0deec53508b19505e3000000

AGING, ADULT DEVELOPMENT, AND WORKMOTIVATION

RUTH KANFERPHILLIP L. ACKERMAN

Georgia Institute of Technology

We describe a framework for understanding how age-related changes in adult devel-opment affect work motivation, and, building on recent life-span theories and re-search on cognitive abilities, personality, affect, vocational interests, values, andself-concept, identify four intraindividual change trajectories (loss, gain, reorganiza-tion, and exchange). We discuss implications of the integrative framework for the useand effectiveness of different motivational strategies with midlife and older workersin a variety of jobs, as well as abiding issues and future research directions.

By 2010, nearly half the U.S. workforce will becomposed of individuals forty-five years or older(Bureau of Labor Statistics, 2002)with similarworkforce age trends occurring in most devel-oped countries. The prevalence of midlife andolder persons in the U.S. workforce stems from aconfluence of several factors, including the ag-ing of the baby boom generation, proportion-ately lower birthrates during the last third of thetwentieth century, and recent economic condi-tions that discourage early retirement. In theUnited States, initial interest in workforce agingtended to focus on the potentially negative ram-ifications for organizational effectiveness. How-ever, as a result of accumulating research evi-dence that refutes the notion of a universalnegative relationship between age and job per-formance (e.g., McEvoy & Cascio, 1989; Waldman& Avolio, 1986), legislative mandates to preventpotential age bias in personnel actions (e.g., theAge Discrimination in Employment Act), andprojections of future manpower shortages in se-lect job sectors, many organizations have under-gone a renaissance in their attitude toward thehiring, retraining, motivating, and retaining ofmidlife and older workers.

Recent reviews of worker age and organiza-tionally relevant outcomes further suggest thatchronological age serves as a proxy indicator

for a broad constellation of age-related pro-cesses that exert diverse and indirect effects onwork outcomes (for reviews, see, for example,Davies, Matthews, & Wong, 1991; Hansson,DeKoekkoek, Neece, & Patterson, 1997; Sterns &Miklos, 1995; Warr, 2001). In particular, recentwork on career stages (e.g., Dalton & Thompson,1986; Greller & Simpson, 1999) and barriers toemployee development among midlife andolder workers (e.g., London, 1998; Simpson, Grel-ler, & Stroh, 2002; Sterns & Doverspike, 1988) sug-gests that age-related changes in motivationalvariables, rather than chronological age or cog-nitive abilities per se, play a key role in success-ful work outcomes for middle-aged and olderworkers. To date, however, relatively little atten-tion has been given to understanding the effectsof aging and adult development on work moti-vation (for exceptions, see Boerlijst, Munnichs, &van der Heijden, 1998; Raynor & Entin, 1982;Warr, 2001). In light of growing organizationalinterest in how best to manage and motivate anolder workforce, the paucity of research in thisarea represents a serious gap in our knowledge.

In this article we extend extant theories ofwork motivation to take into account the dynam-ics of adult development across the life span. Inconfronting the question of age-related changesin motivation, we suggest that contemporary,process-oriented theories of work motivation(e.g., expectancy theories, goal theories) are in-sufficient for identifying the sources of age-related differences in work motivation. Process-oriented formulations usefully delineate thestructure and operation of proximal psychologi-cal variables (e.g., expectancies, self-efficacy)

We thank Peter Warr, Gilad Chen, and the reviewers fortheir valuable comments on earlier versions of this article.We also express deep gratitude to the late Frederick Kanferfor his many contributions to this work. Support for prepa-ration of this article was provided by a grant from the Na-tional Institute on Aging (AG16648).

Academy of Management Review2004, Vol. 29, No. 3, 440458.

440

through which person and situation factors af-fect goal choice and goal striving. They do not,however, address the what and why ques-tions that are inevitably raised in a full accountof how age affects workplace motivation.

For example, it is commonly noted that olderworkers, relative to younger workers, are oftenmore reluctant to engage in new skill trainingand tend to prefer collaborative (versus compet-itive) tasks. Explanations of these phenomenabased on the level of input variables (e.g., self-efficacy) describe the impact of age-relatedchanges on motivational processing but do notexplain why such age-related changes are ob-served. To address this question, we need anintegrative account of work motivation that de-lineates (1) the dynamics of adult developmentin cognition, emotion, personality, and the selfand (2) the pathways through which these dy-namics affect psychological input variables in-volved in motivational processing.

From a practical perspective, the proposedframework enhances the precision of extant pro-cess models and aids prediction about whenand which organizational interventions aremost likely to enhance work motivation and per-formance. For example, as discussed in detaillater, specific age-related cognitive changesmay yield low but realistic self-efficacy judg-ments for skill training that places heavy de-mands on fluid intellectual abilities (e.g., pro-gramming). Interventions that attempt to buildself-efficacy through increased effort in the faceof such person-task realities are likely to beshort lived and counterproductive. In contrast,low self-efficacy judgments for training of skillsthat do not place heavy demands on fluid intel-lectual abilities (e.g., conflict management) maybe productively enhanced through increasedeffort.

The integration of adult development theoryand research with extant process-oriented mod-els of work motivation also offers the opportu-nity to set out a research agenda for workplacemotivation in domains for which there is cur-rently little empirical data. For example, in moststudies of work motivation during the past half-century, researchers have used young adults.Thus, these samples do not allow detection ofpotential changes in motive primacy at laterpoints in life. Similarly, relatively little is knownabout how age-related changes in affect regula-tion influence goal striving. For example, evi-

dence suggesting that older workers use differ-ent emotion regulation strategies, relative toyounger workers, may be used to create morefavorable conditions for effective self-regula-tion. Finally, the use of specific motivationaltechniques, such as assigned goal setting, mayshow age-related differences in effectiveness asa function of developmental changes in self-concept and identity.

In summary, we maintain that aging andadult development represent important butlargely unexplored influences on work motiva-tion. To provide a starting point for theory de-velopment and systematic research efforts inthis area, we begin by placing motivation in thebroader context of life-span research that em-phasizes intraindividual change over the lifecourse. We propose that the effects of aging andadult development in key person characteris-tics, such as cognition and nonability traits, canbe organized usefully into four common themesbased on the nature of change across the lifespan: loss, growth, reorganization, and ex-change. Within each theme we describe repre-sentative research from a variety of domains,including cognitive abilities, personality, affect,interests, and values. Next, we coordinate age-related changes in person characteristics withmotivational processing components to illumi-nate the pathways through which aging andadult development in person factors may affectmotivational outcomes. We also describe impli-cations for the use and modification of manage-rial practices designed to enhance work motiva-tion. In the final sections we consider severalissues not explicitly addressed in our analysisand propose future research directions to ad-dress the workplace issues that an aging work-force will require organizations to consider inthe not-too-distant future.

BASIC TENETS OF AN ADULT LIFE-SPANAPPROACH

Life-span theorists propose that person-environment transactions not only determinethe direction, intensity, and persistence of ac-tion but also help shape the person-situationcontext in which motivation takes place. That is,reciprocal person-environment transactions cre-ate differential trajectories of adult develop-ment and action opportunities. Early work and

2004 441Kanfer and Ackerman

nonwork experiences, such as schooling, maycontribute to work motivation across the lifespan, through their influences on the individu-als development of skills, work beliefs, and joboptions. Similarly, individual differences in mo-tives, such as achievement orientation, duringyoung adulthood may condition initial jobchoices and career paths in a way that attenu-ates or accentuates the need for lifelong learn-ing or skill development. In developmentalmodels, work motivation opportunities and con-straints stem from the confluence of dynamicinternal forces (such as career and job prefer-ences) and external forces (such as changes intechnology and economic conditions).

Life-span researchers generally work withthree basic assumptions: (1) the individualscharacteristics are not fixed before reachingadulthood (where adults are typically definedas individuals over the age of eighteen), (2) adultdevelopment is a sequential and generally or-derly progression, and (3) development in adult-hood is not simply an extension of child andadolescent development. In addition to consid-ering inner changes (e.g., biological, psycholog-ical, and maturational), life-span researchersalso recognize the effects of external forces onindividuals and groups (e.g., sociologicalchanges for specific cohorts within a culturalmilieu).

An appreciation of life-span development re-quires consideration of two different types ofindividual differences: (1) interindividual differ-encesthat is, the differences between individ-ualsand (2) intraindividual differencesthatis, individual change during development. Thebulk of correlational research between traitsand organizational criteria involves investiga-tion of interindividual differences. When sam-ples are homogeneous in age, such correlationsrepresent a snapshot of the relationships for aparticular generational cohort. When samplesare heterogeneous in age, the result can be aconfounding of aging effects, cohort differences,and the trait-criterion relationships that aregenerally the main interest of the investigator.In contrast, researchers traditionally evaluateevidence for intraindividual differences in lon-gitudinal studies, where the same people arefollowed over an extensive time period. Gener-ally speaking, however, longitudinal studies are

not capable of evaluating generational cohortdifferences.1

Research from the life-span perspective em-phasizes observable regularities in (age) groupdifferences, as well as intraindividual changeover the life course. For our current purposes,the question we focus on is how these findingscan enrich extant work motivation formulationswith respect to midlife and older workers. In thenext section we consider the four themes weused to organize the impact of maturational ef-fects on work motivation.

LOSS, GROWTH, REORGANIZATION, ANDEXCHANGE

In the United States, aging is often associatedwith general declineparticularly in cognitiveand intellectual capabilities. Increasingly, how-ever, researchers in a number of domains haveshown that the assumption of general declinewith age is simplistic and misleading. Therehave been numerous applications of life-spanperspectives to intellectual abilities, personal-ity, affect, and theories of the self (e.g., see Lach-man & James, 1997). Such research suggests thatadult development is not simply a matter of de-cline but can better be considered in terms offour distinct patterns of developmentnamely,loss, growth, reorganization, and exchange. Inthis section we highlight major findings forthese development patterns from research in thecognitive abilities, personality, affective, value,self-concept, and vocational interest domains.

1 As noted by a reviewer, age-related changes observed incross-sectional studies are often difficult to disentangle fromcohort effects. Cross-sequential designs, involving longitu-dinal investigation of different cohorts, have many advan-tages over cross-sectional studies but are often impractical.Nonetheless, the accumulation of discriminant and conver-gent validity for the hypothesized change process acrossmethodologies can instill confidence that observed age-related changes occur across cohort groups. In cognitiveaging research, for example, demonstration of changes inage-related knowledge can be obtained through conver-gence of findings obtained in longitudinal studies con-ducted in different cultures (with different cohort experi-ences); multimethod, multiformat studies; and experimental,cross-sectional studies that examine age-related differencesin knowledge across a variety of domains.

442 JulyAcademy of Management Review

Loss

In research on adult intellect over the pastsixty years or so, scholars have well docu-mented a differentiation between two broadkinds of intellectual abilities across the adultlife span: (1) fluid intellectual abilitiescalledGfand (2) crystallized intellectual abilitiescalled Gc (see Cattell, 1943, 1987). The losstrajectory is largely associated with findingsthat indicate a gradual decline in cognitive abil-ities over the life span, particularly those asso-ciated with fluid intelligence (Wechsler, 1944).Fluid intellectual abilities are most associatedwith working memory, abstract reasoning, at-tention, and processing of novel information. Al-though there is some controversy about the ex-act age when Gf peaks, the typical researchfinding is that maximum levels of Gf are usuallyreached in the early twenties (for a review seeSchaie, 1996).

The well-documented findings indicating adecline in Gf over the life course have led manypractitioners to conclude that this is the majorfactor responsible for putative declines in per-formance with age. However, as Poffenbergersuggested sixty years ago:

It would be a safe estimate to say that onesmaximal capacity in essential functions declinesin the neighborhood of 1 to 1 12 percent a yearfrom the age of 25 years to the age of 50 to 55years, with a more rapid fall in the later years. Itmust be remembered, however, that the work ofthe world is seldom done at maximal capacity, foronly in the rare emergency does one call forth hisfull power. The level of achievement that is at-tained at any age is much more likely to be aquestion of interest and incentive than of capac-ity (1942: 65).

Although Poffenbergers observation thatworkers infrequently work at their capacity lim-its is an arguable point, it is clear that as capac-ity declines with age, the cognitive cost ofexerting the same quantity of cognitive re-sources increases. Moreover, this pertains toboth immediate levels of cognitive effort andeffort in terms of extended allocation of timeresources. That is, the prospect of working sev-enty to eighty hours a week may be manageablefor younger workers but typically becomes in-creasingly less attractive for middle-aged andolder adults.

As Poffenbergers statement implies, the det-rimental impact of a decline in fluid intellectual

abilities (essential functions) on job perfor-mance is not univocal. Rather, it must be con-sidered in light of two additional issues: (1) theinfluence of task demands and job knowledgeand (2) the potential compensatory effects of mo-tivation. We consider these issues in greaterdetail later in this article.

Growth

A second themeemphasizing a positive tra-jectory with agestems from theory and re-search focusing on the second kind of intellec-tual abilities: Gc. Crystallized intellectualabilities represent broad aspects of educationalor experiential knowledge (Cattell, 1987). Gc isassociated with general knowledge, extent ofvocabulary, and verbal comprehension. In con-trast to Gf, measures of Gc tend to show increas-ing levels of performance well into middle ageand beyond.

Although Gf is generally well encapsulatedand represented in tests of general mental abil-ity, Gc actually represents an extremely widerange of domain knowledge and is less wellrepresented in standard intelligence tests (Ack-erman, 1996). Literally all domain knowledgeheld by individuals makes up Gc, although tra-ditional measures of Gc typically focus only onknowledge that is well represented by the widerculture. A complete assessment of Gc for adultsof working age would, by necessity, includeboth occupational (job) knowledge and avoca-tional knowledge (such as hobbies, music, art,popular culture, etc).

Ackerman and his colleagues (Ackerman,1996, 2000; Ackerman & Rolfhus, 1999; Beier &Ackerman, 2001, 2003) obtained evidence thatindicates a strong positive association betweenadult age (up to sixty years) and knowledgelevel, when sampling varying knowledge do-mains. Further, they found this relationship isstronger than the association between tradi-tional measures of Gc and age, suggesting thatmiddle-aged adults are more knowledgeablethan young adults, when considering a broadconceptualization of knowledge.

The evidence showing age-related increasesin Gc or domain knowledge across the lifecourse suggests that individuals may compen-sate for declines in Gf by selecting jobs andgoals and by using self-regulatory strategiesthat optimize existing knowledge and skills. As


such, the effect of decline in Gf on job perfor-mance may be partially or fully offset by theindividuals movement into jobs or work rolesthat place high demands on Gc and low de-mands on Gf (see also Baltes & Baltes, 1990).

Reorganization

Mounting evidence for a third pattern of adultdevelopment comes from theorizing and re-search related to changes in the organizationand structure of nonability traits (such as per-sonality, emotion, and affect) across adulthood.In contrast to the previous two themes, whichfocus on normative changes in level of cognitiveabilities, the reorganization theme suggestssome type of discontinuity in adult developmentthat creates a qualitatively different constella-tion of motives for action.

Carstensens (1998) socioemotional theory pro-vides an example of current theorizing in thisarea. According to this theory, motive changesacross adulthood occur as a result of the reorgan-ization of goals around affect. Specifically,Carstensen proposes that the purpose of socialinteraction changes over the life span as a func-tion of a shift in the individuals time orientationemphasizing life lived from birth to life leftuntil death (see Neugarten, 1968). Thesechanges, in turn, influence the development anduse of different emotion regulation strategies(e.g., avoiding conflict, suppression).

According to socioemotional selectivity the-ory, adolescents and young adults seek socialinteractions primarily for their informationalvalue and their ramifications for future opportu-nities. Regulatory strategies aimed at modifyingthe environment or modulating emotion-relatedbehaviors are likely to be more useful than cog-nitive change or reappraisal strategies duringthis life phase. During later adulthood, however,the informational value of social interaction de-clines, as the individual focuses on time left.At this life stage, the motive for social interac-tions shifts from gaining resources to obtainingaffective rewards (emotional satisfaction) andsupporting ones identity. These changed mo-tives for social interaction, in turn, promote theuse of cognitive change and reappraisal strate-gies for modulating the emotion process.

Exchange

Recent work on specific issues in personality,self-concept, interests, and values provides ad-ditional evidence for developments that resultin an exchange in the primacy of motives acrossthe life spanthat is, age-related changes inthe strength of action tendencies or the primacyof motive classes for action. We describe evi-dence for such exchange development patternsbelow.

Personality traits. Trait theorists, such asCosta and McCrae (1988), emphasize the geneticand biological basis for personality structureand argue that traits, by definition, can be ex-pected to remain stable over time and consistentacross situations. Trait research in this traditionhas focused on two questions about the stabilityof personality. The first concerns normativechange in level of personality traits acrossadulthood (Costa & McCrae, 1994). Findingsfrom both cross-sectional and longitudinal stud-ies indicate that there are mean, age-relatedchanges in trait levels across the life span (e.g.,Jones & Meredith, 1996). Extroversion, neuroti-cism, and openness to experience all showmean declines with age so that older adults canbe expected, on average, to be less active, lessanxious, and less open to new experiences thanyounger adults. In contrast, increases in meanlevels of conscientiousness and agreeablenessover the life span have been observed in severalcross-sectional studies. Warr, Miles, and Platts(2001), for example, have shown significant agedifferences in conscientiousness, with olderadults more conscientious but less sociable, out-going, change oriented, and career motivatedthan younger adults.

In a second relevant stream of personality re-search, scholars have examined the age atwhich personality organization stabilizes andpersonality traits can be expected to show in-creasing consistency over time. Bloom (1964), forexample, suggests that stable personality or-ganization occurs around the age of twenty, al-though Costa and McCrae (1988; McCrae &Costa, 1990) suggest that personality traits sta-bilize by age thirty. However, results of a recentmeta-analysis by Roberts and DelVecchio (2000)indicate that personality traits do not stopchanging at a specific period in the life course.As Roberts and DelVecchio state, It appearsthen that traits are mostly consistent in adult-


hood, with some indication that they retain adynamic quality (2000: 17).

Stage-based models. The dynamic qualitythat Roberts and DelVecchio (2000) describe ap-pears most evident during life course transitionsassociated with entry into the workforce andagain during the period typically regarded asmidlife. The increased consistency observedduring midlife is of particular interest. Furtherevidence bearing on this period is provided inrecent work derived from stage models of per-sonality.

Stage-based models of personality describethe normative, sequenced patterns of personal-ity change across the life course. In many mod-els, such as Eriksons (1964) psychosocial theoryof development, each age-related period is as-sociated with a specific developmental task.The successful resolution of the task providesthe basis for an individuals progression to thenext stage in the sequence.

Maslows (1943, 1954) need hierarchy theory isarguably the most well-known stage model inorganizational behavior. Consistent with basictenets of modern life-span approaches, in thisearly theory Maslow suggested that progressionthrough the hierarchy is a function of the adultdevelopment in the context of experience andonly loosely associated with age, such thathigher-order needs for affiliation and self-actualization tend to occur in later adulthood,following accomplishment of lower-order needsduring early adulthood. Although Maslows ap-proach has largely faded from view in organi-zational research, other stage-type approachesto adult personality development have flour-ished (e.g., Neugarten & Hegestad, 1976; Sheldon& Kasser, 2001). Indeed, work drawing on thesemodels, such as Passages, by Sheehy (1976),Seasons of a Mans Life, by Levinson, Darrow,Klein, Levinson, and McKee (1978), and Hall andMirviss (1995) concept of the protean career,have substantially impacted popular thinkingabout the nature and meaning of work, particu-larly with regard to changes in motivation forwork and learning during the midlife period.

Building on Eriksons (1964) theory, McAdamsand his colleagues (see McAdams & de St.Aubin, 1998) have focused on an aspect ofmidlife development that has been relativelyunderstudiednamely, the rise of generativitymotives during midlife. Generativity motives re-fer to a class of tendencies pertaining to caring

for others, parenting, and helping the broadersociety and future generations. Studies provideinconsistent evidence on intraindividual differ-ences in generativity across adulthood, butcross-sectional studies of generativity by McAd-ams, de St. Aubin, and Logan (1993) and byKeyes and Ryff (1998) do show mean differencesin the strength of generativity motives acrossage groups. Their findings indicate that bothyounger and older adults express and displayless evidence of generativity (e.g., providingemotional support to others) than do middle-aged adults.

McAdams (2001) has also suggested that cir-cumstances may lead individuals to expressgenerativity in some but not other life roles.MacDermaid, Franz, and DeReus (1998) haveprovided empirical evidence for this notion in astudy showing the specificity of generativity ex-pressions as a function of context (home versuswork) and role (parent versus worker).

The findings on generativity suggest that per-sonality trait approaches to work motivationmight neglect an important dimension ofachievement motives for midlife workers. In con-trast to achievement motives that typically focuson job performance levels per se, generativitymotives focus attention on the process and col-laborative nature of goal accomplishments. Fur-ther research is needed to evaluate the salienceof these motives among midlife workers in theworkplace and the relationship of generativitymotives to the direction of workplace behavior.The emphasis on collaboration and helping as-sociated with this motive class also suggeststhat (compared to younger workers) midlifeworkers may respond more positively to mana-gerial strategies that emphasize cooperationrather than competition.

Vocational interests. Whereas we can definemotives as dispositions to find a general class ofincentives attractive (Veroff, Reuman, & Feld,1984), vocational interests represent preferenceswith respect to the attractiveness of variousclasses of work/job attributes. To date, therehave been relatively few studies of patterns ofvocational interests across the life span, andmuch of the evidence that exists is considerablydated. Perhaps the largest investigation of thestability of interests was started in the 1930s, byStrong (1955; see also the extensive review byCampbell & Hansen, 1981). Strong noted thatthere was substantial stability of vocational in-


terests across a period of twenty-two years. TheStrong Vocational Interest Blank does not assessunderlying traits per se but, rather, providesprofiles for individuals that indicate the degreeof similarity between the individuals attitudesand preferences and those of job incumbentsacross many different occupations. The reportedtest-retest correlations across eighteen years in-dicated the degree to which individuals wereconsistently aligned with particular jobs andranged from .48 (public administrator) to .79 (en-gineer).

On the surface, these results suggest that vo-cational interests change relatively little acrossthe work life span. However, because the Strongmeasure does not generally take into accountthe level of the job (e.g., line versus supervi-sory levels), it may be that the longitudinal find-ings reported by Strong (1955) reveal consistencyin the kinds of jobs that individuals prefer butdo not take into account changes with respect tointerest in job levels (however, see an ap-proach suggested by Holland, 1959). For exam-ple, individuals who have profiles indicatingconcordance with architects as young adultsgenerally have a similar profile twenty yearslatereven though they may have moved fromjobs that mostly involve extensive drafting tosupervising groups of junior architects.

Findings in the cognitive ability and knowl-edge domains further suggest that there may beimportant dissociations between the vocationalinterests of individuals across the work lifespan, depending on the job demands for Gf andGc. For high Gfdemanding jobs, the expecta-tion is that, in order to protect self-concept (seebelow), individuals will show a transition from apreference for line work as early adults to apreference for supervisory activities (whichhave lower Gf demands). In contrast, for jobsthat have high Gc requirements, individuals canbe expected to show greater stability of interestsacross the work life span. Occupations such asmedical doctor, college professor, lawyer, andso on typically fit the high Gc demand pattern.

Clearly, a missing link in this domain is evi-dence on how vocational interests changeacross adulthood with respect to the nature andrelative importance of various job dimensions,including, for example, power, social interac-tion, task challenge, autonomy, and variety. In areview of the relationship between age andpreferences for various job features, Warr (1997,

2001) suggests that age is likely to be positivelyassociated with increased preferences for phys-ical security, job security, salary, and opportu-nities for skill utilization through late midlife. Incontrast, age should be negatively associatedwith preferences for job variety, feedback, andprovision of external goal assignments. Warr(2001) further suggests that age-related prefer-ences in job features may influence motivationthrough their impact on the valence or antici-pated utility of performance.

Affect and emotion. Researchers investigatingage and affective constructs have focusedbroadly on three aspects: experience, expressiv-ity, and control (Gross, Carstensen, Pasupathi,Tsai, Skorpen, & Hsu, 1997). Gross et al. (1997)suggest that age-related declines in emotionalexpressivity may be limited to negative emo-tions and that aging is associated with improve-ment in emotion regulation, particularly interms of the use of cognitive emotion regulationstrategies. Charles, Reynolds, and Gatzs (2001)longitudinal research findings on positive andnegative affect support these notions and indi-cate that negative affect declines with age butthat positive affect remains relatively stablethrough late midlife. Results obtained byCarstensen, Paspuathi, Mayr, and Nesselroade(2000), in a cross-sectional experience-samplingstudy, further indicate that both younger andolder adults experience positive emotionsthrough age sixty but that older adults also re-port fewer negative emotions than youngeradults. Such findings suggest improved emo-tional functioning in each age group and areconsistent with evidence by Blanchard-Fieldsand Irion (1988) indicating that older adultsmake more frequent use of emotion control strat-egies than younger adults.

Self-concept. The prevailing wisdom in thestudy of self-concept is that, within many cogni-tive and intellectual domains, most individualsare sensitive to changes in their abilities withincreasing age and typically seek to act in waysthat serve to protect their overarching self-concept. That is, with the transition from youngadulthood into middle age and beyond, individ-uals see themselves as possessing lower abili-ties for Gf-related tasks and higher levels ofGc-related knowledge (Ackerman, Beier, & Bo-wen, 2002). Strategies for protecting self-conceptinclude avoidance (such as an aversion to Gf-related activities) and compensation (such as an


increased interest in demonstrating Gc-relatedknowledge).

Protecting self-concept seems to be one of theleading determinants of middle-aged and olderadults avoidance of some kinds of career devel-opment activities (e.g., Maurer, 2001). Within theframework delineated by Dweck and Leggett(1988), it appears that such adults fall victim to aperformance orientation, where they avoid sit-uations that are likely to place them at a disad-vantage in comparison to younger workers. Asalient example of this phenomenon is whenmiddle-aged or older workers are confrontedwith new technology (such as when managerswere confronted with e-mail and computerizedspreadsheets two decades ago). In contrast,such workers may be much more willing to signup for career development activities that buildon their real or perceived strengths (such asdeveloping management potential or interper-sonal skills), since they do not view these activ-ities as demanding Gf-type abilities. Nonethe-less, there is persuasive evidence that olderworkers tend to do more poorly in employeetraining (Kubeck, Delp, Haslett, & McDaniel,1996).

Values. Although the study of values has along history (e.g., see Allport, Vernon, & Lindzey,1951, and Rokeach, 1973), relatively little of thiswork has direct relevance to understanding theimpact of adult development on work motiva-tion. Ryff and Baltes (1976) suggest that as indi-viduals move from midlife to old age, they tendto show a diminution of instrumental values(such as financial security) but an increased ori-entation toward terminal values (such as adesire for a world at peace). Although this hy-pothesis is consistent with other personality andaffective life-span theories that stress recalibra-tion of a lifetime perspective and an increase ingenerativity motives during midlife, a more de-finitive conclusion about age-related changes invalues awaits empirical investigation.

Summary

Theoretical and empirical advances in thelife-span perspective indicate that the charac-terization of adult aging as a process of declineis incomplete. In the intellectual domain, de-cline in Gf occurs in the context of growth in Gc,as reflected in knowledge and skills. In the per-sonality domain, the mean level of some person-

ality traits declines (e.g., openness to experi-ence) while other traits show a mean levelincrease (conscientiousness). Consistent withthese findings, a growing body of work suggestsa reorganization of personality and affect, oftenin some form of discontinuity, around midlife.Characteristics of this period of adulthood in-clude the emergence of generative motives;greater use of emotion control strategies; and apreference for transactions that support positiveaffect, self-concept, and identity. To date, how-ever, we know of no studies that examine thepathways by which these age-related changesin person characteristics influence motivationalprocessing and work outcomes. In the next sec-tion we propose a coordinating framework foridentifying pathways of work motivation influ-ence, and we propose implications of aging andadult development research for motivationaloutcomes.

TOWARD AN EXPANDED FRAMEWORK: WORKMOTIVATION IN A LIFE-SPAN CONTEXT

Consistent with Warr (2001), we use a variantof the expectancy-value formulation to delin-eate the mechanisms and processes involved inwork motivation. We describe how age-relatedchanges and adult development (in the factorsdiscussed previously) may influence motivation,and we highlight consequences for work behav-ior and job performance.

Kanfers Model

Similar to other expectancy theories of moti-vation, Kanfers (1987) model posits that individ-uals allocate personal resources (e.g., effort) to aparticular job or task based on perceptions ofthree functions: the perceived effort-perfor-mance function (see Figure 1a), the perfor-mance-utility function (see Figure 2), and theeffort-utility function (see Figure 3). The effort-performance function describes the form of therelationship that an individual perceives acrossa range of effort levels and the likely perfor-mance outcome associated with levels of effort.As such, the perceived effort-performance func-tion may mirror or differ from an objective effort-performance function (see Figure 1a) that de-scribes the demands of the job or task.Inaccurate perceived effort-performance func-tions can occur, for example, when the individ-


ual has an erroneous self-concept (i.e., eitherbelieving that higher or lower performance willoccur for a particular level of effort) or when the

individual has insufficient information aboutthe impact of his or her effort on the level ofperformance. Performance-utility and effort-utility functions describe the individuals per-ceptions of the form of the relationship betweenthe attractiveness of different levels of perfor-mance and effort, respectively.

We propose that the effects of adult develop-ment on work motivation occur through the im-pact on these three motivational components orfunctions. In general, the perceived effort-performance relationship is determined by taskdemands in the context of an individuals cog-nitive abilities, knowledge, and skills. Nonabil-ity factors, such as traits, motives, interests, andself-concept, exert their primary influence on theperformance-utility and effort-utility functions.We describe the details of this mapping andimplications below.

FIGURE 1Hypothetical Relationships Between Level of

Effort and Level of Performance

(a) Effort-performance function. This function relates toeither actual or perceived relationship between effort andperformance.

(b) Changes in effort-performance functions with age/experience, shown for a task that is primarily associatedwith crystallized intelligence (Gc).

c) Changes in effort-performance functions with age,shown for a task that is primarily associated with fluidintelligence (Gf).

FIGURE 2Hypothetical Performance-Utility Function As a

Function of Age

FIGURE 3Hypothetical Effort-Utility Function As a

Function of Age


Impact of Age-Related Changes in Cognitionon Motivation

We propose that adult life-span patterns forGf and Gc have their primary effect on the effort-performance function and must be considered inlight of the nature of demands imposed by thework or job role. Further, because Gf and Gchave different life-span development patterns,there are distinct implications for issues of workmotivation. These implications are capturedwithin the effort-performance framework (de-scribed by Norman & Bobrow, 1975, as the per-formance-resource function; see also Ackerman,1987, and Fiedler, 1989, for a related approach).Generally, one can map the level of attentionaleffort to the amount of Gf resources allocated toa task. When job tasks are novel and not trivialin difficulty, they will require Gf resources suchthat there is usually a direct performance returnfor increasing allocation of effort (see Figure 1a).

The mapping of job or work role demands interms of Gf and Gc also requires considerationof demand level. Although job analysis is ulti-mately required to determine the demands of aspecific job, the broad organization of jobsaround relative levels of Gf and Gc demandspermits further consideration of age-related ef-fects on the perceived effort-performance func-tion.

Work that involves substantial attentional ef-fort (Gf) and comparatively lower levels of Gc,such as the job of an air traffic controller, placesvery high demands on Gf and also shows themost diminished performance with age (Sells,Dailey, & Pickrel, 1984). By congressional man-date, one cannot even train to become an airtraffic controller after the age of thirty-one. Sim-ilarly, controllers retire relatively young, with amandatory retirement age of fifty-six. Jobs thathave similarly high demands on Gf include do-mains of the physical sciences and mathematics(e.g., see Simonton, 1988). In light of the declinein Gf with age, jobs that place higher demandson Gf than Gc can be expected to become moredifficult for middle-aged workers, and, as a re-sult, workers typically show diminished perfor-mance as they age.

In contrast, for high-knowledge tasks (Gc de-pendent), age-related growth in relevant knowl-edge enables relatively high levels of perfor-mance, even in the absence of high levels ofeffort (this is sometimes called an effort-

insensitive task; see Figure 1b). In contrast tojobs that demand higher levels of Gf, jobs thatplace higher demands on Gc should not becomemore difficult or show diminished performanceat midlife. An example of such a job is teachingelementary school, where performance tends toremain relatively stable across the career andretirement typically occurs past midlife.

Many jobs entail a substantial combination ofGf and domain knowledge for peak perfor-mance. With such jobs, performance is betterdetermined as a joint function of level of Gf anddegree of domain knowledge. Examples in thisbroad category include neurosurgery and finan-cial analysisjobs that often demand relativelyhigh levels of both Gf and Gc. In these jobs, evenas Gf declines with increasing age, concomitantincreases in job knowledge more than compen-sate, yielding superior performance well intomiddle age. Jobs that require a high level ofeducation or training typically fit this mold,such as many medical occupations, or othertechnical fields, such as architecture. However,for job tasks that retain substantial novelty orthat cannot be accomplished by drawing on pre-vious experience, there remains a substantialperformance return for increasing levels of effort(an effort-dependent task). For such tasks, aslevels of Gf decline with increasing age, theslope of the effort-performance function de-creases so that even maximal effort leads to alower level of performance than was possible atyounger ages (see Figure 1c).

A final consideration pertains to the demandsassociated with employee development or train-ing. New job skill training, such as learning touse a new operating system or a new program-ming language, can be expected to initially in-crease demands on Gf and decrease the slope ofthe effort-performance function. As such, olderworkers are more likely to show poorer perfor-mance in these transient environments com-pared to younger workers.

Implications of Age-Related Changes inCognition for Managerial Practice

As noted previously, age-related changes incognition involve trajectories of both loss (in Gf)and gain (in Gc). We posit that the effectivenessof managerial practices to increase effortamong midlife adults depends on the type ofcognitive abilities demanded by the job. Jobs


that are primarily associated with Gf demandsare likely to become increasingly difficult forworkers as they enter midlife. If a worker hopesto maintain performance levels, he or she willneed to compensate for decline in Gf by workingharder and/or longer. Thus, we propose thefollowing.

Proposition 1a: As Gf declines withage (loss), increased effort is expectedto become the first-use strategy bywhich individuals will seek to sustainperformance in jobs that place highdemands on Gf.

Nonetheless, the potential compensatory ef-fect of increased effort in such jobs is a double-edged sword. Depending on the demands of thejob and the extent of cognitive decline, in-creased effort often entails a cost that can even-tually outweigh the benefits to performance.First, increased effort may not fully compensatefor declining abilities. Mathematicians, for ex-ample, may devote more attention and time to aproject as a means of compensating for a de-cline in fluid abilitiesto a point. Second, con-sistent with self-regulation theories of motiva-tion, declining performance in the face ofincreased effort can be expected to have strongnegative consequences for self-efficacy andself-concept. Decline in these variables, in turn,can be expected to reduce work motivation andincrease the likelihood of job withdrawal be-haviors.

For adults performing jobs that demand highlevels of Gf, motivational practices designed toincrease effort may also show short-term posi-tive effects but may fail to sustain motivationover the long term, as the accumulating costs ofmotivational compensation exert negative ef-fects on self-efficacy and/or self-concept.

Proposition 1b: As Gf declines withage (loss), compensatory motivationalstrategies to sustain performance byincreasing effort in high-Gf tasks willbe undermined by negative effects onpsychological factors (e.g., self-effi-cacy) that support motivation.

Our analysis suggests that, in contrast to theusual purpose of implementing motivational in-terventions to enhance performance levels, mo-tivational interventions undertaken with midlifeadults engaged in high-Gf jobs serve princi-

pally to slow performance decline. Althoughwell intentioned, such motivational interven-tions appear to shift the goal focus from promo-tion to prevention. Consistent with findings ob-tained by Higgins and his colleagues (seeBrockner & Higgins, 2001), prevention-focusedgoals exert a panoply of detrimental effects onperformance through their effects on emotionand motivation.

In contrast, an opposite pattern of effects canbe observed for motivational interventionsamong midlife adults performing jobs that aremainly associated with substantial demands onGc (see Figure 1b). Because job experience andage are often inextricably intertwined, themidlife worker will be more skilled than theyounger worker (age-related gain); thus, anyloss of Gf abilities will be compensated for byhigher levels of job knowledge (Gc). Effectiveperformance without substantial investment ofeffort, in turn, is likely to enhance self-conceptand self-esteem. Lawyers, accountants, teach-ers, writers, and other knowledge workersmay show stable or increasing levels of perfor-mance well past middle age. Studies by Hunter(1983) and Schmidt, Hunter, and Outerbridge(1986) provide indirect support for the positiveinfluence of Gc (represented in terms of job ex-perience and job knowledge) on job perfor-mance. This analysis suggests the following.

Proposition 2a: As Gc increases withage (gain), individuals engaged in jobsthat demand high levels of knowledgeand/or experience (high-Gc jobs) areexpected to maintain their performancewithout increased effort.

Proposition 2b: As Gc increases withage (gain), motivational interventionsto increase effort among individualsengaged in jobs that demand highlevels of knowledge and/or experi-ence (high-Gc jobs) are expected toimprove performance as a result of thepositive consequences of performancefor other variables, such as self-concept.

Our hypotheses imply that motivational inter-ventions typically are not needed to support per-formance among midlife adults engaged inhigh-Gc jobs. However, even for high-Gc jobs,organizations often undertake motivational in-


terventions in order to improve performance be-yond current levels. Organizations may, for ex-ample, encourage employee training or sethigher performance standards as a means ofenhancing job effort. But, as we discuss in thenext section, age-related reorganization of mo-tives may lead to different perceptions byyounger and older workers with respect to thereturn on investment of effort.

Proposition 2c: Managers seeking toimprove performance (above currentlevels) among individuals engaged inhigh-Gc jobs will need to providestronger performance incentives tospur the effort of midlife workers, com-pared to younger workers.

At this point, however, there have been nostudies that directly evaluate this hypothesis; itremains an important research question.

Impact of Age-Related Changes in Personality,Self-Concept, and Values on Motivation

The picture of adult development regardingpersonality, self-concept, and values is a com-plex one, given the interplay of many differentsources of variance across an individuals worklife span. Nonetheless, most perspectives pointtoward major reorganization and exchange ofmotives for action. We propose that thesethemes affect the relationship between level ofperformance and its utility (see Figure 2). Foryounger workers, an increasing utility is presentfor increasing levels of work performance. In-creasing performance potentially yields manyextrinsic and intrinsic rewards, such as pay,sense of achievement, recognition, and promo-tion.

As workers age, however, the strength ofachievement motives (demonstration of masteryand excellence compared to others) and open-ness to experience appear to decline, and thestrength of motives related to promoting posi-tive affect and protecting self-concept increases.Older individuals are less threatened by a fail-ure to get promoted, partly because occupa-tional achievement plays a smaller role in theirlives, compared with individuals who are in theearly part of their careers. For older workers,who may have reached the top of their occupa-tional level, higher levels of performance areless likely to be associated with a pay increase

(although the rising use of variable pay systems,to include stock options and bonuses, may sus-tain the performance-pay relationship for someindividuals). Although integrative data are lack-ing in this area, we speculate that changes inage-related goals reduce the utility of perfor-mance.

Proposition 3a: In the absence of workrole changes and/or changes in thecompensation scheme, the attractive-ness of higher levels of job perfor-mance is expected to decline withage.

This proposition suggests that, among olderworkers, work motivation will be less deter-mined by level of performance achievementand, rather, more determined by judgments ofhow much effort is required for requisite perfor-mance (the perceived effort-performance rela-tionship) and the utility of allocating that effort(the effort-utility relationship). In essence, moti-vation in this context is largely a matter of howmuch effort an individual prefers to allocate andthe extent to which preferred effort levels willsatisfy task demands. In high-Gf tasks, motiva-tion among older workers may be dampened asthe discrepancy between preferred effort leveland task demands increases. In mixed-Gf/Gctasks, declines in work motivation may be atten-uated by altering work roles to reduce the Gfdemands of the job.

However, it is important to note that this prop-osition builds on a classic treatment of the per-formance criterion, in which performance typi-cally refers to technical performance (e.g.,number of sales made, products produced [e.g.,see Borman & Motowidlo, 1993]). Although thetechnical dimension of performance has histor-ically been of greatest interest to organizations,there is increasing evidence to support the no-tion that other aspects of performance, such ascontextual performance, also contribute to theaccomplishment of organizational objectives(e.g., Johnson, 2001).

The narrow specification of performance interms of technical achievement may obscure im-portant features of performance that may sup-port the perception of a steeper performance-utility function. Alternative definitions ofperformance that emphasize nontechnical fac-tors, such as helping and job dedication, mayrepresent the primary dimensions for evaluat-


ing performance, as well as activities of partic-ular value to midlife and older workers. Jobs inwhich these features are emphasized, such asteaching, may attenuate the flattening of theperformance-utility function and thereby pro-mote work motivation. For example, perfor-mance goals that explicitly incorporate mentor-ing activities, responsibility to others, or ill-defined problem solving as a dimension ofperformance may increase the attractiveness ofhigher levels of performance among midlife andolder workers. Similarly, jobs in which higherlevels of performance offer the opportunity forpositive affective events and/or strengthenedsense of identity may also preserve steeper per-formance-utility perceptions.

Proposition 3b: Age-related decline inthe performance-utility function ismoderated by the correspondence be-tween the performance criterion andage-related valued outcomes.

Implications of Age-Related Changes inPersonality, Self-Concept, and Values forManagerial Practice

The reorganization of goals and exchange ofmotive primacy associated with midlife alter thework motivation landscape for numerous em-ployees and present a conundrum for many or-ganizations. Among younger workers, deficits inwork motivation associated with the perfor-mance-utility function often can be addressedby modifying the schedule of extrinsic incen-tives, by goal setting, and/or by redesign of workto permit new learning opportunities. The per-formance criterion (e.g., quantity, quality of pro-duction) is rarely adjusted. Among midlife andolder workers, however, shifts in motive primacyand goals require that adjustments be made tothe performance criterion. Adjustments in theperformance criterion, such as when perfor-mance is defined in terms of project manage-ment, innovation, problem solving, or trainingeffectiveness, foster recalibration of perfor-mance levels to shifts in valued outcomes.

For some individuals, recalibration occurs inthe context of organizationally sponsored careerprogression through a series of work rolechanges. Among engineers, for example, entry-level work roles tend to emphasize mastery oftechnical performance, but supervisory and

managerial work roles in the field emphasizeeffective project management. Recalibration ofthis form typically requires high levels of per-formance at each level. For other individuals,recalibration may occur through job changesoutside the organization.

Our analysis suggests that work motivationamong midlife and older workers may be en-hanced via strategies aimed at tailored recon-figuration of work roles and associated perfor-mance criteria. Although theory and research oncareer progression provide indirect support forthis notion (e.g., Dalton & Thompson, 1986), theviability of this strategy for enhancing workmotivation remains an important empiricalquestion.

Impact of Age-Related Changes in Affect andInterests on Motivation

The third component in accounts of motiva-tional processing pertains to the perceived util-ity of effort. Our review of age-related changesin person characteristics suggests that adult de-velopment may affect this function and, in turn,work motivation in two ways: (1) through thereorganization of motives to promote positiveaffect and (2) through change in the intensity ofvocational interests.

A fundamental characteristic of motivationdefined in terms of effort is its potential cost tothe individual. As Kanfer (1987) has noted, re-search on the relationship between task charac-teristics and arousal suggests that the utility ofeffort is highest in the midrange. Low alloca-tions of effort are associated with boredom; highallocations of effort are associated with stress.Findings in the motivational literature (e.g., At-kinson, 1957) provide support for the notion thatyoung adults prefer to engage in tasks with amoderate level of effort.

Research in two areas suggests that the utilityof effort can be expected to decline with age.First, evidence from the affective domain indi-cates that persons in midlife and later are morelikely than young adults to select goals andactivities that promote positive affect. To theextent that effort expenditures are likely to bepositively associated with emotional exhaus-tion, stress, and negative affect (and olderadults show heightened motive strength for pos-itive affect), lower levels of effort are expected to


have higher utility, and higher levels of effort tohave lower utility.

Second, although the interest literature sug-gests that there is substantial stability in thedirection of vocational interests during adult-hood, there may be a change in intensity ofinterest. That is, relatively little direct researchexists on stability and change in the level ofdesired work effort across the life span. How-ever, accounts from the popular literature suggestthat younger adults are more resilient in the faceof relatively high levels of continued effort (e.g., asmedical interns, prepartnership lawyers and ac-countants, etc.), in comparison to middle-aged andolder adults. These descriptions are also consis-tent with recent psychological (Heckhausen &Schulz, 1998; Schulz & Heckhausen, 1996) and eco-nomic (Posner, 1995) accounts of behavior amongolder workers that suggest a reduction in the at-tractiveness of resource allocations to work asworkers reach middle age and beyond. Thus, itappears reasonable to speculate that the utility ofhigher levels of effort is lower among adults inmiddle age and beyond than among youngeradults (see Figure 3).

Proposition 4: The attractiveness ofhigher levels of effort declines with age.

Implications of Age-Related Changes in Affectand Interests for Managerial Practice

The hypothesized impact of age-relatedchanges on the effort-utility function suggeststhat identical extrinsic motivators (such as dif-ficult and specific goals and incentive pay) mayhave a diminished effect on older adults, sincethey must be balanced against an effort con-servation orientation. In contrast to youngeradults, older adults may be less willing to com-mit to performance goals, engage in work be-haviors, or persist at tasks that they perceiveinvolve substantial levels of effort.

From an organizational perspective, this anal-ysis suggests that interventions to reduce effortdemands may be more effective among olderadults than interventions that offer large incen-tives for increased effort. In skill training, for ex-ample, younger adults may perceive highly struc-tured instruction as boring but the provision ofperformance-contingent incentives, such as pay,to be highly motivating. In contrast, among olderadults, highly structured programs that reduce ef-

fort demands may be more motivating than theprovision of incentives for performance requiringextraordinary levels of effort.

SUMMARY AND CAVEATS

In this article we have presented a theoreti-cally based scheme for investigating the effectsof aging, or adult development, on work motiva-tion. We first suggested that age-relatedchanges in person characteristics influencework motivation through their impact on thepsychological variables involved in motiva-tional processing. We identified four broad pat-terns of age-related change in person character-istics over the life courseloss, growth,reorganization, and exchangeand reviewedresearch evidence from the cognitive, personal-ity, affective, vocational interest, and self-concept domains.

Next, we suggested how aging might affectwork motivation by mapping age-relatedchanges to components of Kanfers (1987) expect-ancy-based model of motivational processing.Results of this mapping indicate multiple path-ways through which aging may affect the deci-sion to allocate personal resources to work rolesand jobs. We proposed that age-related decline(loss) in fluid cognitive abilities and age-relatedgrowth in crystallized abilities affect motivationthrough effects on the amount of effort requiredto sustain performance. We further proposedthat the nature of job demands, in terms of themixture and level of demands on fluid cognitiveabilities and crystallized knowledge, mediatesthe effects of age on this component of motiva-tional processing. Age-related changes in mo-tives and interests affect the perceived utility ofperformance and effort. As individuals entermidlife, extrinsic rewards for higher levels ofperformance and achievement lose their lustre,as interest in affirming ones identity and con-cerns for protecting the self-concept increase.Finally, we proposed that the declining salienceof work in older life and the putative reorgani-zation of personality to promote positive affectreduce the attractiveness of higher levels ofwork effort.

We recognize that our integration is incom-plete and does not address a number of issuesthat will require further attention as theory andresearch on this topic proceed. We describe a


sampling of outstanding concerns and direc-tions for future research briefly below.

The Interplay of Age-Related Patterns of AdultDevelopment

Our review of life-span theory and researchindicates that intraindividual change over thelife course is multidimensional. Our organiza-tion of adult development along four trajectoriesis intended to clarify the multiplicity of changepatterns that occur but is not meant to imply auniversal pattern of age-related change. Indeed,age-related decline in Gf may occur in tandemwith age-related gain in Gc, and patterns ofexchange and reorganization in nonabilitytraits occur within this context.

Further, the impact of adult development oncomponents of motivational processing mustalso be considered in terms of situational affor-dances and barriers. Researchers can use theinteractions among patterns of development toidentify workers for whom work motivation maybe most problematic in midlife and beyond.Among professional workers, for example, op-portunities for work role change may sustainhigh levels of work motivation throughout thelife course. In contrast, for service workers per-forming routinized tasks (e.g., custodians), thedearth of opportunities for work role changesmay increase the likelihood of work motivationdeficits in midlife and late adulthood.

The Moderating Influence of Job and WorkRole Demands

Our analysis indicates that jobs that demandhigh levels of Gf resources pose a triple threat tomotivation among older workers. Ever greateramounts of effort are needed to minimize perfor-mance declines as the utilities of performanceand effort allocations also decrease. The resultmost likely will be poorer overall performance,increased stress, or a combination of both. Assome organizations have found (e.g., the mili-tary and air traffic control), the best options areoften to remove the individual from these kindsof jobs (either through a generous early retire-ment program or through reassignment to jobsthat have lower Gf demands, such as adminis-tration or training).

However, it is important to recognize that cog-nitive/intellectual effort represents only one of

many conceivable types of effort. Many jobsplace relatively low demands on Gf (beyond ini-tial training) and may require effort in interper-sonal, rather than technical, domains. Perfor-mance in some service sector and managerialjobs, for example, may depend less on cognitive/intellectual effort and more on interpersonal en-gagement. In these jobs, age-related cognitivedeclines may have little effect on the perceivedeffort-performance function. Age-related changesin personality and affect that emphasize positiveinterpersonal relationships and enhanced emo-tion control over the life span may operate to en-hance work motivation in these jobs, if higher lev-els of performance bolster self-concept and/orincrease the perceived utility of effort. Future re-search to determine the perceived demands asso-ciated with people jobs and age-related effectson motivation in these jobs will have importanttheoretical and practical implications.

Generational Influences

One aspect of life-span development that wehave not considered here is the influence of gen-erational cohort differences. Although such adiscussion goes beyond our current concerns(e.g., see Smola & Sutton, 2002, and Whyte, 1956),we would be remiss if we did not note that themodal goals, values, and the experiences of dif-ferent generations of workers (especially interms of the nature of work and careers) mayalso interact with developmental changes incognitive, affective, and conative determinantsof work motivation. Cohort differences betweenbaby boomers and gen Xers, for example, ineconomic and political conditions, health, mo-bility, educational opportunities, the nature ofwork, demands for caregiving, and attitudes to-ward retirement may also contribute to our pro-posed influences on the perceived utilities ofeffort for work and for job performance.

Interindividual Differences

Our review has focused on normative changesover the life span, but we have not explicitlytaken into account the stability of interindi-vidual differences across the life span and thepotential influence of these differences on workmotivation. For example, although mean levelsof openness to experience may decline with age,we still expect older persons who are higher in


openness to experience to show higher levels ofmotivation for training than older persons lowerin openness to experience. That is, within a co-hort group, we can expect that individual differ-ences in relevant traits will continue to provideuseful information for predicting individual dif-ferences in work motivation.

Our framework suggests that organizationsseeking to enhance motivation among midlifeand older workers must also consider the influ-ence of the person characteristics we have re-viewed in this article (such as self-concept [seeWylie, 1989]) but that have not often been con-sidered in personality-based studies of work be-havior and job performance. For example, olderadults with stronger self-concept in the workdomain may be more responsive to role-basedmotivational interventions than individualswith weaker work-related self-concept.

Employee Development and Work Motivation

The relative reluctance of older workers toengage in new skill learning is problematic forboth individuals and organizations. Our analy-sis of work motivation in the context of agingsuggests that the motivation for training willdecline with age, as declines in cognitive abilityslow learning and the time frame for develop-ment of Gc-type expertise (in which perfor-mance may be sustained with less effort) de-creases. As such, we expect that motivation forthe development of new expertise, whether toprevent skill obsolescence or to develop exper-tise in a second or third career, should be higherearly in midlife, before the individual ap-proaches retirement (e.g., see Kubeck et al., 1996,and Warr, 2001, for a review; also see Posner,1995, for a similar conclusion).

Goal Striving

Our analysis of age-related effects on workmotivation has focused on the decision to allo-cate resources to work, and only briefly hastouched on potential age-related effects on goalstriving. Although theory and research on cog-nitive aging are currently insufficient for under-standing how aging may affect the self-regula-tory processes by which individuals implementtheir goals, we believe this is an important av-enue for future research. Memory deficits asso-ciated with aging, for example, may exert ad-

verse effects on critical components of self-regulation, such as self-monitoring, and sodampen motivation to overcome obstacles dur-ing learning and job performance.

CONCLUDING COMMENTS

To date, research on the ramifications of adultdevelopment for work motivation has been quitesparse. Warr (1992, 1994, 1997, 2001), in his inves-tigations of aging and work behavior, has of-fered a number of valuable suggestions abouthow age-related change may affect work moti-vation and about ways to enhance and sustainwork motivation among older workers. We haveextended this line of theorizing to consider howage-related changes in person characteristicsmay affect motivational processing and out-comes.

Our review and proposed framework suggestthat there is neither theoretical justification norempirical evidence to support the notion of aninevitable and universal decline in work moti-vation with age. We also suggest that work mo-tivation in midlife and later years follows thesame basic principles as work motivation inyoung adulthoodnamely, the allocation of per-sonal resources to work behaviors that build oncompetencies, promote a sense of self-efficacyand self-concept, and offer opportunities for theattainment of desired outcomes. The differencesin work motivation as adults age pertain mainlyto the impact of age-related changes in compe-tencies and motives on motivational-processingcomponents. They also pertain to differential op-portunities for protecting self-concept and to therange of options and incentives provided by or-ganizations and the work environment for at-tainment of reorganized goals. As discussed,age-related changes may enhance, decrease, orhave little effect on work motivation, dependingon work circumstances. We propose that workmotivation among older workers may be en-hanced by organizational strategies and mana-gerial practices that take into account norma-tive patterns of adult development.

Contemporary process-oriented formulationsof work motivation typically emphasize effortand time, rather than knowledge, as the per-sonal resources that an individual allocates todetermine the direction, intensity, and persis-tence of his or her work behavior. For olderworkers, knowledge is typically substantial,


whereas time and cognitive effort are often inshorter supply. For younger workers, just theopposite is true. From an organizational per-spective, the most effective motivational inter-ventions for midlife and older workers may bethose that target the utilization and distributionof relevant knowledge and experience. Alongthese lines, we anticipate that interdependentteam environments may be particularly benefi-cial for facilitating this form of work motivation(e.g., see Drucker, 1994).

We also note that modern theories of workmotivation tend to emphasize intrinsic rewardsrelated to learning and extrinsic rewards re-lated to pay, promotion, and recognition. Amongolder workers, constraints on learning, pay, andpromotion often lessen the value of these re-wards. Rewards related to emergent motives forknowledge utilization, helping, collaboration,and enhancing positive affect have yet to beaddressed in either theory or practice. If theoriesof work motivation suggest that older workersare, as a group, less motivated, perhaps theproblem lies in limitations of our current theo-ries and organization practices, rather than withthe class of individuals.

A theory of work motivation that can accountfor age-related differences must go beyond chro-nological age, per se. In particular, the theorymust provide testable predictions for (age-related) group differences and provide explana-tions for the origins and effects of such differ-ences on motivational processes and outcomes.The development and validation of a viable per-son-oriented theory of work motivation remainformidable goals. We hope that our formulationwill help to move the field further forward in thisdirection.

REFERENCES

Ackerman, P. L. 1987. Individual differences in skill learning:An integration of psychometric and information process-ing perspectives. Psychological Bulletin, 102: 327.

Ackerman, P. L. 1996. A theory of adult intellectual develop-ment: Process, personality, interests, and knowledge.Intelligence, 22: 229259.

Ackerman, P. L. 2000. Domain-specific knowledge as thedark matter of adult intelligence: gf/gc, personalityand interest correlates. Journal of Gerontology: Psycho-logical Sciences, 55B(2): P69P84.

Ackerman, P. L., Beier, M. E., & Bowen, K. R. 2002. What wereally know about our abilities and our knowledge. Per-sonality and Individual Differences, 33: 587605.

Ackerman, P. L., & Rolfhus, E. L. 1999. The locus of adultintelligence: Knowledge, abilities, and non-abilitytraits. Psychology and Aging, 14: 314330.

Allport, G. W., Vernon, P. E., & Lindzey, G. 1951. Study ofvalues: A scale for measuring the dominant interests inpersonality. Boston: Houghton Mifflin.

Atkinson, J. W. 1957. Motivational determinants of risk-taking behavior. Psychological Review, 64: 359372.

Baltes, M. M., & Baltes, P. B. 1990. Psychological perspectiveson successful aging: The model of selective optimizationand compensation. In P. B. Baltes & M. M. Baltes (Eds.),Successful aging: Perspectives from the behavioral sci-ences: 134. New York: Cambridge University Press.

Beier, M. E., & Ackerman, P. L. 2001. Current events knowl-edge in adults: An investigation of age, intelligence andnon-ability determinants. Psychology and Aging, 16:615628.

Beier, M. E., & Ackerman, P. L. 2003. Determinants of healthknowledge: An investigation of age, gender, abilities,personality, and interests. Journal of Personality andSocial Psychology, 84: 439448.

Blanchard-Fields, F., & Irion, J. 1988. The relation betweenlocus of control and coping in two age contexts: Age asa moderator variable. Psychology and Aging, 3: 197203.

Bloom, B. S. 1964. Stability and change in human character-istics. New York: Wiley.

Boerlijst, J. G., Munnichs, J. M. A., & van der Heijden, M. I. J. M.1998. The older worker in the organization. In P. J. D.Drenth, H. Thierry, & C. J. de Wolff (Eds.), Handbook ofwork and organizational psychology, vol. 2 (2nd ed.):183213. East Sussex, UK: Psychology Press.

Borman, W. C., & Motowidlo, S. J. 1993. Expanding the crite-rion domain to include elements of contextual perfor-mance. In N. Schmitt & W. C. Borman (Eds.), Personnelselection in organizations: 7198. San Francisco: Jossey-Bass.

Brockner, J., & Higgins, E. T. 2001. Regulatory focus theory:Implications for the study of emotions at work. Organi-zational Behavior and Human Decision Processes, 86:3566.

Bureau of Labor Statistics. 2002. Occupational outlook hand-book (20022003 ed.). Washington, DC: U.S. Departmentof Labor.

Campbell, D. P., & Hansen, J. C. 1981. Manual for the Strong-Campbell Interest Inventory (3rd ed.). Stanford, CA:Stanford University Press.

Carstensten, L. L. 1998. A life-span approach to social moti-vation. In J. Heckhausen & C. S. Dweck (Eds.), Motivationand self-regulation across the life span: 341364. NewYork: Cambridge University Press.

Carstensen, L. L., Pasupathi, M., Mayr, U., & Nesselroade, J. R.2000. Emotional experience in everyday life across theadult life span. Journal of Personality and Social Psy-chology, 79: 644655.

Cattell, R. B. 1943. The measurement of adult intelligence.Psychological Bulletin, 40: 153193.


Cattell, R. B. 1987. Intelligence: Its structure, growth, andaction. Amsterdam: North-Holland.

Charles, S. T., Reynolds, C. A., & Gatz, M. 2001. Age-relateddifferences and change in positive and negative affectover 23 years. Journal of Personality and Social Psychol-ogy, 80: 136151.

Costa, P. T., & McCrae, R. R. 1988. Personality in adulthood:A six-year longitudinal study of self-reports and spouseratings on the NEO personality inventory. Journal ofPersonality and Social Psychology, 54: 853863.

Costa, P. T., & McCrae, R. R. 1994. Stability and change inpersonality from adolescence through adulthood. InC. F. Halverson, G. A. Kohnstamm, & R. P. Martin (Eds.),The developing structure of temperament and personal-ity from infancy to adulthood: 139150. Hillsdale, NJ:Lawrence Erlbaum Associates.

Dalton, G. W., & Thompson, P. H. 1986. Novations: Strategiesfor career management. Glenview, IL: Scott, Foresman.

Davies, D. R., Matthews, G., & Wong, C. S. K. 1991. Ageingand work. International Review of Industrial andOrganizational Psychology, 6: 149211.

Drucker, P. 1994. Knowledge work and knowledge society:The social transformations of this century. Edwin L. God-kin Lecture, Harvard University, New Haven, CT.

Dweck, C. S., & Leggett, E. L. 1988. A social-cognitive ap-proach to motivation and personality. Psychological Re-view, 95: 256273.

Erikson, E. 1964. Insight and responsibility. New York: Norton.

Fiedler, F. E. 1989. The effective utilization of intellectualabilities and job-relevant knowledge in group perfor-mance: Cognitive resource theory and an agenda for thefuture. Applied Psychology: An International Review, 38:289304.

Greller, M. M., & Simpson, P. 1999. In search of late career: Areview of contemporary social science research appli-cable to the understanding of late career. Human Re-source Management Review, 9: 309347.

Gross, J. J., Carstensen, L. L., Pasupathi, M., Tsai, J., Skorpen,C. G., & Hsu, A. Y. C. 1997. Emotion and aging: Experi-ence, expression, and control. Psychology and Aging, 12:590599.

Hall, D. T., & Mirvis, P. H. 1995. The new career contract:Developing the whole person at midlife and beyond.Journal of Vocational Behavior, 47: 269289.

Hansson, R. O., DeKoekkoek, P. D., Neece, W. M., & Patterson,D. W. 1997. Successful aging at work. Annual review,19921996: The older worker and transitions to retire-ment. Journal of Vocational Behavior, 51: 202233.

Heckhausen, J., & Schulz, R. 1998. Developmental regulationin adulthood: Selection and compensation via primaryand secondary control. In J. Heckhausen & C. S. Dweck(Eds.), Motivation and self-regulation across the lifespan: 5077. New York: Cambridge University Press.

Holland, J. L. 1959. A theory of vocational choice. Journal ofCounseling and Clinical Psychology, 6: 3545.

Hunter, J. E. 1983. A causal analysis of cognitive ability, jobknowledge, job performance, and supervisor ratings. In

F. Landy, S. Zedeck, & J. Cleveland (Eds.), Performancemeasurement and theory: 257266. Hillsdale, NJ: Law-rence Erlbaum Associates.

Johnson, J. W. 2001. The relative importance of task andcontextual performance dimensions to supervisor judg-ments of overall performance. Journal of Applied Psy-chology, 86: 984996.

Jones, C. J., & Meredith, W. 1996. Patterns of personalitychange across the life span. Psychology and Aging, 11:5765.

Kanfer, R. 1987. Task-specific motivation: An integrative ap-proach to issues of measurement, mechanisms, pro-cesses, and determinants. Journal of Social and ClinicalPsychology, 5: 237264.

Keyes, C. L. M., & Ryff, C. D. 1998. Generativity in adult lives:Social structural contours and quality of life conse-quences. In D. P. McAdams & E. de St. Aubin (Eds.),Generativity and adult development: 227263. Washing-ton, DC: American Psychological Association.

Kubeck, J. E., Delp, N. D., Haslett, T. K., & McDaniel, M. A.1996. Does job-related training performance declinewith age? Psychology and Aging, 11: 92107.

Lachman, M. E., & James, J. B. (Eds.). 1997. Multiple paths ofmidlife development. Chicago: University of ChicagoPress.

Levinson, D. J., Darrow, C. N., Klein, E. B., Levinson, M. H., &McKee, B. 1978. The seasons of a mans life. New York:Knopf.

London, M. 1998. Career barriers: How people experience,overcome, and avoid failure. Mahwah, NJ: Lawrence Erl-baum Associates.

MacDermid, S. M., Franz, C. E., & DeReus, L. A. 1998. Gener-ativity: At the crossroads of social roles and personality.In D. P. McAdams & E. de St. Aubin (Eds.), Generativityand adult development: 181226. Washington, DC: Amer-ican Psychological Association.

Maslow, A. H. 1943. A theory of human motivation. Psycho-logical Review, 50: 370396.

Maslow, A. H. 1954. Motivation and personality. New York:Harper & Row.

Maurer, T. J. 2001. Career-relevant learning and develop-ment, worker age, and beliefs about self-efficacy fordevelopment. Journal of Management, 27: 123140.

McAdams, D. P. 2001. Generativity in midlife. In M. E. Lach-man (Ed.), Handbook of midlife development: 395446.New York: Wiley.

McAdams, D. P., & de St. Aubin, E. (Eds.). 1998. Generativityand adult development. Washington, DC: American Psy-chological Association.

McAdams, D. P., de St. Aubin, E., & Logan, R. L. 1993. Gener-ativity among young, midlife, and older adults. Psychol-ogy and Aging, 8: 221230.

McCrae, R. R., & Costa, P. T. 1990. Personality in adulthood.New York: Guilford Press.

McEvoy, G. M., & Cascio, W. F. 1989. Cumulative evidence of


the relationship between employee age and job perfor-mance. Journal of Applied Psychology, 74: 1117.

Neugarten, B. 1968. Middle age and aging. Chicago: Univer-sity of Chicago Press.

Neugarten, B., & Hagestad, G. 1976. Age and the life course.In J. E. Birren (Ed.), Handbook of aging and the socialsciences: 3555. New York: Van Nostrand Reinhold.

Norman, D. A., & Bobrow, D. B. 1975. On data-limited andresource-limited processes. Cognitive Psychology, 7: 4464.

Poffenberger, A. T. 1942. Principles of applied psychology.New York: Appleton-Century.

Posner, R. A. 1995. Aging and old age. Chicago: University ofChicago Press.

Raynor, J. O., & Entin, E. E. 1982. Motivation, career striving,and aging. New York: Appleton-Century.

Roberts, B. W., & DelVecchio, W. F. 2000. The rank-orderconsistency of personality traits from childhood to oldage: A quantitative review of longitudinal studies. Psy-chological Bulletin, 126: 466477.

Rokeach, M. 1973. The nature of human values. New York:Free Press.

Ryff, C. D., & Baltes, P. B. 1976. Value transition and adultdevelopment in women: The instrumentality-terminalitysequence hypothesis. Developmental Psychology, 12:567568.

Schaie, K. W. 1996. Intellectual development in adulthood:The Seattle longitudinal study. New York: CambridgeUniversity Press.

Schmidt, F. L., Hunter, J. E., & Outerbridge, A. N. 1986. Impactof job experience and ability on job knowledge, worksample performance, and supervisory ratings of job per-formance. Journal of Applied Psychology, 71: 432439.

Schulz, R., & Heckhausen, J. 1996. A life-span model of suc-cessful aging. American Psychologist, 51: 702714.

Sells, S. B., Dailey, J. T., & Pickrel, E. W. (Eds.). 1984. Selectionof air traffic controllers. Report no. FAA-AM-84-2. Wash-ington, DC: U.S. Department of Transportation, FederalAviation Administration.

Sheehy, G. 1976. Passages: Predictable crises of adult life.New York: Dutton.

Sheldon, K. M., & Kasser, T. 2001. Getting older, gettingbetter? Personal strivings and psychological maturityacross the life span. Developmental Psychology, 37: 491501.

Simonton, D. K. 1988. Scientific genius: A psychology of sci-ence. New York: Cambridge University Press.

Simpson, P. A., Greller, M. M., & Stroh, L. K. 2002. Variationsin human capital investment activity by age. Journal ofVocational Behavior, 61: 109138.

Smola, K. W., & Sutton, C. D. 2002. Generational differences:Revisiting generational work values for the new millen-nium. Journal of Organizational Behavior, 23: 363383.

Sterns, H. L., & Doverspike, D. 1988. Training and developingthe older adult worker. In J. E. Berren, P. K. Robinson, &J. E. Livingston (Eds.), Fourteen steps in managing anaging work force: 97110. Lexington, MA: Heath.

Sterns, H. L., & Miklos, S. M. 1995. The aging worker in achanging environment: Organizational and individualissues. Journal of Vocational Behavior, 47: 248268.

Strong, E. K., Jr. 1955. Vocational interests 18 years aftercollege. Minneapolis: University of Minnesota Press.

Veroff, J., Reuman, D., & Feld, S. 1984. Motives in Americanmen and women across the adult life span. Develop-mental Psychology, 20: 11421158.

Waldman, D. A., & Avolio, B. J. 1986. A meta-analysis of agedifferences in job performance. Journal of Applied Psy-chology, 71: 3338.

Warr, P. 1992. Age and occupational well-being. Psychologyand Aging, 7: 3745.

Warr, P. 1994. Age and employment. In M. D. Dunnette,L. Hough, & H. Triandis (Eds.), Handbook of industrialand organizational psychology, vol. 4 (2nd ed.): 485550.Palo Alto, CA: Consulting Psychologists Press.

Warr, P. 1997. Age, work, and mental health. In K. W. Schaie& C. Schooler (Eds.), The impact of work on older adults:252296. New York: Springer.

Warr, P. 2001. Age and work behaviour: Physical attributes,cognitive abilities, knowledge, personality traits, andmotives. International Review of Industrial and Organi-zational Psychology, 16: 136.

Warr, P., Miles, A., & Platts, C. 2001. Age and personality inthe British population between 16 and 64 years. Journalof Occupational and Organizational Psychology, 74:165199.

Wechsler, D. 1944. The measurement of intelligence. Balti-more: Williams & Wilkins.

Whyte, W. H., Jr. 1956. The organization man. New York:Simon and Schuster.

Wylie, R. C. 1989. Measures of self-concept. Lincoln: Univer-sity of Nebraska Press.

Ruth Kanfer is professor of psychology at the Georgia Institute of Technology. Shereceived her Ph.D. in psychology from Arizona State University. Her research ad-dresses the influence of personality, motivation, and emotion in skill training andworkplace transitions.

Phillip L. Ackerman is professor of psychology at the Georgia Institute of Technology.He received his Ph.D. in psychology from the University of Illinois at Urbana. Hisresearch concerns ability and trait complex determinants of individual differences inskills and adult intellectual development.


0deec53508b19505e3000000

Documents