behavior in organizations as a function of employee's...

Psychological Bullet in1982, Vol. 91, No. 3, 482-497

Copyright 1982 by the American Psychological Association, Inc.0033-2909/82/9103-0482S00.75

Behavior in Organizations as a Function ofEmployee's Locus of Control

Paul E. SpectorFlorida Mental Health Institute, University of South Florida

Locus of control is an important variable for the explanation of human behaviorin organizations. The nature of the concept, its measurement, and general evi-dence for its validity are discussed. Several hypotheses are presented involvingthe locus of control in an organizational context, and supporting evidence fromapplied studies is reviewed. Specifically, it is suggested that locus of control isrelated to motivation, effort, performance, satisfaction, perception of the job,compliance with authority, and supervisory style. Furthermore, locus of controlmay moderate the relation between incentives and motivation and between sat-isfaction and turnover.

Little attention has been given to individ-ual personality in research on job motivationand satisfaction. For the most part, the ma-jor theories in organizational psychology as-sume that the same basic processes accountfor behavior across all individuals and thatsituational characteristics cause predictablebehavior across people.1 This article at-tempts to demonstrate the usefulness of per-sonality in explaining human behavior inorganizations and focuses on locus of controlas it relates to behavior in organizationalsettings.

The general theory of locus of controlarose from observation and research in clin-ical psychology. Both the measurement andtheory have been refined so that the conceptis heuristically useful. Over two dozen stud-ies of locus of control have been related spe-cifically to attitudinal, motivational, and be-havioral variables in organizational settings.One major task of this article is to integratethe general theory with the organizationalfindings.

The Concept of Locus of Control

People attribute the cause or control ofevents either to themselves or to the externalenvironment. Those who ascribe control ofevents to themselves are said to have an in-

Requests for reprints should be sent to Paul E. Spec-tor, Florida Mental Health Institute, 13301 North 30thStreet, Tampa, Florida 33612.

ternal locus of control and are referred toas internals. People who attribute control tooutside forces are said to have an externallocus of control and are termed externals.

Rotter (1966) and his colleagues devel-oped the concept of locus of control fromRotter's (1954) social learning theory. Ac-cording to Phares (1976), the concept wasdeveloped to explain the seeming tendencyof some individuals to ignore reinforcementcontingencies. Their failure to respond aspredicted to rewards and punishments wasattributed to a "generalized expectancy"that their own actions would not lead to at-tainment of rewards or avoidance of punish-ment. The tendency for internals to believethey can control events and externals to be-lieve they cannot leads to a number of pre-dictions about their behavior that are dis-cussed at length below.

Measurement of Locus of Control

The most widely used instrument to mea-sure locus of control is Rotter's (1966) In-ternal-External (I-E) scale, which consistsof 23 locus of control and six filler items ina forced-choice format. Scores are calcu-lated by summing the total number of ex-

1 Some notable exceptions are Hulin and Blood's(1968) suggestion that subscription to middle class val-ues moderates the job satisfaction-enlargement relationand Hackman and Lawler's (1971) finding that higherorder need strength moderates the relation between jobcharacteristics and satisfaction.

482

EMPLOYEE'S LOCUS OF CONTROL 483

ternally oriented responses for each pair.Thus, scores range from 0-23 with lowscores representing internality and highscores, externality.

Validity of the Concept ofLocus of Control

The first basic question to address con-cerns the basic validity of the locus of controlconcept itself, that is, whether internals per-ceive more often than externals that eventsare due to their own actions and whetherthey have more choices in situations. Roark(1978) reported that among the employeesshe surveyed, internals were more inclinedto attribute to their own'actions the obtain-ing of their present jobs. Hammer and Vardi(1981) found among manufacturing em-ployees that internals were more likely thanexternals to attribute past job changes totheir own initiative. Finally, Harvey, Barnes,Sperry, and Harris (1974) demonstrated ina laboratory study that internals tended toperceive more alternatives in a choice situ-ation than did externals.

Not only do internals perceive greatercontrol, but they may actually seek situa-tions in which control is possible. Kabanoffand O'Brien (1980) described leisure timeactivities of internals and externals alongfive dimensions, including skill utilization(the amount of skill necessary for success)and influence (the amount of personal con-trol involved). They found a small but sta-tistically significant tendency for internalsto engage in leisure activities that requiredgreater skill and allowed more personal con-trol. Julian and Katz (1968) conducted alaboratory study of competitive game be-havior in which subjects were given thechoice of relying on either their own skill oron a more competent opponent. Externalswere more likely to rely on the opponent,and internals preferred to rely on themselves.Kahle (1980) gave laboratory subjects thechoice of a task requiring either luck or skill,As might be expected, externals were morelikely to choose luck, whereas internalstended to prefer skill.

It is interesting that there seems to be aninteractive relation between locus of controland experience. That is, locus of control may

affect behavior, and the consequences of be-havior may in turn affect locus of control.In a laboratory study, Krolick (1979) foundthat internals tended to shift their I-E scoresin an external direction following an expe-rience of failure, but externals did not shifttoward internality following success. Be-cause internals seem to be more sensitive toinformation that is relevant to them (Phares,1976), it would follow that they would bemore sensitive to reward contingencies in alaboratory task situation, especially if thetask is ego involving.

An interesting longitudinal field study byAnderson (1977), however, demonstratedthat shifts in locus of control can occur forexternals as well as for internals. Andersonadministered Rotter's I-E scale to victimsof Hurricane Agnes both 8 months and 42months after the disaster. Each victim wasthe owner of a small business that was ex-tensively damaged by floods. Anderson alsoobtained a measure of business performancefrom a local credit agency both before and42 months after the hurricane. The 102 sub-jects of the study were divided into fourgroups based on whether they were internalor external at the time of the initial 8-monthassessment. The results showed a shift to-ward greater internality for internals whoseperformance improved and a shift towardgreater externality for externals whose per-formance deteriorated. The improved exter-nals did not shift toward internality, and thepoorer internals did not become external.

Although both Krolick (1979) and An-derson (1977) showed locus of control shiftsas a function of experience, their results aresomewhat conflicting. Krolick found a shiftonly for internals in an external direction;Anderson found an internal shift, but no ex-ternal shift, for internals and an externalshift for externals. Andrisani and Nestel(1976) provided further evidence for an in-ternality shift in their longitudinal study ofadult male career mobility. They found thatcareer success led to greater internality fortheir sample, although they did not look atinitial differences in locus of control in theway that Anderson did. As a whole, thesethree studies suggest that locus of controlmay be sensitive to experience, althoughthey fail to define variables that lead to

484 PAUL E. SPECTOR

shifts. Internals, however, may engage inactivities that require greater skill and thatallow more personal control (Julian & Katz,1968; Kahle, 1980), thereby increasing theprobability that their experiences will beconsistent with an internal view of the world.

Phares (1976) summarized findings con-cerning differential behavior by internalsand externals. Specifically, he noted that incontrast to externals, internals exert greaterefforts to control their environment, exhibitbetter learning, seek new information moreactively when that information has personalrelevance, use information better, and seemmore concerned with information ratherthan with social demands of situations.

Several studies linking locus of control tolearning and problem solving have demon-strated superior performance by internals(e.g., DuCette & Wolk, 1973; Ude & Vog-ler, 1969; Wolk & DuCette, 1974). Phares(1968) found that internals were superior toexternals in the use of memorized informa-tion for problem solving even though therewere no differences in acquisition. Appar-ently, internals made better use of infor-mation in a complex problem-solving situa-tion.

Because internals believe in and seek per-sonal control, they should exhibit less con-formity than do externals. Crowne and Liv-erant (1963) found more conformity forexternals in an Asch conformity situation,and Hjelle and Clouser (1970) found thatinternals exhibit less attitude change afterexposure to a persuasive message. Biondoand MacDonald (1971) showed that inter-nals are not only resistant to influence butmay demonstrate psychological reactance(Brehm, 1966) and shift their attitudes inan opposite direction to the influence at-tempt. Internals, however, are not totallyunaffected by social influence. Phares (1976)argued that internals may be even more sub-ject to social influence of an informationalvariety than are externals. He cited as anexample James, Woodruff, and Werner's(1965) finding that individuals who quitsmoking following the Surgeon General'sReport on its ill effects were more internalthan nonquitters.

Locus of control has been shown to be

related to other "paper-and-pencil" mea-sures, raising questions about the viabilityof the concept, at least as measured by Rot-ter's scale. Of particular interest to this dis-cussion are the reported findings concerninganxiety, social desirability, and achievementmotivation.

Locus of control is negatively related toanxiety; that is, externals tend to be moreanxious than internals. Joe (1971) and morerecently Archer (1979a) found the existingstudies quite consistent in demonstrating thisrelation. Ray and Katahn (1968), in a factoranalysis of the items of the I-E Scale, Man-ifest Anxiety Scale, and Test Anxiety Scale,found no factors that crossed scales; theytherefore concluded that anxiety was not ahidden factor in locus of control. Archer,Joe, and Ray and Katahn concluded thatlocus of control and anxiety were distinct butrelated concepts. Organ (1976) reached thesame conclusion but pointed out that "thereremains the problem of specifying the natureof the causal model incorporating these vari-ables" (p. 1097). Archer argued for a com-plex relation in which these two variables are"potentially interactive and multideter-mined phenomena" (p. 619).

The interrelatedness of locus of controland anxiety complicates the interpretationof research findings. For example, anxietymoderates the relation between task com-plexity and performance (Sarason, 1972)—the performance of high-anxiety individualsis impeded on complex tasks and learningtasks but not on simple tasks. It may wellbe that anxiety rather than locus of controlexplains why internals seem to learn betterthan externals. Obviously, additional re-search is needed to ascertain the joint roleof anxiety and locus of control in task per-formance.

Future researchers might do well to in-clude a measure of anxiety with the I-EScale in an attempt to uncover both theirjoint and their interactive effects. A step inthat direction has been taken by Archer(1979b), who split internals and externalsby high and low trait anxiety. He found acomplex interaction involving anxiety, locusof control, and situational characteristics. Itmay well be that in many circumstances lo-


cus of control has full meaning only if theeffects of anxiety are simultaneously consid-ered.

The relation between locus of control andsocial desirability appears to be equivocal.Rotter (1966) claimed that the I-E Scalewas not subject to social desirability bias,but Joe (1971) cited several studies that havefound correlations between the I-E Scaleand both Marlowe and Crowne's and Ed-ward's social desirability scales. At the pres-ent time, it is unclear why some studies havefound the relation whereas others have not.Given the equivocal results relating locus ofcontrol and social desirability, it is difficultto assess the possible confounding effects inthe studies to be reviewed here. It appears,however, that most studies have failed to finda relation between these variables. Further-more, one of the most well-established or-ganizational correlates of locus of control-job satisfaction—has been found to be un-correlated with social desirability (Lewis,Note 1; Smith, Note 2).

Finally, Yukl and Latham (1978) lookedat the relation among goal setting, locus ofcontrol, and achievement motivation. Theyfound that internals set harder goals thanexternals do, but in their sample of factoryworkers I-E was correlated with need forachievement (r = -.55), causing Yukl andLatham to question whether locus of controlor need for achievement was the crucial vari-able. Hartley (1976), however, in measuringI-E, achievement motivation, and job sat-isfaction, found that the combination of bothpersonality variables was related to satisfac-tion, although achievement motivation alonewas not. These results suggest that locus ofcontrol and achievement motivation are infact distinct concepts that may in some casesbe related to the same variables.

Locus of Control in anOrganizational Context

The basic distinguishing characteristicbetween internals and externals, belief inpersonal control, should have direct andpowerful effects on organizations in severalways. First, because internals tend to believethat they can control the work setting through

their behavior, they should attempt to exertmore control than would externals, providedthat control is perceived to lead to desiredoutcomes or rewards. If a situation cannotprovide desired outcomes, the internal shouldnot differ from the external in attempts atcontrol. For some individuals, however, con-trol itself might be rewarding, leading someinternals to attempt control for its own sake.

Giles (1977) provided interesting datafrom an organizational study that supportsthe theoretical role of locus of control in in-dividual action. Female factory workerswere administered a questionnaire includinga shortened version of the I-E Scale and ameasure of satisfaction with higher orderneeds. They were then asked to volunteer fora job enrichment program. Although locusof control did not predict who would vol-unteer, it did moderate the relation betweenneed satisfaction and volunteering. That is,internals with low satisfaction on the currentjob were more likely to volunteer (r = —.44)than were externals (r = —.20). Thus, al-though internals were no more likely thanexternals to volunteer, they were more likelyto take action when they were dissatisfiedwith their current situation.

The attempt of internals to control thework setting might be manifested in manyways. The internal would probably attemptcontrol in the following areas: work flow,task accomplishment, operating procedures,work assignments, relationships with super-visors and subordinates, working conditions,goal setting, work scheduling, and organi-zational policy. The factors on which controlattempts focused would be determined bythe potential rewards each carried and bythe constraints within the organizational set-ting.

Internals, as previously mentioned, per-form better in learning and problem-solvingsituations, apparently because of their betteruse of information (Phares, 1976). It wouldcertainly be expected that internals wouldexert more effort toward collecting relevantinformation in situations where they attemptcontrol. This would lead one to predict betterperformance by internals in training and inperforming tasks that necessitate the use ofinformation. Internals, because of their gen-

486 PAUL E, SPECTOR

eralized expectancies of environmental con-trol, should be easier to motivate. Thus, oneshould find internals more responsive thanexternals if the appropriate performance-re-ward contingencies can be presented.

As discussed above, externals are moreconforming and compliant than internals.Internals look to themselves for direction;externals look to others. Thus, externalsmake more compliant followers or subordi-nates than do internals, who are likely to beindependent and resist control by superiorsand other individuals. This compliance byexternals may seem reminiscent of the sub-missiveness of authoritarians (Adorno, Fren-kel-Brunswik, Levinson, & Sanford, 1950),and in fact, Lefcourt (1966) cited two theses(Holden, 1958; Simmons, 1959) that reporta moderately high correlation (r = .51) be-tween locus of control and the California FScale. One further piece of data suggestingsimilarities between authoritarians and ex-ternals is provided by Goodstadt and Hjelle's(1973) finding that externals tend to use acoercive leadership style in dealing with sub-ordinates (see section on leadership).

Externals, because of their greater com-pliance, would probably be easier to super-vise as they would be more likely to followdirections. Externals, however, might becompliant with the social demands of co-workers as well as with the legitimate au-thority of supervisors, and these co-workerdemands might at times conflict with thoseof management. Thus, externals might com-ply with one or the other depending on theirrelative strength of influence.

The nature of a job within the context oforganizational factors and demands woulddetermine whether an internal or externalwould be best suited. If a job requires com-plex information processing and frequentcomplex learning, internals would be ex-pected to perform better; for simple tasks,however, the performance differential woulddisappear. When tasks or organizational de-mands require initiative and independenceof action, the internal would be more suit-able; when the requirement is for compli-ance, however, the external would be moreappropriate. Finally, for jobs requiring highmotivation, internals would be more likelyto believe that their efforts will lead to re-

wards, especially when they actually do, andthus internals would tend to exhibit highermotivation. Therefore, it would seem thatinternals are best suited for highly technicalor skilled jobs, professional jobs, and man-agerial or supervisory jobs. Externals wouldbe more suited to factory line jobs, unskilledlabor jobs, clerical jobs, and jobs of a routinenature.

Unfortunately, the frequent organiza-tional demand for both compliance and com-plex task performance leads to the obviousconflict that the two characteristics of com-pliance and complex task skill may be an-tithetical. With the increasing complexity ofjobs because of automation, organizationsmight well have to sacrifice compliance forskill. This problem may be especially acutein the military, where the internal's skill andthe external's compliance are necessary formany complex jobs created by modern, so-phisticated equipment. Future organizationsmust by necessity find ways of managingpeople in complex jobs that allow them per-sonal control while achieving organizationalobjectives.

Relations Between Locus of Control andOrganizational Variables

Before proceeding further it should benoted that the discussion in the previous sec-tion was primarily theoretical, although dataexist to validate many of the hypotheses. Thenext major section of the article reviews theresearch on the relation of locus of controlwith several major behavioral variables stud-ied in organizations. For the most part, thesefindings support the theoretical notions justpresented, although they are limited in anumber of ways.

First, almost all of the studies presentedused Rotter's I-E Scale. There was littleattempt to establish convergent validitythrough the use of alternative measures. Sec-ond, most of the studies were cross-sectionaland correlational. This makes it quite dif-ficult to draw strong causal inferences fromthe data. Furthermore, many of the relationsreported as statistically significant were basedon large samples and small correlation coef-ficients. Third, most of the studies relied en-tirely on paper-and-pencil, self-report mea-


sures. The problem of method variancelimits the confidence with which many con-clusions can be stated. Fourth, as mentionedabove, locus of control seems to be relatedto other personality variables, complicatinginterpretation of results. In many investi-gations it is impossible to know whether re-sults can be better explained by reference tovariables that were not included, such asanxiety or social desirability.

Motivation

It might be expected that internals woulddisplay greater job motivation than wouldexternals because they perceive themselvesto have greater control over the environment,It is not that externals are less oriented to-ward valued rewards or personal goals butrather that internals will exert greater effortstoward acquiring rewards or achieving goalsbecause they are more likely to believe theirefforts will be successful. In the specific jobor organizational setting, the internal willexhibit more task-oriented and goal-orientedbehavior, and for that reason will exhibitmore job motivation, although underlyingpersonal motivations may be the same. Jobmotivation that is operationalized in termsof effort and task orientation will appear tobe higher in internals.

Although internals will tend to exhibitgreater job motivation than externals do, jobsettings in which rewards do not follow per-formance will not long show the internal-external differences. As discussed above, in-ternals are sensitive to reinforcement con-tingencies, and when effort on the job doesnot lead to rewards, internals may adopt amore external perspective.

There is research that relates to generaljob motivation, although often in an indirectmanner. Organ and Greene (1974a) studiedthe relation between locus of control and ascale they developed to measure perceivedpurposefulness of behavior on the job. Thispurposefulness measure consisted of 10 itemsconcerned with goal setting, task orientation,meaningfulness of work, and task-relateduse of time. As such, purposefulness is quitedirectly related to motivation in that highscores reflect high job motivation. As ex-pected, locus of control was negatively cor-

related with this scale (r = -.43) in that in-ternals saw their jobs as more purposeful.

Lied and Pritchard (1976) collected dataon both the I-E scale and Mirels and Gar-rett's (1971) Protestant Ethic (PE) Scale.The PE scale is designed to measure howclosely individuals subscribe to a cluster ofvalues reflected by the Protestant Ethic (i.e.,hard work is a virtue and leads ultimatelyto rewards). In validating the scale, Merrensand Garrett (Note 3) found that PE scoreswere positively related to effort (time spentworking) on a repetitive task. Lied and Prit-chard found a correlation (r = -.41) be-tween the I-E and PE scales, suggesting thatinternals are more motivated to work thanexternals are.

Finally, Reitz and Jewell (1979) con-ducted a survey of over 3,000 workers in sixcountries. Their results indicated that locusof control was significantly related to jobinvolvement (r = -.2—.35), with internalsshowing more involvement, again indicatinggreater motivation.

Expectancy Theory

Most of the job motivation studies involv-ing locus of control have been attempts tovalidate expectancy theory hypotheses. Themost popular expectancy theory applicationto organizations has been developed byVroom (1964). This theory proposes twotypes of expectancies, namely, that effortwill lead to good job performance and thatgood performance will lead to rewards. Thefirst is actually the belief in personal effec-tiveness; that is, the individual can performwell if he or she makes the effort. It is inmany ways similar to self-esteem, at leastin terms of self-perceived ability on the job.The second is the belief that good perfor-mance will be rewarded; that is, good per-formers get rewarded. This is similar to abelief in justice in the work world that ismuch like the concept of equity—the personwho provides more inputs (good perfor-mance) receives more outcomes (rewards).Basically, if an individual holds both expec-tancies strongly, he or she will have high jobmotivation and will (within limits of abilityand organizational constraints) perform well.

Theoretically, internals should hold higher

488 PAUL E. SPECTOR

expectancies of both varieties than externalswould. Internals are more likely to believethat their efforts will result in good perfor-mance, and they exhibit stronger beliefs intheir own competence. Therefore, they shouldexhibit greater self-esteem, a hypothesis sup-ported by Lied and Pritchard (1976). Inter-nals should also hold greater expectanciesthat good performance leads to rewards andshould tend to perceive the job situation asmore equitable than externals do, a hypoth-esis for which no data exist. There are sixstudies of expectancy theory that include thelocus of control variable. Before summariz-ing their findings, which are supportive, itshould be noted that methodological prob-lems in operationalizing the necessary con-structs in the theory have prevented thecompletion of its crucial testing. Thus,hypothesized relations between expectanciesand performance have been small and some-times inconsistent (Campbell, 1976).

Szilagyi and Sims (1975) attempted totest the hypothesis that there would be arelation between locus of control and bothtypes of expectancies. Personnel of a uni-versity medical center served as subjects andwere classified into one of five job categoriesor levels. The results showed modest but con-sistent correlations between I-E scores andperformance-reward expectancies across alllevels (r = —.2—.39). The correlationswith effort-performance expectancies weresmaller and reached significance for onlythree of the five levels (r = -.02-—.25). Liedand Pritchard (1976) also studied the rela-tion between locus of control and both ex-pectancies. This time the correlation be-tween I-E and effort-performance washigher than between I-E and performance-reward (r = -.40 and -.20, respectively).

Broedling (1975) reported the relationbetween I-E and Vroom's entire expectancytheory composite score and between I-E andthe various components. The correlation ofI-E with effort-performance was r = —.28,that of I-E with the product of performancereward and valence (perceived value of re-wards) was r = —.39, and that of I-E withthe entire composite was r = -.38. Mitchell,Smyser, and Weed (1975) found that inter-nals held greater effort-performance andperformance-reward expectancies than did

externals, and Evans (1974) found that mo-tivation (product of effort-performanceand performance-reward expectancies) washigher for internals than for externals. Fi-nally, Kimmons and Greenhaus (1976) re-ported that internals showed higher meanperformance-reward contingency scores thandid externals.

Thus, these results are consistent in dem-onstrating the hypothesized relation betweenlocus of control and expectancy. That is, in-ternals hold higher expectancies that effortwill lead to performance and that perfor-mance will lead to rewards. Lawler (1971,p. 177) argued that this tendency makes onlyinternals suitable for pay-incentive systems.That is, internals develop expectancies thatwill lead them to exert greater job effort formonetary or other rewards. Externals do notdevelop these expectancies and thereforeseem oblivious or insensitive to pay incen-tives. Lawler suggested using locus of controlto select internals for jobs with incentive sys-tems.

Job Performance

For at least two reasons one would predictthat internals would perform better on thejob than externals. First, they hold greaterexpectancies that effort will lead to goodperformance and good performance to re-ward. Thus, they exert greater effort in sit-uations where rewards are tied to perfor-mance, and ultimately, greater effort shouldlead to better performance across individu-als. In fact, Lawler (1968) provided evidencethat expectancies of good performance lead-ing to rewards is a causal factor in high jobperformance. Using cross-lagged and dy-namic correlations, Lawler found that self-reported expectancies were related to bothpeer and supervisor ratings of performancein a manner that was consistent with thenotion that expectancies affect performance.Second, as discussed previously, internalsseek more relevant information and performbetter than externals in complex task situ-ations. Again, this should lead to better per-formance by internals for tasks involvingcomplex information and learning, assumingthe internals are motivated to perform.


Several studies support the notion that in-ternals exert greater effort on the job andperform better. Some have investigated per-sonal career effectiveness across several jobs,and others were concerned with more im-mediate job performance.

Career effectiveness is measured over timeby salary increases and promotions. Al-though its relation to performance on spe-cific jobs may not be strong, in a global sense,career effectiveness reflects job performance,and hence one would expect internals to bemore successful in their careers because theyperform better.

Three studies support this contention.Heisler (1974) collected data on 196 gov-ernment employees of all levels. He com-puted an index of effectiveness based on fivevariables: number of promotions, salary in-creases, awards received, current salary, andgrade differential (current job grade levelminus entry job grade level). The correlationbetween this effectiveness index and the I-E scale was found to be modest but signif-icant (r = —.25). In addition, Heisler hy-pothesized that specific beliefs in the skillversus luck nature of the job would moderatethe relation between I-E and effectiveness.This hypothesis was upheld in that employ-ees who believed agency rewards were con-trolled by luck demonstrated a lower I-Eeffectiveness correlation than did employeeswho believed rewards were related to skill(r = -.01 and -.34, respectively).

Valecha (1972) reported the results of a5-year longitudinal study conducted on anational sample of 4,330 men. His resultsshowed that for white men, internals madebetter job progress (increase in job level)than did externals. Finally, Andrisani andNestel (1976) presented results of a 3-yearlongitudinal study of a national sample ofmen. Although the relations were rathersmall, their data showed that scores on I-Ewere related to success at work, as indicatedby occupational attainment and earnings.

Four studies have also investigated therelation between locus of control and im-mediate job effort and performance. Unfor-tunately, the measures of effort and perfor-mance were self- or supervisor ratings, andthe strength of relations found were rathermodest. The results, however, support the

contention that internals exert greater effortand perform better than externals.

Majumder, MacDonald, and Greever(1977) studied the relations among locus ofcontrol and several organizational variablesincluding supervisors' rating of performance.Their sample was composed of 90 rehabili-tation counselors working for a state voca-tional rehabilitation program. Locus of con-trol was correlated with performance (/• =.40), with internals receiving higher ratings.

Broedling (1975) collected I-E scores andperformance ratings for 207 naval personnel.Ratings were made of both effort and per-formance by supervisors, peers, and the sub-jects themselves. Although correlations be-tween I-E and performance were quitesmall, they supported the present hypothesisin that internals tended to score higher onboth effort and performance. Lied and Prit-chard (1976) investigated locus of controland effort among 146 Air Force trainees.They found significant correlations for bothself- and trainer ratings of effort (r = —.39and -.30, respectively). Finally, Hersch andScheibe (1967) studied locus of control andeffectiveness of students working for theConnecticut Service Corps in state mentalhospitals. For 2 of 3 years for which datawere available, supervisors' ratings of per-formance were significantly correlated withI-E (r = -.2 and -.37).

The existing evidence suggests that inter-nals do perform better than externals. In-ternals seem to exhibit greater personal car-eer effectiveness, exert greater effort, andperform better on the job. One should keepin mind, however, that internals will onlydisplay better performance if they perceivethat effort will lead to valued rewards. Inmany situations, internals may hold higherperformance-reward expectancies but notvalue the rewards. Furthermore, if there areno rewards for performance (either imme-diate or long-term career), internals mightnot differ from externals in their perfor-mance-reward expectancies. In addition, theadvantage that internals have over externalsin information seeking and utilization is onlyan advantage in situations involving complexinformation, so simple situations may yieldno internal-external differences. Futurestudies that attempt to confirm the relation

490 PAUL E. SPECTOR

between locus of control and performanceshould take into account the moderating in-fluence of situation complexity and real per-formance-reward contingencies as well asthe contribution of trait anxiety.

Job Satisfaction

Internals should demonstrate greater jobsatisfaction than externals do for at leastfour reasons. First, because internals tendto take action more frequently than externalsdo, the dissatisfied internal is more likely toquit a dissatisfying job. Thus, there wouldbe fewer dissatisfied internals than externals.Second, internals may perform better andreceive the benefits of that performance. Insituations where rewards follow perfor-mance, internals are likely to be more sat-isfied. Third, internals tend to advance morequickly and receive more raises than do ex-ternals. More frequent promotions and sal-ary increases should be expected to lead togreater satisfaction. Organizational level hasbeen shown to be positively correlated withsatisfaction, although direction of causalityhas not been established (Porter & Lawler,1965). Finally, cognitive consistency theorywould predict that individuals who have per-ceived personal control to leave the situationand who choose to stay will tend to reeval-uate the situation favorably to retain con-sistency between their attitudes and behavior(Salancik & Pfeffer, 1978). Internals whoperceive greater control and ability to leaveare more likely to leave in dissatisfying sit-uations. If they perceive the opportunity toleave but do not, they will be under someinternal cognitive pressure to evaluate thejob situation as favorable, thus justifyingtheir behavior. Externals who perceive nooptions are under only external constraintsto remain on the job and feel little pressureto change their job attitudes in a positivedirection.

The research supports the locus of con-trol-satisfaction hypothesis. Internals havebeen found to be more satisfied generallythan externals, although there have beensome interesting interactions involving sat-isfaction with supervision (see section onleadership). One exception is Dailey's (1978)finding that internals were less satisfied withtheir co-workers.

Three studies are reviewed comparing sat-isfaction of internals and externals holdingmanagerial, professional, and high-level jobs.Gemmill and Heisler (1972) collected locusof control and job satisfaction data as partof a larger study of 133 production man-agers. They found a significant correlationof —.27 between these variables. This findingis somewhat confounded by the fact that I-E was correlated (r = —.26) with job level.Organ and Greene (1974b) found a corre-lation of —.36 between I-E and satisfactionfor senior scientists and engineers. Finally,Satmoko (1973) studied satisfaction withMaslow's categories of needs among high-level government employees in Indonesia.Again, internals were found to be more sat-isfied than externals.

Three studies sampled nonsupervisory em-ployees and found comparable results. Mu-noz (1973) found that among New YorkCity policemen, internals were more job sat-isfied than externals. Lester and Genz (1978)found similar results with their two policesamples. Singh (1978) also reported greatersatisfaction among internal nurses.

Two studies mixed employees of differentjob levels and found essentially the same re-sults. Andrisani and Nestel (1976) in theirnational survey found a small, significantcorrelation between I-E and satisfaction(r = -.19). Mitchell et al. (1975) found in-ternals to be more satisfied than externals,although they found supervisors to be moreinternal than nonsupervisors.

Although the findings are quite consistentfor overall satisfaction, the findings for spe-cific job aspect satisfaction were somewhatdifferent. Dailey (1978) reported that for asample of scientists and engineers from 15organizations, internals were less satisfiedwith co-workers than were externals. He ex-plained these results in terms of the greatersocial orientation of externals. Mitchell etal. (1975) and Runyon (1973) investigatedsatisfaction with supervision and found it tobe moderated by leadership style (see nextsection on leadership).

Leadership

There are two sides to the leadership pro-cess that have been studied—the subordi-nate's reaction to his or her supervisor's be-


havior or style and the behavior of thesupervisors themselves. It would be expectedthat locus of control affects both the super-visor's behavior and the subordinate's reac-tion to it.

Runyon (1973) studied the moderatingeffect of locus of control on the relation be-tween supervisory style and satisfaction withsupervision. He administered to 110 hourlymanufacturing employees questionnairescontaining the I-E scale, a single-item mea-sure of satisfaction with supervision, and ameasure of supervisory style of the subject'ssupervisor. As expected, internals were moresatisfied with supervision than were exter-nals under a participative style, and theywere more satisfied with a participative thanwith a directive style. Externals were moresatisfied than internals under a directivestyle, and they were more satisfied with adirective than with a participative style.Mitchell et al. (1975) were partially able toreplicate these findings. In their sample of900 public utility employees of varying lev-els, internals were more satisfied than ex-ternals with supervision regardless of style,although internals were more satisfied withparticipation and externals with direction.

Abdel-Halim (1981) administered theLeader Behavior Description Questionnaire(Stogdill, 1963), the I-E scale, and a mea-sure of intrinsic job satisfaction to lower andmiddle managers. Internals' satisfaction wasunrelated to their supervisors' consideration,but externals reported less satisfaction withlow-consideration than with high-consider-ation supervisors. Evans (1974), however,studied a group of young managers andfound a small but significant relation be-tween I-E and the Leader Behavior Descrip-tion Questionnaire scales of initiation (/• =-.26) and consideration (r = -.24). Appar-ently, internals and externals either perceivesupervisors somewhat differently, or super-visors tend to treat their internal and exter-nal subordinates differently. Thus, it may bedifficult to draw firm conclusions from stud-ies relying on perceptions of supervisory be-havior by subordinates.

Cravens and Worchel (1977) conducteda laboratory simulation of a repetitive job.Subjects worked on a manual task under asupervisor who used either a coercive or anoncoercive style to increase productivity. It

was found that internals complied less withthe coercive supervisor than did externals.There were no differences with the noncoer-cive supervisor,

These results suggest that the appropriatesupervisory style may differ depending onthe subordinate's locus of control. It is rea-sonably clear that the two types of individ-uals prefer different styles and may reactdifferently to them. This fits consistentlywith the aforementioned findings that locusof control correlates with authoritarianism(Holden, 1958; Simmons, 1959).

Two laboratory studies provide data con-cerning the different behavior of internal andexternal leaders. Anderson and Schneier(1978) extensively studied the behavior ofcollege students who had been assigned classprojects in groups. Among their findingswere the following: (a) Internals were morelikely to emerge as group leaders, (b) inter-nal leaders performed better in class thanexternal leaders, (c) groups led by internalsperformed better than those led by externals,and (d) internal leaders were more taskoriented, and external leaders were more so-cially oriented. These results are quite con-sistent with the profile of internals as beingaction oriented and better performers andexternals as being more socially oriented.

Goodstadt and Hjelle (1973) conducteda laboratory simulation of a factory settingto study the relation between locus of controland use of power by supervisors. They as-signed college students as supervisors overwork groups assigned to one of two repetitivetasks. The work groups were composed ofconfederates, one of whom created difficul-ties for the supervisor. Data were collectedon the types of supervisory practices used asa function of I-E scores. It was found thatinternals attempted to use personal persua-sion to a greater extent than did externals,whereas externals used coercion more thaninternals did.

These results again fit with the locus ofcontrol-authoritarianism parallels in thatexternals, who tend to be more authoritar-ian, rely more on coercive means of super-vision. Whether authoritarianism alone ac-counts for these results is an empiricalquestion. The research summarized in thissection, however, suggests that internals andexternals differ in their personal supervisory

492 PAUL E. SPECTOR

styles and in their reactions to the supervi-sory styles of their superiors. Externals seemto prefer supervisors who are directive, andthey themselves rely more on coercion withtheir subordinates. In addition, they seemmore concerned with the social rather thanthe task aspects of the job. Internals, on theother hand, prefer participative approachesfrom their supervisors, rely more on personalpersuasion with their own subordinates, andseem more task oriented and less sociallyoriented.

Perceived Job Characteristics and RoleStrain

From their generalized expectancies, onemight assume that internals and externalswould differ in their perceptions of job char-acteristics as well as in their reactions tothose characteristics. It would be predictedthat because internals perceive more per-sonal control over their environment, theywould tend to perceive the job as offeringmore autonomy; also, because internals aremore sensitive to information in the eviron-ment, they should report more feedback onthe job. Kimmons and Greenhaus (1976)provided support for both of these conten-tions. In their sample of 193 managers, in-ternals reported having more autonomy andreceiving more feedback than did externals.

Role strain and role ambiguity are twoother variables that have been investigated.Here the results with locus of control areinconsistent. Gemmill and Heisler (1972)found a correlation of .31 between theirmeasure of job strain (uncertainty of pro-motion, ambiguity of supervisor's evalua-tions, too heavy workload, too little author-ity, etc.) and locus of control, with internalsperceiving less strain than do externals. Or-gan and Greene (1974b) found a correlation(r = .42) between I-E and a similar measure,role ambiguity. Evans (1974), however, founda negative correlation between I-E and thesame role ambiguity measure that Organand Greene used,

Abdel-Halim (1980) found that locus ofcontrol moderated the relation between roleambiguity and job satisfaction. Specifically,externals were considerably more satisfiedunder low than under high ambiguity. This

tendency was insignificant for internals, whoseemed less affected by task ambiguity.

On the other hand, Batlis (1980) foundno evidence for a moderating effect of locusof control on the relation between role am-biguity or role conflict and job satisfaction.There were small but significant correlationsbetween both role variables and locus of con-trol. There was, however, a nonsignificantrelation between locus of control and jobsatisfaction, making one question the com-parability of these data with other reportedstudies.

Sims and Szilagyi (1976) hypothesizedthat locus of control would moderate therelation between perceived job characteris-tics and job satisfaction. They focused theirinvestigation on job aspects measured by theJob Characteristics Inventory, which in-cludes the following dimensions: variety, au-tonomy, task identity, feedback, interactionswith others, and friendship opportunities.They calculated correlations between eachof these dimensions and job satisfaction sep-arately for internals and externals. Only forautonomy was there a significant differencein the correlations with work satisfaction(r = .39 and .16 for externals and internals,respectively) and with supervisor satisfaction(r = .49 and .15 for externals and internals,respectively). Kimmons and Greenhaus(1976) replicated part of this study with dif-ferent instruments and found that locus ofcontrol did not significantly moderate therelation between satisfaction and autonomyor feedback, although internals had a some-what higher correlation between autonomyand satisfaction than did externals, a trendthat is opposite to that in Sims and Szilagyi'sstudy.

The literature on the relation between per-ceived job characteristics and locus of con-trol is inconsistent and inconclusive. Thestudies included here that reported a relationbetween I-E and job characteristics wereconsistent in showing that there may be dif-ferences in perceptions, but they were in-consistent in demonstrating a direction ofrelation. Likewise, the moderating role oflocus of control in the relation between per-ceptions of and affective responses to the jobis unclear. It would seem potentially fruitfulfor future research to investigate the.relation


between locus of control and perceptions ofthe job in an attempt to clarify these incon-sistencies.

Turnover

The relation between locus of control andturnover is complex, and a consistent cor-relation between locus of control and quit-ting would not be expected. On the one hand,internals tend to take action and thus mightbe expected to quit jobs more readily. Onthe other hand, they tend to be more suc-cessful on the job and more satisfied, factorsassociated with less individual turnover.

Job satisfaction has consistently beenfound to be somewhat predictive of turnover(Mobley, Griffeth, Hand, & Meglino, 1979;Porter & Steers, 1973). The relations found,however, have been modest, suggesting thatdissatisfaction alone does not account forturnover. Locus of control might well mod-erate the relation between satisfaction andturnover in the following way: Externalstend not to take action, and therefore evenif they are dissatisfied they may stay on thejob, at least until environmental factors forcethem to leave. Internals, on the other hand,tend to take action and would be expectedto quit a dissatisfying job. Therefore, thecorrelation between satisfaction and turn-over should be higher for internals than forexternals.

If dissatisfied internals tend to leave jobs,it would follow that conditions leading tosatisfaction would moderate the locus of con-trol-turnover relation. For highly satisfyingjobs, internals would exhibit the same rateof turnover as do externals; for highly dis-satisfying jobs, internals would exhibit moreturnover than do externals.

No data exist to validate these hypotheses,but data are available on the relation be-tween job tenure and locus of control. Harvey(1971) found among chief government ad-ministrators in Canada that locus of controlwas negatively related to tenure; that is, in-ternals held longer tenure on the job. Similarresults were found by Andrisani and Nestel(1976) in a general male population and byOrgan and Greene (1974a) with senior sci-entists and engineers.

Summary and Conclusions

The results of the research summarizedin this article suggest that locus of controlmay be an important personality variable inorganizational research and theory. It maybe useful as a moderator in tests of expec-tancy theory and predictions of turnover,and it may help to explain behavior in anumber of organizational situations. Fur-thermore, on a practical level, locus of con-trol may be useful as a selection device formany specific jobs and settings.

Both laboratory and field studies have es-tablished that the behavior of internals andexternals can differ across situations. In gen-eral, internals tend to believe that they havepersonal control over rewards and events.This generalized belief or expectancy leadsthe internal to take action when action isperceived to lead to rewards that are valuedor desired. Thus, the internal is potentiallymore motivated than the external, who mayseem to ignore the reinforcement contingen-cies in a situation.

Organizational settings in which rewardsare tied to performance, such as piece-ratesystems or incentive systems, would be ex-pected to work well with internals and poorlywith externals. Hence, locus of control couldbe used to select employees who will workunder incentive systems. In order to motivateor at least to control the performance of ex-ternals, directive supervision would probablybe far more effective than incentives. Asshown above, externals tend to be more sat-isfied with directive supervision (Mitchell etal., 1975; Runyon, 1973), to comply morewith demands of coercive supervisors (Crav-ens & Worchel, 1977), and to be more com-pliant with social demands than internals are(Phares, 1976).

The style of supervision preferred by in-ternals and externals seems to be quiteopposite. Internals who tend to be self-motivated prefer participative approaches,whereas the more outward-directed exter-nals prefer directive approaches. Thus, ex-ternals would seem more appropriate fordirective supervisors or situations that de-mand directive supervision. Internals wouldbe more appropriate for participative super-visors or situations requiring participative

494 PAUL E. SPECTOR

approaches. This difference in preferredstyle seems to carry over into the style usedby internals and externals with their ownsubordinates. That is, internals tend to beparticipative supervisors, whereas externalstend to be directive supervisors.

Task or organizational demands often dic-tate the style of supervision that should beused. Tasks that require close coordinationof many people generally require quite di-rective styles. Thus, assembly line workwhere individual worker's tasks are highlyinterdependent, and battlefield operationswhere precise carrying out of orders is es-sential, would be most appropriate for ex-ternals who are more suited for directive su-pervision. Tasks that require initiative andindependent action are more suited for in-ternals, who are in turn best supervised byparticipative supervisors. Situations in whichinitiative is needed but a directive supervi-sory style is adopted may create demandsthat both internals and externals find incom-patible. That is, the internal's initiative andthe external's compliance are necessary, butthese may be qualities that are not oftenfound in the same individual. To summarize,low initiative-high compliance and high in-itiative-low compliance call for externalsand internals, respectively. High compli-ance-high initiative situations create con-flicting demands. Externals in such situa-tions comply but perform poorly whereindependent action is essential. Internals, onthe other hand, are able to perform inde-pendently but become frustrated and un-comfortable with compliance demands.

These contentions are consistent with re-search in education and psychotherapy.There is evidence, although it is somewhatequivocal, that internals prefer and oftenperform better with participative approachesand externals with directive approaches inboth the classroom (Harpin & Sandier,1979; McMillan, 1980) and psychotherapy(Messer & Meinster, 1980; Rostow, 1980).

Another difference between internals andexternals is their ability to handle complexinformation. Internals seem better at col-lecting and processing information and wouldbe better at performing complex tasks. Thistendency is totally independent of intelli-gence (Phares, 1976), suggesting that per-

haps it is motivation that accounts for theperformance differential. Internals wouldseem better suited for tasks requiring com-plex information collection or processing.Thus, not only the compliance and initiativedemands of a task but also its informationcomplexity should be considered in selectingthe optimal person for a given job. Again,there may be inconsistencies in the demandsmade by a particular job. Jobs requiringcomplex information processing and highcompliance might demand qualities that areincompatible. Tasks that are complex in in-formation-processing demands are obviouslybest handled by internals. Where the com-plexity demands conflict with the need forcompliance, one must decide which demandis the most cogent and select internals orexternals on that basis. If complexity de-mands are more important, one should selectinternals, although this may cause problemswith compliance. If compliance is more im-portant, externals would be more appropri-ate, although they will probably performmore poorly than internals. Where compli-ance is essential it might be best to simplifyindividual jobs, and where complexity mustoccur, participative supervisory practicesshould be implemented.

Of course, as discussed previously, thereare considerable methodological limitationsto the research summarized here, and theseconclusions are tentative. This research isprimarily cross-sectional and correlationaland suffers from method variance and toonarrow a scope; that is, it ignores other im-portant variables, specifically anxiety. In ad-dition, most of this research has not at-tempted to delineate complex relationsamong forms of behavior, locus of control,and other variables, both situational and in-dividual. Some exceptions were the studiesthat used locus of control as a moderator ofsupervisory style and job satisfaction.

On the bivariate level, many of the rela-tions described in this article have been wellestablished. What is needed for future re-search are more complex studies that willcontribute to a more thorough understand-ing of these relations. Researchers might dowell to include anxiety in their studies todetermine its interactive effects, as at-tempted by Archer (1979b). Furthermore,


studies that use behavioral variables relatedto personality are needed, as the current re-liance on self-report greatly limits conclu-sions. It is suggested that locus of controlmight prove to be a useful variable for em-ployee selection. The limited data of this re-view indicate that I-E might well be relatedto job performance for at least some jobs,but its utility as a selection device needs em-pirical validation. Certainly, more researchis needed on locus of control and leadershipvariables. The current data are incompleteconcerning supervisory styles and behaviorof internals and externals and reactions byinternal and external subordinates to inter-nal and external supervisors.

One further note is that internals seem tobehave in ways that validate much theoryin organizational psychology. That is, inter-nals respond to reinforcement contingencies(incentive systems) on the job, they seem toprefer participative supervision, they dem-onstrate initiative, and they tend to take per-sonal action on the job. Externals on theother hand seem unresponsive to incentives(they want them but will not necessarilywork harder for them) and prefer directivesupervision. Thus, much organizational the-ory might well be limited to internals.

Reference Notes

1. Lewis, B. A. Effects of social desirability responsebias on the measurement of job satisfaction. Un-published master's thesis, University of South Flor-ida, 1977.

2. Smith, P. C. Personal communication, March 11,1981.

3. Merrens, M. R., & Qarrett, J. B. The Protestantethic personality variable and work behavior. Paperpresented at the meeting of the Eastern PsychologicalAssociation, Philadelphia, April 1974.

References

Abdel-Halim, A. A. Effects of person-job compatibilityon managerial reactions to role ambiguity. Organi-zational Behavior and Human Performance, 1980,26, 193-211.

Abdel-Halim, A. A. Personality and task moderatorsof subordinate responses to perceived leader behavior.Human Relations, 1981, 34, 73-88.

Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J.,& Sanford, R. N. The authoritarian personality.New York: Harper & Row, 1950.

Anderson, C. R. Locus of control, coping behaviors, and

performance in a stress setting: A longitudinal study.Journal of Applied Psychology, 1977, 62, 446-451.

Anderson, C. R., & Schneier, C. E. Locus of control,leader behavior and leader performance among man-agement students. Academy of Management Journal,1978, 21, 690-698.

Andrisani.'P. J., & Nestel, G. Internal-external controlas contributor to and outcome of work experience.Journal of Applied Psychology. 1916,61, 156-165.

Archer, R. P. Relationships between locus of controland anxiety. Journal of Personality Assessment,1979, 43, 617-626. (a)

Archer, R. P. Relationships between locus of control,trai t anxiety, and state anxiety: An interactionist per-spective. Journal of Personality, 1979, 47, 305-316.(b)

Batlis, N. C. Job involvement and locus of control asmoderators of role-perception/individual-outcome re-lationships. Psychological Reports, 1980, 46, 111-119.

Biondo, J., & MacDonald, A. P., Jr. Internal-externallocus of control and response to influence attempts.Journal of Personality, 1971, 39, 407-419.

Brehm, J. W. A theory of psychological reactance. NewYork: Academic Press, 1966.

Broedling, L. A. Relationship of internal-external con-trol to work motivation and performance in an ex-pectancy model. Journal of Applied Psychology,1975, 60, 65-70.

Campbell, J. P. Motivation theory in industrial and or-ganizational psychology. In M. D. Dunnette (Ed.),Handbook of industrial and organizational psychol-ogy. Chicago: Rand McNally, 1976.

Cravens, R. W., & Worchel, P. The differential effectsof rewarding and coercive leaders on group membersdiffering in locus of control. Journal of Personality,1977, 45, 150-168.

Crowne, D. P., & Liverant, S. Conformity under varyingconditions of commitment. Journal of Abnormal andSocial Psychology, 1963, 66, 547-555.

Dailey, R. C. Relationship between locus of control,perceived group cohesiveness, and satisfaction withco-workers. Psychological Reports, 1978, 42, 311-316.

DuCette, J., & Wolk, S. Cognitive and motivationalcorrelates of generalized expectancies for control.Journal of Personality and Social Psychology, 1973,26, 420-426.

Evans, M. G. Extensions of a path-goal theory of mo-tivation. Journal of Applied Psychology, 1974, 59,172-178.

Gemmill, G. R., & Heisler, W. J. Fatalism as a factorin managerial job satisfaction, job strain, and mobil-ity. Personnel Psychology, 1972, 25, 241-250.

Giles, W. F. Volunteering for job enrichment: A test ofexpectancy theory predictions. Personnel Psychology,1977, 30, 427-435.

Goodstadt, B. E., & Hjelle, L. A. Power to the pow-erless: Locus of control and the use of power. Journalof Personality and Social Psychology, 1973, 27,190-196.

Hackman, J. R., & Lawler, E. E. Employee reactionsto job characteristics. Journal of Applied Psychology,1971, 55, 259-286.

Hammer, T. H., & Vardi, Y. Locus of control and car-

496 PAUL E. SPECTOR

eer self-management among nonsupervisory employ-ees in industrial settings. Journal of Vocational Be-havior, 1981, 18. 13-29.

Harpin, P. M., & Sandier, I. N. Interaction of sex, locusof control, and teacher control: Toward a student-classroom match. American Journal of CommunityPsychology, 1979, 1. 621-632.

Hartley, M. P. The relationship of locus of control andneed achievement to job satisfaction (Doctoral dis-sertation, Rutgers University, 1975). DissertationAbstracts International, 1976, 36, 6553A. (Univer-sity Microfilms No. 76-8689)

Harvey, J. H., Barnes, R. D., Sperry, D. L., & Harris,B. Perceived choice as a function of internal-externallocus of control. Journal of Personality, 1974, 42,437-452.

Harvey, J. M. Locus of control shift in administrators.Perceptual and Motor Skills, 1971, 33, 980-982.

Heisler, W. J. A performance correlate of personal con-trol beliefs in an organizational context. Journal ofApplied Psychology, 1974, 59, 504-506.

Hersch, P. D., & Scheibe, K. E. Reliability and validityof internal-external control as a personality dimen-sion. Journal of Consulting Psychology, 1967, 31,609-613.

Hjelle, L. A., & Clouser, R. Susceptibility to attitudechange as a function of internal-external control.Psychological Record, 1970, 20, 305-310.

Holden, K. B. Attitude toward external versus internalcontrol of reinforcement and learning of reinforce-ment sequences. Unpublished master's thesis, OhioState University, 1958. Cited in H. M. Lefcourt, In-ternal versus external control of reinforcement: A re-view. Psychological Bulletin, 1966, 65, 206-220.

Hulin, C. L., & Blood, M. R. Job enlargement, indi-vidual differences, and worker responses. Psycholog-ical Bulletin, 1968, 69, 41-55.

James, W. H., Woodruff, A. B., & Werner, W. Effectof internal and external control upon changes in smok-ing behavior. Journal of Consulting Psychology,1965, 29, 184-186.

Joe, V. C. Review of the internal-external control con-struct as a personality variable. Psychological Re-ports, 1971, 28. 619-640.

Jul ian , J. W., & Katz, S. B. Internal versus externalcontrol and the value of reinforcement. Journal ofPersonality and Social Psychology, 1968, 8, 89-94.

Kabanoff, B., & O'Brien, G. E. Work and leisure: Atask-attributes analysis. Journal of Applied Psy-chology, 1980, 65, 596-609.

Kahle, L. R. Stimulus condition self-selection by malesin the interaction of locus of control and skill-chancesituations. Journal of Personality and Social Psy-chology, 1980, 38, 50-56.

Kimmons, G., & Greenhaus, J. H. Relationship betweenlocus of control and reactions of employees to workcharacteristics. Psychological Reports, 1976, 39,815-820.

Krolick, G. Changes in expectancy and attribution fol-lowing success, failure, and neutral consequences(Doctoral dissertation, Syracuse University, 1978).Dissertation Abstracts International, 1979, 39, 5074B.(University Microfilms No. 79-08546)

Lawler, E. E. A correlational-causal analysis of the re-lationship between expectancy attitudes and job per-

formance. Journal of Applied Psychology, 1968, 52,462-468.

Lawler, E. E. Pay and organizational effectiveness: Apsychological view. New York: McGraw-Hill, 1971.

Lefcourt, H. M. Internal versus external control of re-inforcement: A review. Psychological Bulletin, 1966,65, 206-220.

Lester, D., & Genz, J. L. Internal-external locus ofcontrol, experience as a police officer, and job satis-faction in municipal police officers. Journal of PoliceScience and Administration, 1978, 6, 479-481.

Lied, T. R., & Pritchard, R. D. Relationships betweenpersonality variables and components of the expec-tancy-valence model. Journal of Applied Psychology,1976, 61, 463-467.

Majumder, R. K., MacDonald, A. P., & Greever,K. B. A study of rehabilitation counselors: Locus ofcontrol and attitudes toward the poor. Journal ofCounseling Psychology, 1977, 24, 137-141.

McMillan, J. H. Effect of instructional procedure andpupil locus of control on achievement and attitudes.Psychology in the Schools, 1980, 17, 123-127.

Messer, S. B., & Meinster, M. O. Interaction effects ofinternal vs. external locus of control and directive vs.nondirective therapy: Fact or fiction? Journal of Clin-ical Psychology, 1980, 36, 283-288.

Mirels, H. L., & Garrett, J. B. The Protestant ethic asa personality variable. Journal of Consulting andClinical Psychology, 1971, 36, 40-44.

Mitchell, T. R., Smyser, C. M., & Weed, S. E. Locusof control: Supervision and work satisfaction. Acad-emy of Management Journal, 1975, 18, 623-631.

Mobley, W. H., Griffeth, R. W., Hand, H. H., & Meg-lino, B. M. Review and conceptual analysis of theemployee turnover process. Psychological Bulletin,1979, 86. 493-522.

Munoz, M. Job satisfaction in policemen and its relationto locus of control, ego strength, and performance(Doctoral dissertation, City University of New York,1973). Dissertation Abstracts International, 1973,34, 2946A-2947A. (University Microfilms No. 73-28,369)

Organ, D. W. Locus of control, anxiety, and cognitivefunctioning. Psychological Reports, 1976, 39, 1091-1098.

Organ, D. W., & Greene, C. N. The perceived pur-,posefulness of job behavior: Antecedents and conse-quences. Academy of Management Journal, 1974, 17,69-78. (a)

Organ, D. W., & Greene, C. N. Role ambiguity, locusof control, and work satisfaction. Journal of AppliedPsychology, 1974, 59, 101-102. (b)

Phares, E. J. Differential utilization of information asa function of internal-external control. Journal ofPersonality, 1968, 36. 649-662.

Phares, E. J. Locus of control in personality. Morris-town, N.J.: General Learning Press, 1976.

Porter, L. W., & Lawler, E. E. Properties of organi-zational structure in relation to job attitudes and jobbehavior. Psychological Bulletin, 1965, 64. 23-51.

Porter, L. W., & Steers, R. M. Organizational, work,and personal factors in employee turnover and ab-senteeism. Psychological Bulletin, 1973,50, 151-176.

Ray, W. J., & Katahn, M. Relation of anxiety to locusof control. Psychological Reports, 1968, 23, 1196.


Reitz, H. J., & Jewell, L. N. Sex, locus of control, andjob involvement: A six-county investigation. Academyof Management Journal, 1979, 22, 72-80.

Roark, M. H. The relationship of perception of chancein finding jobs to locus of control and to job searchvariables on the part of human resource agency per-sonnel (Doctoral dissertation, Virginia PolytechnicUniversity, 1978). Dissertation Abstracts Interna-tional, 1978, 38, 2070A. (University Microfilms No.78-18558)

Rostow, C. D. The effect of self- vs. external monitoringand locus of control upon the pacing and general ad-justment of psychiatric inpatients. Behavior researchand therapy, 1980, 18, 541-548. \

Rotter, J. B. Social learning and clinical psychology.Englewood Cliffs, N.J.: Prentice-Hall, 1954.

Rotter, J. B. Generalized expectancies for internal ver-sus external control of reinforcement. PsychologicalMonographs, 1966, 80(1, Whole No. 609).

Runyon, K. E. Some interactions between personalityvariables and management styles. Journal of AppliedPsychology, 1973, 57, 288-294.

Salancik, G. R., & Pfeffer, J. A social information pro-cessing approach to job attitudes and task design.Administrative Science Quarterly, 1978, 23, 224-253.

Sarason, I. G. Experimental approaches to test anxiety:Attention and the uses of information. In C. D. Spiel-berger (Ed.), Anxiety; Current trends in theory andresearch (Vol. 2). New York: Academic Press, 1972.

Satmoko. Belief in the internal and external control ofreinforcement as a determinant of the attitude towardthe work role (Doctoral dissertation, George Wash-ington University, 1972). Dissertation Abstracts In-ternational, 1973, 33, 4005A-4006A. (UniversityMicrofilms No. 73-4022)

Simmons, W. Personality correlates of the James-Phares scale. Unpublished master's thesis, Ohio StateUniversity, 1959. Cited in H. M. Lefcourt, Internal

versus external control of reinforcement: A review.Psychological Bulletin, 1966, 65, 206-220.

Sims, H, P., Jr., & Szilagyi, A. D. Job characteristicsrelationships: Individual and structural moderators.Organizational Behavior and Human Performance,1976, 17, 211-230.

Singh, M. R. G. The relationship of job satisfaction withlocus of control, organizational setting and education(Doctoral dissertation, University of Michigan, 1978).Dissertation Abstracts International, 1978, 38, 684A.(University Microfilms No. 78-13735)

Stogdill, R. M. Manual for the Leader Behavior De-scription Questionnaire—Form XII. Columbus: OhioState University, 1963.

Szilagyi, A. D., & Sims, H. P., Jr. Locus of control andexpectancies across multiple occupational levels.Journal of Applied Psychology, 1915,60, 638-640.

Ude, L. K., & Vogler, R. E. Internal versus externalcontrol-of reinforcement and awareness in a condi-tioning task. Journal of Psychology, 1969, 73, 63-67.

Valecha, G. K. Construct validation of internal-externallocus of reinforcement related to work-related vari-ables. Proceedings of the 80th Annual Convention ofthe American Psychological Association, 1972, 7,455-456.

Vroom, V. H. Work and motivation. New York: Wiley,1964.

Wolk, S., & DuCette, J. Intentional performance andincidental learning as a function of personality andtask dimensions. Journal of Personality and SocialPsychology, 1974, 29, 90-101.

Yukl, G. A., & Latham, G. P. Interrelationships amongemployee participation, individual differences, goaldifficulty, goal acceptance, goal instrumentality, andperformance. Personnel Psychology, 1978, 31, 305-323.

Received May 4, 1981

behavior in organizations as a function of employee's...

Documents