socio-demographic and socio-economic profiles of...

Proceedings of the First Asia-Pacific Conference on Global Business, Economics, Finance and

Social Sciences (AP14Singapore Conference) ISBN: 978-1-941505-15-1

Singapore, 1-3 August 2014 Paper ID: S468

1

www.globalbizresearch.org

Socio-demographic and Socio-economic Profiles of Satisfaction with

Local Public Services in Turkey: An Analysis on Turkey between

2004 and 2012

Sevilay GÜMÜŞ,

Hacettepe University,

Faculty of Economics and Administrative Sciences,

Ankara, TURKEY.

Email: [email protected]

___________________________________________________________________________________

Abstract

Local authorities play a crucial role in providing public goods and services to the provinces

and individuals' satisfaction play an important role in urban politics for this reason. Since

2014 Turkish local elections has showed that not only citizens but also the jurisdiction itself

have effected significantly by the Turkish citizens' choices of local goods and services

providers. It becomes essential to understand the underlying dynamics of individuals' local

public goods and services satisfaction level. The aim of this paper is to reveal the socio-

demographic and socio-economic profiles of satisfaction for local public goods and services

in terms of citizen satisfaction in Turkey. In this paper, multinomial logistic regression and

chi-square test are employed. According to the findings, satisfaction level of municipality

service in Turkey increases when overall life satisfaction, income level satisfaction,

expectations from future, living in cities and unemployment are increases. Moreover, females

and singles are more likely dissatisfied with overall local public services. Also, education and

satisfaction from municipal services have an inverse relationship which means higher

education level is directly proportional to dissatisfaction level of almost for every given

services. Turkish municipalities may adopt their performance strategies, goals and objectives

based on these findings.

__________________________________________________________________________

Key words: satisfaction, local public services, performance-based budgeting, Turkey

JEL Classification: H11, H41, E69, H72

mailto:[email protected]




2


1. Introduction

For a long time, economists thought that the best and maybe the only way of measuring

the value of goods and services was using prices. Especially normative public economists

believed that income and utility have an inverse relationship and marginal utility of income

decreases when income increases because of money's unit of count, reserve of value and

medium of exchange characteristics. Empirical analyses showed that classical economists

were right theoretically if the hedonic adaptations are ignored. Hedonic adaptation is the

psychological process by which people become accustomed to a positive or negative stimulus,

such that the emotional effects of that stimulus are attenuated over time (Frederick &

Loewenstein, 1999; Helson, 1964; Parducci, 1995).

Since economists have apprehended not only money but also individuals’ preferences and

needs affect economic life deeply, utility school has been developed. The welfare economists

of the utility school have a tendency to measure welfare by subjective terms of utility, and

they accept utility as happiness or satisfaction or desiredness. Since satisfaction, as Oliver

James (2007, p.107) states, is taken as an evaluation attitude towards some object or

experience, it should not be wrong to say that it can also be used to evaluate individuals'

attitude towards public goods and services. As a matter of fact that satisfaction has been

extensively used in private sector to appraise customer decisions in managerial decision

making process, but not used in public sector until the last two decades.

For sure, evaluation of public goods and services are more complicated than evaluation of

market goods and services due to its unique characteristics. Although all citizens consume

almost the same quantity of these goods and services, individuals benefit from them in

different levels (Rosen & Gayer, 2008, p.53) and they may not volunteer to pay equal prices.

Also, because of free rider and fiscal illusion problems individuals may not desire to pay them

at all. For these reasons public goods cannot be priced effectively. Since there is no effective

pricing, a great number of performance criteria cannot be employed and it increases the

importance of alternative performance indicators (Basaran et all, 2014, p.107).

Performance is about deploying and managing well the components of the casual models

that lead to the timely attainment of stated objectives within constraints specific to the firm

and to the satisfaction (Lebas, 1995, p.29). Adopting performance totally differentiates the

system from traditional budgeting methods since demands of individuals such as for better

quality service delivery, more transparency and accountability in government spending,

efficient and effective resource use are taken into consideration likely in private budgeting.

Starting from 1980s many countries have attempted to make reforms in their budgeting

system parallel to this main idea.




3


Nowadays most of the countries use performance-based budget system, but it is hard to

find an exact and a single definition of it. For example, Mikesell (1999, pp.185-186)

verbalized that performance budgets link inputs or costs to program activities and goals, while

Dawson states that the performance based budgeting system applies systemic efforts to assess

government activity and enhance accountability for progress and outcomes in achieving

results (1995, p.1). Nevertheless, in literature, four primary characteristics are accepted to

define or describe the system. These constituents can be listed as setting goals, allowing

comparison from expected and actual process by providing data and information, making

adjustments at this time or during future budget preparations and enabling ad-hoc program

evaluations (Young, 2003, p. 12).

Although "performance" seems a very fancy and attractive term for budgeting literature,

Melkers and Willoughby (1998, p.66) acknowledge that accurate measurement of results,

outcomes and impacts is needed. The idea behind this thought is that governments drive by

citizens' desires and needs, but taxpayers will pay for results not for efforts. Once

governments misdesign its goals or objectives in public expenditures and revenues, they

expend energy without meeting requested or desired results which make taxpayers unhappy or

dissatisfied. That's why factors which affect "performance" must be handled very carefully.

Additionally performance is not only about capability, but also about future so past data

can also be used to evaluate future (Lebas, 1995, p.26). Thence, expectation approaches very

widely used in consumer studies are employed by public services studies especially in local

government service provision and urban studies.

In this paper, socio-demographic and socio-economic profiles of satisfaction for local

public goods and services in Turkey will be tried to expound by using multinomial logit

regression analysis and chi-square test. To introduce the subject and to improve readers'

understanding firstly, in Section 2, previous studies are reviewed and possible links are

identified. Data used in the study and analyses which are directly related with the main

hypotheses of this paper are presented in Section 3. Finally, in Section 4 the main findings

will be summarized and some political foresights under the light of our analyses' outcomes for

Turkey will be presented.

2. Literature Review

Citizen surveys have been employed now by an increasing number of city administrators

or other government officials all around the world to measure the outcomes of their service

provision efforts (Herrian & Tomkins, 2012; Miller & Miller, 1991; Stipak, 1980; Van Ryzin,

2006), to understand desires of individuals, to obtain citizen feedbacks, to improve the

services quality as well as the quantity and to legalize their management decisions etc. For

instance, Dayton, Ohio has used citizen surveys annually since 1974 (Stipak, 1980; p.523).




4


Government Accounting Standard Board (1994), International City/Country Management

Association (2002) and National Academy of Public Administration (1999) in the United

States have used and paid a significant attention on municipalities' performance measurement

by these surveys in last two decades.

These public officials employ citizen satisfaction surveys not only for gauging service

effectiveness but also for their administrative performance assessments. These assessments

are a set of specific performance measures related to certain service functions or as indicators

of progress toward a mission in terms of efforts and accomplishments (Kelly & Swindell,

2002, p.610) in performance-based budgeting system. So, evaluation of municipalities' public

services and societies' socio-demographic and socio-economic features relations through these

surveys will also bring a different perspective to demonstrate public spending figures and to

production quantity which may vary from one region to another.

Since nowadays government performance auditing and control mechanisms are more

focused on improving the provision and quality of public programs and services, the

recognition of citizen satisfaction level is increased. On the other hand, satisfaction with

municipal services is produced by two sources: the objective quantity and quality of service

provided, and the subjective perception of such delivery (Fizgerald & Durant, 1980, p.585).

Because this subjective perception reflects the citizens' satisfaction and desiredness levels, it

compromises very valuable information for strategic management held by jurisdiction,

bureaucrats, policy makers and so on (Basaran et al.,2014;114).

Even though most of the studies on the satisfaction field got inspired from Leon

Festinger's theory of cognitive dissonance formed in 1957 and Richard L. Oliver's expectation

disconfirmation theory (1977, 1980) built upon the root principles, urban service studies

generally consider determinants as demographic properties (as studied by Durand, 1976;

Lovrich and Taylor, 1976, Fizgerald, 1980; Kelly & Swindell, 2002; Basaran et all, 2014),

political attitudes (DeHoog, Lowery and Lyons, 1990; Stipak, 1983) and socio-economic

characteristics (Ross, 1983) as it mentioned above.

However, it is impossible to ignore expectation disconfirmation theory since it is directly

and significantly related to the reflection of the relationship between expectation and

satisfaction levels. In theory, former consumption experiences and comparative judgments are

assumed to generate expectations (Oliver, 1980, p. 460), and the difference between prior

expectations and actual performance can be named as expectancy disconfirmation (Erevelles

& Leavitt, 1992, p.104). According to his model, high performance causes positive

disconfirmation while high expectations generate more disconfirmations. These connections

naturally tie expectation, satisfaction and performance to each other. So, it will be a subject of

future works.




5


In this study, along with socio-demographic and socio-economic characteristics;

expectation's relationship with satisfaction will try to be analyzed in terms of Turkish local

public services.

3. Methodology

3.1 Research Questions and Model

As indicated in introduction, we would like to find out which socio-demographic and

socio-economic factors influence Turkish municipalities' performances in the context of local

public goods and services. For this purpose, refuse and waste services, city water services,

public transportation services of Turkish municipalities are chosen to conduct the analyses.

To achieve our objectives, we test seven hypotheses;

H01= Turkish urbanities are more likely satisfied with public services produced by Turkish

municipalities

H02 = Females are more dissatisfied with Turkish municipality services than males

H03= Marital status does affect public service satisfaction significantly

H04= People happy with their life mostly happy with municipality services

H05= People who expects better services in future (hopeful people) are more likely the

people who satisfy with local public services

H06 = Unemployed citizens do not satisfy with municipality services

H07= Satisfaction of income level regardless income level's itself has a positive relation

with municipal services

In brief, the influences of socio-demographic and socio-economic factors on local public

services satisfaction will be analyzed or vice versa in this paper and these hypotheses are

constituted for this purpose.

In order to find answers for the hypotheses, multinomial logit regression and chi-square

test will be applied. Multinomial logistic regression test is conducted since we have

categorical dependent variables and multiple independent variables and we would like to

predict probabilities of different findings of one class on a dependent variable based on set of

independent variables.

Logistic model,

ze1

1)z(f

and its graph has a sigmoid shape as it can be seen Figure 1.

Figure 1




6


5 0 5 10

0.5

11

0

f( )z

5-4 z

This function is well-suited for modeling a probability because the values of f(z) ranges

from 0 to 1 as z varies from to .

For the each service the model below is applied.

p = Pr(Y=1| k1 X,...,X ).

p = )]X...XX(exp[1

1

kk22110

and p̂ = )]Xˆ...XˆXˆˆ(exp[1

1

kk22110

Note: With no predictors, Yn

Y

p̂

n

1i

i

3.2. Data

The source of data used in this study is Life Satisfaction Survey of Turkish Statistical

Institute from 2004 to 2012. The conducted survey comprises questions of happiness,

satisfaction and expectation levels of individuals as well as of individuals' socio-demographic

and socio-economic characteristics. Some of the questions are given in Appendix Table A1 in

detail. Table A2 in Appendix shows characteristics of our sample and Appendix Table A3

represents the summary statistics for variables in regression models.

The number of attendees in this survey is 52,792 (23,739 female and 29,053 male) for the

eight years. In detail, female population is greater than male population each year and an

increase in attendees with higher education can be seen. Although participants are mostly

working in the private sector, unemployment level has not been changed substantially. Since

sample size and subjects are selected very carefully but randomly to reflect all citizens of

Turkey as well as public thought, these details become crucial to monitor changes in society's

structure. So, we would like you to remember that Turkey is a developing country which has

relative economic and political stability existing with one party administration starting from

2002.




7


In used data, life satisfaction and income satisfaction has been scaled from 1 (very

satisfied) to 5 (very dissatisfied), but since only satisfaction and dissatisfaction answers lure

our attention, extreme answers are eliminated. Additionally, even though satisfaction levels of

local public goods and services differ from 1 to 4 as 1 (satisfied), 2 (dissatisfied), 3 (no idea)

and 4 (municipality does not have such a service), we only focus on the affection relation

between satisfied and dissatisfied as it can be seen in the tables. Answers to the questions on

expectations (definition can be found at Appendix) are coded between 1 (very hopeful) and 4

(very hopeless) and codes of socio-demographic or socio-economic replies differ based on

what the question consists of.

4. Findings

As mentioned earlier, in order to test the hypotheses, multinomial logistic regression

model is employed. Satisfaction levels of each municipality services are taken as dependent

variable, and reference category is chosen as the first category, satisfied. We do not take "no

idea" and "municipality does not have such a service" answers into consideration since we

would like to determine satisfaction and dissatisfaction levels.

Results can be seen in Table 1

Table 1

Table 1

B p Std. Error Wald Exp (B) - 2 LL Chi-Square

Intercept -0,979 *** 0,093 111,751 19.935,814 199,474

YEAR_KOD -0,040 *** 0,006 48,137 0,961 19.890,731 154,392

INCOME -0,004 0,012 0,125 0,996 19.787,691 51,351

CITY -0,341 *** 0,038 78,678 0,711 20.274,226 537,886

FEMALE 0,147 *** 0,034 18,553 1,159 19.760,550 24,210

Refuse and Waste SINGLE 0,485 *** 0,082 35,280 1,624 19.794,419 58,079

Services MARRIED 0,400 *** 0,073 29,754 1,492 19.807,127 70,788

(dissatisfaction) DIVORCED 0,022 0,136 0,026 1,022 19.739,839 3,499

SEPARATE 0,261 0,175 2,234 1,298 19.739,800 3,460

EDUCATION 0,127 *** 0,013 100,576 1,136 19.859,704 123,364

UNEMPLOYED -0,143 *** 0,036 15,755 0,867 19.763,591 27,252

H_INCM -0,211 *** 0,031 45,608 0,810 19.784,556 48,216

S_INLIFE -0,271 *** 0,042 42,711 0,762 19.788,100 51,761

EXP -0,346 *** 0,036 91,792 0,708 19.830,424 94,084

a. The reference category is: satisfied. The chi-square statistic is the difference in -2 log-likelihoods between the final model and a reduced model.

The reduced model is formed by omitting an effect from the final model.

The null hypothesis is that all parameters of that effect are 0.

*** P < 0.01 ** P < 0.05

* P < 0.10

Refuse and waste services satisfaction moderately rises each year compared to the

dissatisfaction. Similarly, dissatisfied citizens who live in cities are 0,711 times more than




8


urban satisfied citizens. In other words, people appreciate municipalities' refuse and waste

services in urban societies more than people who live in country side.

Moreover, according to the analysis females are more dissatisfied on these services than

males. If marital status of sample society is taken into account, it is seen that dissatisfaction

level of refuse and waste services is higher than the satisfaction level. Yet, the most

dissatisfied unit is found to be singles whereas divorced subjects are relatively satisfied in this

group.

Unsurprisingly, education and satisfaction are inversely correlated. Accordingly an

increase in education causes a decrease in satisfaction. Based on the refuse and waste services

data, it has been figured out that the service dissatisfied people have 1,136 times greater

educational level than the service satisfied people. On the other hand, individuals who are

satisfied in life are more likely to be and significantly dissatisfied with municipalities' such

services; and who are hopeful from the future on service provisions are mostly satisfied

people.

If the situation is evaluated in terms of income, no major changes due to income level

have been observed. But, when it comes to the satisfaction on income, then it can be observed

that people who are satisfied with their income are less dissatisfied with municipality

services. Lastly socio-economic characteristics and unemployment reflect that unemployed

people are generally happy or satisfied with these services.

Table 2

Table 2


Intercept

-0,983 ***

0,092 114,451 20.380,671 195,953

YEAR_KOD

-0,053 ***

0,006 86,038 0,949 20.373,787 189,069

INCOME 0,000 0,012 0,001 1,000 20.189,261 4,543

CITY

-0,197 ***

0,039 25,561 0,821 20.753,923 569,205

FEMALE

0,171 ***

0,034 25,501 1,187 20.211,154 26,437

City Water SINGLE

0,446 ***

0,081 30,375 1,562 20.234,448 49,731

(dissatisfaction) MARRIED

0,422 ***

0,073 33,790 1,525 20.266,049 81,331

DIVORCED 0,005 0,135 0,002 1,005 20.193,923 9,206

SEPARATE 0,346 ** 0,171 4,080 1,413 20.188,991 4,273

EDUCATION

0,122 ***

0,013 93,626 1,129 20.290,207 105,490

UNEMPLOYED

-0,156 ***

0,036 19,115 0,856 20.209,967 25,250

H_INCM

-0,157 ***

0,030 26,989 0,855 20.213,754 29,036

S_INLIFE

-0,304 ***

0,041 54,583 0,738 20.249,342 64,624




9


EXP

-0,383 ***

0,036 115,718 0,682 20.306,450 121,732

a. The reference category is: satisfied.

The chi-square statistic is the difference in -2 log-likelihoods between the final model and a reduced model.

The reduced model is formed by omitting an effect from the final model. The null hypothesis is that all parameters of that effect are 0.

*** P < 0.01

** P < 0.05

* P < 0.10

For city water, when income level and time are considered, no significant changes in

correlation between satisfaction and dissatisfaction have been detected. Nevertheless, city

water services' satisfaction level exhibits very small changes for each year. As another socio-

demographic factor, living in urban or country side significantly and directly affects

satisfaction from city water services and as it expected people who live in cities are more

satisfied with mentioned service. Also, satisfied urbanites are greater than dissatisfied

urbanites.

Similar to refuse and waste services, females, singles and married subjects'

dissatisfactions are higher than their satisfactions. Although divorced people seem relatively

neutral on city water services, their decisions are not significant at all. Apart from these, it has

been figured out that higher education level is directly proportional to dissatisfaction level.

Life satisfaction level positively affects people's thoughts on city water service. In other

words, people who are happier in their life are more likely to be satisfied with city water

services. Also, hopeful subjects are generally happy and satisfied with these services.

When income level influences satisfaction and dissatisfaction of this kind of services

indifferently, citizens satisfied with their income level seem satisfied with city water services.

Since there are no big differences observed between low income levels and high income

levels in terms of provided services, such a similar pattern in unemployment period is

expected when their low income are taken into consideration at the time. However,

unemployment level does actually affect the satisfaction level from city water services.

Unemployed people are less dissatisfied with city water services.

Table 3 represents public transportation services dissatisfaction levels referenced to

satisfaction levels of these specified services with socio-demographic and socio-economic

characteristics

Table 3

Table 3





10


Intercept -1,437

*** 0,104 190,877 31.790,405 435,497

YEAR_KOD -0,018

*** 0,006 8,775 0,982 31.625,077 270,170

INCOME 0,061

*** 0,013 22,821 1,063 31.455,692 100,784

CITY 0,050 0,048 1,064 1,051 33.202,344 1.847,436

Public FEMALE 0,091 0,036 6,340 1,095 31.375,647 20,739

Transportation SINGLE 0,599 *** 0,087 47,242 1,821 31.454,767 99,859

(dissatisfaction) MARRIED 0,273

*** 0,080 11,614 1,314 31.380,754 25,846

DIVORCED 0,089 0,142 0,397 1,094 31.356,030 1,122

SEPARATE 0,037 0,195 0,037 1,038 31.355,616 0,708

EDUCATION 0,184

*** 0,013 194,909 1,202 31.550,864 195,956

UNEMPLOYED -0,195

*** 0,038 26,584 0,823 31.404,248 49,340

H_INCM -0,386

*** 0,035 120,821 0,680 31.489,575 134,668

S_INLIFE -0,284

*** 0,045 40,541 0,753 31.415,871 60,963

EXP -0,399

*** 0,039 106,626 0,671 31.465,544 110,636

a. The reference category is: satisfied.

The chi-square statistic is the difference in -2 log-likelihoods between the final model and a reduced model.

The reduced model is formed by omitting an effect from the final model. The null hypothesis is that all parameters of that effect are 0.

*** P < 0.01

** P < 0.05

* P < 0.10

Satisfaction of municipalities' public transportation services slightly differentiates year by

year, so it will not be assessed. Dissatisfied subjects who live in cities are 1,051 times more

than urban satisfied citizens. Scilicet, urbanites are more dissatisfied than those who are

satisfied. As it is mentioned in the other two services, dissatisfied female attendees increase,

and female beneficiaries are more dissatisfied with public transportation services than male

beneficiaries. However, for public transportation service, being female is neither a strong nor

a significant factor.

Being single or being married becomes important in terms of public transportation service

satisfaction when the others are not remarkably important. Single and married citizens'

dissatisfaction levels of public transportation services are higher than their satisfaction levels.

In addition, education level still has an inverse relationship with the satisfaction level.

Even though income is not an important indicator of reflecting the link between

satisfaction and dissatisfaction levels of public transportation services of local governments,

satisfaction of income level actually is; and it means that income satisfaction positively

affects public transportation satisfaction.




11


5. Results and Discussion

For the analyzed services, satisfaction level rises each year very moderately compared to

dissatisfaction level, yet this alteration might be ignored since the probabilities are really

close to each other. Also, in both three analyses urbanites generally seem more appreciative of

the services of municipalities than people living in country-sides. However, when it comes to

public transportation, one of the bleeding wound becomes evident. In recent days, many

urbanites complain about public transportation quality and quantity especially in big cities.

And obviously, this survey verifies this information by reflecting "urbanites' dissatisfactions

increase compared to their satisfactions".

Ladies seem more dissatisfied with stated services. This information is enormously

important for Turkish municipalities since in traditional role of Turkish life women control the

needs of the house and form everything related to it. So, female dissatisfaction of the services

is not only a single person's concern but a whole family's for all reasons. Furthermore, very

surprisingly single attendees are mostly dissatisfied with the services. Since the sample

reflects Turkey in general, it becomes even more interesting. Sociologically, Turks believe

that singles are generally happy or satisfied even with little things. There is a quote saying

that "There is no sovereignty like bachelorhood".

It has been figured out that higher education level is directly proportional to

dissatisfaction level for almost every given service. So, if one district has more intellectual

residents than the other, governors must better be more careful since satisfaction can fluctuate

very easily and not serving appropriately may cause a risk on re-election. It is also important

in terms of performance criterion. If residents of these districts are generally intellectual,

administrators may re-design their objectives because basic services such as waste services

may not satisfy residents as a free-theater event. 2014 Local Elections of Turkey was very

instructive in this respect. Because districts where residents are more educated or intellectual

gave their votes to the political parties who place a great emphasis on green field or at least

show their attitudes towards such a tendency. This cannot be a solid fact to explain or support

our main idea, yet can be a proper example for it.

Another sociologic element is life satisfaction. This type of satisfaction level is found to

have a positive relationship with satisfaction of local public services. In other words, people

who are happy and satisfied with their life are less likely dissatisfied with municipality

services. Besides, people who are hopeful for getting better public services will more likely

appreciate municipality public services.

As one of the socio-economic characteristics chosen for this study, income level does not

cause a significant change between satisfaction and dissatisfaction of the stated services.

However, income satisfaction level is a very important gauge for this alteration. Scilicet,




12


people who are happy with their income are more likely to be happier and satisfied with local

public services. Lastly socio-economic characteristic, unemployment, reflects that

unemployed people are generally happy or less dissatisfied with the discussed services.

So, from the constituted hypotheses H02, H04, H05, H07 are verified while H06 is falsified

without a need of any extra comment. H01 can also be confirmed except public transformation

services and H03 may be accepted if "live separately" and "divorced" options are removed

from marital status group.

In the Public Financial Management and Control Law No. 5018 of Turkey, accepted in

2003 and become effective as of 2005, public institutions are forced to form strategic

programs. According to the Law article 9 quotes that “Public institutions form their strategic

plans with participatory methods in order to measure, to monitor and to evaluate development

plans and programs, structured based on the fundamental principles…”. Also, the Law

mentions that all strategic plans and programs must be structured for the future within the

frame of objective goals, gaugeable performance and achievable targets.

Whereas, the Law requires achievable, gaugeable targets, goals and performances we

suggest Turkish municipalities to re-set their strategic performance goals, performance target

etc. by employing life satisfaction survey as an indicator since these surveys reflects citizens

satisfaction and dissatisfaction levels from a certain public goods or services as well as

expectations from the services. If municipalities keep insists to disregard people’s

declarations, they may not be able to meet the expectations and needs of their citizens and

they may confront with not re-electing and other consequences such as unexpected social

movements like Istanbul protests shaken the total political system.

References

Basaran M.A., Basaran A.A., Cetinkaya M., Erkam S. 2014, “Kamu Hizmetlerinde

Performans Denetimi ve Memnuniyet: Asayiş Hizmetlerinde Memnuniyetin Bulanık Mantık

Temelli Analizi”, 29. Türkiye Maliye Sempozyumu Proceeding Book, 16-20 Mayıs 2014,

Antalya, Turkiye.

Dawson, C.S. 1995, Performance Measurement and Budgeting: Relearning Old Truths,

Albany New York Legislative Commission on Government Administration.

Erevelles, S., Leavitt C. 1992. "A Comparison of Current Models of Consumer

Satisfaction/Dissatisfaction"., Journal of Consumer Satisfaction, Dissatisfaction and

Complaining Behavior, Volume 5 No. 10. Pp.104-114.

Fitzgerald M. R, Robert F. D. 1980, "Citizen Evaluations and Urban Management Service

Delivery in an Era of Protest", Public Administration Review, Vol. 40. No 6. Pp. 585-594.




13


James, O. 2007, "Evaluating the Expectations Disconfirmation and Expectations Anchoring

Approaches to Citizen Satisfaction with Local Public Services", Journal of Public

Administration Research and Theory, Volume 19 pp. 107-123.

Kelly, J.M., Swindell D. 2002, "A Multiple-Indicator Approach to Municipal Service

Evaluation: Correlating Performance Measurement and Citizen Satisfaction across

Jurisdiction", Public Administrative Review, Vol. 62, No. 5, 610-621.

Lebas, M.J. 1995, "Performance Measurement and Performance Management", International

Journal of Production Economics, Volume 41, pp 23-35.

Melkers J. and Willoughby K. 1998, "The State of the States: Performance-Based Budgeting

Requirements", Public Administration Review, Vol. 58, No.1 pp.66-73.

Mikesell, J. 1999, Fiscal Administration: Analysis and Applications for the Public Sector (5th

Edition), Fort Worth, TX Harcourt Brace College Publishers.

Oliver, R. L. 1980, "A Cognitive Model of the Antecedents' and Consequences of Satisfaction

Decisions", Journal of Marketing Research, Vol.17 No.4 pp.460-469.

Rosen, H and Gayer T. 2008, Public Finance. (8th Edition), McGraw-Hill International

Edition , NY USA.

Stipak, B. 1980, "Local Governments' Use of Citizen Surveys", Public Administration

Review. Vol. 40, No.5. pp. 521-525.

Young, R.D. 2003, "Performance-Based Budget Systems". Public Policy and Practice. USC -

Institute for Public Service and Policy Research, January 2003 pp.11-24.

Appendix

Table 4

Table A1

Survey Questions for Key Variables

Satisfaction

How satisfied are you with the performance of your local authority's listed services?

REFUSE AND WASTE SERVICES

CITY WATER SERVICES

PUBLİC TRANSPORTATION SERVICES

Expectations

Thinking about the future public services delivery, how hopeful are you?

VERY HOPEFUL

HOPEFUL

HOPELESS

VERY HOPELESS

Table 5

Table A2 TOTAL NUMBERS FOR EACH YEAR




14


year 2004 2005 2006 2007 2008 2009 2010 2011 2012

individual 5513 5748 5372 5425 5371 6266 5858 6351 6887

female 2999 3090 2959 3005 2937 3482 3287 3460 3833

male 2514 2658 2413 2420 2434 2784 2571 2891 3054

city 4552 4717 4361 4343 4358 5256 4962 5404 5821

countryside 961 1031 1011 1082 1013 1010 896 947 1066

single 938 995 843 817 857 1035 870 988 982

married 4132 4291 4125 4149 4050 4643 4439 4768 5227

divorced 77 72 86 93 108 42 26 153 443

spouse died 341 369 296 344 335 397 384 410 232

live separately 25 21 22 22 21 149 139 32 0

illiterate or not go to school 977 1070 976 959 838 1001 884 947 1019

primary school 2373 2443 2234 2307 2171 2532 2348 2503 2545

secondary school 589 609 655 581 667 772 713 810 859

high school or vocational 1045 1083 984 1041 1060 1193 1161 1257 1392

collage or open collage 505 504 495 507 599 256 252 270 306

university 24 39 28 30 36 459 439 497 641

master or doctorate 0 0 0 0 0 52 61 67 125

unemployed 3633 3693 3428 3489 3389 3787 3556 3690 4024

public sector employee 439 427 420 392 416 441 423 420 557

private sector employee 1441 1628 1524 1542 1566 2038 1841 2213 2276

total local public services 5513 5748 5372 5425 5371 6266 5858 6351 6887

Table 6

Table A3

Descriptive Statistics

N Minimum Maximum Mean Std. Deviation

YEAR 52.791 1 9 5,17 2,620

INCOME 52.791 1 14 3,11 1,428

CITY 52.791 0 1 0,83 0,376

FEMALE 52.791 0 1 0,55 0,497

SINGLE 52.791 0 1 0,16 0,364

MARRIED 52.791 0 1 0,75 0,430

DIVORCED 52.791 0 1 0,02 0,143

SEPARATE 52.791 0 1 0,01 0,090

EDUCATION 52.790 1 7 2,75 1,392

UNEMPLOYED 52.791 0 1 0,62 0,486

HAPPY_WITH_INCOME_LEVEL 40.119 0 6 0,45 0,557

SATISFIED_INLIFE 37.310 0 1 0,84 0,371

EXPECTATION 52.791 0 1 0,70 0,460

REF_WASTE 52.791 1 4 1,27 0,526




15


CITY_WATER 52.791 1 4 1,28 0,542

PUBLİC_TRANS 52.791 1 4 1,65 1,002

socio-demographic and socio-economic profiles of...

Documents