19_IntrinsicExtrinsicOrientations

8/12/2019 19_IntrinsicExtrinsicOrientations

1/10

10.1177/0146167203252867 ARTICLEPERSONALITYANDSOCIALPSYCHOLOGYBULLETINMalka,Chatman/INCOMEANDSUBJECTIVEWELL-BEING

Intrinsic and Extrinsic Work Orientationsas Moderators of the Effect of Annual

Income on Subjective Well-Being:

A Longitudinal Study

Ariel Malka

Jennifer A. Chatman

University of California, Berkeley

Income is only weakly associated with both subjective well-being(SWB) and job satisfaction in the United States, a surprisingfinding in light of the importance placed on financial status incapitalistic societies. To explore this further, the authors exam-ined intrinsic and extrinsic work orientations as potential mod-erators of the effects of financial compensation on SWB and jobsatisfaction. Masters of business administration students (N =124) completed measures of work orientation and, 4 to 9 yearslater, reported their current salary, SWB, and job satisfaction.As predicted, individuals high in extrinsic orientation experi-enced higher SWB and job satisfaction to the degree that they

earned more money, whereas those high in intrinsic orientationwere lower on SWB at higher income levels. These findings arediscussed in terms of the Values as Moderators Perspective ofSWB and Cognitive Evaluation Theory.

Keywords: intrinsic motivation; work orientations; annual income;subjective well-being; job satisfaction

Subjective well-being (SWB) refers to peoples cogni-tive and emotional evaluations of their lives (Diener,Suh, Lucas, & Smith, 1999). Because how much moneyone makes determines material comfort and, to a large

degree, social status, some researchers have taken aninterest in how strongly a persons incomeinfluences hisor her SWB, or in other words, whether money can buyhappiness. Research addressing this question indicatesthat there is, indeed, a positive association betweenannual income and SWB, but it is surprisingly smallamong people whose basic needs are generally met(Diener, Sandvik, Seidlitz, & Diener, 1993; Veenhoven,1991). Thus, although earning more rather than less

money does make peoplehappier, the impact of incomeon SWB is minor when basic needs are satisfied.

Job satisfaction refers to peoples affective orienta-tions to their work roles (Lease, 1998). Although it israrely the focal question in organizational research,studies involving the assessment of both job satisfactionand income have tended to show that, similar to SWB,job satisfaction is only modestly influenced by how muchmoney one makes (Boudreau, Boswell, & Judge, 2001;OReilly & Caldwell, 1980). In a sense, this finding iseven more counterintuitive because it implies that

incomenotonly weaklypredicts overall quality of life butalso satisfaction in the life domain with which income isintimately associatedwork. Taken together, these find-ings point to the paradoxical conclusion that even in asociety dominated by capitalismin which goods, ser-vices, and even values are often framed in financial terms(Tetlock, Kristel, Elson, Green, & Lerner, 2000), and inwhich ones status is signaled through ones income andthe associated fruitsoverall happiness and satisfactionwith work are not very closely aligned with financialsuccess.

737

Authors Note:Thanks to Oliver John, Virginia Kwan, Chris Crandall,

and an anonymous reviewer for helpful comments on earlier versionsof this article. We also express our gratitude to the members of theMBA Assessment Project at the Institute of Personality and Social Re-search for collecting the data reported in this study. Correspondenceregarding this article shouldbe sent to Ariel Malka,Instituteof Person-ality and Social Research, University of California, Berkeley, 4143Tolman Hall #5050, Berkeley, CA 94720-5050; e-mail: [email protected].

PSPB,Vol. 29 No. 6, June 2003 737-746DOI: 10.1177/0146167203252867 2003 by the Society for Personality and Social Psychology, Inc.

at UNIV CALIFORNIA BERKELEY LIB on October 1, 2010psp.sagepub.comDownloaded from
http://psp.sagepub.com/http://psp.sagepub.com/http://psp.sagepub.com/http://psp.sagepub.com/


2/10

The modest correlations just described may not tellthe whole story regarding the effects of income on SWBand job satisfaction. Specifically, these weak associationsmaybe masking substantial variationin thestrength, andeven direction, of these effects among different types ofpeople. In other words, there may be some people for

whom SWB and job satisfaction are highly contingent onincome, others for whom SWB and job satisfaction areunrelated to income, andstill others for whom SWB andjob satisfaction are actually negatively contingent onincome. Therefore, a more useful question for SWBresearchers to address might be, For whom can moneybuy happiness? In line with this reasoning, we investi-gated intrinsic and extrinsic work orientations as poten-tial moderators of the effects of annual income on SWBand job satisfaction. Using an archive of data collectedwith masters of business administration (MBA) stu-dents, we tested the long-term implications of these twowork orientations, assessed during the first year of the

MBA program, for the subsequent effects of annualincome on SWB andjob satisfaction, assessed 4 to 9 yearsafter the initial assessment. This design enabled us togauge the enduring impact of work values on the contin-gencies ofSWBandjob satisfactionon financialsuccess.

SWB is conceptualized as having a cognitive appraisalcomponent, life satisfaction, and two emotional compo-nents, positive and negative affect (Diener, 1998; Dieneret al., 1999). Many studies show substantial correlationsamong these components implying that they may, takentogether, represent a superordinate SWB factor (Dieneretal., 1999; see Elliot, Sheldon,& Church, 1997). Job sat-

isfaction also has been studied as both an omnibus con-struct (e.g., OReilly & Caldwell, 1980) and a multifac-eted construct (e.g., Wanous & Lawler, 1972). Becausewe are interested in peoples general appraisals of theirlives and jobs, we treat SWB and job satisfaction as uni-taryglobalconstructs. Intrinsicandextrinsicwork orien-tations represent work-related reinforcement prefer-ences, or tendencies to value specific types of incentivesinherent in thework environment (Caldwell, OReilly, &Morris, 1983; Glynn, 1998). Individuals high in intrinsicorientation (sometimes referred to as expressive orien-tation) value opportunities for intellectual fulfillment,creative self-expression,and the pleasureassociatedwith

task mastery on the job. Individuals high in extrinsic ori-entation (sometimesreferredto as instrumental orienta-tion) primarily value remuneration and view work prin-cipally as a means to attaining remuneration. Althoughthese orientations appear conceptually to be bipolar,instruments assessing individual differences related tointrinsic and extrinsic motivation tend to yield scoresthat are nearly independent (Amabile, Hill, Hennessey,& Tighe, 1994; Pintrich & Garcia, 1991). We, therefore,conceptualize intrinsic and extrinsic work orientations

as independent dimensions of work value, with uniqueinfluences on peoples psychological responses to theirannual income.

We examined intrinsic and extrinsic work orienta-tions as moderators because of their relevance to twotheories that address how motives might influence psy-

chological responses to annual income. The first theory,the Values as Moderators Perspective of SWB, proposesthat resources impactSWBto theextent that they arerel-evant to ones values (Oishi, Diener, Suh, & Lucas,1999). This perspective implies that incomeshouldhavea stronger positive effect on SWB and job satisfactionamong individuals who are highly extrinsically orientedthan among individuals whoarelow on extrinsicorienta-tion. The second theory, Cognitive Evaluation Theory(CET), proposes that excess rewards for conducting analready intrinsically valued task can be experienced ascontrolling and may thereforenegatively influence qual-ity ofmotivationandSWB (Deci & Ryan, 1985). Thisper-

spective implies that for those relatively high in intrinsicorientation, higher income might actually be associatedwith lower SWB and job satisfaction.

VALUES AS MODERATORS OF SUBJECTIVE WELL-BEING

Some SWB research has been guided by the assump-tion that the degree to which specific life outcomes andresources affect SWB depends on how much these lifeoutcomes and resources help people attain their valuedgoals. In one study, daily life events had a greater impacton participants SWB to the degree that the events wererelevant to participants goals (Emmons, 1991). Simi-

larly, Diener and Fujita (1995) found that possessingresources that areusefulfor attainingonesidiosyncraticpersonal goals was associated with higher SWB. Oishiand colleagues proposed the Values as Moderatorsframework to explain such findings (Oishi, Diener,Lucas, & Suh, 1999; Oishi, Diener, Suh, et al., 1999).According to this framework, satisfaction with a specificlife domain only contributes to overall SWB to thedegree that the domain is valued. For example, satisfac-tion with onesachievementsonly translates to high SWBto theextent that one valuesachievement. So, accordingto this framework, daily life events (e.g., getting pro-moted) and resources (e.g., enhanced status) contrib-

ute to domain-specific satisfaction (e.g., achievement-based satisfaction), and domain-specific satisfaction inturn leads to global SWB if the domain of focus is valued(e.g., if one is achievement-oriented). Empiricalresearch supports this reasoning. For example, onestudy showed that being satisfied with a specific lifedomain has a stronger influence on overall SWB to thedegree that the life domain is personally valued (Oishi,Diener, Suh, et al., 1999). Furthermore, cross-culturalresearch suggests that overall SWB is best predicted by

738 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN



3/10

satisfactionwith life domains that arevaluedwithin onesculture (Kwan, Bond, & Singelis, 1997; Oishi, Diener,Lucas, et al., 1999).

If resources and life outcomes influence SWB to theextent that they are relevant to ones values, then extrin-sic orientation, or a persons tendency to value work pri-

marily as a means to attaining money, should moderatethe link between annual income and SWB. Specifically,for individuals high in extrinsic orientation, thereshouldbe a relatively strongpositive association betweenincome and SWB. Furthermore, although job satisfac-tion has not been studied in the context of the Values asModerators framework, it is likely that this domain-specific satisfaction is also differentially contingent onincome depending on an individuals level of extrinsicorientation. Some organizational theorists propose thatgeneral job satisfaction is determined by the extent towhich the work environment satisfies salient individualneeds (Lofquist & Dawis, 1975). If a persons need for

income is especially salient, then he or she should expe-riencea relatively high-paying jobas more satisfying thana relatively low-paying job. Extrinsically oriented individ-uals especially value financial compensation in theirjobs. Wetherefore predict that job satisfaction, similar toSWB, will be more strongly contingent on incomeamong individuals high in extrinsic orientation than itwill be among individuals low in extrinsic orientation.

COGNITIVE EVALUATION THEORY

Because financial compensation is not an intrinsicvalue, the Values as Moderators perspective does not

appear to offer insight regarding how income might dif-ferentially impact SWB depending on level of intrinsicorientation. CET, however, does. CET is a subtheory ofSelf-Determination Theory that addresses the cognitiveprocessesthat mediate theimpact of environmental con-tingencies on human motivation (Deci & Ryan, 1985).Researchers in this tradition note that a large portion ofcompetent human activities, such as those exhibited atwork, are intrinsically rewarding, meaning they are val-ued for their capacity to produce spontaneous internalreinforcement during engagement. In these cases, theopportunity to engage in the activity is rewarding in andof itself and external rewards such as money and praise

are not necessary for a person to initiate and sustain theactivity.

According to CET, rewards do not necessarily pre-clude intrinsic motivation, but presenting rewards forengaging in an already intrinsically appealing activitycan undermine intrinsic motivation. This is becauseintrinsically motivated action is partially rooted in thefundamental psychological need for self-determination,or the needto feel as though ones actions are freelycho-sen. Furthermore, rewards are often presented in a con-

trolling manner, or as intended to coax the individualinto performing the activity, fostering perceptions of anexternal rather than internal locus of causality(DeCharms, 1968). Thus, receiving a reward for doingan activity that one would have done without reward cancause a shift in the attributed cause of the behavior from

an internal to an external source. Perceiving ones taskengagement as instrumental to the acquisition of exter-nal incentives reduces feelings of self-determinationand, consequently, undermines intrinsic motivation.Numerous laboratory experiments attest to the validityof this undermining effect (see Deci, Koestner, & Ryan,1999; cf. Eisenberger & Cameron, 1996).

A target activity must be initially intrinsically interest-ing for rewards to undermine intrinsicmotivation; whena person does not find an activity intrinsically interest-ing, there is, of course, no intrinsic motivation to under-mine. This tenet of CET has been confirmed with bothexperimentalresearch (e.g.,Calder & Staw, 1975) as well

as field research in which individual differences inintrinsicorientation are used as a proxy for initial intrin-sic interest (Caldwell et al., 1983). One field studyshowed that receiving a subsidy to cover educationalexpenses only undermined intrinsic motivation amongMBA students who were highly intrinsically orientedwith respect to school (Caldwell et al., 1983). Unlike stu-dents who were not intrinsically motivated to engage intheir coursework in the first place, students high onintrinsic orientation experienced diminished intrinsicinterest as a consequence of being rewarded. In thepres-ent study, we consider intrinsic orientation a proxy for

initial intrinsic interest.Other field research has explored the larger rele-vance of intrinsic and extrinsic motivation for SWB andjob satisfaction. Kasser and Ryan (1993), for example,found that individuals who ascribed relatively high valueto their extrinsic goal pursuits in comparison to theirintrinsicgoalpursuits had lower levels of SWB. Likewise,Sheldon and Kasser (1998) demonstrated that makingprogress towardextrinsicgoals didnot produce thewell-being benefits associated with progress on intrinsicgoals. With regard to job satisfaction, findings suggestthat intrinsicmotivation is beneficial for job satisfaction,whereas extrinsic motivation is not (e.g., OReilly &

Caldwell, 1980).Individuals high on an intrinsic orientation toward

work view their work-related activity as a way to experi-ence fulfillment and mastery-related pleasure. Theseindividuals have a nontrivial initial level of intrinsicinterest in their work, a prerequisite for the undermin-ing effect. Such undermining of intrinsic motivation inthe workplace could have broader implications for theirSWB and job satisfaction. We, therefore, predict thatintrinsic orientation will moderate the relationship

Malka, Chatman / INCOME AND SUBJECTIVE WELL-BEING 739



4/10

between income and SWB and job satisfaction, respec-tively, such that, compared to individuals low in intrinsicorientation, individuals high in intrinsic orientation willbelesssatisfiedintheirjobsandloweronSWBastheyaremore highly compensated for their work.

A secondary reason for making this prediction has to

do with the factors that influence job choice. Althoughhigh-paying jobs are sometimes chosen on the basis ofintrinsic factors (see Morgan, Isaac, & Sansone, 2001), itseems likely that within a sample of business profession-als, individuals with relatively high income are morelikely to have discounted intrinsic factors when makingtheir job decisions than are individuals with relativelylow income. That is, the actual decision to take a job pri-marily on the basis of financial considerations mightimply to some extentthat lowerweightwasgivento otherfactors such as those pertaining to the intrinsic qualitiesof the job. Research has shown that in choosing a job,peopletrade offvarious jobattributes such as compensa-

tion and potential intrinsic enjoyment of the work (e.g.,Saks, Wiesner, & Summers, 1996; Shapira, 1981).Neglecting intrinsic factors when making job choicescan negatively affect SWB and job satisfaction. Onemight expect that people who hold strong intrinsic val-ues will experience these negative consequences to astronger degree. Although this explanation has intuitiveappeal, it has a less solid theoretical basis than does thefirst explanation (e.g., Morgan et al., 2001). Nonethe-less, this process represents a reasonable alternativeexplanation for the phenomena that we predict.

METHOD

We tested our hypotheses using a sample of MBA stu-dents who completed measures of work orientation dur-ingthefirst yearof theirprogram andreportedtheir cur-rent salary and levels of SWB and job satisfaction 4 to 9years after the initial assessment.

Participants

Participants in this study were 124 (58 women, 66men) MBA graduates of the University of California,Berkeleys, Haas Schoolof Businesswhocompletedtheirdegrees between 1987 and 1992. The data for this studywere collected through the MBA Assessment Project

conducted in the Institute of Personality and SocialResearch (see Craik et al., 2002).Participantscompletedan initial assessment package containing several individ-ual difference measures during the first year of their 2-year MBA program. One hundred ninety-one partici-pants were initially assessed in 1986 (n= 60), 1987 (n=71),1988 (n=24),and1991(n= 36); 151(79%) of theseparticipants completed a follow-up questionnaire in1995. Of these participants, 27 either were unemployedat the time of the 1995 follow-up, returned an incom-

plete 1995 follow-up survey, or had not completed thenecessary measures from the initial assessment. This lefta usable sample of 124 participants who had completedthe relevant measures from the initial assessment atleast 4 years prior to the 1995 follow-up. The mean ageof participants in this sample at the time of the initial

assessment was28. Of the participants, 77% were Cauca-sian, 8% were Asian, 5% were African American, 8%were Latino, and 2% specified Other or did not reporttheir ethnicity. At the time of the 1995 follow-up, 68% ofthe participants were married or cohabitating and 46%of theparticipantshadat least onechild.Annual incomeat this time ranged from $18,000 to $2,000,000, with amedian of $80,000.

Procedure

Participants were recruited via course announce-ments to participate in a weekend Personality and Mana-gerial Assessment Center. Because more students volun-

teered than could be accommodated, participants wereselected from the volunteers in a manner that maxi-mized their demographic representativeness to thebroader MBA cohorts. Participants completed severalindividual difference measures over the course of oneassessment weekend, including the measure from whichthe intrinsic and extrinsic orientation scales used in thisstudy were derived. In 1995, participants were mailed afollow-up questionnaire that they were asked to com-plete and return via U.S. mail.

Initial Assessment Measures: Intrinsicand Extrinsic Work Orientations

During the initial assessment, participants completedmeasures of intrinsic orientation and extrinsic orienta-tion that were derived from two components of a surveyrequesting information about participants jobs, careeraspirations, and life circumstances. In one of these com-ponents, participants were instructed to rate the degreeto which each of a series of statements described a char-acteristicof their idealjob on a 1 (notat all important) to7(extremely important) scale. In the second component,participants were instructed to rate the importance offive kinds of rewards that work can provide using thesame rating scale. Because the instructions for these two

sets of items differed, we transformed individual itemresponses into standard scoresbefore averaging them toform the two work orientation measures. Descriptive sta-tistics reported in this section, however, were computedprior to these transformations.

Weformed each of the two work orientation measuresby identifying items, drawn from these two componentsof thesurvey, whose content wasappropriate to the theo-retical constructs. We formed the extrinsic orientationmeasure by averaging self-reported ratings of the ideal




5/10

job characteristic good salary and benefits and therewardhaving thematerial possessions and lifestyleyoudesire (M= 5.06,SD= 1.03, = .70). We formed theintrinsic orientation measure by averaging ratings of theideal job characteristics work that allows you to makefull useof your abilities, work that stimulates your intel-

lectual growth, the opportunity to contribute to thegrowth and welfare of the organization, opportunitiesto be creative and innovative, andwork that provides afeeling of accomplishment and the reward utilizingand developing your capabilities to the fullest possibleextent and in so doing deriving a sense of personalaccomplishment(M= 5.85,SD= .63, = .63). Althoughtest-retest data arenotavailable for these particular mea-sures, other research shows that individual differencedimensions related to intrinsic and extrinsic motivationare temporally stable (e.g., Amabile et al., 1994).

1995 Follow-Up Measures: Annual Income,

Subjective Well-Being, and Job Satisfaction

Annual income. Participants reported their annualincome for the current year. As expected, the incomedistribution was skewed with extreme positive valuespulling the mean to $106,071 (SD= $183,947). As is typi-cal in compensation research, the log of the current sal-arywasused to prevent extreme positive values from hav-ing a large impact on the results (see Hodson, 1985).

Subjectivewell-being. We assessedthethreecomponentsof SWB posited by Diener (1998): life satisfaction, posi-tive affect, and negative affect. Life satisfaction was mea-sured with the Satisfaction With Life Scale (SWLS)

(Diener, Emmons, Larsen, & Griffin, 1985). This mea-sure consists of five items (e.g., I am satisfied with mylife) that participants rated on a 1 (strongly disagree) to 7(strongly agree) scale (M= 4.88,SD= 1.21,= .85). Noneof the SWLS items pertains to job- or career-related satis-faction. We assessed positive affect and negative affectwith the Positive and Negative Affect Schedule (PANAS)(Watson, Clark, & Tellegen, 1988). Participants wereinstructed to rate 10 positive emotion words (e.g.,excited) and 10 negative emotion words (e.g., afraid) withregard to how they had felt during the past year on a 1(very slightly or not all) to 5 (extremely) scale. Responseswere averaged to form positive affect (M= 3.75, SD= .61,= .88) and negative affect (M= 1.99,SD= .57,= .84)subscales. Because we do not specify differential predic-tions for the individual components of SWB, we formeda composite SWB index by averaging thezscores of thethree components (with negative affect reversed).

Job satisfaction. We assessed participants job satisfac-tion with the items All in all, how satisfied are you withyour current job using a 7-point scale in which 1 indi-cated beingcompletely dissatisfiedand 7 indicated being

completely satisfiedand How close is your current job toyour ideal job, again using a 7-point scale in which 1indicated verydifferent than my ideal joband7 indicated itismy ideal job(M= 4.81,SD= 1.12,= .75).

Control Variables

Research has shown evidence for links between vari-ous demographiccharacteristicsand motivational orien-tation,as well as dimensions related to SWB andjob satis-faction (e.g., Diener et al., 1999; Maurer & Tarulli,1994). Therefore, we obtained reports of participantssex, ethnicity, and age at the time of the initial assess-ment and, along with cohort, included these as controlvariables in the regression analyses. In the follow-upassessment, participants reported (and we included ascontrol variables) their marital status and whether theyhad children because these factors might have implica-tions forSWB(e.g.,Mookherjee, 1997) or theamountofincome necessary for basic sustenance. Finally, during

the follow-up, participants reported the number ofhours they worked per week as well as information rele-vant to their organizational status (the number of levelsin theirorganizationsandthenumberof levelsthey werefrom the top level). An organizational status measurewas computed by dividing the number of levels partici-pants had attained (total number of levels in the organi-zation minus number of levels from the top) by the totalnumber of levels in the organization. Hours worked andorganizational status were then included as controlsbecause each might have implications for work orienta-tion, income, and/or SWB, as well as the relationsamong these variables (e.g., Chatman, Caldwell, &OReilly, 1999; OReilly & Chatman, 1994).

RESULTS

Preliminary Analyses

Sex and cohort analyses. We tested for sex and cohortdifferences in thetwowork orientations, income, job sat-isfaction, composite SWB, and the three components ofSWB. Menhadsignificantly higher incomes thanwomen(r= .25,p< .01) and marginally significant sex effectswere found for the affect variables such that men experi-enced more positive affect (r= .15,p< .10) and women

experienced more negative affect (r= .15,p< .10). Par-ticipants from earlier cohorts were earning moreincome than those from later cohorts (R2 = .06,p= .06),which was expected given the differences in time spenton the job market between the cohorts.

We tested for sex interactions with the work orienta-tionsandincome in the prediction of all of the outcomevariables. We also tested for three-way interactionsbetween sex, the work orientations, and income in pre-dicting the outcome variables. None of the interactions

Malka, Chatman / INCOME AND SUBJECTIVE WELL-BEING 741



6/10

was significant and we therefore collapsed across sex inall of the analyses.

Zero-order correlations. Income had a near-significantpositive association with job satisfaction (r= .17,p< .06)but, of interest, was unrelated to SWB (r= .04,ns). Thisis likely due to the high mean wealth of the sample. Wefound a small positive correlation between intrinsic ori-entation and extrinsic orientation (r= .22,p< .05), sup-porting our contention that these orientations are notbipolar but also suggesting that they are not entirelyindependenta point to which we return in the Discus-sion section. In the regression analyses, we usedresidualized measures of intrinsic and extrinsic orienta-tion, partialling out their shared variance, to test formain effects and to form the cross-products of work ori-entation and income.

Intrinsicorientation waspositively related to SWB(r=

.29,p< .01) but uncorrelated with job satisfaction (r=.08, ns). Extrinsic orientation was uncorrelated withSWB(r=.00, ns) and job satisfaction (r= .12,ns). Intrin-sic and extrinsic orientations were both uncorrelatedwith annual income (rs = .07, nsand .03, ns, respec-tively). Finally, SWB and job satisfaction were highly cor-related (r= .51,p< .001).

Primary Analyses

Overview. To test our predictions, we individuallyregressed SWB and job satisfaction on two hierarchicalregression modelsthe extrinsic orientation as a mod-erator model and the intrinsic orientation as a modera-

tor model. The first two steps were identical across bothof the models: Step 1 included the control variables sex,age, ethnicity (four vectors), cohort (three vectors),marital status, parentversusnot a parent, organizationalstatus, and number of hours worked per week; Step 2included annual incomeandtheresidualizedversionsofintrinsic and extrinsic orientation. In each of the analy-ses, we entered an interaction term at Step 3: the Extrin-sic OrientationAnnual Income cross-product for theextrinsic orientation as a moderator model and the

Intrinsic Orientation Annual Income cross-productfor the intrinsic orientation as a moderator model. Vari-ables were mean-centered prior to the computation ofthe interaction terms. We report the interaction resultsinTable 1. In this section, we reportsignificanteffects for

the work orientations, income, and the interactionterms.

Subjective well-being. The control variables did notaccount for a significant proportion of the variance inSWBatStep1(R2 =.08, ns).Addingthetwo work orienta-tions andannual incomeas predictors at Step2 yielded asignificant increment in variance (R2 = .10,p< .01),which was accounted for by a positive main effect forintrinsic orientation (= .32,p< .01). Participants highon intrinsicorientation tendedto reporthigher levelsofSWB. Testing the extrinsic orientation as a moderatormodel, we added the Extrinsic Orientation AnnualIncomeinteraction term at Step 3 of theanalysis.Addingthis interaction term yielded a significant increment invariance (R2 = .04,p< .05) and as predicted, a positiveinteraction was observed (= .20,p< .05). Computingseparate standardized regression slopes for individualshigh (+1SD) and low (1SD) on extrinsic orientation,we obtained values of .11 for individuals high in extrinsicorientation and .28 for individuals low in extrinsic ori-entation. This confirmed our hypothesis that incomewould have more of a positiveeffect on SWB for individu-als high in extrinsic orientation than for individuals lowin extrinsic orientation. Of interest, income had a stron-

ger negative effect on SWB for individuals low in extrin-sic orientation.

Adding the Intrinsic Orientation Annual Incomeinteraction term also yielded a significant increment invariance in SWB (R2 = .04,p< .05) and a negative inter-action wasobserved ( = .23,p< .05). The standardizedregression slopeswere .41 for individuals high in intrin-sicorientation and .05 for individuals low in intrinsicori-entation, confirming our hypothesis that income wouldhave more of a negative effect on the SWB of individuals


TABLE 1: Standardized Regression SlopesBetween Income and Subjective Well-Being(SWB)/Job SatisfactionVariablesby Intrinsicand Extrin-

sic Orientation Levels

Standardized Regression Slope Between Income and

Work Orientation SWB Job Satisfaction Life Satisfaction Positive Affect Negative Affect

ExtrinsicHigh .11 .32 .16 .09 .00Low .28 .08 .28 .19 .14Interaction beta and significance .20* .20* .22* .14

.07

IntrinsicHigh .41 .02 .39 .35 .17Low .05 .18 .07 .07 .03Interaction beta and significance .23* .08 .23* .21* .07

p< .15. *p< .05.



7/10

high inintrinsicorientationcomparedto individuals lowin intrinsic orientation.

Job satisfaction. At Step 1, the control variablesaccounted for a near-significant proportion of the vari-ance in job satisfaction (R2 = .16,p< .09). Adding the twowork orientations and annual income at Step 2 yielded a

near-significant increment in variance (R2 = .05,p

Documents

19_IntrinsicExtrinsicOrientations