduates acquire more o most orms o experience than rst levelgraduates. Te exception is internships and work placements, on which the o tenmore vocationally oriented rst level students spend more time. At both levels, health

graduates spent a lot o time on the various orms o experience, particularly on study-related work experience and internships. Second level humanities and social sciencesgraduates also send a lot o months on experience outside education, but a largeproportion o this involves non-study-related work experience. Voluntary positionsare particularly common among second level graduates in almost all elds. Secondlevel Humanities graduates gain the most experience abroad.

. . Determinants o study behaviour

o some extent at least, the e ort students expend on their study is likely to dependon their own innate personality. It is however conceivable that experiences gained




in higher education can in uence this, by exposing students to situations in whichthey eel more motivated. o examine this, OLS regression analyses were run withthe three indicators presented in Section . . as dependent variables, and personal

background characteristics, programme characteristics and additional experiences aspredictors.

Table 5.1Relation between personal background characteristics and study behaviour (standardized linearregression coe cients) 4

Did more work thanneeded to pass exams

Strive or highestpossible grades

Study hours

Gender ( emale) 0.038 0.084 0.040Age 0.054 -0.023

At least one parent has HE -0.018 -0.023Had (pre)school-aged child during HE

able . shows the e ects o personal background characteristics on the indicatorso study behaviour. According to all three indicators, women clearly work harder inhigher education than men. Older students worked less hours but did more o tenextra work than younger students, which may indicate a greater degree o efciency or other bene t o their greater li e experience. Interestingly, social background interms o having at least one parent with a higher education degree has no e ect at

all on study hours, and a negative e ect on the other two indicators. Tese negativee ects seem at rst sight counterintuitive, but may in act re ect a lower degree o sel -con dence among students who are so to speak treading new ground in their amily.Such students may eel an extra need to prove that they belong in higher education,

while or those whose parents have already been there it may seem more natural.Having (pre)school-aged children while in higher education has no e ect at all onstudy behaviour as indicated by these three items.

able . shows the e ect o various programme characteristics. Second level studentsstudied longer hours and did more o ten extra work than rst level graduates, but

were no more or less inclined to strive or higher grades. rivially, parttime studentsstudied much shorter hours than ulltime students. Less obvious is the nding thatthis is also re ected in the intrinsic and extrinsic motivation o students. O theother programme characteristics, the degree to which a programme was regarded asdemanding has the strongest e ects. Again, in the case o study hours this is only tobe expected. It is at least a little surprising that such programmes are positively related

. It must be remarked at the outset that we cannot establish with any certainty the causal link involved.It must be remarked at the outset that we cannot establish with any certainty the causal link involved.Nonetheless, in some cases it seems at least plausible that the eature in question promotes motivationand e ort.

. Te results presented in ables . to . are based on the same three regression analyses, so all e ectsTe results presented in ables . to . are based on the same three regression analyses, so all e ectsreported e ects include controls or all other variables. All multivariate analyses in this chapter includecontrols or country, eld and type o HE, gender, age and parents’ education.

. Tis holds or both mothers and athers.



CHAPTER

to the other two indicators. One might as well imagine that students o programmesthat are especially demanding might have their hands ull just getting through therequired study material, and would nd extra work and striving or higher grades

a luxury that they can ill a ord. Te positive e ect o demanding programmes may suggest that students who are challenged by a demanding programme rise to thechallenge by working even harder than they need to get their degree. Te remainingprogramme characteristics have only weak or nonsigni cant e ects, and in one caseeven a weak negative e ect.

Table 5.2Relation between programme characteristics and study behaviour (standardized linear regressioncoe cients)

Did more work than

needed to pass exams

Strive or highest

possible grades

Study hours

2nd level programme 0.059 0.038Part-time programme -0.083 -0.100 -0.277Other programme characteristics:• Generally regarded as demanding 0.137 0.089 0.156• Employers amiliar with content• Freedom to compose own programme 0.026 -0.028• Broad ocus 0.019• Vocational orientation 0.028• Academically prestigious 0.028

able . shows the e ect o various modes o teaching and learning on study behav-iour. In most cases the e ects are only rather weak. A somewhat stronger e ect is seenor the extent to which the teacher as source o in ormation was emphasized on the

willingness to strive or higher grades. Tis eature has no e ect at all on either study hours or intrinsic motivation, suggesting that strongly teacher-centred education may promote a more extrinsic study motivation. A strong emphasis on lectures and onacts and practical knowledge have positive, although not very strong, e ects on allthree indicators.




Table 5.3Relation between modes o teaching and learning and study behaviour (standardized linear regres-sion coe cients)

Did more work than neededto pass exams

Strive or highest possiblegrades

Study hours

Lectures 0.034 0.037 0.035Group assignmentsParticipation in research projects 0.028Work placements/internships 0.031Facts & practical knowledge 0.028 0.029 0.028Theories & paradigmsTeacher as source o in ormation 0.052Problem- or project-based learning 0.032 0.031Written assignments 0.033 0.036

Oral presentations 0.033 0.043Multiple choice exams

able . shows the e ects o experiences gained during higher education. In general,only quite weak e ects are observed. Particularly in the case o study hours thisis surprising; one would expect time spent on one activity to be at the expense o another, so one would expect to observe negative relationships. Tis is not the case. It seems that students nd time or these other activities without this compromisingthe time they spend on studying. Study related work experience has a positive e ect

on intrinsic and extrinsic study behaviour, but or non study-related experience theopposite is true. Spending time in voluntary positions is related to a lower degree o striving or high marks. Tis may re ect a greater degree o sel -con dence amonggraduates who have acquired such experience.

Table 5.4Relation between experiences during higher education and study behaviour (standardized linearregression coe cients)

Did more work than neededto pass exams

Strive or highest possiblegrades

Study hours

Study-related work experience 0.032 0.021Non study-related work experience -0.020 -0.037Voluntary positions -0.026Work placements 0.034Experience abroad

. Work placements (which are included in study hours) even have a positive e ect on the overall hours Work placements (which are included in study hours) even have a positive e ect on the overall hourso study.



CHAPTER

5.3 Higher education as producer of “mobilization” competences

A key idea underlying this whole chapter is that graduates are not entirely at the mercy

o the work situation they nd themselves in or the mobilization o competences,but can actively strive to increase the level o mobilization even when the objectiveconditions are un avorable. In the previous section we saw some indications or thisin the di erent levels o e ort students put into their study. We also saw indicationsthat certain eatures o higher education may stimulate students to apply themselvesmore and to try to get more out o their higher education programme. It may be thatsuch experiences actually oster the acquisition o competences that help graduatesto make the most o their capacities regardless o the objective conditions in whichthey nd themselves. I this is the case, higher education may have a role to play, by ostering such abilities. It is equally conceivable that higher education may play a rolein ostering abilities that are use ul or mobilizing the human resources o others.

As pointed out in Chapter , six competences were singled out in advance on theo-retical grounds as likely to be important or graduates’ ability to mobilize humanresources. Tese are the ability to work under pressure , the ability to use time efciently ,the ability to work productively with others , the ability to mobilize the capacities o others ,the ability to make one’s meaning clear to others , and the ability to coordinate activities .Te rst two competences are thought to be especially important or mobilizing one’sown human resources, and the last our especially or mobilizing the human resources

o others. We re er to these six competences in this chapter collectively as “mobiliza-tion competences”. In Chapter a mixed picture emerged in terms o graduates’ evaluation o their study programme in terms o mobilization competences. On one hand, around one in vegraduates described the ability to work productively with others and the ability toper orm well under pressure as a strong point o the study programme, and ew saw these competences as weak points. In contrast, around one in six graduates regardedthe ability to mobilize the capacities o others as a weak point o the study programme,

with hardly any mentioning this competence as a strong point. Te ability to use time

efciently was more in balance, although it was slightly more likely to be rated astrong point than a weak point. Relatively ew graduates o ered any opinion one way or the other about the study programme in terms o the remaining two mobilizationcompetences, the ability to make one’s meaning clear to others, and the ability tocoordinate activities.

It is reasonable to assume that graduates evaluation o the study programme withrespect to a given competence is related to the extent to which graduates have acquiredthe competence in question during higher education. While we don’t have a measureo this in our data, we do know how highly graduates rate their own competences atthe time o the survey. Figure . shows the percentage o graduates who report thatthe mobilization competences were weak or strong points, split into those who at




the time o the survey report a low, medium or high level o the relevant competence(respectively those who answered - , - and - on the -point scale ranging rom “very low” to “very high”).

Figure 5.5Strong and weak points o mobilization competences, by own level

-40% -30% -20% -10% 0% 10% 20% 30%

l ow own level

medi umown l evel

high own l evel

COORDINAT E ACTIV ITI ES:

l ow own level

medi umown l evel

high own l evel

M AKE M EANING CLEAR:

l ow own level

medi umown l evel

high own l evel

M OBILIZE CAPACITI ESOTHERS:

l ow own level

medi umown l evel

high own l evel

WORK WITH OTHERS:

l ow own level

medi umown l evel

high own l evel

USE TIM E EFFICIENTLY:

l ow own level

medi umown l evel

high own l evel

PERFORM UNDER PRESSURE:

weak point s trong point

As we might expect, the higher graduates rated their own level o competence, themore inclined they are to rate that competence as a strong point and the less likely they are to rate it as a weak point o their study programme. Tat said, it is ar roma one-to-one relationship. Some graduates rated a competence as a weak point eventhough their own level was high, and some rated a competence on which their own

reported level was low as a strong point o the programme. Particularly striking isthe pattern or the ability to mobilize the capacities o others. Even graduates whoreported that their own level was high were much more likely to rate this competenceas a weak point than as a strong point o the programme. It may be that these gradu-ates have acquired most o this competence at work. While we cannot test this suppo-sition directly, we do have some indirect indications. Almost hal o those graduates

who rated their own level on this competence as high but regarded it as a weak pointo their study programme are currently responsible or supervising others in theircurrent work. Tis percentage is considerably higher than the % or the populationas a whole (see Section . . ), which may suggest that at least part o this competencehas been developed at work.



CHAPTER

Te question arises what higher education can do about improving the level o gradu-ates’ competences in these areas. o gauge this, we conducted a series o multivariateanalyses, in which the e ect o various aspects o graduates’ higher education experi-

ences were used as predictors o the graduates’ own level o the mobilization compe-tences. We ocus hereby on some key characteristics o higher education programmes,the main modes o teaching and learning applied, the experiences acquired duringhigher education, the results achieved at the end o the programme, and the behav-iour o students during the programme as independent variables. ables . to .8show the relevant results.

Be ore going in to a discussion o the results in detail, we can make some remarksabout the results in general. First o all, it appears that higher education can makea di erence in terms o generating mobilization competences. Many o the charac-teristics and experiences o higher education have a statistically signi cant e ect onthe acquired level o these competences. Secondly, it should be remarked that theindividual e ects, even when statistically signi cant, are not very large. o give somekind o perspective on what the e ects mean, a graduate who reported that his or herstudy programme is regarded as demanding to a very high degree scores on average aquarter o a point higher (on a seven-point scale) on the ability to make your meaningclear than a graduate who reported that the programme was not at all demanding.It should be stressed that this was one o the strongest e ects observed; most o theother e ects were not as strong. Tirdly, the cumulative e ect o these characteristics

is also not very large. Only a small raction o the total variance in these competencesis explained by these variables. Even taking into account the act that our indicatorsalmost certainly do not cover the ull range o variation in educational experiencesthat might help shape competences, this is disappointing. In sum, we can say thatnot only is there no individual ‘magic bullet’ that on its own guarantees success, evencumulatively the contribution that higher education can make is quite modest.

urning now to the e ects o the various indicators, able . shows that theprogramme characteristic that makes the biggest di erence in most cases is whetheror not the programme was demanding. Tis is not very surprising, one would expect

students o demanding programmes to learn more competencesin general than gradu-ates o less demanding programmes. Vocationally oriented programmes are good atproducing those competences that are thought to be relevant or mobilizing the humanresources o others, while academically prestigious programmes are good at producingcompetences relevant to mobilizing one’s own human resources. Programmes that areamiliar to employers seem to produce somewhat higher levels o ability to per ormunder pressure, but lower levels o ability to coordinate activities. First and secondlevel programmes do not generate signi cantly di erent levels o mobilization compe-tences, with the exception o the level o ability to mobilize the capacities o others,

which appears to be a little lower among second level than among rst level graduates.Freedom to choose and the breadth o ocus do not have any signi cant e ects onmobilization competences.




Table 5.5E ects o programme characteristics on the acquired level o competences relevant or mobiliza-tion (standardized linear regression coe cients)

Ability to:per ormwell underpressure

use timee ciently

workproductivelywith others

mobilizecapacities o others

makemeaning clearto others

Coordinateactivities

2nd level programme

Other programme characteristicsGenerally regarded as demanding 0.026 0.046 0.041 0.026 0.051 0.039Employers amiliar with content 0.030 -0.021Freedom to compose ownprogramme

Broad ocusVocational orientation 0.027 0.028Academically prestigious 0.028 0.031

Table 5.6E ects o modes o teaching and learning on the acquired level o competences relevant or mobi-lization (standardized linear regression coe cients)

Ability to:per ormwell under

pressure

use timee ciently

workproductively

with others

mobilizecapacities o

others

makemeaning clear

to others


LecturesGroup assignments 0.031 0.044 0.030Participation in research projectsWork placements/internshipsFacts & practical knowledge 0.030 0.036 0.032 0.024Theories & paradigms 0.021 0.030 0.031 0.030Teacher as source o in ormationProblem- or project-based learningWritten assignments 0.026

Oral presentations 0.037 0.034 0.052 0.027Multiple choice exams

able . shows e ects o various modes o teaching and learning on the level o mobili-zation competences. Student-centred aspects like groups assignments and oral presen-tations have quite strong e ects on several mobilization competences. Although theseare o ten eatures o project- and/or problem based learning, this mode o teachingand learning has hardly any e ect a ter controlling or these aspects. Interestingly,acts and practical knowledge and theories and paradigms also have rather stronge ects. Tis suggests that, in addition to methods in which students play an activerole, a strong emphasis on theoretical and practical knowledge helps generate compe-tences that are important or mobilizing human resources. We can only speculate



CHAPTER

about the mechanism involved here, but it is conceivable that the possession o agood knowledge base makes it easier or graduates to make the most out their ownand others’ human resources.

able . shows the e ects o various orms o experience gained during higher educa-tion. As we might expect, study-related work experience has an e ect on the devel-opment o several o the mobilization competences. Te e ects are not very stronghowever. Te strongest e ects however are those o positions held in voluntary organi-zations during higher education, especially on the competences thought to be relevantor mobilizing the human resources o others. Non study-related work experience,

work placements and time spent abroad have little or no e ect.

Table 5.7

E ects o experiences during higher education on the acquired level o competences relevant ormobilization (standardized linear regression coe cients)


use timee ciently

workproductivelywith others


makemeaning clearto others


Study-related work experience 0.025 0.029 0.037Non study-related work experience 0.023Voluntary positions 0.022 0.041 0.032 0.049

Work placementsExperience abroad

able .8 shows the e ects o level o study behaviour on the level o the six mobiliza-tion competences. In general, study behaviour showed surprisingly little e ect on thedevelopment o mobilization competences. Tis is doubly surprising, since one wouldexpect a high degree o motivation to work hard and achieve good results to not only to be good or developing competences in general, but in a sense to be a componento the very competences we are looking at here. A ter all, one might assume that ahigh degree o motivation is a prerequisite or mobilizing human resources. Te only

positive exception is a rather strong e ect o a willingness to strive or higher gradeson the ability to use time efciently. Te only other e ect o note is a negative e ecto willingness to do more work than needed to pass exams on the ability to per ormunder pressure. It is unclear what mechanism is involved here, but it may be a caseo reverse causality, whereby students that cannot handle pressure well tend to over-prepare or their exams.




Table 5.8E ects o level o study behaviour on the acquired level o competences relevant or mobilization(standardized linear regression coe cients)


use timee ciently

workproduct-ively withothers


makemeaningclear toothers

coordinateactivities

Study BehaviourAverage study hours• 0.024Strived or higher grades• 0.040Did more work than needed to pass exams• -0.031

. Mobilization o human resources a ter higher education

In this section we will try to put a urther piece o the puzzle into place by describingsome indicators that may be regarded as relevant to the mobilization by graduateso their own resources. We start by brie y describing the extent to which graduatesmobilize their own human resources: are they actively engaged in the labour orce, i so or how many hours and at what level, to what extent do they utilize their capaci-ties in the hours when they are at work and what other activities are they engaged in.In this section we try to provide an impression o this dimension o mobilization. We

then present some indicators o the extent to which graduates are involved in mobi-lizing the human resources o others: are they directly responsible or supervising ormonitoring the per ormance o other sta members, and do they have real strategicdecision-making authority at the level o the organization?

. . Mobilizing one’s own capacities

Labour orce participation and education-job matchChapter reports extensively on labour orce participation and the education-job

match, so we will not dwell long on this here. It is sufcient to mention a numbero the most striking results. Te authors show that around a three-quarters o rstlevel graduates and a slightly lower proportion o second level graduates are currently employed in jobs that match their own level and eld o education. Te remaininggraduates are either unemployed or are employed in jobs or which their own level and/or eld o education is not considered appropriate. Te authors re er to these gradu-ates collectively as “mismatched”. British and Spanish graduates have relatively highshares o mismatches, while relatively ew Finnish and Norwegian graduates samplesare mismatched. Czech and British rst level graduates are quite o ten employed in

jobs that do not match their own eld o education, while Spanish graduates are moreo ten employed or working jobs that match neither their own level nor their owneld. Tis shows that the Spanish sample more o ten than the other samples experi-



CHAPTER

ence the most severe orms o mismatch. Graduates in the Humanities are most likely to be mismatched both in terms o being unemployed and in terms o having employ-ment at an appropriate level and in an appropriate eld. Health and Wel are graduates

are least likely to experience such mismatches.

Working hours Mobilizing one’s own resources is not only a matter o having appropriate employ-ment. Graduates can only mobilize their own capacities in the hours that they actu-ally work, and many graduates work less than a ulltime week. Figure . shows thepercentage o graduates who work part-time.

Figure 5.6Percentage o graduates working part-time, by country (% o graduates in paid employment)

0%

5%

10%

15%

20%

25%


20-32 h.p.w.

< 20 h.p.w .

About % o all graduates work part-time. In general, the highest proportion o part-time work is seen in those countries with a high level o unemployment. Temajor exception to this rule is the Netherlands, where part-time work is known to beparticularly popular as a orm o work-sharing between young parents, and where theunemployment level is low. Most o these Dutch graduates work in “longer” part-time

jobs, with working hours between and hours per week. In countries such as Italy and Spain, where the unemployment level is relatively high and a higher proportion

o graduates work less than hours per week, it seems more likely that part-time work is more o ten involuntary, being the only work that graduates have been ableto obtain.

Tere is little di erence between rst and second level programmes in terms o part-time work (see Figure . ). Tere are however pronounced di erences between eldso study. Arts & Humanities and Education graduates are more likely to work shorter

. Since there is no international standard de nition o ulltime work, any cut-o point we choose willSince there is no international standard de nition o ulltime work, any cut-o point we choose willbe somewhat arbitrary. We adopt a conservative de nition o ull-time work. Based on the assumptionthat a standard working day is no more than 8 hours, anybody working hours or more per week willbe working or more than the equivalent o our standard days. We de ne this or our purposes as aulltime working week.




hours than graduates in other elds. By contrast, only a small proportion (less than%) o all graduates at both levels in Engineering, Manu acturing & Construction

work part-time.

Figure 5.7Percentage o graduates working part-time, by eld and level o education (%)

0 %

10 %

2 0 %

3 0 %

f i r s t l ev e l s e c o nd l ev e l

Utilization o knowledge and skills Although we have established that a large proportion o graduates work or longhours in jobs matching their education, this provides no guarantee that graduates’

capacities are sufciently utilized. It is o ten assumed that overeducation impliesunderutilization. However, in recent years, there has been an increasing awarenessthat, although overeducation is likely to be related to underutilization, the two are inact quite distinct (see e.g. Allen & Van der Velden ). By no means all overedu-cated workers ail to utilize their capacities and, conversely, some adequately educated

workers are less than satis ed about the extent to which their knowledge and skillsare utilized in their work. Such discrepancies may be due to the act that graduatesare in act more or less able than their level o education suggests or, alternatively, tothe act that the requirements o the job in terms o knowledge o skills is di erentrom what one would expect rom the ormal level o education required. Since it is

actual mobilization o graduates’ own capacities we are interested in, we need a moredirect indicator than overeducation. Figure .8 shows the proportion o graduates percountry who report that they utilize their capacities to a high or very high extent.8

8. Answer or on a ve-point scale ranging rom (not at all) to (to a very high extent) to the question Answer or on a ve-point scale ranging rom (not at all) to (to a very high extent) to the question”o what extent are your knowledge and skills utilized in your current work?“



CHAPTER

Figure 5.8Utilization o knowledge and skills, by country (%)

0%

20%

40%

60%

80%

100%


In general, as one would expect, there is quite a strong correspondence betweenthe degree o overeducation and the degree o skill utilization in a country. Tecountries that are shown in Chapter to have high levels o mismatches are gener-ally speaking the countries with the lowest levels o skill utilization and vice versa.First level programmes show slightly lower levels o skill utilization than secondlevel programmes (see Figure .9). Again mirroring the results in Chapter , Arts& Humanities graduates show a relatively low degree o utilization, and Health &

Wel are graduates a high degree.

Figure 5.9Utilization o knowledge and skills, by eld and level o education (%)

0 %

2 0 %

4 0 %

6 0 %

8 0 %

10 0 %

F i r st l ev e l S ec o n d l ev e l

Tese results suggest that overeducation is indeed related to skill utilization. Inorder to con rm that this also applies at the individual level, Figure . shows thepercentage o graduates that report high levels o utilization, by categories o educa-tion-job match.




Figure 5.10Utilization o knowledge and skills, by education-job match (%)

0%

20%

40%

60%

80%

100%

Higher levelappropriate

Same levelappropriate

Lower level of tertiaryeducation appropriate

Below tertiary levelappropriate

Own or related field appropriate Other/no field appropriate

Figure . indeed con rms the expected relation. In every country, the highestpercentage o skill utilization is seen among graduates working at the same or a higherlevel, decreases somewhat or graduates working at a lower level o tertiary educa-tion, and is lower still or graduates working below tertiary level. At each level wealso observe a strong relation with the horizontal match: graduates working in jobsor which their own or a related eld is most appropriate report much higher levelso skill utilization than those working in jobs or which another eld or no particulareld would have been more appropriate. Although the expected relation between

education-job match on one hand and skill utilization on the other is con rmed,there are a number o points that are worthy o note. Firstly, although the utiliza-tion level is high or graduates working in jobs or which at least their own level andtheir own eld are most appropriate, it is well below %. One might be inclinedto dismiss this as random noise, but additional analyses con rm that the level o skillutilization within this group is clearly associated with higher values on outcome vari-ables like job satis action and income. More striking is the act that about one in vegraduates working below tertiary level and outside their own eld nonetheless reporta high degree o skill utilization. Tis result is once again validated by a strong rela-tion with outcome variables. Although nding employment that matches one’s own

level and/or eld o education obviously increases one’s chances o utilizing one’s ownknowledge and skills capacities, many overeducated graduates nonetheless manage tomobilize their own human resources in this respect. Given this variation in skill utili-zation within categories o education-job match (over which higher education andgraduates will have little i any direct control), it is o interest to identi y actors thathave an impact on it. We will return to this point later in the chapter.

Other activities Although the ocus o this chapter is mainly on mobilization o human resources within the world o work, it is important not to lose sight o the act that gradu-ates can also put their capacities to use in other areas. Figure . shows the degree



CHAPTER

o participation in the our weeks preceding the survey in activities other than paid work.

Figure 5.11Participation in past 4 weeks in activities other than work, by country (%)

0%

20%

40%

60%


education or training family care or childrearing voluntary work

* Education/training not asked in Belgium-Flanders.

Given the act that the survey was conducted around ve years a ter graduation, asurprisingly large proportion (about a third) o all graduates are involved in some kindo education or training. Tere are pronounced di erences between countries, with

around % o Czech graduates engaged in urther learning, compared to aroundone in ve in Estonia and less than a quarter in Italy and Norway. A slightly lowerproportion o graduates were involved in amily care, ranging rom more than hal inFrance, to less than % in the UK. Te proportion o graduates doing some kind o voluntary work is lower than that engaged in the other two classes o activity, but ataround % is still substantial. Almost % o Swiss graduates do voluntary work, andeven in the country with the lowest percentage (Estonia) more than 8% are engagedin this type o work.

Additional analyses (not shown here in detail) reveal that participation in training is

hardly related to labour orce status. It is highest among parttime workers, suggestinga kind o dual status incorporating study and work, and lowest among those not inactive employment, indicating that training and paid work are not generally speakingsubstitutes or each other, and may even be to some extent complementary. In contrast,there are clear indications that amily care and voluntary work are substitutes or paidemployment. Family care is very common among those not in the labour orce andrelatively rare among ull-time workers (although more than a quarter o ulltime

working graduates still take on some caring duties). Unsurprisingly, ull time workersparticipate less in voluntary work than graduates working less hours or not at all.




. . Mobilizing capacities o others

We have established that most graduates are airly success ul at mobilizing their own

capacities. Ideally, we would like to nd out whether the same applies to mobilizingcapacities o others. In the case o one’s own capacities, we had a very direct indicatoro the degree to which these are actually utilized at work. In the case o mobilizationo the capacities o others, things are less straight orward. On one hand, we havequite a lot o indicators o the ormal role graduates play in the organizations in

which they work. We know whether graduates are responsible or supervision and/orquality control with respect to the work o others, and the extent to which graduatesbear strategic decision making authority in their organization. In other words, weknow whether graduates are involved in mobilizing the capacities o others. However,the impact o these things on the actual per ormance o other workers – the directmeasure o this kind o mobilization - takes place “o stage” as it were. Although wecan probably assume that, in general, employers assign such responsibilities to people

who they eel are best suited to them, we need to keep in mind that we may be missingvariance inhow well graduates are ul lling these duties.

Formal responsibility or other sta members Te simplest indicator o the role played in mobilizing others, and the one mostcommonly encountered in labour market research, is whether or not a person isresponsible or supervising others. Although this indicator is ar rom useless, it

clearly has its limitations. Te label ‘supervisor’ is used to describe a multitude o roles, ranging rom a simple “ rst among equals” role in a teamworking situation topositions o great authority and responsibility. Nor does additional in ormation helpus much, since genuine authority gures in many organizations may only have one ortwo other key gures working under them, while lower level managers on the work-oor may ‘supervise’ the work o tens or even hundreds o unskilled workers. A key question is that o control over the quality o per ormance o others. Figure . there-ore supplements in ormation on the proportion o graduates who supervise others

with data on the proportion o graduates who report a high degree o responsibility or assessing the work o others.

It is clear rom Figure . that only a minority o graduates are responsible or mobi-lizing others, even based on these rather minimalist indicators. About a third o gradu-ates supervise other workers, and only a quarter are responsible or assessing the work o others. Tis suggests that supervision may indeed o ten be a rather per unctory task

without much substance. Nonetheless, at the aggregate level o countries, there is aclear relation between the two. Estonian and UK graduates bear supervisory respon-sibility most o ten, and UK graduates are also most o ten responsible or assessingothers’ work. In contrast, German graduates score rather low on both indicators. Temain exception to the general pattern is ormed by French graduates, who rarely assessthe quality o others, but quite o ten supervise.



CHAPTER

Figure 5.12Responsibility or other sta members, by country (%)

0%

20%

40%

60%


Supervises others Assesses others' work quality

A little unexpectedly, rst level graduates are slightly more likely to supervise othersthan their more highly quali ed second level graduates (see Figure . ).9 Tis di er-ence is probably attributable to the act that second level graduates are much morelikely to work as autonomous pro essionals than their rst level peers. Engineeringand Agriculture graduates o ten have such responsibilities. Education and Arts &Humanities graduates are less likely to do so.

Figure 5.13Percentage o graduates who supervise others, by eld and level o education (%)

0 %

2 0 %

4 0 %

6 0 %


Strategic decision making authority Regardless o whether they actually work with others, graduates who play a strongrole in setting goals and/or deciding strategies or their organization will thereby alsoin uence the mobilization o their co-workers. Although we have indicators o therole o graduates in both setting goals and deciding strategies or their organization,these are highly correlated. o avoid unnecessary repetitions, Figure . shows the

9. o avoid cluttering things, the percentage o graduates who assess others’ work quality is not shown.Te overall pattern or this indicators is similar to that or supervision.



CHAPTER

Figure 5.15Strategic decision making authority,* by eld and level o education (%)

0 %

10 %

2 0 %

3 0 %

4 0 %

5 0 %


Note *: Unweighted mean o percentage in small, medium and large organizations

5.5 Determinants of utilization of own capacities

Having established, in section . , that higher education has only a modest capacity toin uence the level o mobilization competences, we may wonder whether it can real-istically make a contribution to increasing the actual level o mobilization. We do this

by way o a series o multivariate regression analyses. In this section we look at deter-minants o utilization o own capacities. O course a key point hereby is to establish whether the competences relevant to mobilizing human resources have the expectede ects. It is important to remark at this point that although we have earmarked theabovementioned six abilities as mobilization competences on theoretical grounds, wecannot be certain in advance that these are the only, or indeed even the most impor-tant, competences that play a role in mobilizing human resources. For this reason,

we also include clusters o the competences representing pro essional expertise, unc-tional exibility and innovation and knowledge management. In addition, because wecannot be sure that all o the e ects o higher education occur through competences,

we include the same set o higher education characteristics and characteristics as wereincluded above as predictors o competences. Such characteristics may in uence utili-zation directly, by making graduates better at getting the most out o themselvesin difcult situations, but also indirectly, by improving graduates’ chances o beingselected or jobs that are well matched to their abilities. In addition to competencesand higher education characteristics and experiences, we also include some indicatorso experiences gained outside higher education and some characteristics o the organi-zations in which graduates currently work that may have an e ect on their ability tomobilize their own human resources. Te results are shown in ables .9 to . .

Looking rst to the results as a whole, we can say that a limited number o predictorshave very strong e ects, while most predictors have little or no e ect. All in all, the




model explains about one eighth o the total variance in utilization. Encouragingly,a large part o this is accounted or by competences and higher education character-istics and experiences. Tis suggests that, although higher education only explains a

relatively small proportion o mobilization competences, it has a meaning ul e ecton the actual utilization o one’s own capacities. Nonetheless, it is clear that utiliza-tion is in uenced more by actors outside our model than by the indicators we haveincluded. A large part o this is o course the match between one’s own education andthat regarded as appropriate or the job. Tis has not been included here because it isconceptually so closely intertwined with utilization, which would mask a lot o thee ects o our predictors. More interesting, especially in the case o the e ects o highereducation characteristics, is the pattern o e ects within categories o mismatch. I education has a role to play, this may be mainly in those un ortunate but inevitablesituations where graduates nd themselves in employment not matching their educa-tion. Especially then graduates are likely to bene t rom having competences thathelp them get more out o themselves. For this reason, the e ects o competences andhigher education characteristics and experiences are shown separately or di erentcategories o mismatch. urning to the e ects o individual predictors, able .9 shows the e ects o di erentkinds o competences.

Table 5.9Utilization o capacities, by own level o competences (standardized linear regression coe cients)

own leveland eld

own level,other eld

lower level,own eld

lower level,other eld

all employees

Mobilization competenceAbility to per orm well under pressure 0.031 0.058 0.043Ability to use time e ciently -0.098Ability to work productively with others0.026Ability to mobilize the capacities o othersAbility to make your meaning clear 0.035 0.068 0.031Ability to coordinate activities -0.023 0.100

Clusters o other competencesPro essional expertise 0.099 0.202 0.120Functional fexibility -0.036 -0.063 -0.065Innovation and knowledge management0.050 0.060

Te results shown in able .9 are surprising in several respects. First o all, the compe-tences that were thought to be important or mobilizing one’s own resources – theability to per orm well under pressure and the ability to use time efciently - haverelatively little e ect. Te ability to per orm under pressure does help somewhat in

jobs matching the graduates own eld, but the ability to use time efciently has nopositive e ect at all and even a strong negative e ect in jobs that match neither the



CHAPTER

graduates’ level nor their eld. We might speculate that in such jobs, which presum-ably place ew demands on graduates’ specialized abilities, being able to organize one’stime efciently may only exacerbate the problem.

At least as surprising as the relative absence o e ects o the competences that wereexpected to be especially relevant is the strong e ect o the cluster pro essional exper-tise. Tis is the only competence or cluster to show a signi cant e ect when gradu-ates’ work does not match their education. Even more remarkable is the nding thatthe e ect is strongest among graduates working in jobs or which their own or ahigher level but a di erent eld is regarded as most appropriate. It would seem thatpossessing a high level o pro essional expertise enables graduates to better utilize theircapacities in general, even (or especially) when their work doesn’t match their training.Competences related to innovation and knowledge management also improve mobi-lization, although not when the job doesn’t match one’s education. Te quite strongnegative e ects o unctional exibility could be a case o reversed causality: gradu-ates who are not in a position to do what they are good at may need to become moreexible.

able . show the e ects o several programme characteristics. Table 5.10Utilization o capacities, by programme characteristics (standardized linear regression coe cients)

own level andeld

own level,other eld

lower level,own eld


all employees

2nd level programme 0.033 0.094 0.043

Other programme characteristicsGenerally regarded as demanding 0.020Employers amiliar with content 0.062 0.069 0.073 0.089Freedom to compose own programmeBroad ocusVocational orientation 0.063 0.093 0.075Academically prestigious 0.030 0.038

Higher education characteristics have some residual e ects on utilization a ter control-ling or competences. Graduates o second level programmes are more success ul inutilizing their knowledge and skills than graduates o rst level programmes, althoughthis only holds in jobs that match the graduates’own level. Te amiliarity o employers

with the content o the programme has quite strong e ects, as does vocational orien-tation and to a lesser extent academic prestige. It may be that these e ects work moreindirectly, by increasing the chance that graduates nd their way to employers whoknow what they are capable o , than directly, by enhancing graduates’ abilities toget the most out o themselves. Tis is consistent with the nding that the e ects o these characteristics are largely con ned to graduates working in jobs matching their




own eld. Having graduated rom a demanding programme has no signi cant overalle ect on utilization a ter controlling or competences (on which, as we saw earlier, ithas a rather strong e ect).

Table 5.11Utilization o capacities, by modes o teaching and learning (standardized linear regression coe -cients)

own leveland eld

own level,other eld

lower level,own eld


all employees

Lectures 0.043 0.033Group assignmentsParticipation in research projectsWork placements/internships 0.097 0.029

Facts & practical knowledge 0.038 -0.073Theories & paradigmsTeacher as source o in ormationProblem- or project-based learning 0.071Written assignmentsOral presentations 0.101Multiple choice exams

Modes o teaching and learning have ew residual e ects a ter controlling or compe-tences (see table . ). Te e ect o work placements may, like vocational orientation

and amiliarity o employers with the content o the study programme, be indirect,increasing the chances that graduates are employed by organizations that are gearedto their speci c knowledge and skills. Such an e ect is however not plausible orlectures, which also show a positive e ect. It is not clear what mechanism underliesthis e ect.

As able . shows, experiences either be ore, during or a ter higher education havelittle e ect on utilization. Only study-related work experience during higher educa-tion and work experience since graduation have signi cant positive e ects, but theseare quite small. Te only other signi cant e ect in able . is that o initial search

duration, which decreases the level o utilization.



CHAPTER

Table 5.12Utilization o capacities, by experiences be ore, during and since higher education (standardizedlinear regression coe cients)

own leveland eld

own level,other eld

lower level,own eld


all employees

EXPERIENCES BEFORE HEStudy-related work experienceNon study-related work experience

EXPERIENCES DURING HEStudy-related work experience 0.028Non study-related work experienceVoluntary positionsWork placements

Experience abroad

EXPERIENCES AFTER HEWork experience 0.075 0.023Initial search duration -0.020 -0.034

Table 5.13Utilization o capacities, by study achievement and motivation, cultural and social capital andparental role (standardized linear regression coe cients)

own leveland eld

own level,other eld

lower level,own eld


all employees

Study achievement and motivationRelative grade 0.022 0.037Did more work than needed to pass exams 0.020Strived or higher grades 0.027

Cultural and social capitalQuality o social network 0.077 0.073 0.063 0.124 0.090At least one parent has HE 0.021

Parent o young child (<5yrs)FatherMother -0.039 -0.033

As able . shows, study achievements, in the orm o grades, has a signi cant, butnot very large e ect on utilization. Study motivations, indicated by the extent to

which graduates were prepared to do more work than needed to pass exams and/or tostrive or the highest possible grades, has no overall signi cant e ect on later utiliza-tion, although a very weak e ect is seen or graduates working in jobs matching theirown level and eld o education. Social capital, in the orm o a good social network,has a strong positive e ect on utilization – also in non-matching jobs – but cultural




capital, in the orm o having at least one highly educated parent, only seems toimprove utilization o one’s knowledge and skills very slightly. Both these e ects may be due to an increased chance o nding employment in which one can utilize more

o one’s capacities rather than through an increased ability to get more out o onesel .However, we should remark that the measure o social capital, like the dependentvariable, re ers to the situation as it was when graduates completed the questionnaire.In contrast to the other predictors described so ar, which re er to the situation duringor even be ore higher education, we cannot plausibly claim that this relation is causal,only that graduates who have a good social network utilize their capacities more onaverage than graduates with a less use ul network. Having at least one child under veyears has a small but signi cant negative e ect, however only or women.

able . shows the e ects o various characteristics o the organization in which grad-uates work and the market in which the organization operates. Like social network,the situation described by these variables is concurrent with that or the dependentvariable, so the e ects should be regarded as descriptive rather than causal. As such,there is little point in considering the results separately or di erent categories o education-job match. We sufce with the results or all employed graduates.

Table 5.14Utilization o capacities, by organization and market characteristics (standardized linear regressioncoe cients)

Sel -employed 0.060Public/non-pro t sector 0.122

SCOPE OF OPERATIONS (re .: local):RegionalNationalInternational

SIZE OF ORGANIZATION (re .: <10 employees):10-49

50-99100-249 -0.021

250-9991000 or more

MARKET CHARACTERISTICSStrong competitionCompetition based on quality 0.030Unstable demand -0.020

CHANGES IN ORGANIZATIONWork tasksReorganization -0.033Large-scale layo s -0.038

Organization at the ore ront o innovation 0.073Extent to which per ormance is monitored



CHAPTER

As one might expect, the sel -employed are better able to utilize their own capacitiesthan those who work or others. Working in the public or non-pro t sector has aneven stronger positive e ect. Tere is little evidence that working in larger organiza-

tions, and/or in organizations with a national or international scope, allows graduatesany more or less opportunities to utilize their capacities than they have in smallerand/or more locally oriented organizations. O the market characteristics, competi-tion based on quality (as opposed to price) has a small positive e ect on utilization.Instability in demand is associated with lower levels o utilization. Organizationalchange in the orm o reorganizations and/or large scale layo s, have a negative e ecton utilization, although this is not very large. Graduates working in organizations thatare more at the ore ront when it comes to introducing innovations have more oppor-tunities to utilize their knowledge and skills than graduates who work in organizersthat tend to ollow rather than set the trend. Tere is no signi cant e ect o moni-toring o per ormance, suggesting that, in general, graduates neither abuse the greaterdegree o reedom, nor make use o it to put their capacities to better use.

5.6 Determinants of mobilization of others’ capacities

As mentioned above in Section . . , in the case o mobilization o the capacitieso others, we only have indicators o whether graduates are involved in mobilizingthe capacities o others, and not o how well they are doing so. Tis means that

any observed e ects o educational and background characteristics work throughincreasing or decreasing the chance that employers assign such responsibilities tograduates, and not necessarily through making graduates better or worse at actually ul lling such roles. Since we assume that employers will be more inclined to assignsuch responsibilities to people who they eel are best suited to them, educational andbackground characteristics may act as signals (or help to promote other characteristicsthat in turn are seen as signals) o suitability.

In this section, the results o three multivariate analyses will be presented. Tedependent variables are supervision (logistic regression analysis o chance that one is a

supervisor), assessing quality o others (OLS regression analysis o the extent to whichone is responsible or this) and strategic decision making authority (OLS regressionanalysis o the mean o the two underlying variables). Most o the same independentvariables are included as in the analysis o utilization. ables . to . show theresults o these analyses.

In all three cases, higher education variables and competences account or -8% o thetotal variance in the dependent variables. Particularly competences have quite strongand consistent e ects (see able . ). As we might expect, the ability to mobilizethe capacities o others has the strongest e ects on all three indicators, especially thechance that one supervises others. A similar pattern holds or the ability to coordi-nate activities and the ability to per orm well under pressure, but the e ects are not




as strong. Contrary to expectations, the ability to make your meaning clear has noe ect on any o the three outcomes, and the ability to work productively with otherseven has a negative e ect on supervision and strategic decision making authority. Tis

seems to suggest a rather hierarchical attitude towards leadership. Surprising as well isthe nding that the ability to use time efciently signi cantly reduces the chances thatone is assigned a leadership role in terms o the three indicators.

Table 5.15Mobilization o capacities o others, by own level o competences (regression coe cients)10

supervise others quality controlstrategic decision-making authority

Mobilization competencesAbility to per orm well under pressure 0.133 0.055 0.025

Ability to use time e ciently -0.066 -0.050 -0.054Ability to work productively with others -0.107 -0.047Ability to mobilize the capacities o others 0.228 0.139 0.094Ability to make your meaning clear -0.023Ability to coordinate activities 0.198 0.079 0.078

Clusters o other competencesPro essional expertise 0.187 0.058 0.049Functional fexibility 0.026 0.081Innovation and knowledge management 0.029

As was the case or utilization, the competence cluster representing pro essional exper-tise has strong e ects. Tis is not so surprising: one would expect employers to putsomeone who knows his or her stu in charge. In contrast to what we saw or utiliza-tion, unctional exibility has a positive e ect on quality control and strategic deci-sion making authority. Innovation and knowledge management has only a rather

weak e ect on strategic decision making authority.

Table 5.16Mobilization o capacities o others, by programme characteristics (regression coe cients)

supervise others quality control strategic decision-

making authority2nd level programme 0.153

Other programme characteristicsGenerally regarded as demandingEmployers amiliar with contentFreedom to compose own programme -0.077Broad ocusVocational orientationAcademically prestigious

. Te coefcients presented in ables . to . are logistic regression coefcients in the case o “super-vising others” and standardized linear regression coefcients in the other two columns.



CHAPTER

Programme characteristics have little e ect on leadership (see able . ). It seemsthat employers look or direct signals o leadership traits in terms o competencesrather than relying on characteristics o the programmes graduates have ollowed.

Graduates o second level programmes are slightly more o ten employed as supervi-sors, while graduates who reported a high degree o reedom in composing their ownstudy programme were less likely to be supervisors.

Table 5.17Mobilization o capacities o others, by modes o teaching and learning (regression coe cients)

supervise others quality control strategic decision-making authority

LecturesGroup assignments

Participation in research projects 0.021 0.041Work placements/internshipsFacts & practical knowledgeTheories & paradigms 0.019

Teacher as source o in ormationProblem- or project-based learning 0.031 0.026

Written assignmentsOral presentationsMultiple choice exams 0.019

Te reliance o employers on competences rather than educational proxies to assignleadership roles is urther borne out by able . . Although several modes o teachingand learning have signi cant e ects on quality control and strategic decision makingauthority, the e ects are quite weak. A strong emphasis on participation in researchprojects and on problem- or project based learning has a positive e ect on both indi-cators, while emphasis on theories and paradigms and – curiously – multiple choiceexams increase the chances that one is responsible or strategic decision making.



CHAPTER

(the interaction e ect with gender) the e ect o having one or more young children isuni ormly negative, while or men (the main e ect), it is consistently positive.

Finally, able . shoes the e ects o organization characteristics. Being sel -employedtrivially increases the extent o strategic decision making authority, and only slightly less trivially the extent to which one is responsible or quality control. Graduates

working in the public or non-pro t sector are less likely to supervise and/or assess the work o others. Size and scope o organizations have the expected (trivial) e ects onstrategic decision making authority. More interestingly, there is also a negative e ecto size on supervision and quality control. It is not immediately obvious why theseorganizations would be less likely to let higher education graduates supervise or assessothers.

Table 5.20Mobilization o capacities o others, by organization characteristics (regression coe cients)

supervise others quality control strategic decision-making authority

Sel -employed 0.038 0.190Public/non-pro t sector -0.318 -0.085

SCOPE OF OPERATIONS (re .: local)Regional -0.036National -0.075

International -0.102

SIZE OF ORGANIZATION (re .: <10 employees)10-49 -0.10450-99 -0.258 -0.035 -0.116100-249 -0.136250-999 -0.391 -0.1321000 or more -0.464 -0.025 -0.142

MARKET CHARACTERISTICS

Strong competition 0.027Competition based on quality 0.026Unstable demand 0.019

CHANGES IN ORGANIZATIONReorganization 0.395 0.072 0.047Large-scale layo s -0.030

Organization at the ore ront o innovation 0.051 0.043 0.082Extent to which per ormance is monitored 0.033 -0.059




Graduates working in organizations experiencing strong competition, and those working in organizations experiencing unstable demand are more likely to superviseand/or assess the work o others. Tere is however no e ect on strategic decision

making authority. By contrast, the more competition is based on quality, the moreresponsibility graduates bear or strategic decision making.

Te degree o stability o the organization and its environment seems mainly to work in the graduates’ avour in terms o their being assigned leadership roles. We saw already that unstable demand increases graduates’ role in quality control. Further,graduates in organizations that have undergone a reorganization since they started

working there are much more likely to be supervisors and to bear responsibility orquality control and/or or strategic decision making. Tis is consistent with the ideathat supervisors survive. However, graduates in organizations that have experiencedlarge-scale layo s are given less strategic decision making authority. Innovations inproduct or service, or in knowledge or methods, appear to provide organizations witha reason to assign higher education graduates more leadership responsibility. Finally,the only characteristic to show opposing e ects is the extent to which a graduatesown per ormance is monitored. Graduates or whom this is the case are in turn morelikely to be responsible or controlling the quality o the work o others. Tis may suggest that the degree o control or monitoring may to some extent be a structuralcharacteristic o organizations as a whole. In contrast, graduates whose work is closely monitored bear less strategic decision making authority.

. Conclusions

In this chapter we looked at several indicators o mobilization o human resourcesduring and a ter higher education and attempted to shed some light on the actorsthat promote or inhibit such mobilization. We ound evidence that European studentsare somewhat economical with the e ort they put into achieving good study results inhigher education. Tey work slightly less than hours per week on their study, butonly a minority reports doing substantial extra work above what was required to pass

their exams. Students appear to be more extrinsically than intrinsically motivated: tothe extent that they put in extra e ort, they want to see this rewarded in the orm o higher grades. Tere are substantial di erences between countries, with Dutch gradu-ates putting in the least e ort and Spanish graduates putting in the most accordingto the indicators used.

I students don’t work as hard as they might on their study, this does not mean thatthey are idle. On average students put in almost months during their study onother activities, mainly paid employment. Again, we see strong di erences betweencountries, with Spanish graduates doing least and Dutch graduates the most. Tisresult would appear to suggest that there is a trade-o between study and extra-curricular activities, but multivariate analyses reveal that the relation between the




bear a high degree o strategic decision-making authority, although in medium andsmall organizations this proportion drops to about a quarter and a th respectively.

Surprisingly, the degree o mobilization o own capacities appears to be more strongly in uenced by one’s own level o pro essional expertise than by speci c mobilizationcompetences. Tere are relatively ew residual e ects o higher education characteris-tics and experiences a ter competences have been taken into account. However, one’ssocial network appears to be a good predictor o all orms o mobilization o humanresources, suggesting that knowing the right people can help get one into demanding

jobs with real authority. Several characteristics o the organizations graduates work in and the context in which it is located have signi cant e ects on mobilization.Private sector employees are less likely to utilize their own capacities, but more likely to play some kind o leadership role in the organization. A similar split is observedor reorganizations, which have negative e ects on utilization, but positive e ects onmobilization o others. Working in an organization which is at the ore ront in termso innovation has a positive e ect o all orms o mobilization.

Re erences

Allen, J. & R. van der Velden ( ), Educational Mismatches Versus Skill Mismatches: E ectson Wages, Job-related raining, and On-the-job Search, Ox ord Economic Papers, 3, -

. Asselberghs, K., Batenburg, R., Huijgen, F. and Witte, M. de ( 998).Kwalitatieve structuur vande werkgelegenheid in Nederland . Deel I . [Te qualitative structure o employment in theNetherlands. Part IV.] Te Hague: OSA.

Cohn, Elchanan, and Shahina P. Khan. 99 . “Te Wage E ects o Overschooling Revisited.”Labour Economics,( ): - .

Duncan, G. and S. Ho man ( 98 ), ‘Te Incidence and wage E ects o Overeducation’,Economics o Education Review,, -8 .

Hartog J. and H. Oosterbeek ( 988), ‘Education, Allocation and Earnings in the Netherlands:Overschooling?’,Economics o Education Review , , 8 - 9 .

Hersch, 99 ; Education Match and Job Match, Te Revieuw o Economics and Statistics, ( ):- ).

Sicherman, Nachum. 99 . ““Overeducation” in the Labour Market.” Journal o Labour Economics 9( ): - .

M.S.M. van Smoorenburg & R.K.W. van der Velden ( ), Te raining o School-leavers:Complementarity or Substitution? In: Economics o Education Review, 19, , pp. - .



Chapter International Dimensions o HigherEducation and Graduate Employment

Ulrich Teichler

. The growing relevance o international dimensions

“Internationalisation” and “globalisation” tend to be named as key issues in Europe when changes in the relationships between higher education and the world o work areunder consideration. Te technological and economic dynamics are o ten describedas globalisation, whereby an increasing proportion o graduates is active in organi-sations which are globally and interconnected and have to use oreign language tocommunicate with persons rom di erent cultures, to build up in-depth knowledgeon other countries and to represent their organisation abroad. Li e in general becomesmore international with growing migrant populations as well as more and more grad-uates opting or careers in other countries. Universities historically were among the

most international organisations anyway, and in recent years many curricular re ormsare undertaken to prepare students better both or the globalising world in generaland or possible international careers in speci c (c . Huisman and van der Wende

; Altbach ; eichler ; Knight ). Last but not least an increasingnumber o students – many o them mobilized by the ERASMUS Programme o theEuropean Union - opt or temporary student mobility, whereby most o them expectthat this will help to cope more success ully with increasing international dimensionso the world o work and other spheres o li e and also believe that this leads to careerenhancement in general (see eichler , Bracht et al. ).

In the ramework o the REFLEX study on graduate employment and work, atten-tion was paid to mobility in the li e course: the country o origin and the country they lived, studied and worked at di erent li e stages. Tis does not only allow usto analyse patterns o mobility over these li e stages, but also to examine how inter-nationally mobile persons di er rom non-mobile ones in the early stages o career.

With respect to a growing international character o work assignment, the REFLEX study addressed the role oreign language pro ciency plays. Tus, it is possible bothto identi y the pro essional areas in which high levels o oreign languages are requiredand to examine how careers o those highly versatile in oreign languages di er romthose less pro cient in this respect. Moreover, as similar questions were posed in thepredecessor project CHEERS, it is possible to analyse changes rom the graduates



International Dimensions o Higher Education and Graduate Employment

About percent o the respondents lived in a country di erent rom that o graduationat the age o 16 . Actually, the majority o graduates surveyed which had come romother countries did not move to the country o study as a young child or during their

period o school education, but rather later or the purpose o study.

Table 6.1Mobility Prior to Study by Country (percent)

ES IT FR CH AT DE NL BE UK NO FI CZ EE TotBorn abroad 2 1 3 10 5 5 5 1 11 4 2 1 0 4Mother born abroad 2 2 11 25 10 5 7 2 18 5 1 3 0 7Father born abroad 1 1 10 23 9 5 7 2 18 4 1 3 0 7Abroad at age o 16 1 0 2 6 4 3 2 1 9 2 1 1 1 2

About twice as many o the REFLEX graduates’ parents as the REFLEX gradu-ates themselveswere born in another country , as able shows as well. Te numbero respondents living in the country o graduation since birth whose parents wereimmigrants is about as high as the number o respondents who were born in anothercountry and came to the country o graduation either as migrants as children ormoved to this country or the purpose o studying the ull degree programme.

Available educational statistics show that about percents o the students in theEuropean countries analysed during the later hal o 99 s had been oreign students

(c . Kelo, eichler and Wächter ; UNESCO ). It does not come as a surpriseto note that the proportion o oreigners among graduates according to the REFLEX survey is slightly lower. On the one hand, some short-term mobile students areincluded in the educational statistics who eventually graduate in the home country.On the other hand, the proportion o students not success ully graduating seems tobe higher among oreign students than among home students.

. . Mobility during the course o study

percent o the graduates responding in the REFLEX study reported that they spenta period abroad, i.e. di erent rom the country o graduation, during the course o study or purposes o study and/or work. Actually,

percent spent a period o study abroad and percent a period o work abroad during their course o study.

Tus, about two percent spent both periods o study and o work abroad. Otherstudent and graduate surveys suggest that most o the students reporting work abroadactually spend an internship abroad, i.e. working experience linked to their study.

able . shows substantial di erences o mobility during the course o study by country o graduation. But even in countries where temporary periods abroad or

●

●



CHAPTER

study and work were a less requent choice than on average in Europe in the latterhal o the 99 s, i.e. Spain, Italy, the United Kingdom, Estonia and Norway, a periodabroad or study or work is by no means an exception. Te proportion ranges even in

these countries rom to 9 percent.

able . shows as well that many graduates rom the Czech Republic ( %) andFinland ( %) had spent a period o work abroad during their course o study. Inreverse, ew Italian and Spanish graduates experienced an internship or another periodo work while still having been enrolled. Actually, those going abroad during theircourse o study or the purpose o study or work spent on average (arithmetic mean) a period o months or study and months or work in another country.

Table 6.2

Mobility During the Course o Study by Country (percent)ES IT FR CH AT DE NL BE UK NO FI CZ EE Total

Abroad during studyor study and/or work 16 17 36 30 36 30 30 28 19 19 32 27 17 26Spending time abroadduring higher educationor study 12 15 31 27 30 29 26 24 15 17 24 13 12 21Spending time abroad duringhigher education or work-related reasons 3 3 9 6 10 5 7 6 5 4 14 17 7 7Living abroad during studyprogramme 0 0 1 1 0 0 1 1 1 1 0 0 0 1Duration o time abroad Months or study

Arithm mean 7 5 6 7 7 8 6 6 10 8 6 6 8 7Median 6 4 4 6 6 6 5 5 6 6 5 4 4 6

Months or work-related reasonsArithm mean 7 7 6 7 8 4 5 3 9 8 6 5 6 6Median 4 3 3 5 5 3 4 2 5 6 3 4 4 4

Tese data by and large con rm results o other studies previous studies which hadocussed on temporary student mobility in Europe. Among the Western Europeanrespondents o the CHEERS survey who had graduated ve years earlier, 8 percenthad spent some time during the course o study abroad in order to study or to work (Jahr and eichler ), i.e. about two-thirds as many as among the respondents o the REFLEX survey. Te available data underscore the act observed in other studentsurveys as well that temporary student mobility was clearly on the rise in Europe overthe 99 s.

It should be noted that no Europe-wide statistics are available o the proportion o students spending a period abroad during their course o study (see the overview onavailable statistics in Kelo, eichler and Wächter ; data available or Germany



CHAPTER

Table 6.3International Mobility During the First Five Years A ter Graduation by Country (percent o thosegraduating in their home country)

ES IT FR CH AT DE NL BE UK NO FI CZ EE TotalMobility a ter graduation 19 18 21 28 27 16 18 32 21 10 18 24 24 22Abroad or study 8 10 6 12 7 5 4 9 6 5 4 7 13 7Abroad or work 11 11 17 19 23 13 16 28 18 7 15 18 14 16Living abroad when startingrst employment 3 2 5 4 9 2 2 3 8 1 3 3 2 4Living abroad ve yearsa ter graduation 2 3 3 6 9 2 1 3 7 1 2 2 4 3Working abroad ve yearsa ter graduation 2 3 4 2 9 3 1 4 7 1 3 2 2 3Duration o time abroad

Months or study Arithm. mean 8 9 9 8 11 8 6 9 8 10 7 8 11 9Median 6 3 4 4 5 4 3 5 4 9 4 5 6 5

Months or work-related reasonsArithm. mean 14 10 10 9 16 10 9 8 12 14 13 7 11 11Median 6 4 3 4 7 5 5 2 6 9 6 4 6 5

Kind o mobilitya ter graduation

Work long(more than one year) 5 2 5 6 9 3 4 5 7 3 5 3 5 5Work short(not more than one year) 8 7 11 13 12 9 11 20 10 4 9 15 9 11

Available labour market statistics suggest that only about three percent o the highly quali ed labour orce in Europe are nationals o other European countries. Te nd-ings o the REFLEX survey seem to be in line with these data; however, the REFLEX data do not suggest that intra-European pro essional mobility among highly quali ed persons is increasing substantially in recent years .

O all graduates surveyed in the REFLEX study who worked abroad ve years a ter

graduation, percent had chosen Germany, percent the United Kingdom, percent Switzerland and 9 percent the U.S. as destination. Tis pattern is clearly distinct rom that o graduates surveyed in the CHEERS study o whom - percenteach had gone to the United Kingdom, the U.S. and Germany.

As able . shows,Germany is the exceptionally requent destination or graduates rom Austria( % o graduates rom Austria universities working abroad in Germany) andthe Netherlands ( %),Similar priorities can be noted or graduates rom France going to Switzerland( %), rom Belgium to the Netherlands ( %) and rom Norway to Sweden ( %),

while graduates rom other countries opted or a broader range o destinations.

●

●




Table 6.4Major Countries o Work Abroad Five Years A ter Graduation by Country o Graduation (percent;only those graduating in their home country)

Country o employment ES IT FR CH AT DE NL BE UK NO FI CZ TotalGermany 12 9 16 22 43 0 32 7 2 0 12 20 17United Kingdom 17 16 5 21 4 10 7 21 0 9 12 34 12Switzerland 0 9 31 0 15 16 0 0 6 0 8 3 11United States 16 9 0 16 6 18 12 7 15 35 4 4 9Netherlands 2 0 3 7 3 2 0 33 0 10 6 6 7France 13 9 0 15 2 6 2 16 13 0 0 2 6Belgium 6 6 5 5 3 5 11 0 2 0 4 8 4Luxembourg 2 0 13 0 1 15 0 2 0 0 0 0 4Sweden 0 3 0 0 0 3 4 0 0 34 23 1 3Spain 0 10 4 3 0 2 0 2 0 0 6 1 3Ireland 6 3 2 0 0 0 4 0 12 6 7 0 2Canada 2 3 8 0 0 3 3 0 3 0 0 3 2Italy 8 0 0 5 3 0 0 0 7 0 0 1 2Denmark 1 5 0 0 1 0 0 0 0 0 9 0 1China 0 2 3 0 1 2 0 0 2 0 0 0 1Liechtenstein 0 0 0 3 5 0 0 0 0 0 0 0 1Austria 2 3 0 2 0 3 0 0 0 0 4 1 1Norway 3 0 0 0 0 0 0 0 6 0 4 0 1Australia 0 0 0 0 2 0 3 0 5 0 0 0 1United Arab Emirates 0 0 0 0 2 0 0 2 0 0 0 0 1Russia 0 1 0 0 1 0 3 0 3 0 0 1 1Thailand 0 0 0 0 0 6 4 0 0 0 0 0 1Slovakia 0 0 0 0 0 0 0 0 0 0 0 10 1Romania 0 0 3 0 1 0 0 0 0 0 0 0 1South A rica 0 0 0 0 3 0 0 0 0 0 0 0 1Portugal 8 0 0 0 0 0 0 0 0 7 0 0 1Other 2 13 5 0 7 10 15 10 24 0 2 5 8

Foreign graduates ( oreign nationals at the time o graduation in the CHEERS study,

graduates born abroad in the REFLEX study) show clearly distinct patterns o border-crossing mobility a ter graduation. Te CHEERS study already had shown that themajority o oreign graduates either returned to their home country or li e and work or got employed in a third country.

Among employed REFLEX graduates born in a country di erent rom that o gradu-ation,

percent were employed in their country o birth, and percent were employed in a third country

about ve years a ter graduation. Obviously, persons graduating abroad are – unlikethose graduating in their home country – not solely preoccupied with the career inthe country o graduation, but rather opt or subsequent border-crossing mobility inlarge numbers.

●

●



CHAPTER

. The impact o international mobility during the course o study and earlycareer on employment and work

. . The mani old relevance o early international mobility

Various studies undertaken prior to the REFLEX study have shown that mobility prior to study, temporary study and other study-related temporary activities abroadas well as studying the whole degree programme in another country are likely to leadgraduates rom European institutions o higher education to international careersand well as to work assignments in their home country or the country o graduationdi erent rom their home country that require visible international competencies.International experiences prior to and during the course o study are obviously valu-able assets both or mobility a ter graduation and or job roles characterized by visibleinternational dimensions.

Prior ndings are less conclusive, though, whether international experience prior toor during the course o study will lead to more high- ying careers in general. Forexample, surveys o ormer ERASMUS students suggest that they perceive advantages

with respect to insertion to the world o work as compared to their non-mobile peers,but that they do not reach a higher status a ew years a ter graduation (see Maiwormand eichler 99 ; eichler ; Bracht et al. ). Tere ore, it is interesting to

compare the careers o internationally experienced REFLEX respondents to those notmobile prior to and during the course o study.

. . Mobility during the course o study and early career

Graduates surveyed in the REFLEX study who had been internationally mobiletemporarily during the course o their study had a smoother transition to employ-ment in some respects than ormerly non-mobile students. Teir job search period

was somewhat shorter ( .8 months on average as compared to . months o the non-

mobile ones), and their overall period o unemployment during the rst years a tergraduation was clearly shorter on average.

It is interesting to note as well that graduates who had been mobile during the courseo study changed employers somewhat more o ten during the rst ve years a ter grad-uation than graduates who had not been mobile. Actually, ormerly mobile studentshad . employers on average as compared to . o those who had not been mobile.




. . The distinct pro le o those mobile during or shortly a ter the course o study

It is widely assumed that persons opting or a period o study abroad are a selectgroup. As a consequence, an eventual more success ul career o ormerly mobilestudents might not be the result o the study abroad experience but rather due todi erences in socio-biographic background and study between the ormerly mobileand the ormerly non-mobile students. It is o ten argued that ormerly mobile studentso ten come rom amilies where the ather and possibly the mother have a highereducation degree and have an above-average income and status. Moreover, ormerly mobile students are o ten depicted as highly motivated and energetic persons, i.e.persons with values and abilities which might turn out to be use ul on the job anyway.However, one cannot take this widely claimed arguments as well ounded – amongothers because the ERASMUS programme, the single most important institutionaldriver or temporary mobility o students within Europe, clearly intends to serve abroad range o students with respect to socio-biographic background, country andeld o study, as well as academic ability. For example, studies undertaken suggest thatERASMUS only seems to be socially selective i compared to all youth in Europe, butnot i compared to all youth enrolled in higher education.

In any event, a study aiming to analyse the impact o mobility during the course o study on the graduates’ career is well advised not merely to compare the careers o

ormerly mobile students to ormerly non-mobile students, but to take into consid-eration a possible selectivity e ect on this relationship. For example, the abovemen-tioned somewhat shorter job search period o ormerly mobile students than those

who had not been mobile during the course o study turns out to be signi cant whencontrolled by various socio-biographic dimensions as well as by various dimensionso study behaviour.

In the ramework o the REFLEX study, the impact o early mobility on subsequentcareer is analysed somewhat di erently rom prior studies. Te logic o the REFLEX study, undertaken ve years o graduation, is to ask about the impact o all prior

experiences on the competences and on the employment and work situation ve yearsa ter graduation. Tere ore, it is appropriate to examine the impact o both the inter-national experience during the course o study and o shortly a ter the course o study (i.e. prior to the biographic moment o the survey) on the employment and work situ-ation ve years a ter graduation. Similarly, we have to ask context rst: How do thestudents who had been mobile during the course o study and/or shortly a ter gradua-tion di er rom those students who had not been mobile at all up to the moment thesurvey was undertaken?

Actually, o the REFLEX respondents who had graduated in the country where they were born,

percent had been temporarily mobile only during their course o study,●



CHAPTER

percent had international experience both during their course o study andduring the rst ew years a ter graduation, and had been internationally mobile only during the rst ew years a ter gradu-

ation.able . provides an overall overview how respondents having been mobile duringor shortly a ter the period o study di er rom those not mobile as ar as their socio-biographic background and their course o study are concerned. Tose who had beenmobile:

come more rom amilies with parents graduating rom higher educationprogrammes,have been over-proportionally enrolled in Humanities and less than average inEducation as well as in Health and Wel are,have been requently enrolled second level programmes (i.e. those highereducation programmes that provide direct access to PhD programmes,

were more active in student organisations or other voluntary organisations,andhave participated more requently in internships or other work experienceduring the course o study.

Surprisingly – in contrast to overrepresentation o women in the ERASMUSprogramme – the percentage o women among the ormerly mobile persons is nothigher in the REFLEX survey than among the non-mobile persons. Also, those

mobile during and shortly a ter the course o study do not di er in the weekly timeand intensity they had spend on their study rom those who had not been mobile. Asked to rate their success during the course study, ormerly mobile studentsconsidered their grades as compared to other students o their programmeshigher than ormerly non-mobile students.

Te CHEERS study had shown that persons who had been mobile during the courseo study rated their general competences upon graduation only marginally higherthan those who had not been mobile. A signi cant higher level was only observed ora minority o general dimensions addressed. As one might expect, however, ormerly

mobile students rated them by ar superior in respect o dimensions directly relatedto international experience, e.g. oreign language pro ciency and understanding o other cultures.

●

●

●

●

●

●

●

●




Table 6.5Socio-Biographic Background and Study by International Mobility During and Shortly A ter theCourse o Study (per cent; only those graduating in their home country)

International mobility TotalNo mobility During A ter Both

GenderMale 39 37 53 47 41Female 61 63 47 53 59

Father or mother with higher education degreeISCED5+6 38 49 44 51 42

Field o studyEducation 11 7 5 5 9Humanities 10 18 12 20 12Social sciences 31 30 30 33 31Law 7 6 6 5 7Natural sciences 6 5 10 8 6Mathematics 4 3 5 2 4Engineering 18 19 26 19 19Health 13 12 7 7 11

Level o study programmeSecond level programme 49 63 55 68 54

Study behaviourStudy-related work experience during highereducation 44 53 44 54 46

Non-study related work experience duringhigher education

47 58 53 57 50

Participated in work placement/internships 57 62 50 56 57Position in student or othervoluntary organizations 19 32 27 36 23Average grade in study programmeAverage grade higher than average 50 54 57 61 52

Te REFLEX survey suggests, as able . indicates, similarly that those mobileduring the course o study perceive their general competences about ve years a ter at

most in a ew areas and marginally higher than those not having studied abroad orsome period. Te study shows, however, that international experience shortly a tergraduation is slightly more strongly linked to competences ve years a ter graduationthan international experience during the course o study.




Table 6.7International Mobility During the Course o Study and Shortly A ter Study and Select Aspects o Employment and Work Five Years A ter Graduation (percent; only home graduates living in homecountry ve years a ter graduation)

No mobility Only duringstudy

Only a terstudy

Both

Permanent employed in 2005 80 77 77 73Full-time employed in 2005 (35+) 78 80 83 80Appropriate (sub)level o education current job relative to highest sublevel currently attained

Higher level 11 11 12 11Same level 73 75 75 75Lower level o tertiary education 7 8 7 9Below tertiary level 9 7 6 5

Use ulness o study programmeProvided a good basis or current work tasks 52 50 49 52

Occupational positionOther 3 2 2 2Clerks 5 3 3 3Associate pro essionals 22 17 17 16Pro essionals 63 68 67 67Manager 7 9 10 12

Job characteristicsOpportunity to learn new things 63 67 68 69

New challenges 54 58 62 60High earnings 27 27 29 29Good career prospects 36 36 41 42Social status 38 41 41 44Job security 66 62 57 54

Size o the organisation1-49 35 29 28 3250-999 38 41 37 391000+ 27 30 35 29

Working in innovative organisations regarding

Product or service 45 47 52 52Technology, tools or instruments 39 40 47 45Knowledge or methods 50 50 54 54

Scope o operations o organizationLocal 24 15 11 11Regional 26 25 15 16National 24 25 21 21International 26 35 53 52

Gross monthly income job 2005( ulltime employed graduates)

Arithmetic mean (Euros) 2,511 2,878 2,725 2,873Median (Euros) 2,300 2,612 2,600 2,746



CHAPTER

Te last nding, again, supports the view that international experience and learningis very important horizontally, i.e. as a preparation or work either abroad or or work at home which is characterized by visible international components.

As already pointed out, the data presented in able . cannot be viewed as directimpact measures because they might be in uenced by other actors such as country o graduation, eld o study, kind o degree, etc. Some o these actors are controlled inthe multiple regression analysis presented in able .8. wo dependent variables arepresented: income and use ulness o the study programme as a basis or per ormingcurrent work tasks. Te multiple regression analysis was per ormed separately orcountries and study programmes within countries ( rst and second level programmes)in order to check i the relevance o international mobility during study dependson country and level o degree. wo models were developed and tested. In the rstmodel, only eld o study (dummies) and international mobility (dummy variable)

were included. In the second model also relevant bio-graphic variables (gender, schoolper ormance) and aspects o study behaviour (work experiences, activities in studentorganisations) were additionally included.

able .8, rst, indicates that the links between mobility and the employment and work situation ve years a ter graduation are mostly not spurious. An impact is visibleeven i socio-biographic and educational actors are controlled which might be avour-able or employment and work. Second, mobility during the course o study and

shortly a terwards contributes more to higher income than to a close link between thestudy programme and current work tasks. Tird, signi cant links between mobility and career cannot be observed in all countries and types o study programmes.

As regards the ormer we note that mobility has no signi cant career impact inFinland, Estonia, the Netherlands, Switzerland and the United Kingdom.Mobility during the course o study contributes avourably to income o thosegraduating rom second level programmesin the Czech Republic, France, Italy and Spain according to both models o analysis. Tese same holds true orNorwegian graduates only or the rst model and only with a lower level o signi cance and or Belgian graduates only according to the second model o

analysis. As regards rst level degrees, only graduates rom German Fachhochschulenreport a signi cantly higher income as a consequence o mobility during thecourse o study.Mobility during the course o study has signi cantly increased the perceiveduse ulness o the study programme or per orming current work tasks only orGerman university graduates, or Czech graduates o other study programmesand, on a lower level o signi cance or graduates o master-equivalentprogrammes in France.

Altogether, we note that the general career impact o mobility during the course o study or shortly a terwards is in some cases considerable. However, some e ects atleast are con ned to speci c countries.

●

●

●

●




Table 6.8The Pro essional Impact o International Mobility During Study by Country and Type o StudyProgrammes (signi cant regression coe cients; OLS)

Model 1 Model 2Income Use ulness o study

programmeIncome Use ulness o study

programmeIT First level

Second level ++ ++ES First level

Second level ++ ++FR First level

Second level ++ + ++AT First level

Second level +

DE First level ++ ++Second level ++

NL First levelSecond level

UK First levelSecond level

FI First levelSecond level

NO First level ++ ++Second level +

CZ First level ++ --Second level ++ ++

CH First levelSecond level

BE First levelSecond level ++

EE First levelSecond level

+ signi cant on the 5 % level++ signi cant on the 1% level.

. Foreign language pro ciency

Te REFLEX study addressed only oreign language pro ciency as a dimension o internationally relevant competences and job requirements. Tere ore, the ollowinganalysis will ocus on the di erences o employment and work between those gradu-ates with high and those with low oreign language requirements and those with highand low language pro ciency.



CHAPTER

International experience is a key asset or acquiring oreign language pro ciency: percent o those mobile prior to their study period rated their ability to

write and speak in a oreign language as high as compared to percent o

those not mobile prior to study.emporary mobility during the course o study or shortly a terwards is astronger actor in this respect. 8 percent o those going abroad during thecourse o study or purposes o study or work or shortly a ter graduation ratedtheir oreign language pro ciency as high as compared to percent o thosenot mobile during the course o study or shortly a terwards.

Jobs requiring a high level o oreign language pro ciency are as a rule the “better jobs” according to a large range o indicators o employment and work success. Asable 9 shows, jobs requiring a high level o oreign language pro ciency notably arecharacterized by more requent

high social status ( % as compared to %)positions as managers or pro essionals ( 9% as compared to 9%),good career prospects ( % as compared to %) andopportunities to learn ( % as compared to 9%).

In addition, jobs requiring a high level o oreign language pro ciency are char-acterized by considerably higher wages (€ , as compared to € , per monthon average). In contrast, jobs requiring high language pro ciency are some morerequently part-time.

Altogether, job characteristics di er more strongly according to oreign languagerequirements than according to actual language pro ciency o the graduates. Additional analysis – not shown here – reveals that careers are superior or those whose jobs require a higher level o oreign language pro ciency and who have also aacquired a high level o oreign language pro ciency.Foreign languages play a varying role according to economic sector:

On the one hand, oreign language requirements are pronounced in theproduction sector o the economy.On the other hand, oreign languages are ar less than average needed in healthand social work.

In the production sector, clearly the “better” jobs require by ar more o ten a high leveloreign language pro ciency. In the health and social work sector, the jobs requiringhigh oreign language pro ciency are also “better” in various respects, but the di er-ences are substantially smaller.

●

●

●

●

●

●

●

●




Table 6.7Foreign Language Pro ciency and Select Aspects o Employment and Work Five Years A terGraduation (percent; only home graduates living or working at home ve years a ter graduation)

Language

competences

Language

requirementsLow High Low High

Permanent employed in 2005 80 77 79 78Full-time employed in 2005 79 80 78 83Appropriate (sub)level o education current jobrelative to highest sublevel currently attained

Higher level 12 10 10 12Same level 72 75 71 76Lower level o tertiary education 7 8 8 7Below tertiary level 9 7 10 5

Appropriate (sub)level o education current job relative tostudy programme

Higher level 15 15 13 18Same level 71 73 71 73Lower level o tertiary education 5 5 6 4Below tertiary level 9 7 10 5Occupational positionOther 3 2 4 1Clerks 6 4 6 3Associate pro essionals 22 17 22 17Pro essionals 62 68 62 69Manager 7 9 7 10

Job characteristicsOpportunity to learn new things 61 67 59 73New challenges 53 60 51 66High earnings 25 29 24 32Good career prospects 34 36 31 42Social status 35 41 34 45Job security 65 61 64 60

Size o the organisation1-49 32 28 32 2850-999 35 36 35 361000+ 32 36 33 37

Working in innovative organisations regardingProduct or service 43 52 42 56Technology, tools or instruments 36 44 34 50Knowledge or methods 48 54 46 59

Scope o operations o organizationLocal 27 16 27 12

Regional 27 21 29 16National 24 22 25 20International 22 42 19 52

Gross monthly income job 2005 Low High Low HighArithm.mean 2414 2775 2454 2750Median 2193 2500 2242 2500



CHAPTER

. Internationally mobile careers compared to home careers

As already reported, about percent o the REFLEX respondents worked abroad ve

years a ter graduation. Tis held true ortwo percent o those who studied all the time and graduated at their homecountry,ve percent o those who studied temporarily abroad and graduated in theirhome country, and9 percent who graduated in a country di erent rom their location o birth.

By and large, those working abroad di er to a lesser extent rom those working athome as ar as various dimensions o their employment and work are concerned (seeable .8) than those in jobs requiring a high level o oreign pro ciency rom thoserequiring little or no oreign language pro ciency ( able . ).

Actually, those working abroad stated ar more o ten than the pro essionally non-mobile ones that their job is characterized by

good career prospects ( % as compared to %) andopportunities to learn ( % as compared to 9%),

working in organisations innovative with respect to technology, tools or instru-ments ( % as compared to %).

In contrast, graduates working abroad, as compared to graduates at home, reported,as compared to graduates working in the country o graduation, only somewhat morerequently managerial and pro essional positions, a high status, good job prospects,

new challenges and opportunities to learn new things. For example, 8 percent o those working abroad ve years a ter graduation were in managerial and pro essionalpositions as compared to percent o those working in the home country.Tose working abroad are not consistently in a better employment and work situa-tion than those working in their home country. Te ormer had a longer process o transition to employment as well as less requently permanent jobs and perceived lessrequent a high level o job security. Finally, no di erences between both groups exist

with respect to work autonomy and regarding links between knowledge and work assignment.

●

●

●

●

●

●




Table 6.8International Career and Select Aspects o Employment and Work Five Years A ter Graduation(percent; only those who graduated in their home country)

Home AbroadPermanent employed in 2005 79 68Full-time employed in 2005 (35+) 80 84

Appropriate (sub)level o education current job relative tohighest sublevel currently attained

Higher level 11 14Same level 73 70Lower level o tertiary education 8 9Below tertiary level 8 8Appropriate (sub)level of education current job relative tostudy programme

Higher level 15 23Same level 72 63Lower level o tertiary education 5 6Below tertiary level 8 8

Occupational positionOther 3 4Clerks 5 2Associate pro essionals 19 16Pro essionals 65 69Manager 8 8

Job characteristicsOpportunity to learn new things 64 74New challenges 57 66High earnings 27 44Good career prospects 35 45Social status 38 44Job security 63 52

Size o the organisation (3 cat)1-49 31 2450-999 36 29

1000+ 34 47Working in innovative organisations regarding

Product or service 47 55Technology, tools or instruments 40 55Knowledge or methods 51 60

Scope o operations o organizationLocal 22 7Regional 24 13National 23 12International 32 68



CHAPTER

It should be added that graduates working abroad ve years a ter graduation earnedabout one tenth more per month than those working at home. In terms o income perhour, the ormer earned about EURO as compared to EURO o the latter. It is

difcult to establish the extent to which this is due to speci c allowances or oreignerstaking care o hardships o working abroad or due to a “real” income enhancement.

Table 6.9Monthly Income o Those Working at Home and Those Working Abroad by Country and Field o Study (EURO)

Country Home Abroad Diference Index Di %IT 1662 2320 658 140 40ES 1487 2115 628 142 42FR 2213 2967 754 134 34

AT 2683 3269 586 122 22DE 3684 5081 1397 138 38NL 2401 2962 561 123 23UK 2756 2895 139 105 5FI 2576 3054 478 119 19NO 3661 3799 138 104 4CZ 874 1051 177 120 20CH 4281 3466 -815 81 -19Field o study1 Education 2186 2201 15 101 1

2 Humanities 2158 1930 -228 89 -113 Social sciences 2569 3266 697 127 274 Law 2690 3576 886 133 335 Natural sciences 2483 2814 331 113 136 Mathematics 3050 4522 1472 148 487 Engineering 2760 3719 959 135 358 Medicine 2690 2793 103 104 4

Actually, highest income advantages o those working abroad are reported, able .9

shows, by graduates rom Spain ( %), Italy ( % and Germany ( 8 ). In contrast,hardly any di erence is stated by those rom the United Kingdom and Norway. Swissgraduates working abroad even have on average a 9 percent lower income than Swissgraduates working at home.

Finally, the REFLEX study con rms the ndings o previous studies that graduates who had studied abroad, who had acquired internationally competences and whoactually work abroad a ter graduation di er most strongly rom the non-mobile ones“horizontally”, i.e. in terms o visibly international job assignments. In the REFLEX study, this is demonstrated in able .8 by the act that more than twice as many graduates working abroad than those working at home are employed in an organisa-tion with an international scope ( 8% A compared to %).




. Concluding observations

Te choice o an institution o higher education and the choice o the country o work

has remained or the cohort o those graduating rom European institutions o highereducation around very much a national a air. Only our percent had graduatedin country di erent rom their nationality or their living environment, and only threepercent worked ve years a ter graduation in a country di erent rom that o theinstitution o higher education they graduated rom.

But internationalisation o competences and job requirements is widely spread. Morethan a quarter o the graduates reported that they spent a period abroad during thecourse o study either or purposes o study or or the purpose o work, whereby

work is o ten related to study (internships or similarly). And even a larger proportiono graduates take over job assignments where high oreign language pro ciency isrequired. Te in ormation provided by the REFLEX study and other studies suggestthat persons internationally experienced prior to graduation or shortly a ter are clearly more likely to be internationally mobile and are clearly more likely to take over jobs athome which require international competencies. Tis con rms a strong “horizontal”link between international learning and experience on the one hand and international

work on the other hand.

Tere are “vertical” links as well, i.e. between international experience and career

success, though less close and less consistent. In some respects, work abroad and work requiring visible international competencies are positively rewarded in terms o statusand desirable work tasks. Also, temporary study abroad is eventually awarded o ten by components o a more attractive career. But these “vertical” advantages are less requentand smaller. Tey do not hold true in all respects: International careers might requirelonger periods o transition to work and are o ten connected with higher employmentrisks. Moreover, some o the advantages visible at rst are spurious, because inter-national careers and international job requirements are more requent in economicsectors and occupational groups which have an above-average status. Finally, amongthose with international experience prior to study some are migrants experiencing

unequal opportunities in their country o study and work. But altogether, acquisitiono international experiences and competencies as well as choice o job requiring inter-national competencies and possibly international pro essional mobility are on averagesomewhat more highly rewarded than other study and career options.

Re erences

Altbach, P.G. ( ). International Higher Education. Re ections on Policy and Practice.Boston:Center or International Higher Education.



Chapter Winners and Losers

Liv Anne StørenClara Åse Arnesen

. Introduction

Te preceding chapters look at the outcomes o higher education in European coun-tries, each rom a particular perspective. In this chapter we adopt a more holistic view,attempting to determine to what extent particular groups o graduates can be identi-ed as “winners” or “losers” in the labour market. Analyses o labour market successesand ailures normally ocus on predicting objective measures such as unemployment,over-education and wages. Tis chapter will also analyse such actors, which implicitly treat participation in higher education as an economic investment on which bothindividual graduates and societies as a whole hope to recoup a satis actory economicreturn. Given the huge sums invested in higher education, this ocus on economicreturns to education is understandable and legitimate. However, it is important to

recognise that there are other ways o looking at success o graduates in the labourmarket. What i graduates strive or other things than secure employment with highearnings and succeed in reaching those other goals? Are those graduates not also“winners”? More generally, what makes someone a winner (loser) in one dimensiondoes not necessarily imply that he or she is a winner (loser) in other dimensions as

well.

In this chapter we will look at determinants o success and ailure on both objectiveand subjective measures. Te indicators o objective success or ailure are the employ-ment situation – have graduates managed to secure paid work, and i so does this

match their own attained level and eld o higher education? – and the wages earned.Te subjective measures concern work values and the realization o these values, and job satis action. We will explore to what extent the objective and subjective indicatorshave similar predictors. In other words; we will explore to what extent these di erentindicators overlap or not. Moreover, we will explore to what extent objective successpredicts success in the subjective dimensions. Attention will be paid throughout thechapter to country di erences in terms o success or ailure and how these di erencesmight be explained, as well as possible gender di erences and di erences by eldsand types o higher education. Be ore presenting the empirical results, we will brie y outline the oundations on which our analyses are based.



CHAPTER

Di erent orms o education–job (mis)match will be studied as indicators o labourmarkets success or ailure; both the most extreme orm o mismatch such as unem-ployment, and vertical educational mismatch which re ers to the lack o correspond-

ence between the level o the education acquired and the level required in the job. Also a third orm o possible mismatch will be studied, which we call horizontalmismatch; that is working in a job matching one’s own level but not one’s own eld o education. Tis may be a exible and rewarding way o labour market adaptation, orit might be a situation that is more or less orced upon the individual and represents akind o mismatch with possibly negative consequences on wages, realization o work orientations or job satis action. Te identi ying o horizontal mismatch is o specialinterest or our ourth orm o mismatch; those being both vertically and horizontally mismatched. Tis re ers to graduate persons holding jobs like or instance taxi-driveror shop assistant. In addition to labour market match or mismatch, our second objec-tive measure o the extent to which the graduates are success ul iswages .

Di erent theories have di erent explanations o success and ailures in the labourmarket. According to the assignment theory (Sattinger 99 ) the existence o labourmarket phenomena as . i. unemployment and over-education can be motivated aslabour market responses to the problem o assigning workers to jobs. Both individualsand jobs can be ranked in terms o skills. Individuals will be ranked according tothe skill level they possess and jobs in accordance with the skill level they require.I the there are more skilled workers than there are complex jobs, some individuals

will end up in jobs or which they are overquali ed. Tis implies that persons withno-matching jobs will be overeducated, have lower productivity and receive lower wages.

Problems in the educations-work transitions are o ten explained by the search theory (Hammermesh and Rees 98 ), which among other things points to lack o in orma-tion. Hartog ( ) also points to that due to the search process and to imper ectin ormation unemployment and over-education may exist temporarily and be a resulto “waiting room e ect”. We will examine the labour market situation – years a tergraduation, and thus we suppose that the waiting room e ect will have minor impact.

However, initial problems might have long-lasting e ects; thus, the theories o “statedependence” may be relevant contributing to explanations o the occurrence o unem-ployment and over-education. According to such theories (Heckman & Borjas 98 ;Heckman 98 , Andress 989) unemployment experiences early in the career may have negative e ects at a later stage. Persons with a previous period o labour marketproblems such as unemployment may have an increased risk o being unemployed orhaving a job or which they are over-educated as a result o a sel -en orcing process.

Although over-education is an issue that has been approached in several papers theextent to which over-education or other types o mismatch varies by type o educa-tion is studied quite seldom in the literature. In this chapter mismatch by eld o study and other characteristics o study programme will be taken into account. Green



Winners and losers

and McIntosh ( ) nd that business and management studies, as well as socialsciences, have the highest over-quali ed graduate rates. Heijke et al. ( ) examinethe role o ‘generic’ or ‘vocational’ educations in the transition to the labour market.

Tey ound, among other things, that those with vocational competencies more o tenhad an occupation within own educational domain than those with generic compe-tencies, urther, that there was a negative wage e ect o having work outside owndomain (horizontally mismatched).

Also in the human capital perspective over-education may exist; it may or instanceresult rom a choice because the low-level job is a good investment opportunity (Hartog ). Te human capital theory (Becker 9 , Mincer 9 ) states that aperson will be paid in accordance with his/her human capital, but in case o over-education the graduate is not paid according to his/hers potential marginal products(Green et al. 999, ). Empirical research shows that overeducated workers receivelower wages than workers with the attained level o education (Hartog ), andempirical research has also shown that human capital actors account or only a parto the wage variation, and or instance it does not account or the gender di erencesin wages ound in many studies. In addition to human capital, also social capital(Bourdieu, 98 ; Coleman, 99 ) may have an in uence on labour market opportuni-ties. Tis re ers to resources situated in social networks.

A job has both a pecuniary and non-pecuniary rewards in the labour market and the

graduates do not only strive or secure work or high wages. Mathios ( 989) argues .i. that when analysing wage di erentials among highly educated persons, one shouldtake into account the non-pecuniary actors o a job as well. Te analysis o realizationo work orientations which will be undertaken in this chapter is one way to take intoaccount non-pecuniary actors.

. Labour market situation – match or mismatch

We will investigate the labour market situation among the graduates by the use o a

variable we have called “mismatch”. Tis variable is based on the respondent’s sel -assessment o his/her job in relation to his/hers education. Sel -assessment is viewed asthe best available 1 measure concerning the measurement o education-job mismatch(Hartog ).Te graduates are grouped into ve categories, ranked in ascending order o presumedseverity o mismatch:

. A job analyst might do a better job, but sel -assessment is the most economic method and it probably as valid as job-analyses because the content o jobs change aster than the available instruments orstandard classi cations o jobs.

. See Hartog ( ), Allen and van der Velden ( ) and van der Velden and van Smoorenburg ( 99 )or a discussion o methods concerning the measurement o skills and education-job (mis)match.

. See urther de nition in Appendix .



CHAPTER

Employed with relevant work (that is: no mismatch).Horizontally mismatched,that is working in a job matching one’s own level butnot one’s own eld o education.

ertically mismatched,that is working in a job matching one’s own eld butnot one’s own level o education.Both vertically and horizontally mismatched. Unemployed .

We will investigate the labour market situation at the time o the survey by educa-tional level and because o this we will use the educational level the graduateshad achieved at the time o the survey. We use the label “ rst level” or graduates

who have – years o higher education (equivalent to bachelors in some coun-tries) not providing direct access to doctorate . We use the term “second level” orgraduates with years o more higher education providing direct access to doctorate .

Te number o observations in the analyses below re ers to those who belong to thelabour orce , that is being employed, or being unemployed and seeking or a job. Tepercentage o the respondents who belong to the labour orce is 9 , varying rom 9per cent in Finland, 9 per cent in Czech Republic, Austria and UK, 9 per cent inEstonia, 9 per cent in France and Italy and 9 per cent in Germany, to 9 per cent inSwitzerland and Spain, 9 per cent in the Netherlands and Norway and 98 per centin Belgium.

O those who are in the labour orce, per cent are unemployed (weighted averageor countries), 73 per cent hold relevant employment and the rest are either vertically mismatched (9 per cent), or both vertically and horizontally mismatched ( per cent),or horizontally mismatched (8 per cent) according to the de nition above. Teseshares di er a lot by country and level o education, see below.

. . Labour market situation by country, education level and eld o study Te second level graduates are somewhat more o ten mismatched than rst level grad-uates, but this applies only to vertical mismatch, which involves mostly lower level

tertiary jobs in the case o secondly level graduates, but mostly jobs below tertiary levelor rst level graduates. Te share o unemployed is the same in both cases ( able . ).Figure . and . show how this varies between countries. Tose in relevant work arenot included in the graphs, to acilitate comparison o the o ten small proportions inthe other categories. Table 7.1Per cent mismatch, total sample, by education level

Horizontallymismatched

Verticallymismatched

Both horizontally andvertically mismatched

Unemployed

First level 10 5 6 4Second level 7 11 6 4

. .

.

4..



Winners and losers

Figure 7.1First level graduates. Mismatch at the time o the survey

2 46

8

16

56

12

69

3

1618

8 65

3

1

910

4

5

8

6

63

3 6 3 4

1

1

44

6

6

13

811

2

4 5 4

15

33 8

2

9

4

3

0

5

10

15

20

25

30

35

40

45

50

55

N o r w

a y I t a l y

F i n l a n

d

G e r m

a n y

E s t o n

i a

A u s t r

i a

S w i t z e

r l a n d

N e t h e

r l a n d s

F r a n c

e

B e l g i u m S p

a i n

C z e c

hR e p .

U n i t e d

K i n g .

P e r c e n t

Unemployed

Hor. + vert. m.m.

Vertic. m.m.

Horis. m.m.

Figure 7.2Second level graduates. Mismatch at the time o the survey

6 710

3

107

98 8 9

63

6

7 4

6

13

5 109 13

10

1624

17

26

45

45

55 4

7

7

8

6

12

17

3 5

22 4

57

2 7

42 8

5

0

5

10

15

20

25

30

35

40

45

50

55

F i n l a n d

S w i t z e

r l a n d

C z e c

hR e p .

N o r w a y

A u s t r i a

G e r m a n

y

F r a n c e

B e l g i u

m I t a

l y

N e t h e

r l a n d s

E s t o n i a

S p a i n

U n i t e d

K i n g .

P e r c e n t

Unemployed

Hor. + vert. m.m.

Vertic. m.m.

Horis. m.m.

Both among the rst and second level graduates the British and Spanish samples havehigher shares that are mismatched than most o the other samples, and this appliesalso to the Belgian and Dutch samples, however to a less extent. Te Finnish andNorwegian samples are among those with the lowest percentages who are mismatchedon both levels, ollowed by Germany and Austria. Else, the country di erences vary between the two levels. Among the rst level graduates both the Italian graduates andEstonian graduates have low shares who are mismatched and the Czech are amongthose with the worst situation, while the opposite is the case or second level gradu-ates.



CHAPTER

Te type o mismatch di ers a lot in the country samples with lowest shares with agood match. Te Czech and the British rst level graduates are quite o ten horizon-tally mismatched, while the Spanish more o ten are unemployed and are rather o ten

both horizontally and vertically mismatched. Tis shows that the Spanish samplemore o ten than the other samples experience the most severe orms o mismatch.Te British and Czech samples o rst level graduates also have high shares being bothvertically and horizontally mismatched. Tis indicates that their high shares being(only) horizontally mismatched may imply labour market problems.

We see that a relatively high proportions o the second level graduates experiencevertical mismatch. Tis might be due to the act that at part o them have takenurther education and graduated as second level graduates during the period –

( ). I those graduates are overeducated, this may indicate that they hold thesame type o position that they held be ore taking the urther education. able .shows whether late achievement o second level degree has an impact o the mismatchvariable.

Table 7.2Labour market situation among rst and second level graduates. Total sample o 1 countries

First level graduates Second level graduatesOriginally rst levelgraduates; second leveldegree obtained a ter

re erence year*

Second level degreeobtained in re erence year*

Unemployed 3.7 5.5 4.4Horizontally and verticallymismatched

5.5 8.5 6.0

Vertically mismatched 5.0 28.8 8.6Horizontally mismatched 9.9 4.1 7.7Relevant work 75.8 53.1 73.4N* (=100%) 9728 1304 12139

* The year in which the higher education degree re erred to in the questionnaire was obtained (in most countries1999/2000).

able . shows that – among other things – the act that a part o the second leveldegree graduates have obtained this degree during the – ( ) period, doesnot have a major impact on the total results or the second level group shown inFigure . . Te reason is that the group obtaining second level degree during the

– period is small; only ten per cent o the second level group had obtainedthis degree through urther education during the – period. However, wese a great di erence concerning the share being over-educated or their job betweenthose who originally had a second level degree and the (originally) rst level graduates

who later have achieved a second level degree. Tis means that the returns to educa-tion or the latter group do not (yet) ully correspond to their investment in urther



Winners and losers

education. We will later see whether this also applies to their wages. In multivariateregressions we will take into account whether or nor the degree was obtained throughurther education or whether the graduates have the same degree at the time o the

survey as the one they originally had in the re erence year or time o graduation. Tis will be done both in the wage regressions and in the mismatch regressions.

Figure . below shows how the labour market situation varies by eld o study. Techart shows that Humanities and arts on the one hand and Health and wel are on theother constitute the extreme points. Te shares being (only) vertically mismatched(over-educated) di er very little by eld o study; it is the combination o the di erentorms o mismatch versus holding relevant employment that di ers. Te results inFigure . also indicate that being horizontally mismatched may represent somethingnegative (a real mismatch), because those elds that have the highest share being hori-zontally mismatched (Humanities, Services, Social Science and Science), also havethe highest shares beingboth vertically and horizontally mismatched and/or beingunemployed. Figure 7.3Mismatch by eld o study

2 3

7 6 610 8

10 11

15139 9

9 10 8

10 11 9 9

710

25

45

5

5 5 79

8

12

3

2

33 6

3 5

74

4

7

0

5

10

15

20

25

30

35

40

45

50

55

H e a l t h&W e

l f .

C o m p

u t i n g

E n g i n e e

r i n g

E d u c a

t i o n L a w

B u s i n e s

s

A g r i c u l t u r

e&V e t .

S c i e n c e ,r

e s t

S o c i a ls c

i . ,r e s t

S e r v i c e s

H u m a

n i t i e s

P e r c e n t

Unemployed

Hor. + vert. m.m.

Vertic. m.m.

Horis. m.m.

. Te point o time the (originally) rst level degrees graduated with a second level degree might alsobe o interest; i.e. it is possible that those who graduated with a second level degree late, or instancein or , are most vulnerable or being mismatched, and also that this might in uence thecountry di erences. We have checked this, and the results were that year o graduation has no e ect onthe probability o being vertically mismatched, and it also shows that the country di erences are nota ected by the control or year o graduation.

. We use the ISCED broad elds o study; however, two o the elds are rather broad and comprise ahigh share o the graduates. Tis is Social science, business and law, and Science. We have extractedsome o the more vocational groups rom this broad categorization. Tis is Business and management( rom Social science) as well as Law. From Science we have taken out the group who has graduated inComputing so that they constitute an own category.



CHAPTER

. . Which actors increase the probability o a good match? Above we have seen the results o bivariate relations between education level and eldo study and the mismatch variable, based on weighted averages. Tere are a lot o

individual variables that are important or the chance o experiencing mismatch suchas unemployment and over-education. In this section we will explore the e ects o such variables, controlling or the country di erences. Tis will be done by presentingabridged results o multinomial logistic regression models Te results will be illus-trated in graphs based on estimations o the regression results. Te dependent vari-able is the mismatch-variable described above, which have ve di erent outcomes.Te “re erence category” in these regressions is “holding relevant work”, and we haveinvestigated the probability o being in one o the other our categories relative toholding relevant work. As independent variables we have included demographic vari-ables, educational background variables ( eld o study, level, grades, vocational study,prestigious study programme, urther education) and variables related the graduates’

working career both during education and a ter graduation, as well as parents’ educa-tion and indicators o social network. Te ull results are posted on the project website(www.re exproject.org, see ables W . and W . ).

Te multivariate analyses con rm that the general pattern o (uncontrolled) di er-ences by eld o study shown in Figure . remains a ter controlling or other vari-ables, and that most o the di erences are signi cant; those educated in Computing,Engineering, Health and wel are and Education have the best match, and the situ-

ation seems to be least avourable or those educating in Humanities, Social science(except Law and Business and management) and Services. Tis is in line with thendings o Green and McIntosh’s ( ) study o British graduates when it comesto Social science but not with regard to Business and management. Te e ects o selected other variables are shown in Figures . – . below.

. Te estimations are made according to the ormulaez j

P= ---------------- + ∑ ez j

where Z = the intercept plus the e ects o the control variables (Z = B + B X +B X ....), and j is an expression o the di erent outcomes on the dependent vari-able (the logit has j – di erent sets o parameters).



CHAPTER

Figure 7.5Mismatch by gender, social network and grades. Estimated probabilities

0,12 0,120,13 0,12 0,12 0,12

0,09 0,10 0,08 0,090,07

0,11

0,120,13

0,090,12

0,10

0,14

0,05

0,06

0,05

0,05

0,04

0,05

0,00

0,05

0,10

0,15

0,20

0,25

0,30

0,35

0,40

0,45

Males Females Yes No Yes No

Gender Social network Gades better than average

P r o

b a

b i l i t y

Unemployed

Hor. + vert.m.m.

Vertical m.m.

Horiz. m.m

Re erence category: Dutch males (except the second column), second level degree in Social science, average age andaverage amount o work experience a ter graduation. In the frst our columns: average grades. All columns: value 0 onall other variables in the equations.

Figure . shows that the di erence between male and emale graduates is very small,

though the e ect o gender is signi cant. Females have somewhat higher risk o beingunemployed and over-educated than males (all other things kept constant). Alsohaving a use ul social network has only a small impact, although it does reduce therisk o being both horizontally and vertically mismatched. Whathas a clear e ect isgrades. Tose who report having better grades than their ellow students clearly haveless risk o being vertically mismatched or both horizontally and vertically mismatchedthan those who do not report this.

Our next issue is the e ects o unemployment experiences and work experience. We have controlled or both the number o months that the graduates have reported

having had paid employment a ter graduation, the number o unemployment spellsand the duration o unemployment. Di erences in the amount o work experienceare partly due to urther education, but this is also controlled or. Tus, Figure .shows the net e ect o work experience and unemployment experience. One thirdo the graduates reported some unemployment experience, with the lowest shares inNorway, Estonia and the Netherlands and the highest share in Spain ( per cent!).

. Te estimates in Figure . and . above is based on a model where all independent variables exceptunemployment experience is included, the estimates in Figure . is based on an extended modelincluding control or number o months with unemployment experience and the number o timesunemployed.



Winners and losers

Te net e ect o employment experience may be seen as an e ect o acquired humanand social capital. We consider the possible e ect o unemployment experience as anindicator o state dependency (see the introductory part).

Figure 7.6Mismatch at the time o the survey by work experience and unemployment experience a ter gradu-ation. Estimated probabilities

0,13 0,11 0,12 0,12 0,120,09

0,090,09 0,09 0,10 0,10

0,11

0,09 0,11 0,100,13 0,14

0,240,02

0,060,03

0,050,07

0,14

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,70

65 months workexper.

35 months workexper.

No unemploym.exper.

1 time unempl, 2months

1 time unempl, 8months

4 times unempl,total 8 months

Varying work exp. No unemploym. exp. Average work experience (52 months)

P r o b a b i l i t y

Unemployed

Hor. + vert. m.m.

Vertical m.m.

Horiz. m.m.

Re erence group or the estimates: Dutch male with second level degree in Social science, average age and grades.Else; value 0 on all other variables in the equation, except the variables mentioned in the graph.

Tere is an e ect o both the amount o work experience (irrespective o unemploy-ment spells), and o the duration and number o unemployment spells, and the latterseems to have the greatest e ect. Te two rst columns in Figure . show that the risk o being unemployed is only per cent among those with months o work experi-ence versus per cent among those with months o work experience (estimatedpercentages based on no unemployment spells, with all other variables held constant).Te last our columns show the estimated percentages or di erent numbers and totaldurations o unemployment spells (based on average work experience). Tose with

only one unemployment spell with the duration o two months have per cent pointshigher risk o being unemployed at the time o the survey ( against ) than those withnone unemployment spells, and the risk o being both vertically and horizontally mismatched is versus per cent respectively. I the unemployment spell lastedlonger, or instance 8 months as exempli ed in the graph, the risk o being unem-ployed at the time o the survey increases and this also applies, moderately, to therisk o being both vertically and horizontally mismatched. However, the number o times one has been unemployed has an independent e ect. Tis is displayed in thelast column, which shows very high risk o being unemployed or both horizontally and vertically mismatched.8 It is important to note that only the most severe orms

8. Te example in this column is moderate; the mean duration o total unemployment o those who havebeen times unemployed is . months, but they represent only a very small group. Among those



Winners and losers

. Wages

Wages are the pecuniary reward o being employed. When comparing wages across

countries, it is important to take into account that it is not only the wage level thatdi ers, but also the cost o living. It might be meaningless to compare wages acrosscountries without taking these di erences into account. In order to do cross country comparisons, we have converted the wages to purchasing power parity (PPP) tocorrect or the di erences in costs o living. One has to keep in mind that this type o adjustment is ar rom per ect because o the difculty in nding “baskets” o goodsand services that are strictly comparable across countries. Nonetheless, using even animper ect PPP corrections provides a much better basis or comparing wages acrosscountries than no correction at all.

Figure . shows the hourly wages converted to PPP and Euro or each country.Te average hourly wage or all countries is . Euro, but varies still a lot betweencountries.

Figure 7.7Hourly wage wages converted to PPP and Euro

9.710.110.2

13.914.0

14.714.915.115.1

16.8

18.2

18.9

14.0

7.8

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

20.0

C z e

c hR e

p u b l i c

E s t o n

i a S p

a i n I t a l

y

F r a n c

e

F i n l a n

d

A u s t r

i a

N e t h e

r l a n d s

U n i t e d

K i n g d o

m

B e l g i u m

N o r w

a y

G e r m

a n y

S w i t z e

r l a n d

A l lc o u

n t r i e s

Broadly speaking, the countries can be divided into three groups, those with thehighest wages, the middle group and those with the lowest wages. Te gure showsthat graduates rom Switzerland, Germany and Norway have the highest wages. Itis not surprising that graduates rom Switzerland and Norway is on the top, but itis surprising that Germany is that high. As expected graduates rom Italy, Spain,Estonia and the Czech Republic have the lowest wages. Graduates rom the CzechRepublic earn less than hal o what graduates rom Switzerland do and might beconsidered as a loser on this dimension. Te middle group consists o graduates romBelgium, UK, the Netherlands, Austria, Finland and France.



Winners and losers

(a ter control or human capital actors) and whether or not there are gender di er-ences in work orientations and the realization o these orientations.

Figure 7.8Wage di erences between countries, estimated and observed

0

-38,6 -38,1

-9,1 -8,7 -8,9

0

4,3

8,3

14,4

-34,7

26,7

22,2

13,1

1,71,1

-1,1

-6,3-6,8

-31,3-32,4

-47,7

21,9

2,7

-30,8

-48,7-55

-45

-35

-25

-15

-5

5

15

25

35

C z e c

hR e p u

b l i c

E s t o n

i a S p

a i n I t a l

y

F r a n c

e

F i n l a n

d

A u s t r

i a

N e t h e

r l a n d s

U n i t e d

K i n g d o m

B e l g i u m

N o r w a

y

G e r m

a n y

S w i t z e

r l a n d

Observed Estimated

Te nding that women have lower wages than men is also the case or our gradu-ates (see gure .9). Te column “observed” shows that emales on average earn

per cent less than males. Tat is quite a big di erence. It is interesting to notice thatemales receive lower wages than males in all countries but the gender gap varies acrosscountries. Te di erences are smallest in Switzerland and Belgium ( and per centrespectively) and greatest in Estonia and France ( 8 and per cent respectively). Itis important to keep in mind that the gures mentioned re er to the actual observeddi erence and do not take into account that males and emales might have di erentlevel o education or di erent eld o study. Generally emales have shorter educa-tion than males, are concentrated in elds o study that pay less and might have less

work experience. Tis might explain some o the wage di erences between males andemales. Te column “estimated” in gure .9 re ers to the gender gap a ter control-

ling or these and other actors that might in uence wages.

We see that even a ter controlling or actors that might in uence the wages, emalesstill receive signi cant lower wages than males, about per cent, when we look atall countries together. Tis means that a ter controlling or other variables that mightin uence wages, the initial gender di erences in wages has been reduced by one thirdbut must still be characterized as high. Signi cant lower wages among emales apply to all countries. Te most striking result o controlling or relevant other variablesis that the country di erences in gender gap are reduced dramatically. In countries

with a small observed gender gap, controlling or other variables makes little di er-ences. In some o these countries (Belgium, UK, Czech Republic and Italy), the esti-mated gender gap is even somewhat higher than the observed gender gap. In contrast,



CHAPTER

controlling or other variables makes a big di erence in most o the countries wherethe observed gender gap was large. Te Nordic countries, which o ten are consideredas leading countries when it comes to equal opportunity policy have an estimated

gender wage gap around the average or all countries. Te highest estimated gendergap is ound in Estonia. Although the gender gap in earning varies between countries,

we can conclude that women in general might be considered as wage losers and menas wage winners.

Figure 7.9Wage di erences between males and emales by country

12,4

15

10,39,9

10,510,7

11,9

10,4

7,3

10,4

4,2

7,5

4,8

9,6

20,1

18,2

16,416,3

13,413,0

9,49,38,7

8,0

6,65,9

4,8

14,9

0

5

10

15

20

25

F r a n c

e

E s t o n

i a

F i n l a n d

N o r w a

y

A u s t r

i a

G e r m

a n y

I t a l y

C z e c h

R e p u

b l i c S p

a i n

U n i t e d

K i n g d o m

N e t h e

r l a n d s

B e l g i u m

S w i t z e

r l a n d

A l lc o u

n t r i e s

Estimated Observed

. . Education and wages Wages di er both between di erent levels o education and between di erent elds o study. Previous research has ound that there is a tendency that pro essionally orientedelds o study such as Business and Engineering have the highest wages whereas thosein “so ter” elds o study as Humanities have lower earnings (Finnie and Frenette

). Tis will also be examined below.

Another assumption in accordance with the human capital theory is that thelonger education the graduates have, the higher is the wages supposed to be. We will now examine to what extent this applies to our graduates as well. Figure .shows the average di erence in wages between rst and second level degree graduatesas a percentage o the wages o second level degree graduates.Again the column“observed” is based on the uncontrolled average wages while the column “estimated”is based on the regression analyses.

. Te educational level re ers to the level in 999/ . We have not taken into account whether thegraduate hade nished a second level or PhD/specialist degree during the years rom to the timeo the survey in the column “observed”. However, in the column “estimated” in ormation on urthereducation is used as explanatory variables.



Winners and losers

I we look at the column “observed” we see, as expected, that the wage level is highestamong second level graduates when we look at all countries together as well as in mosto the countries. On average, graduates with a rst level degree have per cent lower

wages than those with a second level degree but the di erences vary a lot across coun-tries. Te largest di erences between rst and second level degree graduates are oundin France ( per cent) and Norway and Finland ( per cent). However, in Germany,

Austria and Estonia graduates with a rst level degree have surprisingly the highest wages. In Italy, Switzerland and Belgium the wages among second level degree gradu-ates are only slightly higher than among rst level degree graduates. A ter controllingor other variables – including whether or not one has gone on to complete a higherlevel degree a ter – the country di erences become somewhat smaller, but theoverall di erential remains about the same. Te initially large di erential in Franceand Norway is considerably reduced, while the apparent anomaly o higher wagesor rst level graduates in Estonia, Germany and Austria disappears (in the case o Germany is even reversed) a ter controlling or other variables.

Figure 7.10Wage di erentials between rst and second level degree graduates

-5,3-6,1

-8,1

-14,1

-16,8

9,2

4,74,5

-1,7

-5,9

-9,7

-19,9

-24,9

-8,1

-19,8

-16,4-15,5

-10,3

-1,3-2,8

-8,5

-15,2

-19,8

-9,9

-30

-25

-20

-15

-10

-5

0

5

10

15

A u s t r

i a

G e r m

a n y

E s t o n

i a

S w i t z e

r l a n d

I t a l y

B e l g i u m

U n i t e d

K i n g d o m

C z e c

hR e p u

b l i c S p

a i n

N e t h e

r l a n d s

F i n l a n d

N o r w a

y

F r a n c

e

A l lc o u

n t r i e s

Estimated Observed

Having undertaken urther education increases a person’s human capital and we areinterested in to what degree it is re ected in the wages. Te regression analyses indi-cate that having undertaken urther education is re ected in the wages. Graduates

with rst level degree who have completed a second level degree have in general a wage gain o per cent a ter controlling or among other things whether they areexperiencing labour market mismatch. Tis means that she/he has a wage almost inline with those with a second level degree rom I we do not control or labourmarket mismatch, the wage gain is smaller (8 per cent) and the originally rst levelgraduates who had achieved a second level degree earn on average per cent less thanthose with a second level degree rom . Te reason or the lower wage gain is thatthose who have completed a second level degree as urther education to greater extent



CHAPTER

are newcomers in the labour market and exposed to mismatch. Graduates in mostcountries who have completed a second level degree as urther education experiencea wage gain when controlled or mismatch. Te exceptions are Italy, Austria, United

Kingdom and Estonia. Second level degree graduates who have completed a PhDdegree have also, when looking at all countries together, a wage gain o 9 per centa ter controlling or among other things whether they are experiencing labour marketmismatch. Tis is also the case in Italy, Austria, Finland the Czech Republic andEstonia. Te wage gain in these countries varies between 9 and per cent. I we donot control or labour market mismatch, there is in general no wage gain. Te reasonis that those having completed a PhD even are more newcomers in the labour marketthan graduates with rst level degree who have completed a second level degree.

As mentioned earlier, wages di er between di erent elds o study. Previous researchhas shown that there is a tendency that pro essionally oriented elds o study such asBusiness and Engineering have the highest wages whereas those in “so ter” elds o study as Humanities have lower earnings (Finnie and Frenette ). We have examinedto what extent this is the case or our graduates as well. Again the column “observed” isbased on the uncontrolled average wages while the column “estimated” is based on theregression analyses. Only signi cant results are re erred in the “estimated” column.

I we look at the column “observed” in gure . we see that in general, graduatesin Computing, Engineering and Science have the highest average wages compared to

graduates in Social science, while graduates in Agriculture, Education and Humanitieshave the lowest. Tese results are mainly in line with results rom previous researchand indicate that graduates in Computing, and Engineering are wage winners andgraduates in Agriculture and Education wage losers. However, the results are notuniversal but vary across countries (see results on the project website, able W. . ).

Te column “estimated” shows that what eld o study a person graduates rom, hasimpact on his/her wage. Tere are, when looking at all countries together, big wagedi erences between di erent elds o study even a ter controlling or other actors.Te wage winners when looking at all countries together are graduates in Business

and Computing (they have and per cent higher wages than graduates in Socialscience) and the losers seem to be graduates in Agriculture and Humanities who haverespectively and per cent lower wages. However, the result is not universal as itvaries somewhat between di erent countries which elds o study are wage winnersand wage losers.

Te regression analyses also indicate that those graduating rom a prestigious study programme in general have higher wages than those who are not graduating romsuch programmes ( per cent). However this is not the case in Austria, Germany and the Netherlands where there are no signi cant e ect. Te wage gain or the resto the countries varies between and per cent (the Czech Republic and Estoniarespectively).



Winners and losers

Figure 7.11Wage di erentials between elds o study

-10,6

-6,2

11,8

7,8

-14,4

-10,2

-6,6

-1,8

0,01,21,8

5,4

-4,3

13,7

-9,0

3,6

9,0

-20

-15

-10

-5

0

5

10

15

20

AgricultureEducationServiceHumanitiesHealthSocialscience, rest

LawBusinessScience, restEngineeringComputing

Estimated Observed

. . Mismatch and wages One important objective o this chapter is to study whether those experiencing labourmarket mismatch also are losers on other outcome indicators as or instance wages.Figure . shows the wage di erentials between graduates experiencing labour marketmismatch and graduates in relevant work.

Figure 7.12Wage di erentials between mismatched graduates and graduates in relevant work

-23,5

-10,0

-2,9

-11,4 -11,6

1,6

-25,0

-20,0

-15,0

-10,0

-5,0

0,0

5,0

Horizontal and vertical mismatch Vertical mismatch Horizontal mismatch

Observed Estimated

Tose graduates who are both horizontally and vertically mismatched seem really tobe losers when we look at the uncontrolled average or all countries. Tey have onaverage wages that are per cent lower than those in relevant work. A ter control-ling or the other actors that have impact o wages, the wage gap is reduced to per cent, which is still considerable, and about the same as or those who are only vertically mismatched. Both groups can there ore be regarded as losers in this respect.



CHAPTER

Interestingly, controlling or other variables makes little di erence or those who areonly vertically or only horizontally mismatched. Te di erential is increased slightly in the case o vertical mismatches, and even switches rom a small negative to a small

positive di erential in the case o horizontal mismatches. Consequently, the lattergroup cannot be regarded as losers in this respect.

Te general pattern o e ects is replicated in most countries, with some small di er-ences. Te lower wages or those being both horizontally and vertically mismatchedcompared to those in relevant work apply to most countries except or Italy, UK, theCzech Republic, Belgium and Estonia where there is no signi cant e ect. Te wageloss or the rest o the countries varies between 8 (Finland) and 9 per cent (Norway).Te negative e ect on wages o being vertically mismatched applies to all countriesexcept or Italy and Estonia and indicates that vertically mismatched persons in mostcountries are losers. Te wage loss varies between (Switzerland) and (Finland). Insome countries like Finland, Norway, the Czech Republic and Switzerland the positiveestimated e ect o being only horizontally mismatched was substantial, these gradu-ates earning between and per cent more than those not experiencing any kindo mismatch (see Re ex website, able W. . ). In contrast, horizontally mismatchedgraduates earned 8 per cent less than relevantly employed graduates in Spain.

Another simple way to study the relation between being mismatched and wages isto see to what extent the wage distribution depends on the status on the mismatch

variable. We have distributed the graduates in each country on three di erent wagegroups, depending on whether they were among the per cent with the lowest wages,among the middle per cent or among the top . able . shows the simple rela-tion between mismatch and wage group.

Table 7.3Mismatch and wages (bottom 25 percent, middle 50 per cent and top 25)

Total Bottom 25 per cent Middle 50 per cent Top 25 per centHorizontal mismatch 100.0 21.3 53.9 25.0Vertical mismatch 100.0 35.2 47.2 17.6

Horizontal and vertical mismatch 100.0 49.8 41.2 9.0No mismatch 100.0 22.7 50.8 26.5Total 100.0 25.4 50.1 24.4

Te table shows that hal o those being both horizontally and vertically mismatchedhave wages among the bottom per cent whereas a little above one third o those verti-cally mismatched all in the same group. Tese groups are also less requent observedamong the top per cent. Tis indicates that the groups are wage losers. Tosehorizontally mismatched do not deviate much rom those who are not mismatchedand the results support the conclusion that the horizontally mismatched might not beconsidered as wage losers.



Winners and losers

. .4 emporary jobs and wages emporary jobs are o ten considered to be bad jobs because they tend to pay less andbecause workers in temporary jobs are less satis ed with their job than workers in

permanent jobs (OECD ). However, among persons with higher education pres-tigious jobs as research ellow and researchers are in most countries based on tempo-rary contracts with a moderate wage level, indicating that temporary jobs might bequite heterogeneous and not necessarily bad. Figure . shows to what extent thereare wage di erentials between those in temporary and permanent job in our sample.Te graph shows that both the uncontrolled (“observed”) and controlled (“estimated”)

wage is higher among those in permanent jobs compared to those in temporary jobs.Tis is the case in all countries, but the size o the di erence di ers between coun-tries. Te uncontrolled average or all countries shows that those in permanent jobsearn per cent more than those in temporary jobs. Te average wage gap is almostunaltered a ter controlling or other actors that might a ect wages, indicating thatthose in temporary jobs might be considered as wage losers. However, again the di er-ences between countries are somewhat reduced a ter controlling or other variables.

As gure . shows, the size o the wage di erential between those in permanent andtemporary jobs varies a lot between countries with no signi cant wage di erences inEstonia to per cent in France.

Figure 7.13Wage di erentials between persons in temporary and permanent jobs. Per cent

14,5

16,1

10,510,7

13,413

25,2

22,6

17,9

12,811,711,3

7,9

5,35,24,5

22,9

20,218,7

15,5

11,7

10,310,1

19,0

18,6

10,3

14,1

0

5

10

15

20

25

30

F r a n c

e

S w i t z e

r l a n d

C z e c

hR e p u

b l i c

G e r m

a n y

F i n l a n d

N o r w a

y S p

a i n

A u s t r

i a

B e l g i u m

N e t h e

r l a n d s

U n i t e d

K i n g d o

m I t a l y

E s t o n

i a

A l lc o u

n t r i e s

Estimated Observed

. . Other actors that have impact on wages Grades above the average increase the wage by per cent, and having graduated roman academically prestigious study programme increase the wage with somewhat above per cent. Having parents with higher education increases the wage with about percent point, whereas social network seems to have a small negative impact, and voca-tional oriented study has a rather small positive impact. Tese are net e ects or thetotal sample a ter control or all other variables, and the e ects vary across countries.



Winners and losers

Later in this chapter we will see whether such a dichotomization gives meaning when we examine the response to questions o work values in the REFLEX survey, and we will see to what extent these types o values di er between our country samples, and

between emales and males. We will also investigate the outcome o the work orienta-tion (job characteristics), and nally we will see whether this in uences the graduates

job satis action.

.4. Factor analyses o work values Work orientations (values) are examined in the REFLEX questionnaire by the useo ten questions, all with scale – re erring to the extent to which the respondentattached importance to the ten items o work orientations (the items are shown inable . ). In additional questions the respondents were asked to what extent theseorientations (then understood as job characteristics), apply to current work. Tismakes questions. Given the act that we will look at both country and genderdi erences, we will see how the values may be clustered into a smaller set o items. Wehave done this by the use o actor analyses, c . able . .

Table 7.4Work values, results o actor analysis

Values Career (Factor 1) Pro essional (Factor 2) Social values (Fac. 3)1 Work autonomy -0,018 0,565 0,0872 Job security 0,384 -0,123 0,5323 Learn new things 0,177 0,754 0,0884 High earnings 0,820 0,025 0,0675 New challenges 0,310 0,735 -0,1326 Good career prospects 0,743 0,319 -0,0617 Leisure activities 0,130 0,010 0,6918 Social status 0,609 0,089 0,2389 Use ul or society -0,121 0,441 0,51510 Combine work with amily 0,018 0,092 0,758

Note: The Czech sample is not included in the analyses, because o lack o in ormation on some o the items. Onlyobservations with valid response on all the ten items are included.

Te actor analysis o work values clearly distinguishes three types o work orienta-tions, that is: Factor : Career and status orientation ( 9 per cent o variance ), Factor: Pro essional/innovative ( exible) orientation ( . per cent) and Factor : Socialoriented values ( amily, security and altruistic values) per cent, which sums up to. per cent total variance explained.

Te way the values are clustering ts quite well to the characterization o values basedon Inglehart et al. ( ) and Wang ( 99 ) mentioned above. Factor “Innovative/pro essional” contains the values “work autonomy”, “new challenges” and “learnnew things” and thus it covers the “sel -expression-values”/”intrinsic” values. Factor (earnings, career) is comparable to “survival values”/”extrinsic” values. We see Factor

. Per cent o variance based on rotation sums o squared loadings.



Winners and losers

When it comes to Factor (pro essional/innovative values) the country di erencesare much smaller; such orientations are common values that seemed to be shared by the vast majority o the respondents. Only the Austrian and Swiss sample score clearly

above the average on Factor . Te results with regard to Factor do not con rm thatrespondents rom (all) the most sel -expression countries according to Inglehart et al.’scultural map score above the other countries, neither do they con rm that Italy andSpain score below. However, Inglehart’s cultural map was con rmed by the resultsconcerning career orientations.

What about Social values? Te clearest di erence in Figure . is the di erencebetween the Spanish and British samples, the Spanish sample scoring especially high,and the British are scoring rather low. Tus, the Spanish sample distinguishes onesel by high scores on both actor and actor . Tis might be interpreted in the way thatthe put more weight on the “extrinsic” (survival) values (we asses the social valuesorientation as a combination o extrinsic and intrinsic values) than the other country samples and thus con rming the cultural map o Inglehart et al. (ibid) mentionedabove.

Tere are gender di erences in addition to country di erences in work orientations.Tis will be illustrated below.

Figure7.15Career orientations. Mean actor scores, males and emales by country

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Spa in It al y F ran ce Swi tz er lan d Au st ri a Germa ny Ne th er la nd s B el gi um Un it edKing.

No rwa y F in la nd E st on ia

M e a n

f a c

t o r s c o r e s

FemalesMales

Te overall tendency is that that the country di erences apply both to males andemales. Both males and emales in the Estonian, Spanish (and next) Italian samples(low-income countries) score above average on the actor scores or career orienta-tions. In many, but not all countries, the males score above the emales. Tis is clearest



Winners and losers

Figure 7.17Social/ amily oriented values. Mean actor scores, males and emales by country

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Spa in It aly Fr anc e Sw it ze rla nd A us tr ia Ge rma ny Ne the rla nds Belgium Un it ed Kin g. Nor w ay Fin la nd Es to nia

M e a n

f a c t o r s c o r e s

FemalesMales

.4. Tree types o “winners” and “losers” Te results o actor analyses concerning work values bring about the questions

whether there are three types o winners/losers concerning whether the work valuesare realized in current job; that is being a winner on the career dimension, the pro es-sional/innovative dimension and the social-values dimension. Tis is investigated by analysing the graduates’ response to the corresponding ten items on job character-

istics; or in other words, the realization o the work values. o be a winner/loser onone dimension, one must have expressed that the cluster o items is important or very important, and that the items in question have either been realized to a airly highextent (winners) or hardly or not at all (losers).

Table 7.5Types and numbers o winners and losers. Per cent o total response

Total, winner Career etc 21.5Total, winner Social values 29.5Total, winner Pro essional/innovative 61.9Total winner on all three dimensions 7.2

Total winner on two o the dimensions 34.6Total winner on at least one dimension 73.7Total loser on allthree dimensions 0.4Total loser ontwo dimensions 3.3Total loser on at leastone dimension 15.6Number o observations, weighted* 18885

* Only observations with valid response to all the ten items on both the set o work values questions are included.The results are based on a weight that gives all the country samples the same size (that is 20000 graduates) andN is based on this weight.

. Most o the respondents ound at least one o the items connected to one o these three dimensionsimportant or very important. O those who had answered all the questions concerning work values,8 per cent ound the career values important (at least one o the career items), 9 per cent ound thesocial values dimension (at least one o the items) important, and 98 per cent ound the pro essional/innovative dimension important. Only . per cent did not nd any o the dimensions important, and9 per cent ound all the three dimensions important.



CHAPTER

Te precise de nition o winners and losers is urther explained in Appendix . able. shows the total distribution o the three types o winners and losers.

Almost three quarters o the sample are winners on at least one o the three dimen-sions according to our de nition. Only per cent are winners on all three dimensions.Most o the “winners” are winners on the dimension “pro essional/innovative” (new challenges, learn new things, work autonomy), while the career dimension (high earn-ings, good career prospects and social status) has the lowest share ( . per cent).

Tere are ew losers; 8 per cent is not in any o the loser groups, and only approxi-mately a hal per cent is categorized as belonging to a loser group on all three dimen-sions. Tis means that more or less the total sample are either winners or “neutrals” onat least one dimension. Tose who are neither losers nor winners, are categorized as“neutral”, and large proportions o the graduates are in this group; this is depicted inthe charts below. Especially on the career-dimension and the social values-dimension,there are large proportions o “neutrals”.

.4. Realization o work orientations by gender Te gender distribution o the three groups di ers somewhat. Tere is a clear over-

weight o emales among the winners in the social-values group, and a slight over- weight o males among the winners on the career-dimension, se Figure . 8.

Figure 7.18Realization o work values. Winners/losers on three dimensions, by gender

2824

63 63

25

34

6163

32 31

71

62

11 136 6 4 4

0 %

10 %

20 %

30 %

40 %

50 %

60 %

70 %

80 %

90 %

100 %

Males Females Males Females Males Females

Car eer orientations Pr ofessional/innovative Social/family values

P e r c e n

t

Loser

Neutral

Winner

Note: The percentages in each dimension are based on the number o respondents who expressed that one o theitems in the dimension was important or very important. This applies to all corresponding graphs below.

. Tese gender di erences are statistically signi cant a ter control or relevant background variables (seeRe ex webstie ables W. . -W. .9, and based on such regressions the probability o being a career-

winner is estimated to be per cent point less among emales than males (the same as in bivariaterelationship in Figure . 8).



Winners and losers

Te gender di erence realization o career orientation is not very large, but noteworthy since it is measured among respondents who nd this dimension impor-tant or very important (8 per cent o the respondents). Tis implies that the large

wage di erences depicted in Figure .9 is probably not due to gender di erences incareer orientations, although this might be o some importance, but is most probable(among other things) due to discriminatory actors. Another nding that underpinsthat gender di erences in wages is not caused (or at least not only) by gender di er-ences in career orientations, is that such orientations seem to have much larger impacton male graduates’ wages than on the wages o the emale graduates. Tis is revealedin additional wage analyses (re . web-site…) where we have controlled orthe e ect onwages o the subjective measure “ nding high earnings important or very important” . Teresult is illustrated in the graph below.

Figure 7.19The e ect on wages o nding high earnings important, by gender.

0,0 %

1,0 %

11,4 %

-3,4 %

-6,0 %

-4,0 %

-2,0 %

0,0 %

2,0 %

4,0 %

6,0 %

8,0 %

10,0 %

12,0 %

14,0 %

High earnings not important High earnings are important

P e r c e n

t

Males Females

Males not interested in highearnings are the reference

group (0,0).

Te graph indicates that those who are striving or high wages doobtain higher wages;

among men the wage increase is per cent compared to those who do not answer thathigh earnings are important. Tis might o course also be a result o the respondentstrying to avoid cognitive dissonance; those who have relatively low wage report thatearnings are not important because their wages are not high. Anyway, the interestingresult is the gender di erence. Females seem to gain less than males rom striving orhigh wages. And, we see that the wage di erence between males and emales who arenot particularly interested in high earnings is only about per cent, while the wagedi erence between males and emales whodo nd this important is about per cent.Te result gives no support to the assumption we suggested earlier in the chapter; thatemale graduates’ work orientations may explain gender di erences in wages.



CHAPTER

.4.4 Realization o work orientations by country and eld o study Our next issue is country di erences, and then we will have to look at one dimensionat the time. We start with country di erences when it comes to being a winner/loser

on the career-dimension.

Table 7.6Winners/losers by country. Realization o career orientations, pro essional orientations and socialvalues orientations.

ES IT FR CH AT DE NL BE UK NO FI EECareer orientations (N=15680)Winner 34 19 22 22 27 20 24 24 31 21 18 43Neutral 50 60 64 68 61 67 68 69 59 69 68 49Loser 16 21 15 10 12 13 8 7 10 10 14 8

Pro essional/innovative orientations (N=18602)Winner 56 51 55 67 73 65 60 63 61 68 67 64Neutral 33 41 38 28 23 30 32 32 33 28 28 32Loser 11 9 7 5 5 6 7 5 6 4 5 4Social values orientations (N=18344)Winner 36 23 34 24 27 21 33 30 24 40 34 35Neutral 56 69 62 72 70 76 66 66 73 58 63 62Loser 8 8 4 4 3 3 1 4 3 2 3 2

Note: Those who did not nd the orientation in question important, are excluded rom the calculation.

Tree o the most career-oriented samples, the Estonian, Spanish and the Britishsamples, have higher shares o winners on the career-dimension than the other country samples although the spanish graduates are more o ten losers on this dimension thangraduates in most other countries (see able . ). As we have seen rom Figure .the Spanish and the Estonians are not among the wage winners compared to theother country samples. Tey might anyhow experience being winners because they compare their situation with the situation o persons in their home country withouthigher education. A similar e ect might apply in the three high income countriesSwitzerland, Germany and Norway, which do not stand out with high percentages

that report being career winners (in act, the percentages are rather low). Graduates inthese countries might compare themselves with other persons in their home country who are making a career in non-academic occupations. Further, the country sample with ewest winners and most losers is the Italian sample.

Te country di erences concerning the realization o pro essional/innovative orien-tations are less striking. We see that the Italian and Spanish samples more o ten areamong the losers also on this dimension, and that the Italian, Spanish and Frenchsamples more seldom than the others report being winners on this dimension, while the

Austrians are clear winners, ollowed by the Norwegians, Swiss and Finnish graduates.



Winners and losers

Te extent to which social/ amily orientated values are realized di ers a lot betweenthe countries. Tere are really ew losers in all the countries on this dimension;however, neither are the winners in a majority (see the strict de nition in Appendix

). Germany has ewest winners, tightly ollowed by Italy. Tis implies that the Italiangraduates rather seldom are among the winners on all the three dimensions. TeNorwegians have the highest share o winners, ollowed by Spain. Tis means thatthe Spanish sample has high shares o winners on the two dimensions where they also scored very high concerning the graduates’ values; the career-dimension andthe social-values dimension. Te Norwegians are also among winners on two o thedimensions where they scored above average when it comes to the worth o the values,(namely) the pro essional/innovative orientations and the social/ amily values.

Te realization o the work orientations might also di er according to eld o study,see able . .

Table 7.7Winners/losers by eld o study. Realization o career orientations, pro essional orientations andsocial values orientations.

EDU HUM SOCRest

BUS LAW SCIRest

COMPUT ENG AGR+VE HE+WEL SER TOT

Career orientations (N=15680)Winner 21 21 26 34 31 23 25 26 21 23 26 26Neutral 63 67 62 57 58 64 65 63 65 65 61 62

Loser 16 13 12 9 11 12 10 11 14 12 13 12Pro essional/innovative orientations (N=18602)Winner 64 61 64 60 64 67 68 63 61 63 57 63Neutral 30 32 30 33 31 28 27 32 37 31 36 31Loser 6 7 6 7 5 5 6 5 3 6 7 6Social values orientations (N=18344)Winner 48 31 31 25 28 27 23 21 26 37 38 30Neutral 51 66 65 71 67 68 74 74 71 60 59 66Loser 2 4 4 4 5 5 3 5 3 3 3 4

Note: Those who did not nd the orientation in question important, are excluded rom the calculation.

Te two winner-groups when it comes to career orientations are those educated inBusiness and in Law. Te di erences by eld o study apply more to the winner- thanthe loser aspect.

Tose educated in Computing are not among the winner groups on the career-dimen-sion. Tis might have something to do with expectations. According to the wageanalyses in the previous section this group is among the wage-winners in almost allthe countries.



CHAPTER

When it comes to the pro essional/innovative dimension there are small di erencesby eld o study. Te di erences by eld o study are larger concerning the chanceo being a winner or loser on the social/ amily-values oriented dimension but again,

the di erences concern the chance o being a winner more than being a loser. Tosegraduating in Education are more o ten winners on the “social” dimension than theother groups , and those educated in Computing, Engineering and Business are moreseldom winners. Graduates in Education (the teachers) obviously have had goodpossibilities to combine work and amily obligations.

Who are most o ten/most seldom winners concerning eld o study when the threedimensions are taken together? Graduates inHumanities and Agriculture and veteri-nary are among those who most seldom are winners on the career dimension, and they are average on the pro essional/innovative and average or below average on the social-values dimension. In most countries the wages o these two groups are relatively low,and we have also seen that graduates in Humanities have a large risk o experiencingall kinds o mismatch (and those educated in Agriculture/Veterinary are average whenit comes to the chance o good match). Tese two groups are groups that most clearly tend to lose or be in the “neutral” groups, and most seldom among the winners.

.4. Realization o work orientations: which actors increase the probability o being awinner? Te actors predicting success on the career-variable are not always the same as the

actors predicting success on the pro essional/innovative variable or social-values vari-able, and vice versa, neither are they always the same as those variables that havecontributed in predicting education-work match or high wages in the previous anal-yses. Tis was shown above when it comes to eld o study, and below we will look at other predictors. Overall, this implies that being a winner on the labour market isnot one-dimensional, but contains many dimensions. Although; the e ects point inthe same direction or some o the variables. We will start with commenting on theresults o variables measuring characteristics o study programme, grades and whetheror not the graduate has a use ul social network, c . Figure . . In this graph and inthe two graphs to ollow we have omitted the category “neutrals”, which there ore

constitutes the rest.

. Additional analysis also shows that the social values dimension is particularly important or thisgroup.



Winners and losers

Figure 7.20The e ect o study programme characteristics, grades and social network on the probability o being a winner/loser. Three dimensions o work orientations

0,28

0,27

0,30

0,26

0,65

0,69

0,63

0,61

0,31

0,29

0,26

0,24

0,03

0,02

0,02

0,03

0,06

0,07

0,07

0,07

0,07

0,08

0,08

0,08

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8

Prestigous edu.

S o c

i a l / f a m

i l y v a

l u e s Vocational study

Prestigous edu. P r o

f e s s

i o n a

l / i n n o v a

t i v e

o r i e n

t a t i o n s

Vocational study

Prestigous edu. C a r e e r o r i e n

t a t i o n s

Vocational study

Probability

Winner Loser

Good grades

Social network

Good grades

Good grades

Social network

Social network

The re erence group or the estimations in Figure 7.26 is Dutch males with average age, educated in social science,with average grades ( rst three rows within in each o the three dimensions, in the orth row the graduate hasgrades above the average). The estimates or the re erence groups are not shown in the graph due to lack o space, but is emphasised in the text.

Concerning the probability o being a career winner, prestigious education has thelargest positive e ect, increasing the per cent rom 23 per cent (the re erence group), to per cent. Next comes social network, and then vocational study, whereas “grades

above the average”, which also is signi cant, has the weakest e ect. Having gradesabove the average increases the probability o being a career winner with only some-

what more than one per cent point. All these variables mainly a ect the probability o being a winner, not being a “loser” (relative to being in the “neutral” group). Tepercentage being a loser in the re erence group concerning the career-dimension is 9

per cent , and having graduated rom at prestigious study programme thus decreasesthe share being a loser by per cent points.

When it comes to the probability o being a winner on the pro essional/innovative dimension, social network has the largest e ect, increasing the probability o beinga winner rom 59 per cent (the re erence group)to 9 per cent. Next comes graduatingrom a prestigious study programme ( per cent). Also both good grades and beingeducated in a vocationally oriented study programmes has positive e ects, howevernot large. And, also or the pro essional/innovative dimension, these variables mainly a ect the probability o being a winner, not the risk o being a loser. Te re erencepercentage or being a loser on the pro essional/innovative dimension is approxi-



CHAPTER

mately 8 per cent,thus prestigious study programmes decrease the probability o beinga loser on this dimension by approximately per cent points.

Te results regarding the third dimension, social/ amily oriented values, are very di erent rom the rst two mentioned. Being graduated rom a vocational orientedstudy is the only study programme characteristic variable that has a signi cant positivee ect on being a winner on this dimension, and the e ect is quite small. It increasesthe probability o being a winner rom 27 per cent (the re erence group)to per cent.

And none o the variables a ect the risk o being a loser on this dimension. Te re er-ence percentage or being a loser on the social/ amily oriented dimension is2.6 per cent.

We have also investigated whethereducational level has an e ect. Compared to secondlevel graduates being a rst level graduate increases the risk o being a loser on thecareer dimension but it has no signi cant e ect on the probability o being a winner(relative to belonging to the “neutral” group) on this dimension. Te same goes orthose who have become PhDs/specialists; they have an increased risk o being loser onthe career variable, but there is no signi cant e ect on the chance o being a career-

winner (relative to being in the “neutral” group). Te results are di erent or thepro essional/innovative dimension. Tose who have obtained a PhD/specialist degreehave, not surprisingly, an increased chance o being a winner on the pro essionaldimension. Correspondingly, those who had obtained a PhD/specialist degree have

reduced chance to be a winner on the social/ amily-values dimension. Tese graduates will have interesting, demanding work, and, urther education to become PhD or aspecialist is not easy to combine with amily tasks and leisure activities.

.4. Winners and losers by job characteristics and labour market situation What also might be o importance and interest is whether and how the chances o being a winner/loser on these three dimensions is related to characteristics o thegraduates’ labour market situation and job characteristics. Tis is taken into accountin additional analyses (see Re ex website, able W. . -W. . ) and results based onthese analyses are presented in Figures . and . .



Winners and losers

Figure 7.21The e ect o mismatch on the probability o being a winner/loser. Three dimensions o work orien-tations

0,18

0,21

0,15

0,63

0,39

0,30

0,20

0,14

0,13

0,03

0,03

0,04

0,06

0,15

0,27

0,08

0,13

0,18

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7

No mismatch S o c

i a l v a

l u e s

Vertical mismatch

Vert. + hor. mismatch

No mismatch P

r o f e s s

i o n a

l o r i e n

t .

Vertical mismatch


No mismatch

C a r e e r o r i e n

t a t i o n s

Vertical mismatch


Probability

Win Lose

The re erence group or the estimations in Figure 7.21 is Dutch males, educated in social science, with averageage and average grades, median wage (14.2 Euro, PPP, per hour), and work in private sector in a permanent job;else the re erence person has value O on all the other variables in the equation except the variables shown in thegure. What varies is match/mismatch.

Being both horizontally and vertically mismatched has a large negative e ect on theprobability o being a winner or all the three winner-categories; and a correspond-ingly positive e ect on the risk o being a loser on the career-dimension8 and thepro essional-dimension. Tis re ects that being a “loser” on the objective measureso labour market adaptation, also to a very large extent implies being a “loser” on the

subjective indicators.

Also being (only) vertically mismatched strongly reduces the chance o being a career- winner and a winner on the pro essional dimension. For the third dimension, thesocial-values dimension, the results di er when it comes to vertical mismatch (over-education). Being vertically mismatched has no negative e ect on the probability o being a winner on this dimension, rather a small positive e ect. Te cause might bethat some o these graduates have pre erred a less demanding work situation becausethey nd that this makes it easier to combine work with amily tasks. Te main picture

8. In the regressions on which the estimations in Figure . are based we have also controlled or wages.Tis reduces the e ect o being mismatched on the career-dimension, but has minor impact on theother two dimensions. Te e ect o wages is shown in the Figure . .



CHAPTER

is that not being mismatched according to the objective measures highly increases theprobability o success in realizing the graduates’ (subjective) work orientations. In thenext graph we will look at di erences in the probability o being a winner/loser on the

three dimensions concerning other aspects o work.

Figure 7.22The e ect o job characteristics on the probability o being a winner/loser. Three dimensions o work orientations

0,19

0,36

0,36

0,37

0,65

0,62

0,60

0,64

0,16

0,18

0,14

0,23

0,03

0,01

0,01

0,01

0,05

0,05

0,06

0,04

0,09

0,08

0,12

0,06

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7

Temporary job, median wage S o c

i a l v a

l u e s Low wage, permanent j.

Temporary job, median wage P r o

f e s s

i o n a

l o r i e n

t .

Low wage, permanent j.

Temporary job, median wage

C a r e e r o r i e n

t a t i o n s

Low w age, permanent j.

Probability

Win Lose

High wage, permanent j.



Medium wage, permanent j.



The re erence group or the estimations in Figure 7.22 is Dutch males, educated in social science, with average ageand average grades, work in public sector and has relevant employment; else the re erence person has value 0 onall the other variables in the equation except the variables shown in the gure. What varies is permanent/tempo-rary job, and wages.

Te main purpose o Figure . is to illustrate the e ect o wages as well as o havingpermanent job versus having a temporary job. Te graph re ers to graduates who all work in public sector , and since the preceding graph re erred to graduates working inprivate sector, we also have the possibility to look at the di erence between publicand private sector, and we will start with this. In the graph above we see that theprobability o being a career winner i you have a permanent job with median wagein public sector is 8 per cent. Corresponding estimate or a person working in private sector is per cent (Figure . ), thus working in private sector increases the prob-ability o being a career winner (controlled or wages) with only per cent points(and the risk o being a loser is 8 per cent both in public and private sector). Workingin private sector also has a signi cant e ect o being a pro essional-winner, but thee ect is very small ( per cent versus per cent). When it comes to being a winner



Winners and losers

on the social-values dimension, the e ect is much greater. Only 8 per cent o thosein private sector (median wage, permanent job) are winners on the social-dimension(Figure . ), but the share is twice as high ( per cent) i one works in public sector

(Figure . ). Tis re ects, most probably, that public sector o ers better possibili-ties or combining work and amily tasks through better arrangements or maternity leaves, exible/reduced work hours etc.

Wages has a great e ect on the probability o belonging to the group that we havede ned as being a career-winner based on their response to the questions on realiza-tion o career orientations. Tis is not surprising, what might be surprising is that thee ects are not larger. I you belong to the high income group (PPP, that is controlledor country di erences in costs o living), the probability o being a career winner isonly per cent, and i you belong to the low-income group the probability is percent. Te “low wage-group” in the estimates in Figure . is assigned the limit orthe value or the bottom quartile on the wage distribution (that is . Euro per hour,ppp) and the “high wage-group” the corresponding value or the top quartile (thatis . Euro). High wage also has some impact on the probability o being a winner/loser on the pro essional dimension, but not a very large impact, and the impact onthe social-values dimension is negligible.

Also having permanent versus temporary job has signi cant e ects (positive or beinga career-winner, negative or being a winner on the pro essional dimension!), butthe e ects on these dimensions are small compared to the e ect on the social-values

dimension (the latter: per cent versus 9 per cent).Summing up; the most important thing or being a winner on the pro essionaldimension is not being vertically mismatched or both horizontally and vertically mismatched, and having use ul social network. Te most important thing or beinga winner on the career dimension is wages and prestigious study programme, and themost important or being a winner on the social-values dimension is to work in publicsector and hold permanent work. In the next section we will see to what extent suchactors have impact on di erences regarding job satis action.

. Job satis action

Farag and Allen ( ) point out that “there are a number o actors or dimensionso work orientations and their realisation which may need to be taken into account

when looking at the determinants o an individual’s job satis action”. In the sectionbelow we will take quite a lot actors into account when we are analysing job satis ac-tion, with special attention to issues discussed and analysed in previous sections, suchas mismatch and work orientations. Farag and Allen compared Japanese and Dutchstudents. Below graduates rom countries are studied, and detailed discussion o

which actors are determinant or eacho the country samples cannot be analysed.But rst we will show the bivariate relationship between country and job satis action.



CHAPTER

Tose who are satis ed with their job according to our de nition are those have tickedo value or on a ve point scale o work satis action.

Figure 7.23Per cent satis ed with their job, by country

63

58

6872

74

6968

73

65

74

65

72 71

0

10

20

30

40

50

60

70

80

90

100

S p a i n I t a l

y

F r a n c

e

S w i t z e

r l .

A u s t r

i a

G e r m

a n y

N e t h e

r l a n d s

B e l g i u m

U n i t e d

K i n g d .

N o r w a

y

F i n l a n d

E s t o n

i a

C z e c h

R e p u

b l i c

P e r c e n

t

In Figure . Czech Republic is included, but will not be included in the results o analyses shown below, because we will take into account the response to work orien-tations, and there is a lack o in ormation on these questions in the Czech response.

Overall, the per cent reporting job satis action was 9, and thus the Czech sample wassomewhat above the average, together with Austria, Norway, Belgium, Estonia andSwitzerland. Italy and Spain have the lowest shares, the same two country samples

which are not among the winners neither in the labour market analyses, the wageanalyses and the realization o work orientations analyses, with the exception o theSpanish sample being a winner on the subjective social-values-dimension and thecareer-dimension.

One o Farag and Allen’s ( ) ndings was that “intrinsic and social orientations were more important as determinants o overall satis action than (other) extrinsic

orientations”. Tis was in accordance Maslow’s hierarchy o needs, which, as Faragand Allen ( ) put it, “leads to the expectation that intrinsically motivated indi-viduals will tend to be more satis ed with their work than extrinsically motivatedindividuals, since intrinsic motivations will only arise once the (lower order) extrinsicneeds have been sufciently satis ed.“ Below (Figure . ), we will illustrate the e ectso being winners/losers on the three dimensions described above on the probability o being satis ed with the job. Te graph is based on separate analyses or the three low-income country samples (Estonia, Italy and Spain) and or the nine high or mediumincome countries. Te reason why the countries are separated into two groups is tosee whether the intrinsic values are less important or job satis action in the low income countries than in the high/medium income countries.



Winners and losers

Figure 7.24Estimated probability o being satis ed with the job. E ects o being a winner/loser on the career-dimension, the pro essional dimension, and the social-values dimension

0,63

0,22

0,52

0,20

0,47

0,26

0,71

0,16

0,58

0,21

0,55

0,23

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

Win-prof./innov.

Lose-prof/innov.

Win-career Lose-career Win-socialvalues

Lose-socialvalues

P r o b a b i l t y

Estonia,Spain,Italy

The rest 9countries

The re erence category or the estimates in the graph has relevant work and median income (that is 9.5 and 15.Euro per hour (ppp converted) in the low and medium-high income countries respectively) , is a male, Dutch (in

high /medium income countries) or Italian (in low income countries) with average age, educated in Law, works inthe private sector in a permanent job.

In both type o countries, those who are winners on the pro essional/innovativedimension are most o ten satis ed with their job, ollowed by winners on the career-dimension, and close behind winners on the social-values dimension. However, boththe pro essional dimension and the social values-dimension are more important or

job satis action in the nine medium or high income countries than in the three low income countries (c . the di erence between the winner and loser categories is largestin the high or medium income countries), whereas winning or losing on the career-

dimension has more or less the same e ect in the two types o countries. Tis con rmsthat the intrinsic values are (somewhat) more important in the high or mediumincome countries than in the low income countries.

Do we nd di erences concerning job satis action between low and high incomecountries also when it comes to other aspects o work? Tis is shown in Figure . .



CHAPTER

Figure 7.25Estimated probability o being satis ed with the job. E ects o job characteristics¨

0,21

0,31

0,41 0,41

0,55

0,38

0,45

0,270,30

0,35

0,43

0,48

0,41

0,45

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

Both horizontal+ vertical m.m.(private sector)

Verticalmismatch only

(private s .)

Horizontalmismatch only

(private s .)

Private sector,relevant empl.

Public sector,relevant empl.

PPPwageLOWPPPwageHigh

P r o

b a

b i l i t y

Estonia,Spain,Italy

The restcountries

The re erence group are respondents who are not in any winner nor loser category (“neutrals”), and except or thelast two categories have median wage. The re erence wage in the category “PPPwageLow” is 7.0 Euro or the lowincome countries and 12.2 Euro in the other nine countries (which represents the limits or the 25 per cent withlowest income in the two types o countries respectively). The re erence wage in the last category (high wage)is 12.4 Euro and 18.7 Euro (which represents the limits or the 25 per cent with highest income respectively). There erence group works in private sector (all columns except the column or “public sector, relevant employment”),and is otherwise the same as in Figure 7.25.

We see that wage has larger impact on job satis action in the low income countriesthan in the other nine countries.

Being mismatched has the largest (net) impact, especially being both horizontally and

vertically mismatched, and this re ers to both types o countries ( versus per centin the low income countries, and versus per cent in the high or medium incomecountries). Also being (only) vertically mismatched has a large negative e ect on jobsatis action. Being only horizontally mismatched has a negative e ect in the nine highor medium income countries (taken together, versus per cent), but it has noe ect in the three low income countries.

Above we have seen that those who work in private sector somewhat more o ten thanthose who work in public sector are winners on the career-dimension and the pro es-sional-dimension, but as regards the social-values dimension the opposite was theact. From Figure . we see that those who work in public sector somewhat moreo ten are satis ed with their work irrespective o being a career-winner, pro essional- or



Winners and losers

social values-winner. Tis re ers to both types o countries, but we seethat working in public sector has the largest positive e ect on the probability o being satis ed with work in the low income countries . Te cause might be that secure work in public sector is

more important in the low income countries than in the high income countries, andanother explanation might be that working in private sector results in a wage gain inthe high income countries but the opposite seems to be the case in the low incomecountries (additional analyses indicate this, see Re ex website, able W. . ).

Finally, we will return to the country di erences (c . Figure . ) with regard to jobsatis action. Te country di erences depend to a large extent on country di erencesin labour market situation. Te regression analyses (see Re ex website) show thata ter control or such variables, the country di erences are reduced or changed, orinstance are the Spanish sample very o ten satis ed with the job when all other actorsare kept constant. However, still the Austrian sample has the largest share being satis-ed with their job. Other noteworthy results are that the e ect o gender is minor,and that there are no di erences between the levels o education, with one exception:PhDs/specialists are more o ten satis ed with their jobs than rst and second levelgraduates. Also the di erences between elds o study are small, with one exception; graduates in Education are more o ten satis ed with their work than the other groups .Tis applies to both types o countries, but especially in the high or medium incomecountries graduates in Education are more o ten than other groups satis ed withtheir job. Also this can be interpreted as an e ect o the importance o the intrinsic

values in the high or medium income countries. Graduates in Education are wage-losers compared to graduates in other elds o study. However, since most graduatesin the high or medium income countries have relatively high wages (also those inEducation), the intrinsic orientations are rather important or job satis action in thesecountries, and social values (like amily values, doing something use ul or society) iso special importance or graduates in Education.

Overall, both the subjective measures o being a winner/loser on the three dimen-sions o work orientations and the objective measures o labour market situation andreturns to education and (over-education/mismatch; wages) are highly relevant or

job satis action in both types o countries. However, wages are less important andintrinsic values more important in the high or medium income countries than in thelow income countries.

. Summary and conclusions

In the table below we have summarized the result o the di erent regression analyses.

Only when it comes to wages there is a clear and quite large gender di erence, inavour o males. Else we see that no elds o study are clear winners or losers on alldimensions, but graduates in Education are winners when it comes to job satis ac-



CHAPTER

tion and are also winners on other parameters except wages and career orientations.Tere is also a tendency that winners on career orientations and/or wages (Business,Computing and Engineering) are not winners on other, more subjective dimensions.

Tose elds o study that most seldom win and more o ten than other elds tend tolose on several dimensions are Humanities and Agriculture and veterinary.

We have also ound that the human capital related actors are signi cant both or theobjective measures (mismatch/match and wages) and or the subjective measures o realization o work orientations, except or being a winner on the social-values dimen-sion.

“State dependence” (previous unemployment experiences) has a clear e ect on the risk o being mismatched ve–six years a ter graduation, and given the act that mismatchhas a clear negative e ect on the job satis action and the chance o being a winneron two o the dimensions or work orientation, the transition problems seem to havelasting negative e ects or a considerable group o graduates. Te table above does notshow the size o the e ects, and it should be mentioned that the di erent predictorshas di erent impact on the various dimensions:• Te most important thing or being a winner on the pro essional dimension is not

being vertically mismatched, and having use ul social network.• Te most important thing or being a winner on the career dimension is wages and

prestigious study programme •

Te most important or being a winner on the social-values dimension is to work in public sector and hold permanent work

Another general nding is that although we see groups that are success ul on someindicators and less success ul on others there is a clear relationship between the objec-tive and the subjective indicators. Tose who experience ailures on the objectiveindicators are quite o ten not success ul on the subjective indicators.

Te e ects o belonging to the di erent country samples are not included in the tableabove, and the reason is that the country di erences to a large extent re er to macro

economic conditions and to di erences in the transition phase; and the country di erences is highly in uenced by the type and number o control variables includedin the regressions. However, the country di erences may be summarized like this:Te Italian sample is most seldom among the winners on all the parameters; both theobjective and the subjective indicators. Te same is the case or the Spanish sampleconcerning the objective measures, but they are among the winners on the some o the subjective indicators. Te Norwegian sample is the one being success ul on mosto the parameters, speci cally the objective measures, but overall also on the subjec-tive measures, and also the Swiss and Austrian samples scores high on several indica-tors, the Swiss especially on wages and labour market match, the Austrian on jobsatis action and the realization o pro essional/innovative work values.



Winners and losers

T a

b l e 7 . 8

E e c t s o v a r i o u s p r e d i c t o r s o n t h e s i x d e p e n d e n t v a r i a b l e s R e l e v a n t

w o r k

W a g e s

W i n -

c a r e e r

W i n - p r o e s s i o n a l

W i n - s o c i a l - v a l u e s

J o b

s a t i s a c t i o n

L e v e l a n d u r t h e r e d u c a t i o n

F i r s t l e v e l v s . ( o r i g i n a l l y ) s e c o n d l e v e l

–

+

S e c o n d l e v e l a s u r t h e r e d u c a t i o n , ( v e r s u s o r i g i n a l l y s e c o n d

l e v e l )

–

( - )

+

( – )

P h D a s u r t h e r e d u c a t i o n ( v s . s e c o n d l e v e l )

–

+

–

+

O t h e r u r t h

e r e d u c a t i o n

( + )

( + )

F i e l d o s t u d y ( v s . s o c i a l s c i e n c e , r e s t )

E d u c a t i o n

+

–

–

+

+

H u m a n i t i e s

–

–

–

C o m p u t i n g

+

+

–

–

S c i e n c e , r e s t

+

–

( – )

E n g i n e e r i n g

+

+

( – )

–

A g r i c u l t u r e a n d v e t e r i n a r y

–

–

–

–

H e a l t h a n d

w e l a r e

+

–

S e r v i c e

( - )

–

L a w

+

–

B u s i n e s s a n d m a n a g e m e n t

+

+

–

–

F e m a l e s ( v s . m a l e s )

( – )

–

–

( + )

+

( + )

G r a d e s a b o v e a v e r a g e

+

+

+

+

( – )

.

V o c a t i o n a l o r i e n t e d s t u d y

+

( + )

( + )

+

( + )

.

G r a d u a t e d r o m a p r e s t i g i o u s e d u c a t i o n

+

+

+

+

.

P a r e n t s H E

+

( – )

.

S o c i a l n e t w

o r k

( + )

( - )

+

+

.

R e l e v a n t w o r k d u r i n g s t u d y

+

+

.

W o r k e x p e r i e n c e a t e r g r a d u a t i o n

+

+

.

.

.

.

U n e m p l o y m e n t s p e l l s

–

.

.

.

.

.

D u r a t i o n o u n e m p l o y m e n t s p e l l s

–

.

.

.

.

.

M i s m a t c h ( v e r s u s r e l e v a n t w o r k )

H o r i z o n t a l m i s m a t c h

.

( + ) * *

–

V e r t i c a l m i s m a t c h

.

– * *

–

–

+

–

H o r i z o n t a l a n d v e r t i c a l m i s m a t c h

.

– * *

–

–

–

– –

P u b l i c s e c t o r ( v s . p r i v a t e )

.

.

–

–

+

+

P e r m a n e n t j o b ( v s . t e m p o r a r y )

.

+ * *

+

–

+

W a g e p e r h o u r s ( P P P )

.

.

+

+

+

R e a l i z a t i o n

o w o r k o r i e n t a t i o n s

W i n - c a r e e r

.

.

.

.

.

+

W i n - p r o e s s i o n a l

.

.

.

.

.

+ +

W i n - s o c i a l v a l u e s

.

.

.

.

.

+

L o s e - c a r e e r

.

.

.

.

.

–

L o s e - p r o e s s i o n a l

.

.

.

.

.

–

L o s e - s o c i a l - v a l u e s

.

.

.

.

.

–

E m p t y c e l l m e a n s n o t s i g n i c a n t ; + m e a n s i n c r e a s e d p r o b a b i l i t y ; – m e a n s d e c r e a s e d ( i n b r a c k e t s , o n l y s m a l l e e c t s ) ; . ( d o t ) m e a n s n o t i n c l u d e d i n t h e m o d e l . R e s u l t s m a r k e d * * b a s e d o n

M o d e l 2 , a l l o t h e r r e s u l t s o n M o d e l 1 o T a b l e W

. 7 . 5

o n t h e R e f e x w e b s i t e .



CHAPTER

Wages are less important and “intrinsic”/”sel -expression” values more important or job satis action in the high or medium income countries than in the low incomecountries. However, overall, both the subjective measures o being a winner/loser

on the three dimensions o work orientations and the objective measures o labourmarket situation and returns to education and (over-education/mismatch; wages) arehighly relevant or job satis action in both low income countries and in high/mediumincome countries.

Re erences

Andress. H.-J. ( ) Recurrent unemployment – Te West German experience: An explora-tory analysis using count data models with panel data.European Sociological Review ( ).

– . Allen, J., Meng, C. & R. van der Velden ( ). REFLEX Country study. ROA, MaastrichtREFLEX Country study. ROA, MaastrichtROA, Maastricht

University (dra t - - ). Allen, J. & R. van der Velden ( ). Educational mismatches versus skills mismatches: e -Educational mismatches versus skills mismatches: e -

ect on wages, job satis action, and on-the-job search.Ox ord Economic Papers.( ): –.

Allen, J. & R. de Vries ( ). Determinants o skill mismatches: the role o learning environment.the match between education and job and working experience.Paper or the LM.NE Con-erence. Amsterdam. – . November .

Allen, J. & R. van der Velden ( ). Te Role o Sel -Assessment in Measuring Skills.Paper orthe ransition in Youth Workshop. Valencia. – . September . Arthur, L., Brennan, J. & E. De Weert ( ). REFLEX – Te Qualitative study (dra tREFLEX – Te Qualitative study (dra t

. . ). Atkinson, J. ( ). Manpower strategies or exible organisations, Personell Management , Au-

gust, pp. – .Becker, G. ( ). Human Capital , New York: National Bureau o Economic Research.Bourdieu, P. ( ), Te orms o capital. In: J. G. Richardson (Ed.), Handbook o Teory and

Research or the Sociology o Education.New York: Greenwood Press.Coleman, JS ( ). Social Capital in the Creation o Human Capital, American Journal o

Sociology , Supplement, – .Finnie, R. & M. Frenette ( ). Earning di erences by major eld o study: evidence rom

three cohorts o recent Canadian graduates, Economics o Education Review , ( ): –.

Farag, S. & J. Allen ( ). Japanese and Dutch Graduates’ Work Oirenetations and Job Satis-action. In: Higher Education and Work: Comparison Between Te Japan and the Netherlands. Research Report No. . Te Japan Institute o Labour.

Finnie, R. & M. Frenette ( ). Earning di erences by major eld o study: evidence romthree cohorts o recent Canadian graduates, Economics o Education Review , ( ): –

.Gervasi, L.O. ( ). Finding a suitable job: Te e ect o the institutional context on sel -perceived

over-education.DemoSoc Working Paper : . Universitat Pompeu Fabra. Barcelona.



CHAPTER

Wang, G. . ( ). A comparative study o extrinsic and intrinsic work values o employees in the United States and Japan.Lewiston: Edwin Mellen Press Ltd.

Wood, R.G., Corcoran, M. E. & P. N. Courant ( ). Pay Di erences among the Highly

Paid: Te Male-Female Earnings Gap in Lawyers Salaries, Journal o Labor Economics ,( ): – .



Chapter Conclusions and policy implications

Rol van der VeldenJim Allen

. General conclusions

In general one may say that higher education graduates in most o the European coun-tries are well in the labour market (see Chapters and ). Despite the deep-rooteddi erences between the national higher education systems, similarities in outcomesare more striking than di erences. Although only a small proportion o graduatesend up in an elite position, the majority ul l a role in jobs that require generalistor specialist tertiary-level training. A high proportion o the human capital that isproduced in higher education appears to be put to good use in the world o work. Teunemployment rate is generally low, and almost three quarters o all graduates indicatethat their knowledge and skills are sufciently used. Tat said, there is still some roomor improvement, particularly or the more than one out o our working graduates

who indicate that their competences are insufciently used. Apparently, employers donot make ully use o the human capital that is at their disposal. Moreover, there arehere arecountries and elds o study where graduates nd it especially difcult to nd a goodposition. O course part o the variation across countries and elds o study is due todi erent national economic conditions or di erences in the demand or graduates inspeci c elds o study, but this is not always the case. Apart rom the ‘usual suspects’(humanities, Southern-European countries), the UK stands out as a country wheregraduates - even ve years a ter graduation - nd it difcult to get a job in which theirskills are ully utilised. Tis might be related to the act that the UK higher educationsystem is much less linked to the world o work than many continental higher educa-

tion systems.

Most studies o the transition rom higher education to work look at objectiveoutcomes, such as employment chances, wages, type o work contract etc. However,graduates may have very di erent goals to strive or: high earnings may be importantor some, but others may place more value on jobs that are challenging and give theopportunity to develop one’s skills. In this report we looked at subjective indicators –the extent to which graduates realized the things that they themselves nd importantin work – in addition to objective indicators o success. It is interesting to note thatthe same actors that determine objective success o ten also a ect subjective success(see Chapter ). Winners on di erent dimensions o ten have above average grades,have access to a good social network, have acquired relevant experience during higher



Conclusions and policy implications

Although most graduate jobs require that one has been educated in a certain area o study, strict exclusivity in this respect is only the norm in the case o the ‘old pro es-sions’. Moreover, most pro essionals nowadays are no longer ully autonomous in

their work, but typically work in organisations in which they are mutually dependenton others. Even the ‘old pro essionals’ rarely have complete control over their ownactivities in the sense o being sel -employed and/or ree o supervision. Tis high-lights the importance o the second most important core competence that graduatesneed to possess: the ability to mobilise one’s own and other’s human resources.

Te mobilization o human resources is important in all kinds o pro essions (seeChapter ) and is a ter pro essional expertise the second most important core compe-tence in terms o predicting success in the labour market. It increases the chance o nding employment – essential i one is to mobilise any human resources at all -, andhas a positive impact on the wages (see Chapter ). When discussing the mobilizationo human resources, it is important to make a distinction between mobilising one’sown resources and mobilising the resources o others (see Chapter ). Most highereducation graduates have been rather success ul in mobilising their own humanresources: most are employed in a more or less ulltime capacity and are able to utilisemost o their skills, even when working in a job that does not require a higher educa-tion degree. Not surprisingly, less graduates play a direct role in mobilising the humanresources o others. Nonetheless, a considerable proportion o graduates do play sucha role, or example by supervising and or assessing others or bearing strategic and/or

decision-making authority or their organisation.Te role o exibility as a core competence seems to be less clear. Although highereducation graduates are exposed to some orms o external exibility like job mobility,unemployment spells, temporary employment contracts (see Chapter ), this seems tobe mainly a temporary phenomenon in the initial transition rom higher educationto work. Although unemployment spells clearly have a negative e ect on the latercareer, other orms o external exibility may o ten be more an opportunity than athreat. Most graduates have reached a stable and satis actory position within - yearsa ter leaving education (see Chapter ), and the small group that is still in tempo-

rary employment consists not only o ‘losers’ in the labour market, but also containsthe privileged and well-trained group o graduates working in Ph.D. or equivalentspecialist trainee programs. In this sense, we nd little evidence that the labour marketor higher education graduates is very insecure or precarious.

However, this only re ers to external exibility. Te opposite is true or internal orunctional exibility. Functional exibility - the ability to cope with changes in the

work environment - does play an important role in the pro essional li e o graduates.Many graduates have already been aced with important changes in their work tasksor with changes in their work environment (see Chapter ). Competences related tounctional exibility do not appear to be directly rewarded in the labour market, butthey do play a role in protecting graduates when coping with changes at work. Being



CHAPTER

very exible in the sense o being prepared to take on work outside one’s own speci carea o training can in act hamper the possibility to ully utilise all o one’s skills as– by de nition – only a part o these skills will be put to use in any job.

Te ourth core competence that we distinguished – that o innovation and knowledgemanagement - also plays a somewhat uzzy role. Although innovation and knowledgemanagement are generally considered key actors driving economic growth, possessinginnovative skills does not always lead to labour market success. In act, such skills arenegatively related to employment chances and have no e ect on earnings (see Chapter). In contrast, being engaged in innovative activities (as opposed to possessing a highlevel o innovative skills) is clearly rewarded (see Chapter ). Apparently, innovativecompetences are only rewarded when they are directly linked with actual innovativeactivities. Although innovation as such takes place more o ten in large organisations,higher education graduates working in small organisations are more likely to actually play a role in introducing such innovations. A certain amount o autonomy is neededto create an environment in which innovation can take place. Being engaged in inno-vative activities is not only related to the typical innovative competences, but also withother competences like communication skills.

Te analysis in chapter makes clear that innovation is not solely related to the typicalhard-core R&D jobs in the private sector, but is important in other jobs and othersectors as well. For example, teachers play an important role in the innovation o

knowledge and methods, even though most would probably not think o them as coreinnovators. Innovation can thus be seen as important and widespread.

International experiences are widespread (Chapter ). More than a quarter o thegraduates reported that they spent some time abroad or study or work, and evenmore indicate that they work in an organization with an international scope o opera-tions and/or require a high level o oreign language pro ciency. Given this highexposure to international in uences, it is worrying that oreign language pro ciency is most o ten cited as one o the weak points o the study program. Spending timeabroad or study or work during or a ter higher education has a positive e ect on the

transition to the labour market. It not only has a positive impact on the chance o being internationally mobile a ter graduation and the chance o obtaining work thatrequires international competences (see Chapter ), but is related to higher wages ingeneral (Chapter ).

In our survey, we ound evidence that the demands in the areas o pro essional exper-tise, unctional exibility, innovation and knowledge management, and mobilisa-tion o human resources are more or less universal (Chapter ). Te required level isrelatively high, with little di erence in general between the di erent competences,although there are some di erences between the countries. Although the supply o competences in these areas is also rather high, at an individual level supply does notalways match demand. Some % o the graduates indicate that their own compe-




tence level is lower than what is required o them in the job and around % that theircompetence level exceeds the requirements. Although these percentages may seemlow, we should note that they may have serious consequences. Shortages can make it

more difcult or graduates to adequately per orm their job, while surpluses may beindicative o work situations that ail to get the most out o graduates.

When looking at the extent to which higher education prepares graduates well or the world o work, it is important to distinguish between the shorter and longer term.In the short term goal, we expect higher education to provide graduates with a goodbasis or starting work. In the longer term, higher education should provide a basisto acquire additional knowledge and skills on the job, and or career development ingeneral. With respect to both the short and long term goals, only to % o thegraduates indicated that their study program clearly succeeded in providing a goodbasis, while - % indicated that their study program clearly ailed to do so. Tis isparticularly worrying as providing a good basis to start working and to develop yourcareer may be considered as key goals o higher education.

Interestingly, graduates were most satis ed over higher education in terms providinga good basis or personal development ( % on average). In contrast, only % indi-cated that their higher education program provided a good basis or developing entre-preneurial skills. Clearly, developing entrepreneurial skills is one o the weak points o the higher education system all over Europe.

Given that many graduates are less than satis ed about the preparation they havereceived in higher education, the important question is what it might do to improvethis. Higher education in general is not considered to be very demanding. Only slightly more than hal o graduates indicate that their higher education program wasvery demanding. Tis ailure to provide students with a challenge is probably one o the reasons why so ew graduates ( %) indicated that they did more work than wasstrictly required to pass the exams. Especially the Netherlands stands out as a country

where students have an ‘easy li e’: less than a third o Dutch graduates indicated thattheir study program was demanding.

What can higher education do to give their students a better start in the world o work? What are the characteristics o the programs that are success ul in this respect?In analyzing this we can distinguish between two di erent unctions o education:the skills production unction (the role o education in providing their students withrelevant competences) and the allocation unction (the role o education in ensuringthat graduates nd appropriate work). Although both goals are clearly connected,they are by no means the same, nor are the characteristics that make programs ef-cient in achieving either one o these two goals. Graduates may have a high level o competence and still nd it difcult to nd a job in which they can ully utilize thesecompetences. Moreover, some higher education characteristics may help graduates to



CHAPTER

nd relevant work, although they do not in themselves have an e ect on the acquisi-tion o skills.

It is clear that ollowing a demanding program is good or developing competences,but it does not necessarily lead to a strong position on the labour market. Followinga program with which employers are amiliar mainly has a strong e ect on allocation,but only a weak e ect on the development o pro essional expertise and no e ect onthe development o competences in the other areas. Tis means that these programsdo not necessarily produce or select more competent graduates, but they are by ar thebest in ensuring that they nd a good job. Te e ect o ollowing academically pres-tigious programs is related to both unctions: they select or produce more competentgraduates, but they also serve as a signal to uture employers, thus helping to have asmooth transition and enter elite positions. Vocationally oriented programs are goodor developing pro essional expertise and are very strong in providing a good basis toenter the labour market and develop the career.

Modes o teaching and learning also play a role. Te level o competence in all coreareas is most strongly related with stressing theories and paradigms. Written assign-ments and oral presentations are also related to higher levels o competence in all ourareas. Group work and participation in research is related to somewhat higher levels o competence in all areas except pro essional expertise, while project and problem basedlearning is related to a higher level o innovation and knowledge management. Most

o these characteristics also a ect the evaluation o the program. Stressing acts andpractical knowledge, stressing theories, internships, giving lectures all help to preparestudents or the working li e, mainly because they provide an important means toacquire pro essional expertise.

Apart rom experiences in higher education, other learning experiences are just asrelevant. ime spent on relevant work experience has a positive e ect on competencedevelopment and all labour market outcomes. Te same holds or having a positionin a student or voluntary organisation and or experience abroad. However time spenton non-relevant work experience has no e ect at all, apart rom increasing the chance

to nd a job.

Having a high relative grade has a pronounced e ect on helping people to get intobetter jobs, and serves as a clear signal to uture employers. Surprisingly, indicators o study behaviour (like working hard and study hours) hardly a ect these outcomes orsometimes even have an adverse e ect. Tis is in line with the e ects we noted earlieron ollowing a demanding program. Although working hard is probably one o thebest ways to develop your competences, we see no direct reward in the labour market.

Working hard is not rewarded in itsel , but signalling this in the orm o higher gradesis.




. Policy implications

When it comes to policy implications we would like to distinguish the ollowing main

stakeholders: the European commission, national governments, employers, highereducation institutions and students.

European commission

International graduate surveys o er important insights into the changing Europeanhigher education systems: they should be repeated at -year intervals Te analyses in this report make clear that a wealth o in ormation can be extractedrom surveys like REFLEX and its predecessor CHEERS. In many respects country di erences are not always as large as is o ten assumed, especially given the act thatthe population comprises graduates o the pre-Bologna regime. Nonetheless, there aresome deep-rooted di erences, both between systems o higher education and betweentypes o study program within systems that are clearly related to the e ectiveness o higher education programs in preparing graduates or the labour market. Te resultsin this report highlight these di erences, and provide an indication o their relation

with quality. Building on insights obtained in the rst international survey CHEERScarried out in 999, the REFLEX project has developed the methodology and instru-ments needed or repeating these surveys on a more regular basis. Tis enables the

monitoring and evaluation o the outcomes o the Bologna process and other re orms with respect to the labour market. We recommend that the European Commissiontakes the lead in ostering such ollow-ups.

Although higher education is increasingly internationally oriented, this does not keep pace with the even more rapid trend toward globalisationMany graduates work in an environment that is strongly internationally oriented.Despite the act that many students have ollowed part o their study program abroad,higher education graduates all over Europe indicate that oreign language pro ciency is one o the weak points o their study program. Te European commission should

do more to oster international exchange in higher education, as well as activitiesdesigned to strengthen oreign language pro ciency, e.g. by co- nancing study programs o ered in a oreign language.

National governments

Strengthen the core orientations in higher educationTe results seem to suggest that both vocationally and academically oriented highereducation have their own distinct value in preparing graduates or the labour market.In act, the more higher education study programs emphasize the development o pro essional expertise in either o these orientations, the more success ul they are.



CHAPTER

National policies should aim to strengthen both academic and vocational highereducation.

Di erent national solutions are needed to solve problems It is interesting to see that despite the many di erences that existed in the Europeanhigher education systems (recall that the graduates in the REFLEX survey were nota ected by the Bologna re orms), the overall impression is that the country di er-ences in terms o outcomes are not overwhelmingly great. Tis suggests that di erentnational equilibriums may exist and that solutions that work in one country cannotsimply be exported to another. Tis does not mean that there are no countries in thedanger zone. Italy, France and Estonia represent countries where a relatively largeshare o the graduates experiences some serious shortages in their competences. Many French graduates also experience a surplus o competences, indicating that in thatcountry many graduates are either over- or under-quali ed.Noteworthy is that apart rom the Southern-European countries, the UK stands outthe Southern-European countries, the UK stands outas a country where graduates nd it difcult to nd a job that ully utilises their skills.

Although ve years a ter graduation, the unemployment rate o the UK graduates isaverage, their share o holding a lower level job and/or a job in which they cannotully utilize their knowledge and skills is much higher than in most o the other coun-tries. It is not clear whether this is caused by the weaker link between higher educationprograms and speci c areas o employment in the UK or with the act that most UK higher education graduates have ollowed programs that are much shorter in duration

than most programs in continental Europe. But the act that UK graduates have notbeen able to catch up in the rst years a ter graduation and more o ten indicate thattheir study program did not provide a good basis to start working, to learn on the jobor to per orm current work tasks deserves serious attention.

Encourage relevant rather than non-relevant work experience during higher educationIt is clear that rom a macro perspective spending time on non-relevant work duringhigher education should be discouraged. It distracts students rom their study andcon ers no bene ts at all in the long term. From an individual point o view, non-rele-From an individual point o view, non-rele-vant work may be needed to cover costs o living while studying. Te policy implica-

tion is that the national systems o student grants/loans should be such that studentscan pay enough time to their study.

External exibility is not always bad Being exposed to external exibility in the orm o multiple changes o employer iso ten regarded as undesirable. Te analyses in this report make clear that where thisis accompanied by spells o unemployment, this can have some damaging e ects onthe later career, but that external exibility per se is not harm ul, and can even bea source o urther skill development. Having a temporary contract in the rst jobis not harm ul, and mobility can o ten better be regarded as an opportunity than athreat. National policy should oremost be ocused on promoting a smooth transition



CHAPTER

Higher education institutes

Study programs should be more demanding One o the prime goals o higher education should be to optimally develop the talentso students. As ‘time on task’ is the best predictor o learning outcomes, this impliesincreasing the study load and creating a culture in which hard work and striving orexcellence is valued and rewarded. Te results show that only slightly more than hal o the graduates indicated that their program was (highly) demanding. Tis percentagedi ers strongly between the di erent countries. Especially the Netherlands stand outas a country where students o ten indicate that their study program was not very demanding.

Study programs should ocus on strengthening pro essional expertise In the past decade we have seen a shi t rom stressing speci c competences to ocusmore strongly on generic competences. However, developing pro essional expertiseprovides the main basis or entering the world o work and developing one’s career,even when working in jobs outside the domain o the eld o study. A basic rationale A basic rationaleor higher education is to impart pro essional expertise, and given the relatively low percentage o graduates indicating that this is a strong point o their study program,higher education institutes should strengthen this urther.

Student-centred methods may work, but don’t ignore the value o knowledge

We saw that student-centred methods like project and problem based learning havea positive e ect on providing graduates with a good basis to enter the labour market,their urther career and – interestingly – they seem to be the modes o teaching mostassociated with developing entrepreneurial skills. However, there is no clear relation

with developing high levels o competence in most areas (except the area o innova-tion and knowledge management). Our preliminary conclusion is that new methodsmay work, but old methods should not be orgotten. Tere is a tendency in educa-tion to think that knowledge in itsel is not important anymore, as technologicaldevelopments seem to render knowledge and skills obsolete soon a ter graduates havele t higher education. However, theories, acts and practical knowledge are essential

components to develop expertise in any area, and it is this pro essional expertise thatis most clearly associated with labour market success.

Assessment drives learning as well Te design o the curriculum and the modes o teaching are not the only ways toa ect learning. As educational research makes clear, assessment drives learning as

well. In this respect, using written assignments or oral presentations are a better way to develop competences and provide a good basis or entering the labour marketand developing a pro essional career, than using multiple choice exams which merely seems a good way to test the short-term memory capacities o students rather than a

way to develop deep-rooted insight.




Give credits or relevant work experience Work experience closely related to the eld o study or holding positions in student orother organisations clearly has a positive e ect on the development o relevant skills.

Higher education institutes could oster this by giving credit points to students whoper orm such relevant work. Tis would encourage students to engage in relevant

work instead o non-relevant work activities.

Don’t overestimate the positive e ect o internships and work placements Graduates who ollowed a program that stressed internship or work placement weremore positive in their evaluation o the program providing a good basis to start

working. However, we ound no e ect on the development o competences, nor did we nd any e ect on current employment chances or earnings. Tis seems to indicatethat its role is mainly in providing a smooth allocation to jobs, rather than in devel-oping pro essional expertise.

Students

Follow your interest and talent Although graduates rom some elds o study (such as Humanities and Agricultureand veterinary) nd it more difcult to enter the labour market and acquire a good

job, this by no means indicates that these elds o study should be avoided. For all

elds o study we nd that two thirds (or more) o the graduates are satis ed with their job, and this also applies to the two elds mentioned (Humanities and Agricultureand veterinary studies). Moreover, we nd only small di erences between elds o study in the percentages o graduates who regret the choice o their program. In ourview, students should primarily ollow their own interest and talent when choosing astudy program in higher education. In ormation about labour market prospects cano course play a secondary role in helping students choose between programmes they are equally interested in.

Acquire relevant experience outside higher education

Our ndings show that acquiring work experience that is related to the study programis bene cial or the later labour outcomes. Te same holds or holding a position instudent or other voluntary organisations (e.g. chair, committee member) or spendingtime abroad or study and/or work. Tese experiences have a positive e ect on thedevelopment o skills and serve as a signal to uture employers. Although many students are engaged in non-relevant work to cover the costs o living, it is ar betterto ocus on relevant work experience. Non-relevant work does not pay o in the longrun, and – i it leads to an extension o the study duration - it might be better to rely on study loans.



CHAPTER

Te relevance o a good network Having a good social network helps one nd a job that matches one’s education.Tis network does not only relate to amily, riends and teachers but extends to other

contacts as well or example contacts acquired through work experience. Especially these pro essional contacts may play a role in providing in ormation about job oppor-tunities and support in nding a job.



Appendix

First and second level programmes per country

First level Second levelItaly Diploma universitario LaureaSpain Diplomatura LicenciaturaFrance Licence DEA

Maîtrise DESSDiplôme d’école spécialisée (santé, art, architecture, journalisme, in rmier, etc.)

Diplôme d’école d’ingénieurs

Diplôme d’école supérieure de commerceCerti cat de la onction publique (ex: CAPES, etc.)Diplôme d’Etat de docteur en médecine, pharmacieou odontolo

Austria Mag./Mag.a oder Dipl.Ing./Dipl.Ing.in (FHS) Mag./Mag.a oder Dipl.Ing./Dipl.Ing.in (Univ.)Dr./Dr.in

Germany Diplom Fachhochschule, Diplom I anGesamthochschule

Diplom Universität, Diplom II an Gesamthochschule

Bachelor MagisterLA Grund- und HauptschulenLA RealschulenLA GymnasienLA SonderschulenLA Berufiche SchuleLA SontigesSonstiges StaatsexamenKirchlicher AbschlussKünstlerischer/musischer AbschlussMaster

The Netherlands HBO WO doctoraal (drs., mr. o ir.)WO opleiding tot basisarts, tandarts o apotheker

The United Kingdom Bachelor MasterFinland AMK-tutkinto Masterintutkinto tai vastaava

Norway 3-4 års hogskoleutdanning Hoved ag/høyere grads embetseksamen

The Czech Republic Bachelor MasterSwitzerland Masters Fachhochschule Masters UniversityBelgium-Flanders Hoger onderwijs van cycli (lange type) Universitair onderwijs - licentiaat o ingenieur

Universitair onderwijs - artsEstonia Bakalaureuseõpe Arsti-, hambaarstiõpe

Rakenduskõrgkooli ja ülikooli diplomiõpe Integreeritud õpe (proviisor, loomaarst jne.)Kõrghar. eeldav -a. õpetajakoolitus, interniõpeMagistriõpe, kutsemagisterMagistriõpe, teadusmagister



Appendix

Appendix To what extent innovative activities are related to a speci c working environment?

Role in innovationo product

Role in innovationo technology

Role in innovationo knowledge

Dependent variable: innovation index B S.E. Signi . B S.E. Signi . B S.E. Signi .

Are you responsible or:setting goalsor your own work

0,297 0,019 0,000 0,208 0,020 0,000 0,357 0,018 0,000

Are you responsible or:deciding howyou do your own job

0,191 0,021 0,000 0,164 0,022 0,000 0,219 0,021 0,170

How closely is your per ormancemonitored by your own supervisor

0,012 0,011 0,252 -0,018 0,011 0,107 0,015 0,011 0,014

To what extent does your current workdemand more knowledge and skillsthan you can actually o er

0,025 0,015 0,089 0,058 0,016 0,000 0,039 0,016 0,000

To what extent are knowledge andskills utilized in your current work?

0,153 0,017 0,000 0,197 0,018 0,000 0,276 0,017 0,316

Public sector -0,336 0,033 0,000 -0,196 0,034 0,000 0,034 0,034 0,000

Female -0,327 0,032 0,000 -0,756 0,033 0,000 -0,324 0,033 0,102

Italy 0,046 0,073 0,531 0,398 0,076 0,000 0,124 0,075 0,828

Spain -0,313 0,073 0,000 0,281 0,077 0,000 0,016 0,075 0,021

France -0,262 0,078 0,001 0,288 0,082 0,000 0,185 0,080 0,000

Austria -0,293 0,076 0,000 -0,182 0,082 0,026 -0,311 0,079 0,000

Germany -0,338 0,073 0,000 -0,412 0,080 0,000 -0,308 0,074 0,642

United Kingdom -0,084 0,076 0,270 0,053 0,081 0,513 0,037 0,079 0,019

Finland 0,205 0,071 0,004 0,468 0,074 0,000 0,176 0,075 0,017

Norway 0,195 0,070 0,005 0,104 0,074 0,161 0,175 0,073 0,000

Czech Republic 0,044 0,070 0,537 0,510 0,073 0,000 0,560 0,077 0,003

Switzerland -0,200 0,073 0,006 -0,077 0,078 0,323 -0,225 0,075 0,486

Belgium -0,303 0,078 0,000 0,075 0,082 0,363 -0,056 0,080 0,000

Estonia 0,214 0,088 0,015 0,492 0,091 0,000 0,399 0,096 0,000

Re : Netherlands

Intercept -2,46 0,119 0,000 -2,69 0,127 0,000 -2,99 0,121

Statistics Cox & SnellR2

NagelkerkeR2

Cox & SnellR2

NagelkerkeR2

Cox & SnellR2

NagelkerkeR2

1 0,071 0,095 22280 0,075 0,103 0,099 0,135



Appendix

Appendix

De nition o mismatch as used in Chapter

Employed with relevant work;that is person not belonging to one o the ourgroups below.Horizontally mismatched (and not vertically mismatched).Tis re ers to persons

who gave an answer to the question “What eld o study do you eel is most appropriate or this work ?” that indicated that their work did not correspond totheir own or a related eld.ertically mismatched (and not horizontally mismatched). Tis group is over-educated (over-quali ed) and the de nition re ers to those who gave an answerto the question “What type o education do you eel is most appropriate or this work? ” (“type” is re erring to “level ” according to the response options in thequestionnaire) that indicated a level below their educational level. We havetaken into account the act that some have acquired a higher educationallevel a ter their graduation in 999/ (as masters/second level graduatesor PhDs/specialists). First level graduate/bachelors who had taken urthereducation and have become masters/second degree graduates and hold a

job that corresponds to the rst level/bachelor level are regarded as vertically

mismatched, and vice versa or master /second degree graduates who haveobtained a PhD/specialist degree.Both vertically and horizontally mismatched. Unemployed . Tis re ers to respondents who answered that they were notcurrently employed and who reported that they actively had tried to obtainpaid work in the past weeks, or who reported that they were awaiting theresults o earlier job applications.

.

.

.

4..



Appendix

Appendix

De nition o winners and losers according to the graduates’ response to thequestions on work values and job characteristics (realization o work values), asused in Chapter

For all the ten work values items a variable was created according to whether ornot the item was important or the respondent. Value (important) + (very important) on a scale rom – were recoded as important (assigned value ,else ).Losers- and winners-variables were created or each o the ten items o jobcharacteristics (to what extent the work values apply to current work). I therespondent had value on the variable mentioned above, that is nds the itemimportant, and value or on corresponding item or job characteristics,he/she was coded as a loser on this variable. I the respondent nds the itemimportant and value or on corresponding item or job characteristics, he/she was coded as a winner on this variable.

From the results o the actor analyses o work values we knew that the work valuesclustered into three dimensions, allowing us to identi y three groups that are career-oriented, pro essional-oriented and “social values”-oriented. Te next step was then:

. Tree new variables were created “lose/win-career”, “lose/win-innovative”, and“lose/win-social”, all with three values; value =lose, value =win, value 9=neutral,the latter as the re erence category to be used in multinomial regression. Tesevariables were created according to the ollowing:

Based on step and a respondent was categorized as a winner on the “lose/ win-career ”-variable i she/he had value on (at least) two o the three job-characteristic variables “win-earnings”, “win-career-prospects” or “win-social-status”, and she/he was categorized as “loser” on the “lose/win-career ”variablei he/she had value on (at least) two o the variables “lose-earnings”, “lose-career-prospects” or “lose-social-status”. Else, the respondent was categorized

as neutral.Like verse values were assigned on the “lose/win-innovative” -variable accordingto the response to the three job-characteristic variables that concern autonomy,new-challenges or learn-new-things.Finally, values were assigned in the same way on the “lose/win-social” -variableaccording to the respondent’s answers to the our job-characteristic variablesthat concern job security, leisure activities, do something use ul or society andcombine work and amily. (Te coding o “ lose/win-social” -variable was basedon the respondent being a winner/loser respectively on three o the our itemscovered by this dimension.)

.

.

●

●

●



Appendix

Multinomial logistic regressions or each o the three winner-situations (dimensions) were run. For each o the regressions respondents who ound one o the three ( our)items connected to the particular dimension important were selected.

general results of the reflex project

Documents