Click here to load reader

Richer Blanchard Vallerand_Work Turnover

  • View

  • Download

Embed Size (px)

Text of Richer Blanchard Vallerand_Work Turnover

  • Publishing Ltd


    8640 Guilford Road, Suite 200, Columbia, MD 21046 # Phone (410) 290-3870 # Fax (410) 290-8726 # Email [email protected] #

    Date: October 15, 2002

    To: Robert J. VallerandLaboratoire de recherche sur le comportement socialDpartement de psychologieUniversit du Qubec, MontralP.O. Box 8888, Station Centre-VilleMontral, Qubec H3C 3P8Canada

    Re: Article entitled A motivational model of work turnover

    Enclosures (to be returned via fax):

    X Page proofs Manuscript (copy edited)

    (1) The above article is to be published in:

    Journal of Applied Social Psychology(Note: The volume, issue, and page numbers will be determined after we receive your corrected proofs.)

    Your manuscript has been copy edited to conform to the style and presentation of APA journals as listed in the PublicationManual of the American Psychological Association. After printing the attached file, please read the proofs carefully and markyour corrections (including typographical errors), if any, directly on the pages. Changes to abbreviation, hyphenation, orpunctuation that do not conform to APA style will not be accepted.

    (2) Please proofread and return to this office at your earliest convenience, your article will not be placed into an issue until after wereceive your corrected proofs. If you have any questions, please call 410-290-3870. Fax your changes to 410-290-8726.

    (3) It is assumed that you may wish to order a quantity of reprints of your article and that such quantity will also include therequirements of your co-author(s). The price for the reprinting of your article (which includes all costs of delivering the reprintsto you) is as follows:

    Order submitted with page proofs Order submitted after receipt of page proofs

    100 copies $ 172.80 100 copies $ 424.00 200 copies $ 233.00 200 copies $ 482.00 300 copies $ 293.20 300 copies $ 540.00 Over 300 copies, please call. Over 300 copies, please call.

    All orders for reprints must be prepaid in U.S. dollars through a U.S. bank to: Bellwether Publishing, Ltd. (FEIN 52-1814747).


    We would greatly appreciate your responses to the following. Please assist us by listing the names of the relevant services that you use mostin your research/teachings.

    Citation Index

    Abstracting service

    Computer database

  • Journal of Applied SociCopyright 2002 by V.

    A Motivational Model of Work Turnover

    SYLVIE F. RICHER, CLINE BLANCHARD, AND ROBERT J. VALLERAND1Laboratoire de Recherche sur le Comportement Social

    Universit du Qubec MontralMontral, Qubec, Canada

    The purpose of this article was to propose and test a motivational model of work turnover.The model posits that feelings of relatedness toward work colleagues and feelings of com-petence jointly and positively affect self-determined work motivation, which in turn facili-tates work satisfaction but prevents emotional exhaustion. Moreover, work satisfaction andemotional exhaustion respectively lead to negative and positive effects on turnover inten-tions. Finally, over tialumni from a schocomponents of the mses (with EQS; Benlight of the relevant

    Work turnover ceorganization. Indeedvidual who leaves thnificant and increaseinformation, it appea

    Over the years, minstance, empirical wfaction (DeCottis & McNeilly, 1995; Thojob stress (Parasuramto leave (Chen, Hu(McElroy, Morrow, 1995) in predicting t

    This research unWhile several motivrelated outcomes (BHand, & Meglino, 1

    1Correspondence conde recherche sur le comptral, P.O. Box 8888, [email protected]


    al Psychology, 2002, 32, ??, pp. 1-26. H. Winston & Son, Inc. All rights reserved.

    me, turnover intentions translate into turnover behavior. A total of 490ol of administration completed a questionnaire assessing the various

    otivational model. Results from structural equation modeling analy-tler, 1992) supported the motivational model. Results are discussed inliterature, and future research directions are proposed.

    rtainly represents one of the most important issues for any, the money and time invested in hiring and training an indi-e organization is lost forever. In addition, such costs are sig- as we move up the organizational hierarchy. In light of thisrs that work turnover deserves scientific attention.uch research has focused on turnover in the workplace. Forork has documented the role of variables such as job satis-

    Summers, 1987; Irvine & Evans, 1995; Orpen, 1995; Russ &mas & Hafer, 1995), perceptions of control (Spector, 1986),an, 1982; Summers, DenisiA, & DeCottis, 1989), intentions

    i, & Sego, 1998; Tett & Meyer, 1993), and absenteeism& Fenton, 1995; Mitra, Jenkins, & Gupta, 1992; Somers,urnover behavior.derscores the role of motivational variables in turnover.

    ational frameworks have been proposed with respect to job-andura, 1991; Carver & Scheier, 1981; Mobley, Griffeth,

    979), one theory that seems particularly useful with respect

    cerning this article should be addressed to Robert J. Vallerand, Laboratoireortement social, Dpartement de psychologie, Universi du Qubec, Mon-ion Centre-Ville, Montral, Qubec, H3C 3P8 Canada, H3C 3P8.

    Copy editorChange "Universie" to "Universite"

    Copy editorDelete the comma between "Quebec" and "H3C 3P8"


    to work turnover is self-determination theory (SDT; Deci & Ryan, 1985, 1991).Indeed, much research over the past 25 years has shown that the theory is able topredict persistent behavior in several life domains (Vallerand, 1997). Further-more, recent investigzational field (Deci1993). Thus, SDT sorganizations. Howeusefulness of SDT w

    The purpose of ttive motivational mo1991) and a recent exrand & Ratelle, in pthat such an undertareal-life test of the vSecond, the proposecerning future applie


    In accord with recently proposed thWhile the model extstudy, we limit oursethe existence of a mcomes of motivationproximal determinatence and relatedneslead to various cogntional sequence hasincluding persisten(DVallerand et al., 1tier, & Curry, 2001).

    The motivationalItE posits the followand feelings of relateence self-determinedhypothesized to posemotional exhaustioexpected to have negtions. Finally, turnoveover behavior over timmotivational model w




    Eations have provided support for its relevance to the organi-, Connell, & Ryan, 1989; Ilardi, Leone, Kasser, & Ryan,hould provide a cogent explanation of turnover behavior inver, no research to date has demonstrated empirically theith respect to work turnover.his article thus consists in presenting and testing an integra-del of work turnover based on SDT (Deci & Ryan, 1985,tension of it, the hierarchical model (Vallerand, 1997; Valle-

    ressB), as well as relevant organizational research. We feelking can yield benefits on two counts. First, it provides aalidity of SDT and the hierarchical model in the workplace.d motivational model should lead to potential insights con-d advances in the work environment.

    Motivational Model of Work Turnover

    SDT, Vallerand (1997; Vallerand & Ratelle, in pressC)e hierarchical model of intrinsic and extrinsic motivation.ends SDT on several counts, for the purpose of the presentlves to the proposition that the model makes with respect tootivational sequence integrating the determinants and out-. Specifically, the hierarchical model posits that certain keynts (e.g., intrinsic task rewards and perceptions of compe-s) influence intrinsic and extrinsic motivation, which in turnitive, affective, and behavioral consequences. This motiva- been supported with respect to different consequences,t behavior in various life domains and in educational

    997) and sports settings (Sarrazin, Vallerand, Guillet, Pelle-

    sequence as applied to work turnover is presented in Figure First, intrinsic job rewards, feelings of job competence,dness toward work colleagues are expected to positively influ- work motivation. Second, self-determined work motivation isitively affect work satisfaction, but to negatively influencen. In turn, work satisfaction and emotional exhaustion areative and positive influences, respectively, on turnover inten-r intentions are hypothesized to eventually translate into turn-e. The empirical evidence for each of the components of the

    ill be presented.


    Central to the m1991) postulated thesic and extrinsic maccomplishing onesences while engaginthey like to learn new

    On the other hanin order to receive sonly as a means to aproposed (Deci & Rexternal, introjectedlated when the sourcsay that they engagebenefits can be seenthe individual has oinducement to engagindividual to engage they would feel guiltmotivated out of ideand is valued as beiwork on a Friday nigplay identified regul

    In addition to intgested that a third mcomplete account of

    Figure 1. Motivational


    GSelf-Determined Work Motivation

    odelF is the construct of motivation. Deci and Ryan (1985, existence of three major types of motivation; namely, intrin-otivation, and amotivation. Intrinsic motivation refers to work for the inherent satisfaction and pleasure one experi-g in it. For instance, individuals might go to work because things related to their jobs.

    d, extrinsic motivation refers to engaging in a given activityomething external to the activity. The activity then servesn end. Three major types of extrinsic motivation have beenyan, 1985, 1991; GRyan, Connell, & Deci, 1985); namely,, and identified regulation. Individuals are externally regu-e of control is outside the person. For instance, workers who in their line of work because of the money or for the fringe as being externally regulated. With introjected regulation,nly partially internalized previous external pressure or

    e in the activity. It is the internal pressure that now leads thein the activity. For instance, workers who go to work becausey if they did not can be said to be introjected. Finally, whenntified regulation, the behavior is performed out of choiceng important by the individual. Thus, workers who stay atht because they have chosen to complete a project would dis-ation.rinsic and extrinsic motivation, Deci and Ryan (1985) sug-otivational concept is necessary in order to provide a more human behavior. This concept is termed amotivation and

    model of work turnover.


    refers to the relative absence of motivation (neither intrinsic nor extrinsic; seeKoestner, Losier, Vallerand, & Carducci, 1996; Vallerand, 1997). Individuals whoare amotivated engage in the activity without a sense of purpose and do not see arelationship betweeninstance, amotivatednot sure if its worth helplessness (Abram

    Deci and Ryan (1cussed earlier shoulding to Deci and Ryamotivation and identtion are external regarea and in other lifeof motivation and tBrire, Lachance, Rirand, 1990; PelletierConnell, 1989; VallThus, self-determineidentified regulationVallerand, 1997, andself-determined moti

    On the D

    Cognitive evaluaposits that feelings otant determinants ofwill affect these feelent life contexts hasself-determined motiself-determined moDeci, 1996; Richer &1978). Other researcalso be influenced b1986; Reeve & Decifeedback from the swhich in turn lead tfeedback leads to decself-determined motfeelings of relatedne

    Deci, Ryan, and Davey, & Ryan, 199 their actions and the consequences of such behavior. For individuals might say that they go to work, although theyreit. Such a state of amotivation can eventually lead to learnedson, Seligman, & Teasdale, 1978; Seligman, 1975).985) also proposed that the different types of motivation dis- be considered on a self-determination continuum. Accord-

    n, the most self-determined types of motivation are intrinsicified regulation. The least self-determined types of motiva-ulation and amotivation. Much research, both in the work domains, has supported the existence of the different types

    heir position on the self-determination continuum (Blais,ddle, & Vallerand, 1993; Blais, Sabourin, Boucher, & Valle-, Vallerand, Green-Demers, Blais, & Brire, 1995; Ryan &erand & Bissonnette, 1992; Vallerand et al., 1989, 1993).d motivation refers to high levels of intrinsic motivation and, and low levels of external regulation and amotivation (see Vallerand et al., 1997, for a more elaborate discussion onvation).

    eterminants of Self-Determined Work Motivation

    tion theory (Deci & Ryan, 1985, 1991), a subtheory of SDT,f autonomy, competence, and relatedness represent impor- motivation. Thus, according to this theory, any event thatings will also influence motivation. Much research in differ- indicated that when feelings of autonomy increase, so doesvation, while when feelings of autonomy are decreased, so istivation (e.g., Fortier, Vallerand, & Guay, 1995; Reeve & Vallerand, 1995; Zuckerman, Porac, Lathin, Smith, & Deci,h (although not as plentiful) has shown that motivation cany feelings of competence (e.g., Harackiewicz & Larson,

    , 1996; Vallerand & Reid, 1984, 1988). Thus, positive verbalupervisor produces an increase in feelings of competence,o self-determined motivation. On the other hand, negativerements in feelings of competence, which in turn undermine

    ivation. However, we still know very little about the role ofss on work motivation.their colleagues (Deci et al., 1989; Ilardi et al., 1993; Kasser,2; Riordan & Griffeth, 1995; Ryan, 1995) proposed that


    feelings of relatedness should also represent an important determinant of motiva-tion and well-being in the work context. Previous research has looked at the rela-tions between interpersonal relationships and various work outcomes. Forinstance, Piotrkowskdict physical and psyGriffeth (1995) demnities in the workplaaddition, Miller andleaving his or her orcolleagues, friends, studies underscore thleagues might have obeing, and turnoverhypothesized that fquences through thei

    It is important to motivation. Hackmavariety, task significamotivation. It is propself-determined motself-determined moconstructive feedbac& Oldham, 1976; LMayfield, 1991).


    Because self-dettioning (Deci, 1980;determined motivatinon-work contexts (Vsupports this hypothbeen associated withIlardi et al., 1993; Kexhaustion (Blais et son also showed thrmotivational orientataddition, these authwork satisfaction, wh

    Moreover, organibetween work satisfaPorter & Steers, 197

    Hi (1985) found that conflict among work colleagues can pre-chological problems among workers. As well, Riordan and

    onstrated that employees perceptions of friendship opportu-ce were related to job satisfaction and job involvement. In

    Labovitz (1973) found that the probability of an individualganization is directly related to the proportion of esteemedand contacts who have already left the organization. Thesee potential role that feelings of relatedness toward work col-n motivation and, in turn, its influence on satisfaction, well-

    intentions. In line with Deci and Ryan (1985, 1991), it iseelings of relatedness produce important affective conse-r impact on self-determined work motivation.underscore that intrinsic job rewards can also influence workn and Oldham (1974, 1976) proposed a model in which skillnce, and feedback represent important determinants of workosed that jobs that provide those elements are likely to fosterivation. While past research on this issue has not assessedtivation per se, much research has revealed that providingk can significantly enhance work motivation (e.g., Hackmanee & Graham, 1986; Stepina, Perrew, Hassell, Harris, &

    quences of Self-Determined Work Motivation

    ermination is associated with enhanced psychological func- Ryan, Deci, & Grolnick, 1995), it is hypothesized that self-on should lead to positive consequences. Research both inallerand, 1997) and in the work domain (Blais et al., 1993)esis. In the work context, self-determined motivation has high levels of work satisfaction (Harigopal & Kumar, 1982;eaveney & Nelson, 1993) and with low levels of emotionalal., 1993; Pedrabissi & Santinello, 1991). Keaveney and Nel-ough structural equation modeling that an intrinsic workion represents an important buffer against stress at work. Inors showed that intrinsic motivation positively influencesich in turn negatively determines turnover.zational research has shown that a negative correlation existsction and personnel turnover (Brayfield & Crockett, 1975H;3; Vroom, 1964). Similarly, a meta-analysis conducted by


    Irvine and Evans (1995) also revealed a strong negative relationship betweenwork satisfaction and turnover intentions (for similar conclusions, see alsoHellman, 1997; Netemeyer, Johnston, & Burton, 1990; Rush, Schoel, & Barnard,1995). Finally, in a ldemonstrated that emGoolsby, and Rhoademotional exhaustio1993; Wright & Croand emotional exhau

    Finally, the link documented in the gVallerand et al., 199intentions and actuPrussia, & Griffeth,1982; Tett & Meyer,revealed that womenchild predicted whewas more likely to oIn sum, most employchoice to do so will m

    The purpose of tturnover (Figure 1).rewards, feelings of jleagues are expecteSelf-determined worsatisfaction, but to nefaction and emotioninfluences, respectivhypothesized to everesearch has repeatenant of motivation (design, feelings of au


    A total of 500 alutreal area participateongitudinal study involving nurses, Firth and Britton (1989)otional exhaustion could predict absenteeism, while Singh,

    s (1994) showed through structural equation modeling thatn predicts turnover intentions (see also Lee & Ashforth,

    panzano, 1998). It would thus appear that work satisfactionstion can both lead to turnover intentions.between behavioral intentions and behavior has been welleneral psychological literature (Ajzen & Fishbein, 1980;

    7; see also Ajzen, 1988) as well as with respect to turnoveral behavior (Chen et al., 1998; Hom, Caranikas-Walker, 1992; Irvine & Evans, 1995; Mitchel, 1981; Parasuraman, 1993). For instance, a study by Granrose and Kaplan (1994)s intentions to return to work following the birth of their firstther or not they did return to work. Moreover, the behaviorccur for women who felt they had control over the situation.ees who intend to leave their jobs and who feel they have the

    ost likely quit their occupations (Mobley et al., 1979).

    The Present Study

    he present study is to test the motivational model of work The model proposes the following sequence: Intrinsic jobob competence, and feelings of relatedness toward work col-d to positively influence self-determined work motivation.k motivation is then hypothesized to positively affect workgatively influence emotional exhaustion. In turn, work satis-al exhaustion are expected to have negative and positiveely, on turnover intentions. Finally, turnover intentions are

    ntually translate into turnover behavior over time. Becausedly supported the role of feelings of autonomy as a determi-see Vallerand, 1997) and to reduce the complexity of thetonomy were not included in the present study.


    mni from a school of administration from the Greater Mon-d in this study. However, we rejected 10 questionnaires, 6


    that were not duly completed and 4 that came from people who were retired, for atotal of 490 final participants (181 women, 309 men). This represents a 24% par-ticipation rate from the initial sample of 2,093 individuals who were asked to par-ticipate. The mean aworking at their currof 43.88 hours per worked in the privatOf the 490 participa10% as clerical wor3% as associates, an


    Feelings of relaVallerand, 1998) is (i.e., feelings of beinstandardized Cronbaleagues (5 items, a =ability of the scale (Rhad to indicate the exinterpersonal relatiothe stem In my relinclude supported,ranging from 1 (do nfor the 10 items was

    Feelings of compassessed through a sscale is made up of ttence toward work. Tnot agree at all) to 7this scale was .76.

    Intrinsic job rewamade up of three iteintrinsic job rewardsjob, and skill variety.This job is one whegets done. The stand

    Work motivation.tory (Blais et al., 199SDT (Deci & Ryan, tionnaire, we used ansures intrinsic mot

    Ige of the participants was 38.92 years. Participants had beenent job for an average of 73 months. They worked an averageweek. Slightly more than half of the participants (53%I)e sector, while 37% were in the public or parapublic sector.nts, 41% worked as managerial staff, 28% as professionals,kers, 8% as president or vice-president, 4% as technicians,d 6% had some other function.

    tedness. The Feelings of Relatedness Scale (Richer &a 10-item scale that assesses the dimensions of acceptanceg understood and listened to by colleagues; 5 items, with achs alpha [a] of .91 for this study) and intimacy with col- .92 for this study). Research supports the validity and reli-icher & Vallerand, 1998). In the present study, participants

    tent to which each of the items corresponded to their currentnships with their work colleagues. The various items followationship with my work colleagues, I feel . . . and items close, and so forth. Items are completed on a 7-point scaleot agree at all) to 7 (very strongly agree). Cronbachs alpha.92.etence. Feelings of competence in the work domain werecale adapted from Losier, Vallerand, and Blais (1993). Thishree items assessing feelings of efficacy, ability, and compe-he scale was assessed on a 7-point scale ranging from 1 (do

    (very strongly agree). The standardized Cronbachs alpha of

    rds. Intrinsic job rewards were assessed through a scalems taken from Hackman and Oldham (1976). Thus, three were assessed; namely, task significance, feedback on the For instance, the following item measured task significance:re a lot of other people can be affected by how well the workardized Cronbachs alpha of this scale was .48.

    Participants completed the Blais Work Motivation Inven-3). This scale assesses the different constructs postulated by

    1985, 1991). In light of the high number of items in the ques- abridged version of the scale. This version of the scale mea-ivation, identif ied regulation, external regulation, and


    amotivation toward work. Each item represents a possible reason why workers goto work. These reasons are scored on a 7-point Likert scale ranging from 1 (donot agree at all) to 7 (very highly agree). Four items assess each of the four moti-vational constructs (e.g., For the satisfwork; = .90), idenhave chosen in orde(e.g., For the differamotivation (e.g., work; = .92).

    As in previous stRyan, 1987; Ryan &mined motivation iindexes using indivmultiple indicators oself-determined moweighted scores andvational subscales unweights were assignplacement on the seRyan & Connell, 19motivation, intrinsicweights of +2 and +regulation items, becof motivation, werefour items for each using the following f- [external regulationmation on the self-dthis measure (involvi

    Work satisfactiontion Scale (Blais, Lathree items that assetype of work correspon a 7-point Likert sThe standardized Cr

    Emotional exhaudimensions of profeWe thus used this sJackson, 1986; see Canadian translationexhaustion subscale mentioned (thus, a total of 16 items): intrinsic motivationaction I experience while I try to meet the challenge of mytified regulation (e.g., Because it is the type of work that Ir to reach my career goals; = .86), external regulationent fringe benefits associated with my work; = .74), andI dont know; I dont think I have what it takes to do this

    udies (e.g., Blais et al., 1990; Fortier et al., 1995; Grolnick & Connell, 1989; Vallerand & Bissonnette, 1992), a self-deter-ndex was derived by computing four separate autonomyidual items of the subscales. These four indexes served asf the latent construct of self-determined work motivation. Ativation index consists of a summation of specif ically

    is used to integrate the information from the different moti-der one score. In line with previous studies using the index,ed to the motivational items according to their respectivelf-determined continuum (Grolnick & Ryan, 1987, 1989;89). Because they are considered self-determined forms of motivation and identified regulation items were assigned1, respectively. On the other hand, amotivation and externalause they are conceptualized as less self-determined forms

    assigned weights of -2 and -1, respectively. As there wereof the motivational subscales, four indexes were computedormula: {[2 (intrinsic motivation) + identified regulation] + 2 (amotivation)]]} (see Vallerand, 1997, for more infor-etermination index). The standardized Cronbachs alpha ofng the four indexes) was .93.. Work satisfaction was assessed through the Work Satisfac-chance, Forget, Richer, & Dulude, 1991). It is made up ofss general satisfaction toward work (e.g., In general, thisonds to what I want to do in my life). The scale is measuredcale ranging from 1 (strongly disagree) to 7 (strongly agree).onbachs alpha for this scale was .86.stion. Emotional exhaustion represents one of the threessional burnout proposed by Maslach and Jackson (1986).ubscale from the Maslach Burnout Inventory (Maslach &Blais, Richer, Lachance, & Dulude, 1991, for the French of the scale). Four items were selected from the emotional(e.g., I feel burnt out from my work and I feel used up at

    Copy editorDelete one of the brackets after the equation; it should appear as )]}


    the end of the work day). This scale is completed on a 7-point scale rangingfrom 0 (never) to 6 (each day). The scale showed a standardized Cronbachsalpha of .87.

    Turnover intentiothree items that askplanned to look for nfor new jobs outsidassessed on a 7-poindardized Cronbachintention scale.


    The alumni assoMontreal area was selected randomly fstamped addressed eaddresses. A letter wstudy, how we obtainof three money prizeter also indicated thaeither part-time or fbecause they were sfrom the 2,093 quesof 24%.

    Participants wereof data collection in sent to participants their current job statquestionnaire. Partictheir jobs, and in botpleted two questionsWhy are you still osample response itewere not occupying versus obligation acthe same job you weI didnt have any chopoint scale ranging fthese two questions enced with respectassessed 1 year earlins. Turnover intentions were obtained from responses toed whether respondents thought about leaving their jobs,ew jobs over the next 12 months, and would actively searche the f irm (ODriscoll & Beehr, 1994). Responses weret scale ranging from 1 (not at all) to 7 (extremely). A stan-s alpha of .91 was obtained from the three-item turnover

    ciation from a School of Administration from the Greatercontacted and provided a list of 2,250 names that wererom their 28,000 members. A questionnaire and a returnnvelope were sent to each of those individuals at their workas also included that explained the general purpose of theed their work addresses, and the possibility of winning one

    s ($250, $150, or $100) by participating in the study. The let-t in order to participate in this study, individuals had to workull-time. A total of 157 questionnaires were returned to usent to the wrong address. We received 500 questionnairestionnaires that reached their destinations, for a response rate

    contacted again through the mail 1 year after the first waveorder to assess turnover behavior. A short questionnaire wasin which they were asked to answer questions pertaining tous. A total of 241 participants completed and returned thisipants were asked to indicate whether or not they had lefth cases why. Those who were occupying the same job com- of choice versus obligation according to the following stem:ccupying the same job you were occupying last year? A

    m is Because I wanted to keep my job. Participants whothe same job completed two questions pertaining to choicecording to the following stem: Why are you not occupyingre occupying last year? A sample response item is Becauseice, I had to change job. Responses were assessed on a 7-

    rom 1 (do not agree at all) to 7 (very strongly agree). Thus,served to assess the level of choice that participants experi- to the behavior. Because participants intentions wereer, various factors out of their control might have influenced

  • 10 RICHER ET AL.

    participants behavior since the assessment of their intentions. Thus, such achoice measure might be particularly useful in explaining the strength of theintentionsbehavior link. Ajzen (1988) used a similar strategy in his theory ofplanned behavior.

    Data Analyses

    Two models werturnover behavior. Tthe number of particlight of the number ever, we tested a secexhaustion, work sabridged model alloturnover behavior w1990).

    The adequacy oftion modeling withobserved variables wate and multivariatethus tested with stan(ML) method of estiforms reasonably wesample size is largewith the present data

    The EQS (Bentlmodel fit. Herein, wBentler-Bonett non-tion of the differenceperfect fit) consistinthe variables. Conseis an adequate represalso a function of sathe chi-square valuein lack of fit as estimbaseline model whe1990, 1992). In addithe lack of fit of the ment per degree of f& Bonett, 1980). Thgo out of this range .90 cutoff value are ge tested. The first model involved all variables except forhis variable was not included in the overall model becauseipants who provided that information was not sufficient inof parameters included in the model (Bentler, 1990). How-ond model involving self-determined motivation, emotionalatisfaction, and turnover intentions and behavior. Thiswed us to determine whether the model can predict actualhile respecting the parameter/participants ratio (Bentler,

    the different models was assessed through structural equa- the EQS program (Bentler, 1992). The raw data of theere used as database for the analysis. Inspection of univari- normality proved satisfactory. The different models weredardized coefficients obtained from the maximum likelihoodmation. A growing body of research indicates that ML per-ll when the data are multivariate normally distributed and the enough (e.g., Chou & Bentler, 1995), which was the, 1992) program provides different indexes to ascertain

    e used chi square, comparative fit index (CFI), and thenormed fit index (NNFI). The chi-square statistic is a func- between the model examined and a saturated model (with ag of all possible sources of variance and covariance amongquently, a nonsignificant chi square indicates that the modelentation of the sample data. Note, however, that chi square ismple size: the larger the number of participants, the highers. On the other hand, the CFI assesses the relative reductionated by the noncentral chi square of a target model versus are all of the observed variables are uncorrelated (Bentler,tion, the NNFI compares the lack of fit of a target model tobaseline model. Thus, NNFI estimates the relative improve-reedom of the target model over the baseline model (Bentlere CFI index varies between 0 and 1, whereas the NNFI can(i.e., > 1). Typically, models with a CFI and NNFI below theenerally rejected (see Bentler & Bonett, 1980).


    Turnover Intentions

    The correlation ctosis of the 22 observcoefficients of the invariables, as well asmodel, 2(200, N = 4cant, the other measumodel. The NNFI anfor 36% of the varian

    As predicted, thecompetence, as wellThe beta weights forrespectively. Resultstively related to emosatisfaction ( = .80)

    Figure 2. Results of thwork turnover. All coefresidual was constrainefollowing formula: vaFalck, & Carlson, 1995


    oefficients, means, standard deviations, skewness, and kur-ed variables are shown in Table 1. Figure 2 displays the path

    tegrated model, the coefficients associated with the observed the measurement errors of the first model; for the overall90) = 592.19, p < .001. Although the chi square was signifi-res of goodness of fit provided support for the hypothesizedd CFI were .93 and .94, respectively. The model accountedce in turnover intentions.

    results show the importance of feelings of relatedness and of as intrinsic job rewards as predictors of work motivation. these three motivational determinants were .22, .18, and .42, also reveal that self-determined work motivation was nega-tional exhaustion ( = -.37) but positively related to work. It was also found that emotional exhaustion was positively

    e structural equation modeling analysis of the motivated model officients were standardized and are significant at z > 1.96. The errord at lower boundY. The measurement error was fixed by using the

    r (1 - ), .99 (1 - .91) = .09 (Bollen, 1989; Wang, Fisher, Siegal,).

  • 12 RICHER ET AL.

    Table 1

    Correlation Matrix, Means, Standard Deviations, Skewness, and Kurtosis for the


    Task characteristics1. Signification2. Feedback3. Variety

    Feelings of compet4. Comp15. Comp26. Comp3

    Feelings of relatedn7. AcceptationX

    8. IntimacyWork motivation (I

    9. Index110. Index211. Index312. Index4

    Emotional exhausti13. EE114. EE215. EE316. EE4

    Work satisfaction (S17. Satif118. Satif219. Satif3

    Turnover intentions20. Int121. Int222. Int3


    X1 2 3 4 5 6 7 8 9

    .21 .27 .22

    ence (Comp).04 .27 .12 .07 .16 .09 .58 .08 .10 .07 .43 .54

    ess.15 .20 .13 .36 .27 .19 .08 .13 .04 .21 .16 .12 .61

    ndex).18 .19 .31 .37 .23 .20 .35 .17 .21 .26 .28 .36 .27 .23 .38 .25 .74.19 .14 .25 .30 .16 .13 .33 .14 .75.20 .21 .27 .36 .20 .17 .37 .20 .78

    on (EE).07 -.01 .09 -.18 -.12 -.07 -.26 -.09 -.26.08 .05 .12 -.16 -.14 -.07 -.12 -.08 -.19

    -.02 -.03 .02 -.22 -.13 -.09 -.24 -.19 -.37.05 -.04 .07 -.27 -.15 -.03 -.26 -.13 -.32

    atif).22 .22 .34 .31 .19 .13 .38 .25 .64.19 .32 .28 .35 .16 .12 .38 .26 .66.13 .17 .30 .30 .18 .16 .40 .18 .52

    (Int)-.05 -.18 -.16 -.14 .00 -.00 -.30 -.20 -.35-.04 -.18 -.18 -.14 -.04 .02 -.29 -.19 -.33-.01 -.13 -.09 -.08 .04 .07 -.27 -.14 -.276.25 5.32 5.76 5.71 5.93 5.91 4.71 3.40 9.001.11 1.40 1.29 1.05 0.83 1.07 0.94 1.14 5.38

    -2.24 -1.18 -1.42 -1.36 -0.63 -1.54 -0.62 -0.07 -0.795.93 1.08 1.89 3.13 0.45 3.76 0.77 -0.47 0.65


    Model Variables

    10 11 12

    .75 .81 .83

    -.21 -.26 -.24-.19 -.15 -.16-.36 -.34 -.34-.29 -.34 -.33

    .64 .61 .68

    .62 .62 .67

    .48 .48 .51

    -.36 -.38 -.42-.29 -.32 -.35-.22 -.28 -.297.58 8.31 8.51 15.43 5.26 5.12 1

    -0.11 -0.19 -0.33 0-0.84 -0.71 -0.60 013 14 15 16 17 18 19 20 21 22

    .62 .57 .61 .69 .62 .64

    -.21 -.10 -.27 -.22 -.26 -.15 -.35 -.30 .78 -.20 -.13 -.24 -.23 .60 .63

    .31 .24 .38 .37 -.46 -.54 -.34

    .28 .20 .31 .30 -.40 -.52 -.35 .81

    .27 .17 .28 .32 -.31 -.47 -.31 .72 .81

    .87 2.83 1.89 1.38 5.29 5.46 5.24 2.65 2.79 1.99

    .52 1.64 1.61 1.47 1.48 1.36 1.34 1.80 2.08 1.71

    .86 0.12 0.68 1.23 -1.23 -1.32 -1.10 1.01 0.89 1.72

    .02 -0.91 -0.31 0.94 0.97 1.34 0.94 -0.06 -0.64 1.80

  • 14 RICHER ET AL.

    related to turnover irelated to turnover in

    Turnover Behavior

    Supplementary aafter the first wave oparticipated in thisabridged model. In temotional exhaustiocurrent job, and inte1 year later; for the oand CFI were .96 anthe other measures osized model. The ressignificant. Moreoveto persist and actual

    The relationshipwas further examineon SDT (Deci & Ry

    Figure 3. Path diagrammodel of work turnovententions ( = .24), while work satisfaction was negativelytentions ( = -.48).

    nalyses were conducted in order to predict turnover 1 yearf data collection. Because the number of participants who

    phase of the study was smaller (n = 241), we tested anhis model, work motivation predicted work satisfaction andn, the latter two variables predicted intentions to leave thentions were hypothesized to predict actual turnover behaviorverall model, 2(86, N = 241) = 164.30, p < .001. The NNFId .97, respectively. Although the chi square was significant,f goodness of fit provided strong support for the hypothe-ults are presented in Figure 3. All hypothesized links werer, analyses reveal a positive relationship between intentionsturnover behavior ( = .28). between turnover intentions and actual turnover behaviord while taking into consideration the notion of choice. Basedan, 1985, 1991), it was hypothesized that the link between

    of the estimated structural model of the compressed motivationalr. All coefficients were standardized and are significant at z > 1.96.


    motivation and behavior would be especially strong for individuals who felt thatthey had the choice to stay or to quit their jobs. However, the turnover intentionsand behavior link should be much weaker in the absence of choice. Two groupswere formed based occupying the same median (4.6). Correturnover and actualresults reveal a low intentions and actualior. However, the regroup that felt they h

    This study testedjob rewards, feelingscolleagues positivelyter positively influenexhaustion. In turn, wto have negative andFinally, over time, tuturnover behavior. Rsupport for the hyponumber of theoretica

    On the Determinants

    A first implicatiotors of work motivatiof competence and owork motivation. Tmented with respectand self-determineVallerand & Reid, 19et al., 1989; Deci & tence play a causal ro

    Several authors causal role in motivaGoodenow, 1992; Rpresent study represeness toward colleagupositive ones feelinon the degree of choice participants felt they had towardjob or changing their occupation by splitting workers at thelation coefficients of the relationship between intentions to turnover were obtained for the two groups. As expected,and nonsignificant correlation (r = .06) between turnover behavior for the group that had little choice in their behav-sults reveal a moderately high correlation (r = .48) for thead the choice to leave or to stay.


    a motivation model of work turnover positing that intrinsic of job competence, and feelings of relatedness toward work influence self-determined work motivation and that the lat-ces work satisfaction, but negatively influences emotionalork satisfaction and emotional exhaustion are hypothesized positive influences, respectively, on turnover intentions.rnover intentions are expected to eventually translate intoesults from structural equation modeling analyses providethesized model. We believe that the present findings have al implications.

    of Work Motivation

    n of the present findings is that an important set of predic-on has been uncovered. Indeed, in the present study, feelingsf relatedness were important predictors of self-determined

    hese findings are in line with those that have been docu- to the existence of a link between feelings of competenced motivation in other contexts (e.g., Losier et al., 1993;84). These results provide empirical support for SDT (DeciRyan, 1985, 1991), which proposes that feelings of compe-le in self-determined work motivation.

    have postulated that feelings of relatedness should play ation (e.g., Connell & Wellborn, 1991; Deci & Ryan, 1991;yan, 1995; Ryan, Stiller, & Lynch, 1994). However, thents the first empirical demonstration that feelings of related-es are positively related to work motivation. Thus, the moregs of relatedness toward work colleagues, the more self-

  • 16 RICHER ET AL.

    determined ones work motivation. These results provide additional support forSDT (Deci et al., 1989; Deci & Ryan, 1985, 1991), which proposes that feelingsof relatedness play a causal role in self-determined work motivation.

    The present findjob rewards (Hackm(Deci & Ryan, 1985influence of intrinsicthe emphasis was plation (Kanfer, 1990). tor between intrinsictypes of job rewardstask identity, task sirelationships with tidentified regulationmight lead to both th

    Influence of Work Mo

    The motivationadirect influence on evariables are expecthypothesized to tranModels 1 and 2 (Figfindings corroboratemotivation and wordetermined the workKumar, 1982; Ilardi also shown that theprobability of experiSantinello, 1991). Thtion, and turnover inthat emotional exhauBritton, 1989; Singhturnover intentions (the present study is work satisfaction anlead to turnover inten

    Finally, based onship between turnovempirical evidence fintended to leave thetant to measure turnings also reveal a significant relationship between intrinsican & Oldham, 1974, 1976) and self-determined motivation, 1991). Interestingly, these findings put an emphasis on the job rewards on work motivation, whereas in past research,ced on the influence of intrinsic job rewards on job satisfac-

    The present results reveal that motivation can act as a media- job rewards and job satisfaction. Future research on the five proposed by Hackman and Oldham (1976)skill variety,gnificance, autonomy, and feedbackand their respectivehe different types of motivation (i.e., intrinsic motivation,, extrinsic motivation, and amotivation) is encouraged as iteoretical and applied advances.

    tivation on Turnover Intentions and Behavior

    l model proposed herein posits that work motivation has amotional exhaustion and work satisfaction. In turn, these twoed to influence turnover intentions. Finally, intentions areslate into actual turnover behavior. The combined results ofures 2 and 3) provide support for these propositions. These past research. Thus, the link between self-determined workk satisfaction has been well documented: the more self- motivation, the more satisfied the individual (Harigopal &et al., 1993; Keaveney & Nelson, 1993). Other research has more self-determined the work motivation, the lower theencing emotional exhaustion (Blais et al., 1993; Pedrabissi &e relationship between emotional exhaustion, work satisfac-tentions has also been tested in the past. It has been shownstion has a positive influence on turnover intentions (Firth & et al., 1994), while work satisfaction is negatively related toIrvine & Evans, 1995; Keaveney & Nelson, 1993). However,the first to show within the confines of the same study thatd emotional exhaustion originate from work motivation andtions. Azjen and Fishbeins (1977) model, we tested the relation-er intentions and actual turnover behavior. Results provideor Azjen and Fishbeins proposition in that the more peopleir jobs, the more they actually did. We feel that it is impor-over intentions because incorporating behavioral intentions


    in the motivational model might best reflect the actual process through whichpeople come to implement behavior. Bad working conditions do not always leadto actual turnover. However, they might lead to intentions that will eventuallygrow into behavior those of Vallerand determined motivatilead to actual behaviappear that using intdict behavior, especiroad.

    It should be notemore complete underachieved by taking irespect to staying vebetween intentions andid not have a choicedid have a choice. Tmight still decide to smoney, because theywords, because they dmentioned that to ththat choice is a crucbetween turnover insequence. In fact, it intentions to be tranmight have importantions for the work do

    Limitations and Futu

    Although the premodel of work turnoof the present studywell, the return rate wvalidity purposes togroup of participantthrough an objectivethe reduced numbercompressed model outcomes (i.e., satisfbehavior was tested.number of participa(Vallerand et al., 1997). These findings are in accord withet al. and Sarrazin et al. (2001) that also found that self-on leads to intentions of behavioral persistence that in turnor in education and sport settings, respectively. It would thusentions in the motivational sequence allows us to better pre-ally if it is measured several months or even years down the

    d, however, that in line with Deci and Ryan (1985, 1991), astanding of the turnover intentionsbehavior relationship wasnto consideration the employees perceptions of choice withrsus leaving their jobs. Results indicate that the relationshipd behavior was quite low (r = .06) when people felt that they, but it was much stronger (r = .48) when they felt that they

    hus, employees might have intentions to leave their jobs, buttay because no other jobs are available, because they need the have acquired seniority within the actual company; in otheront have any other choice (Mobley et al., 1979). It should be

    e best of our knowledge, these findings are the first to showial variable to consider when the goal is to study the link

    tentions and behavior within the purview of a motivationalappears that choice might play the role of a catalyst allowingslated effectively into behavior. This new function of choicet theoretical implications for SDT, as well as applied implica-main. Therefore, future research on this issue appears fruitful.

    re Research

    sent results provide support for the proposed motivationalver, certain limitations should be noted. First, all participants were alumni from the same school of administration. Asas rather low (24%). It would thus be important for external

    replicate the present findings with a more representatives. Second, future studies should examine turnover behavior measure, rather than a self-report measure. Third, because of of participants for the second wave of data collection, a

    assessing the link between motivation and the job-relatedaction and emotional exhaustion) including actual turnover It would be important to replicate this study with a greaternts in order to bring further support for the overall model,

  • 18 RICHER ET AL.

    including both motivational determinants and outcomes, as well as turnoverintentions and behavior.

    In sum, the present study provides support for the proposed motivationalmodel of turnover arelatedness and of cotion. In addition, thiwork consequences might influence turnand actual turnover wvational sequence dleads to a better undebehavior and paves tains to the fact thatincluded in the preseables related to turnothese variables (e.g.,been able to predict strategy is ill advisedent conceptual and model becomes emppredict outcomes. Howell as explain the nBecause we seek expables that are includhowever, that we nevpredict turnover; namables have been incturnover (e.g., IrvinSantinello, 1991) anIndeed, a lack of selence emotional exhavalidated a concepturetical background (Sliterature, we believeagainst competing mbelieve that this strat

    Abramson, L. Y., Selin humans: Critiq49-74.

    Jt work. These findings underscore the role of feelings ofmpetence as determinants of self-determined work motiva-s study has shown how motivation might influence certain(e.g., satisfaction and emotional exhaustion), which in turnover intentions. Finally, the link between turnover intentionsas supported. Overall, the present study showed how a moti-erived from elements of SDT and the hierarchical modelrstanding of the motivational processes involved in turnover

    the way for exciting future research.A final limitation per- several variables known to be related to turnover were notnt study. Cotton and Tuttle (1986) found over 20 such vari-ver. While it is probably true that by incorporating some of

    commitment, job involvement) in our model we would haveadditional variance in turnover, we nevertheless feel that this. Indeed, if we bring together a host of variables with differ-

    theoretical underpinnings, we create a situation where theirical rather than conceptual in nature. Empirical models canwever, conceptual models predict the outcome (turnover) as

    ature of the psychological processes leading to the outcome.lanation as well as prediction, it is thus crucial that the vari-ed in a model make sense conceptually. It should be noted,ertheless incorporated in the model two variables known toely, work satisfaction and emotional exhaustion. These vari-

    luded because past research does support their relation toe & Evans, 1995; Keaveney & Nelson, 1993; Pedrabissi &d because they make sense conceptually within the model.f-determined motivation at work can lead people to experi-ustion and a lack of satisfaction at work. Now that we haveal model that includes variables derived from a strong theo-DT and the hierarchical model) and from the organizational that future research should be geared at testing this modelodels, as suggested by Hom et al. (1993J). Eventually, we

    egy should lead to both theoretical and applied benefits.


    igman, M. E. P., & Teasdale, J. (1978). Learned helplessnessue and reformulation. Journal of Abnormal Psychology, 87,


    Ajzen, I. (1988). From intentions to actions. In I. Ajzen (Ed.), Attitudes, person-ality, and behavior (pp.112-145). Chicago, IL: Dorsey Press.

    Ajzen, I., & Fishbein, M. (1980). Understanding attitudes and predicting socialbehavior. Englew

    Bandura, A. (1991).Behavior and Hu

    Bentler, P. M. (1990cal Bulletin, 107,

    Bentler, P. M. (1992CA: BMDP Stati

    Bentler, P. M., & Bothe analysis of co

    Blais, M. R., Brire(1993). Linventavation Inventory]

    Blais, M. R., Lachanber). Lchelle dScale]. Poster prelaK recherche en

    Blais, M. R., Richer,dation de la verstion of the FrenPoster presented recherche en psy

    Blais, M. R., Sabourvational model ochology, 59, 1021

    Bollen, K. A. (1989)John Wiley & So

    Brayfield, A. H., & performance. Psy

    Carver, C. S., & SchNY: Springer-Ve

    Chen, X., Hui, C., &turnover: Concepof Applied Psych

    Chou, C. P., & Bentmodeling. In R.issues and applic

    Connell, J. P., & WelA motivational aSroufe (Eds.),




    Nood Cliffs, NJ: Prentice-Hall. Social cognitive theory of self-regulation. Organizationalman Decision Processes, 50, 248-287.). Comparative fit indexes in structural models. Psychologi- 238-246.). EQS: Structural equations program manual. Los Angeles,stical Software.nett, D. G. (1980). Significance tests and goodness of fit invariance structures. Psychological Bulletin, 88, 588-606., N. M., Lachance, L., Riddle, A. S., & Vallerand, R. J.ire des motivation au travail de Blais [The Blais Work Moti-. Revue Qubcoise de Psychologie, 14, 185-215.ce, L., Forget, J., Richer, S., & Dulude, D. M. (1991, Novem-e satisfaction globale au travail [The Work Satisfactionsented at the annual convention of the Socit qubcoise depsychologie, Trois-Rivires, Qubec, Canada. S., Lachance, L., & Dulude, D. M. (1991, November). Vali-ion franaise de linventaire de burnout de Maslach [Valida-ch-Canadian version of the Maslach Burnout Inventory].at the annual convention of the Socit qubcoise pour laL

    chologie, Trois-Rivires, Qubec,, S., Boucher, C., & Vallerand, R. J. (1990). Toward a moti-f couple happiness. Journal of Personality and Social Psy--1031.. Structural equations with latent variables. New York, NY:ns.Crockett, W. H. (1955M). Employee attitudes and employeechological Bulletin, 52, 396-424.eier, M. F. (1981). Attention and self-regulation. New York,rlag. Sego, D. J. (1998). The role of organizational citizenship intualization and preliminary tests of key hypotheses. Journalology, 83, 922-931.ler, P. M. (1995). Estimates and tests in structural equation H. Hoyle (Ed.), Structural equation modeling: Concepts,ations (pp. 37-55). N, CA: Sage.lborn, J. G. (1991). Competence, autonomy, and relatedness:nalysis of self-esteem processes. In M. R. Gunnar & L. A.The Minnesota Symposium on Child Psychology: Self-

  • 20 RICHER ET AL.

    processes in development (Vol. 22, pp. 43-77). Hillsdale, NJ: LawrenceErlbaum.

    Cotton, J. L., & Tuttle, J. M. (1986). Employee turnover: A meta-analysis andreview with impl55-70.

    Deci, E. L. (1980). TDeci, E. L., Connell

    organization. JouDeci, E. L., & Ryan,

    human behavior.Deci, E. L., & Ryan

    in personality. In(pp. 237-288). Li

    DeCottis, T. A., & Scedents and cons40, 445-470.

    Firth, H., & Brittonnursing staff. Jou

    Fortier, M. S., Valleschool performanPsychology, 20, 2

    Goodenow, C. (1992and the study of

    Granrose, C. S., & The relationship Psychology, 24, 8

    Grolnick, W. S., & experimental andand Social Psych

    Grolnick, W. S., & Rself-regulation anogy, 81, 143-154

    Hackman, J. R., & OSurvey. Journal o

    Hackman, J. R., & OTest of a theory250-279.

    Harackiewicz, J. M.,supervisor feedbSocial Psycholog

    Harigopal, K., & KuA prediction. Ma

    Oications for research. Academy of Management Review, 11,

    he psychology of self-determination. Washington, DC: Heath., J. P., & Ryan, R. M. (1989). Self-determination in a workrnal of Applied Psychology, 74, 580-590. R. M. (1985). Intrinsic motivation and self-determination in New York, NY: Plenum., R. M. (1991). A motivational approach to self: Integration R. Dienstbier (Ed.), Nebraska Symposium on MotivationO

    ncoln, NE: University of Nebraska Press.ummers, T. P. (1987). A path analysis of a model of the ante-equences of organizational commitment. Human Relations,

    , P. (1989). Burnout, absence and turnover amongst Britishrnal of Occupational Psychology, 62, 55-59.rand, R. J., & Guay, F. (1995). Academic motivation andce: Toward a structural model. Contemporary Educational57-274.). Strengthening the links between educational psychology

    social contexts. Educational Psychologist, 27, 177-196.Kaplan, E. (1994). Returning to work following childbirth:between intentions and behavior. Journal of Applied Social73-896.Ryan, R. M. (1987). Autonomy in childrens learning: An individual difference investigation. Journal of Personalityology, 52, 890-898.yan, R. M. (1989). Parent styles associated with childrensd competence in school. Journal of Educational Psychol-

    .ldham, G. R. (1974). Development of the Job Diagnostic

    f Applied Psychology, 60, 159-170.ldham, G. R. (1976). Motivation through the design of work:. Organizational Behavior and Human Performance, 16,

    & Larson, J. R. (1986). Managing motivation: The impact ofack on subordinate task interest. Journal of Personality andy, 47, 287-300.mar, R. R. (1982). Company satisfaction among managers:nagerial Psychology, 3, 69-75.


    Hellman, C. M. (1997). Job satisfaction and intent to leave. Journal of SocialPsychology, 137, 677-689.

    Hom, P. W., Caranikas-Walker, F., Prussia, G. E., & Griffeth, R. W. (1992). Ameta-analytical sJournal of Applie

    Ilardi, B. C., Leone, visor ratings of job satisfaction aPsychology, 23, 1

    Irvine, D. M., & Enurses: Integratin246-253.

    Kanfer, R. (1990). Mogy. In M. D. Duorganizational pssulting Psycholog

    Kasser, T., Davey, J.,discrepancies in Psychology, 37, 1

    Keaveney, S. M., & Intrinsic motivatiwithdrawal. Jour

    PKoestner, R., & Mction. In L. A. Pe(pp. 527-548). N

    Koestner, R., Losier,introjected formtheory. Journal o

    Lee, R., & Ashforth,Toward an integr

    Lee, R., & Graham, acteristics on themen and women

    Losier, G. F., Vallerade lchelle des(PCDV) [Constlife domains]. Sc

    Maslach, C., & JacksConsulting Psych

    McElroy, J. C., Mormance as predict81-98.

    Ptructural equations analysis of a model of employee turnover.d Psychology, 77, 890-909.D., Kasser, T., & Ryan, R. M. (1993). Employee and super-

    motivation: Main effects and discrepancies associated withnd adjustment in a factory setting. Journal of Applied Social789-1805.

    vans, M. G. (1995). Job satisfaction and turnover amongg research findings across studies. Nursing Research, 44,

    otivation theory and industrial and organizational psychol-nnette & L. M. Hough (Eds.), Handbook of industrial andychology (2nd ed., Vol. 1, pp. 75-170). Palo Alto, CA: Con-ists Press.

    & Ryan, R. M. (1992). Motivation and employeesupervisora psychiatric vocational rehabilitation setting. Rehabilitation75-187.Nelson, J. E. (1993). Coping with organizational role stress:onal orientation, perceived role benefits, and psychologicalnal of the Academy of Marketing Science, 21, 113-124.Clelland, D. C. (1990). Perspectives on competence motiva-rvin (Ed.), Handbook of personality: Theory and researchew York, NY: Guilford. G. F., Vallerand, R. J., & Carducci, D. (1996). Identified ands of political internalization: Extending self-determinationf Personality and Social Psychology, 70, 1025-1035. B. E. (1993). A further examination of managerial burnout:ated model. Journal of Organizational Behavior, 14, 3-20.W. K. (1986). The combined effects of job and family char- job satisfaction, job involvement, and intrinsic motivation ofworkers. Journal of Organizational Behavior, 12, 341-363.nd, R. J., & Blais, M. R. (1993). Construction et validation perceptions de comptence dans les domaines de vieruction and validation of the Perceived Competence scale inience et Comportement, 23, 1-16.on, S. E. (1986). Maslach Burnout Inventory. Palo Alto, CA:ologists Press.

    row, P. C., & Fenton, J. B. (1995). Absenteeism and perfor-ors of voluntary turnover. Journal of Managerial Issues, 7,

  • 22 RICHER ET AL.

    Miller, J., & Labovitz, S. (1973). Individual reactions to organizational conflict.The Sociological Quarterly, 14, 556-575.

    Mitchel, J. O. (1981). The effect of intentions, tenure, personal, and organiza-tional variables o24, 742-751.

    Mitra, A., Jenkins, Gtionship between879-889.

    Mobley, W., Griffethanalysis of the tu

    Netemeyer, R. G., Joand role ambiguPsychology, 75, 1

    ODriscoll, M. P., &and uncertainty aof Organizationa

    Orpen, C. (1995). Tbetween commuStudies, 40, 60-6

    Parasuraman, S. (19multivariate anal

    Pedrabissi, L., & Sanun, chantillon difessionnelQ. Revu

    Pelletier, L. G., Valle(1995). Loisirs etique des loisirs ences du Compor

    SPiotrkowski, C. S. (1(1986), Family lif(Eds.), In support

    Porter, L. W., & Steein employee turn

    Reeve, J., & Deci, Eintrinsic motivati

    Richer, S. F., & Valleordinates intrins135, 707-722.

    Richer, S. F., & Vallesentiment dappaFeelings of Relat48, 129-137.



    Sn managerial turnover. Academy of Management Journal,

    . D., & Gupta, N. (1992). A meta-analytic review of the rela- absence and turnover. Journal of Applied Psychology, 77,

    , R., Hand, H., & Meglino, B. (1979). Review and conceptualrnover process. Psychological Bulletin, 86, 493-522.hnston, M. W., & Burton, S. (1990). Analysis of role conflictity in a structural equations framework. Journal of Applied48-157. Beehr, T. A. (1994). Supervisor behaviors, role stressors,s predictors of personal outcomes for subordinates. Journall Behavior, 15, 141-155.he effect of organizational commitment on the relationshipnication quality and managerial attitudes. Psychological4.82). Predicting turnover intentions and turnover behavior: Aysis. Journal of Vocational Behavior, 21, 111-121.tinello, M. (1991). Motivations et stress professionnel chez

    nstituteurs: comparaison de deux indicateurs de malaise pro-e Europenne de Psychologie Applique, 41, 137-142.rand, R. J., Green-Demers, I., Blais, M. R., & Brire, N. M.

    t sant mentale: les relations entre la motivation pour la pra-et le bien-tre psychologiqueR. Revue Canadienne des Sci-tement, 27, 140-156.985). Research interest group presentation. Cited in Galinskye and corporate policies. In M. W. Yogman & T. B. Brazelton of families. Cambridge, MA: Harvard University, R. M. (1973). Organizational, work, and personal factorsover and absenteeism. Psychological Bulletin, 80, 151-176.. L. (1996). Elements of the competitive situation that affecton. Personality and Social Psychology Bulletin, 22, 24-33.rand, R. J. (1995). Supervisors interactional styles and sub-ic and extrinsic motivation. Journal of Social Psychology,

    rand, R. J. (1998). Construction et validation de lchelle durtenance sociale (SAS) [Construction and validation of theedness scale]. Revue Europenne de Psychologie Applique,


    Riordan, C. M., & Griffeth, R. W. (1995). The opportunity for friendship in theworkplace: An underexplored construct. Journal of Business and Psychology,10, 141-154.

    Rush, M. C., Schoel,the public sector:Behavior, 46, 17-

    Russ, F., & McNeillyturnover intentiomance. Journal o

    Ryan, R. M. (1995). Journal of Person

    Ryan, R. M., & Conntion: Examining and Social Psych

    Ryan, R. M., Deci, Ethe self: Their re& D. J. Cohen (E(Vol. 1, p. 618-65

    Ryan, R. M., Stiller, to teachers, pareself-esteem. Jour

    Sarrazin, P., Valleranvation and dropManuscript subm

    Seligman, M. E. P. (1San Francisco, C

    USingelis, T. M. (19self-construals. P

    Singh, J., Goolsby, Jconsequences of tives. Journal of

    Somers, M. J. (1995An examination Behavior, 16, 49-

    Spector, P. E. (198studies concernin39, 1005-1016.

    Stepina, L. P., Perre(1991). A compaand reward domSocial Behavior a


    U W. A., & Barnard, S. M. (1995). Psychological resiliency in Hardiness and pressure for change. Journal of Vocational39., K. M. (1995). LiksT among satisfaction, commitment, and

    ns: The moderating effect of experience, gender, and perfor-f Business Research, 34, 57-65.The integration of behavioral regulation within life domains.ality, 63, 397-428.ell, J. P. (1989). Perceived locus of causality and internaliza-reasons for acting in two domains. Journal of Personalityology, 57, 749-761.. L., & Grolnick, W. S. (1995). Autonomy, relatedness, and

    lation to development and psychopathology. In D. Cicchettids.), Developmental psychopathology: Theory and methods5). New York, NY: John Wiley & Sons.J. D., & Lynch, J. H. (1994). Representations of relationshipsnts, and friends as predictors of academic motivation andnal of Early Adolescence, 14, 226-249.d, R. J., Guillet, E., Pelletier, L. G., & Curry, F. (2001). Moti-out in female handballers: A 21-month prospective study.itted for publication.975). Helplessness: On depression, development, and death.A: Freeman.94). The measurement of independent and interdependentersonality and Social Psychology Bulletin, 20, 580-591.. R., & Rhoads, G. K. (1994). Behavioral and psychologicalboundary spanning burnout for customer service representa-Marketing Research, 31, 558-569.). Organizational commitment, turnover, and absenteeism:of direct and interaction effects. Journal of Organizational58.6). Perceived control by employees: A meta-analysis ofg autonomy and participation at work. Human Relations,

    w, P. L., Hassell, B. L., Harris, J. R., & Mayfield, C. R.rative test of the independent effects on interpersonal, task,ains on personal and organizational outcomes. Journal ofnd Personality, 6, 93-10.

  • 24 RICHER ET AL.

    Summers, T. P., DeNisiV, A. S., & DeCottis, T. A. (1989). Attitudinal and behav-ioral consequences of felt job stress and its antecedent factors: A field studyusing a new measure of felt stress. Journal of Social Behavior and Personal-ity, 4, 503-520.

    Tett, R. P., & Meyerturnover intentioings. Personnel P

    Thomas, M., & Hafeinvolvement andfull- and part-tim

    Vallerand, R. J. (199motivation. In M(Vol. 29, pp. 271

    Vallerand, R. J., & Bstyles as predicto60, 599-620.

    Vallerand, R. J., Forttence in a real-lifout. Journal of P

    Vallerand, R. J., PellVallires, E. F. (1sic, extrinsic, anMeasurement, 52

    Vallerand, R. J., PellVallires, E. F. (1tion in educationAcademic Motiv53, 159-172.

    Vallerand, R. J., & Rhierarchical modself-determinatioNY: University o

    Vallerand, R. J., & Rtence on intrinsicSport Psychology

    Vallerand, R. J., & Rtive verbal feedbJournal of Behav

    Vroom, V. H. (1964)Wang, J., Fisher, J. H

    ence of measure


    W, J. P. (1993). Job satisfaction, organizational commitment,n, and turnover: Path analyses based on meta-analytic find-sychology, 46, 259-293.r, J. C. (1995). The multiplicative interaction effects of job

    organizational commitment on the turnover intentions ofe employees. Journal of Vocational Behavior, 46, 310-331.7). Toward a hierarchical model of intrinsic and extrinsic. Zanna (Ed.), Advances in experimental social psychology-360). New York, NY: Academic Press.issonnette, R. (1992). Intrinsic, extrinsic, and amotivationalrs of behavior: A prospective study. Journal of Personality,

    ier, M. S., & Guay, F. (1997). Self-determination and persis-e setting: Toward a motivational model of high school drop-ersonality and Social Psychology, 72, 1161-1176.etier, L. G., Blais, M. R., Brire, N. M., Sencal, C. B., &989). The Academic Motivation scale: A measure of intrin-

    d amotivation in education. Educational and Psychological, 1003-1019.etier, L. G., Blais, M. R., Brire, N. M., Sencal, C. B., &993). On the assessment of intrinsic, extrinsic, and amotiva-: Evidence on the concurrent and construct validity of theation scale. Educational and Psychological Measurement,

    atelle, C. F. (in pressW). Intrinsic and extrinsic motivation: Ael. In E. L. Deci & R. M. Ryan (Eds.), The motivation andn of behavior: Theoretical and applied issues. Rochester,f Rochester Press.eid, G. (1984). On the causal effects of perceived compe-

    motivation: A test of cognitive evaluation theory. Journal of, 6, 94-102.eid, G. (1988). On the relative effects of positive and nega-ack on males and females intrinsic motivation. Canadianioral Science, 20, 239-250.. Work and motivation. New York, NY: John Wiley & Sons.., Siegal, H. A., Falck, R. S., & Carlson, R. G. (1995). Influ-ment errors on HIV risk behavior analysis: A case study


    examining condom use among drug users. Structural Equation Modeling, 2,319-334.

    Wright, T. A., & Cropanzano, R. (1998). Emotional exhaustion as a predictor ofjob performance 486-493.

    Zuckerman, M., Porimportance of selality and Social Pand voluntary turnover. Journal of Applied Psychology, 83,

    ac, J., Lathin, D., Smith, R., & Deci, E. L. (1978). On thef-determination for intrinsically motivated behavior. Person-sychology Bulletin, 4, 443-446.

  • 26 RICHER ET AL.


    A Denisi or DeNisi? See References and make the necessary corrections.B Is Vallerand &

    provide the dC Is Vallerand &

    provide the dD Should this b

    sary correctioin your text (ithere are five

    E What do you F What modelG Not in the ReH 1975 or 1955I You should p

    here. Please mJ Hom et al. (1

    or the correctK de la or poL de la or poM 1955 or 1975N Provide the cO The volumes

    the volume nuP Not cited in yQ Provide the ER Provide the ES Unless you ac

    the ReferencePiotrkowskimake the neccitation. In adpage numbers

    T What does LU Not cited in yV DeNisi or DeW Is this still in

    publication anX Shouldnt thisY Should this b

    Please correc Ratelle still in press? If it has been published already, ate of publication.

    Ratelle still in press? If it has been published already, ate of publication.e Vallerand, Fortier, & Guay, 1997? Please make the neces-ns, since you have not already cited Vallerand et al. (1997) .e., you should list all authors at the first text mention if or fewer authors).mean by It here? Please clarify your meaning.? Please clarify your meaning.ferences. Provide the missing reference.? See References and provide correct date.rovide numbers of participants in addition to percentages

    ake the necessary corrections throughout this paragraph.993) is not in the References. Provide the missing reference date.ur la? Please correct.ur la? Please correct.? See query H and provide correct date.ity of this series are individually titled and numbered. Provide mber and title.our text. Please cite or delete from the References.nglish translation for this article title in brackets.nglish translation for this article title in brackets.tually consulted Piotrkowski (1985), you should not list it in s. Instead, you should mention it in a text reference (e.g.,

    , 1985, as cited in Galinsky, 1986). In any case, you must essary corrections to this reference entry and to your text dition, you must provide Galinskys initial(s), as well as the for the chapter.iks mean? Please correct.our text. Please cite or delete from the References.nisi? See query A and make the necessary corrections. press? If it has been published already, provide the date of d the page numbers for the chapter. be Acceptance, rather than Acceptation? Please correct.

    e at the lower boundary, rather than at lower bound? t.

  • Annotations from 20P0117R2.pdf

    Page 1Annotation 1; Label: Copy editor; Date: 10/15/2002 4:16:20 PMChange "Universie" to "Universite"

    Annotation 2; Label: Copy editor; Date: 10/15/2002 4:17:05 PMDelete the comma between "Quebec" and "H3C 3P8"

    Page 8Annotation 1; Label: Copy editor; Date: 10/15/2002 4:20:05 PMDelete one of the brackets after the equation; it should appear as )]}

    A Motivational Model of Work TurnoverSylvie F. Richer, Cline Blanchard, and Robert J. VallerandThe purpose of this article was to propose and test a motivational model of work turnover. The model posits that feelings of relatedness toward work colleagues and feelings of com petence jointly and positively affect self-determined work mot...A Motivational Model of Work TurnoverFigure 1. Motivational model of work turnover.

    Self-Determined Work MotivationOn the Determinants of Self-Determined Work MotivationConsequences of Self-Determined Work MotivationThe Present StudyMethodParticipantsQuestionnaireProcedureData Analyses

    ResultsTurnover IntentionsTable 1Correlation Matrix, Means, Standard Deviations, Skewness, and Kurtosis for theModel VariablesFigure 2. Results of the structural equation modeling analysis of the motivated model of work turnover. All coefficients were standardized and are significant at z > 1.96. The error residual was constrained at lower boundY. The measurement er...

    Turnover BehaviorFigure 3. Path diagram of the estimated structural model of the compressed motivational model of work turnover. All coefficients were standardized and are significant at z > 1.96.

    DiscussionOn the Determinants of Work MotivationInfluence of Work Motivation on Turnover Intentions and BehaviorLimitations and Future Research

    ReferencesAUTHOR QUERIES