goal setting and task performance: 1969-1980 · goal intensity has not been studied as such,...

Psychological Bulletin1981, Vol. 90. No. I, 125-152

Copyright 1981 by the American Psychological Association, Inc.0033-2909/81/9001-0125SOO. 75

Goal Setting and Task Performance: 1969-1980

Edwin A. LockeCollege of Business and Management and

Department of PsychologyUniversity of Maryland

Lise M. SaariDepartment of PsychologyUniversity of Washington

Karyll N. ShawCollege of Business and Management

University of Maryland

Gary P. LathamGraduate School of Business Administration

University of Washington

A review of both laboratory and field studies on the effects of setting goals whenperforming a task found that in 90% of the studies, specific and challenging goalslead to higher performance than easy goals, "do your best" goals, or no goals.Goals affect performance by directing attention, mobilizing effort, increasingpersistence, and motivating strategy development. Goal setting is most likely toimprove task performance when the goals are specific and sufficiently challenging,the subjects have sufficient ability (and ability differences are controlled), feed-back is provided to show progress in relation to the goal, rewards such as moneyare given for goal attainment, the experimenter or manager is supportive, andassigned goals are accepted by the individual. No reliable individual differenceshave emerged in goal-setting studies, probably because the goals were typicallyassigned rather than self-set. Need for achievement and self-esteem may be themost promising individual difference variables.

In this article we summarize research re-lating to (a) the effects of setting varioustypes of goals or objectives on task perfor-mance and (b) the factors (other than thegoals themselves) that influence the effec-tiveness of goal setting.

Ail-encompassing theories of motivationbased on such concepts as instinct, drive, andconditioning have not succeeded in explain-ing human action. Such theories have beengradually replaced by more modest and lim-ited approaches to motivation. These ap-proaches do not presume to explain all mo-tivational phenomena; their domains aremore restricted. The study of goal setting isone such limited approach.

The concept of goal setting falls within thebroad domain of cognitive psychology andis consistent with recent trends such as cog-nitive behavior modification (Meichenbaum,1977). The present interest of researchers ingoal setting has two sources, one academic

Preparation of this manuscript was supported by Of-fice of Naval Research Contract N00014-79-C-0680.

Requests for reprints should be sent to Edwin A.Locke, College of Business and Management, Universityof Maryland, College Park, Maryland 20742.

and the other organizational. The academicsource extends back in time from Ryan(1970) and G. Miller, Galanter, and Pri-brani (1960), through Lewin, to the Wurz-burg School and the associated concepts ofintention, task, set, and level of aspiration(see Ryan, 1970, for a summary). The or-ganizational source is traced from Manage-ment by Objectives programs, now widelyused in industry (see Odiorne, 1978, for asummary), back to the Scientific Manage-ment movement founded by Frederick W,Taylor (1911/1967). These two strains ofthought converge in the more recent workof Locke (1968), Latham (Latham & Yukl,1975b), and others who have studied theeffects of goal setting on task performance.Goal setting is also an important componentof social learning theory (Bandura, 1977),which has become increasingly influential inrecent years. Even the literature on organi-zational behavior modification can be inter-preted largely within a goal-setting frame-work (Locke, 1977).

Research on goal setting is proliferatingso rapidly that recent reviews (Latham &Yukl, I975b; Locke, 1968; Steers & Porter,1974) are now outdated. To provide a longer

125

126 LOCKE, SHAW, SAARI, AND LATHAM

term perspective than just the last 6 years,our review includes research published since1968. Studies that are explicitly clinical andsocial-psychological in nature are not in-cluded (for a detailed review of the latter,see Fishbein & Ajzen, 1975).

review here. We also examine the mecha-nisms by which goals affect action, the ef-fects of feedback, participation, and moneyon goal-setting effectiveness, the role of in-dividual differences, and the determinantsof goal commitment.

The Concept of Goal Setting

A goal is what an individual is trying toaccomplish; it is the object or aim of an ac-tion. The concept is similar in meaning tothe concepts of purpose and intent (Locke,1969). Other frequently used concepts thatare also similar in meaning to that of goalinclude performance standard (a measuringrod for evaluating performance), quota (aminimum amount of work or production),work norm (a standard of acceptable be-havior defined by a work group), task (apiece of work to be accomplished), objective(the ultimate aim of an action or series ofactions), deadline (a time limit for complet-ing a task), and budget (a spending goal orlimit).

Earlier attempts by behaviorists to reduceconcepts like goal and purpose to physicalevents have been strongly criticized (e.g., seeLocke, 1969, 1972). Goal setting might becalled "stimulus control" by a modern be-haviorist, but the key question then becomes,What is the stimulus? If it is only an as-signed goal (an environmental event), thenthe importance of goal acceptance is ignored;an assigned goal that is rejected can hardlyregulate performance. If goal acceptance isconsidered relevant, then the regulatingstimulus must be a mental event—ultimatelythe individual's goal. The environment, ofcourse, can influence goal setting as well asgoal acceptance, an issue that is dealt within some of the recent research,

The basic assumption of goal-setting re-search is that goals are immediate regulatorsof human action. However, no one-to-onecorrespondence between goals and action isassumed because people may make errors,lack the ability to attain their objectives(Locke, 1968), or have subconscious con-flicts or premises that subvert their consciousgoals. The precise degree of association be-tween goals and action is an empirical ques-tion that is dealt with in the research we

Goal-Setting Attributes'

Mental processes have two major attri-butes, content and intensity (Rand, 1967).The content of a goal is the object or resultbeing sought. The main dimensions of goalcontent that have been studied so far arespecificity or clarity (the degree of quanti-tative precision with which the aim is spec-ified) and difficulty (the degree of profi-ciency or level of performance sought). Theterms task difficulty and goal difficulty areoften used interchangeably, but a distinctionbetween them can be made.

A task is a piece of work to be accom-plished. A difficult task is one that is hardto do. A task can be hard because it is com-plex, that is, requires a high level of skill andknowledge. For example, writing a book onphysics is a harder task than writing a thank-you note. A task can also be hard becauseit requires a great deal of effort: digging thefoundation for a pool takes more effort thandigging a hole to plant a flower seed.

Since a goal is the object or aim of anaction, it is possible for the completion of atask to be a goal. However, in most goal-setting studies, the term goal refers to at-taining a specific standard of proficiency ona task, usually within a specified time limit.For example, two individuals are given thesame task (e.g., simple addition), but one isasked to complete a large number of prob-lems within 30 minutes, and the other, asmall number. The harder goal would beachieved by expending greater effort andattention than would be expended to achievethe easy goal. Harder goals, like hardertasks, also can require more knowledge andskill than easier goals (e.g., winning a chess

' Our view of what constitutes a goal attribute differsfrom that of Steers and Porter (1974) who, for example,called participation an attribute of goals. We treat par-ticipation as a mechanism that may affect goal contentor goal acceptance.

GOAL SETTING AND TASK PERFORMANCE 127

tournament vs. coming in next to last). Tosummarize the distinction between the terms,goal difficulty specifies a certain level of taskproficiency measured against a standard,whereas task difficulty refers simply to thenature of the work to be accomplished.

Although greater task difficulty shouldlead to greater effort (Kahneman, 1973;Kaplan & Rothkopf, 1974; Shapira, Note1), the relation of task difficulty to perfor-mance is problematic. If more work is trans-lated into a goal to get more done, taskdifficulty may be positively related toperformance (Sales, 1970). On the otherhand, if harder tasks require more ability orknowledge, most people will, at least ini-tially, perform less well on them, even if theytry harder, than they would on easier tasks(e.g., Shapira, Note 1).

An experiment by Campbell and Ilgen(1976) demonstrated that the distinctionbetween task and goal difficulty has prac-tical utility. They manipulated both dimen-sions independently, On chess problems dif-ficult goals led to better performance thaneasy goals; training subjects on hard prob-lems (tasks) led at first to poorer perfor-mance but later to better performance thantraining subjects on easier problems (tasks).Presumably the harder goals led to greatereffort than the easier goals, and training onthe harder chess problems led to the acqui-sition of more skill and knowledge thantraining on easier ones.

Although there has been extensive re-search on the effects of goal specificity anddifficulty on performance, little attention hasbeen paid to two other dimensions of goalcontent: goal complexity (the number andinterrelation of the results aimed for) andgoal conflict (the degree to which attainingone goal negates or subverts attaining an-other).

The second attribute of goals, intensity,pertains to the process of setting the goal orof determining how to reach it. Intensitywould be measured by such factors as thescope of the cognitive process, the degree ofeffort required, the importance of the goal,and the context in which it is set. Goal in-tensity may be related to goal content; forexample, a more intense psychological pro-cess is needed to set complex goals and to

figure out how to attain them than the pro-cess needed to set and attain simple goals.Goal intensity has not been studied as such,although a related concept, goal commit-ment, has been measured in a number ofexperiments.

Relation of Goal Dimensionsto Performance

Goal Difficulty

In an earlier review of the goal-setting lit-erature, Locke (1968) found evidence for apositive, linear relation between goal diffi-culty and task performance (assuming suf-ficient ability), and more recent studies havesupported these findings. Four results inthree experimental field studies demon-strated that harder goals led to better per-formance than easy goals: Latham andLocke (1975) with logging crews; Yukl andLatham (1978) with typists; and a simulatedfield study by Bassett (1979). In a separatemanipulation, Bassett also found that shortertime limits led to a faster work pace thanlonger time limits.

Twenty-five experimental laboratory stud-ies have obtained similar results with a widevariety of tasks: Bavelas (1978), with a fig-ure-selection task; Bavelas and Lee (1978)in five of six experiments involving brain-storming, figure selection, and sum estima-tion tasks; Campbell and Ilgen (1976) withchess; Hannan (1975) with a coding (creditapplications) task; LaPorte and Nath (1976)with prose learning; Latham and Saari(1979b) with brainstorming; Locke andBryan (1969b) with simple addition; Locke,Cartledge, and Knerr (1970) in four studies,three with reaction time and one with simpleaddition; Locke, Mento, and Katcher (1978)with perceptual speed; London and Oldham(1976) with card sorting; Masters, Furman,and Barden (1977) in two studies of 4- and5-year-old children working on a color dis-crimination task; Mento, Cartledge, andLocke (1980) in two experiments using aperceptual speed task; Rothkopf and Bil-lington (1975) and Rothkopf and Kaplan(1972) in more complex prose-learning stud-ies than that of LaPorte and Nath (1976);and Sales (1970), using anagrams. In Sales's


study, task rather than goal difficulty wasmanipulated by means of varying the workload given to the subjects. Presumably sub-jects developed implicit goals based on theamount of work assigned to them. Ness andPatton (1979) also found that a harder taskled to better weight-lifting performance thanan easier task when subjects were deceivedas to the actual weights.

Four studies found conditional2 supportfor a positive relation between goal difficultyand performance. Becker (1978) with anenergy conservation task, Erez (1977) witha clerical task, and Strang, Lawrence, andFowler (1978) with a computation task, allfound that only subjects who had high goalsand who received feedback regarding theirperformance in relation to those goals per-formed better than subjects with low goals.This pattern of results seems also to havebeen present in Frost and Mahoney's (1976)first study using a reading task (see theirTable 1, p. 339). Subjects with high andmoderately high goals who apparently re-ceived frequent feedback performed betterthan those with average goals, whereas theopposite pattern was obtained for subjectsgiven no feedback during the 42-minutework period (interaction p = . 11; / tests werenot performed).

Six experimental laboratory studies foundno relation between goal level and task per-formance. Bavelas and Lee (1978) allowedonly 15 minutes for an addition task andgave subjects no information either beforeor during the task of how fast they neededto go to attain the goal. Frost and Mahoney(1976) found negative results with a jigsawpuzzle task, although their range of goaldifficulty was limited: from medium to hardto very hard (actual probabilities of successwere .50, .135, and .026, respectively). Thesame narrow range of difficulty (very diffi-cult to moderately difficult) may explain thenegative results of Oldham (1975) using atime sheet computation task. Moreover, notall subjects accepted the assigned goals inthat study, and it is not clear that ability wascontrolled when Oldham (1975, pp. 471-472) did his post hoc analysis by personalgoal level. Organ (1977) also comparedmoderate goals with hard goals using an ana-gram task. However, since no group average

even reached the level of the moderate goal,the hard goal may have been totally un-realistic.

The fifth negative study, by Motowidlo,Loehr, and Dunnette (1978), using a com-plex computation task, examined the goaltheory-expectancy theory controversy. Goaltheory predicts that harder goals lead to bet-ter performance than easy goals, despitetheir lower probability of being fully reached.In contrast, expectancy theory predicts (otherthings being equal) a positive relation be-tween expectancy and performance, the op-posite of the goal theory prediction. Motow-idlo et al. found a positive relation betweenexpectancy and performance, which is inagreement with expectancy theory. One pos-sible confounding factor is that the subjectsin the Motowidlo et al. study did not maketheir expectancy ratings conditional upontrying their hardest to reach the goal or towin (pointed out by Mento et al., 1980,based on Yates & Kulick, 1977, among oth-ers). Thus, low expectancy ratings couldmean that a subject was not planning to ex-ert maximum effort, whereas high ratingswould mean the opposite. This would yielda spurious positive correlation between ex-pectancy and performance. Furthermore,Motowidlo et al. did not provide their sub-jects with feedback regarding how close theywere coming to their goals during task per-formance. (The importance of this factor isdocumented below.) The two studies byMento et al. (1980), which avoided the er-rors of the Motowidlo et al. study and in-corporated other methodological improve-ments, found the usual positive relationbetween goal level and performance and norelation between expectancy and perfor-mance.

Forward and Zander (1971) used goalsset by groups of high school boys on a team-coding task as both independent and depen-dent variables. Success and failure as wellas outside pressures were covertly manipu-

2 Partially or conditionally supportive studies weredistinguished from nonsupportive studies as follows: Astudy was called partially supportive if the treatmentwas significant for one subsample of the full sample ofsubjects or for one of several experimental treatmentsor criteria. If an entire sample or study found no sig-nificant effects, it was called nonsupportive.


lated to influence goal setting, which oc-curred before each trial of the task. Underthese somewhat complex conditions, goaldiscrepancy (goal minus previous perfor-mance level) was either unrelated or nega-tively related to subsequent performance.

The results of the experimental studieswere, to varying degrees, supported by theresults of 15 correlational studies. Andrewsand Farris (1972) found that time pressurewas associated with high performance amongscientists and engineers. Hall and Lawler(1971), with a similar sample, found no re-lation between time pressure and perfor-mance but found a significant relation be-tween both quality and financial pressure(implied goals?) and work performance.Ashworth and Mobley (Note 2) found a sig-nificant relation between performance goallevel and training performance for Marinerecruits. Blurnenfeld and Leidy (1969), inwhat also could be called a natural field ex-periment, found that soft-drink servicemenwho were assigned higher goals servicedmore machines than those assigned lowergoals. Hamner and Harnett (1974) foundthat subjects in an experimental study ofbargaining who expected (tried?) to earn ahigh amount of money earned more thanthose who expected (tried?) to earn lessmoney. Locke et al. (1970), in the last oftheir five studies, found a significant corre-lation between grade goals on an hourlyexam and actual grade earned.

The majority of the correlational studiesfound only a conditional positive relationbetween goal difficulty and performanceand/or effort. Carroll and Tosi (1970) founda positive relation only for managers whowere mature and high in self-assurance;Dachler and Mobley (1973) found it onlyfor production workers (in two plants) withlong tenure (1 or 2 years or more); Dossett,Latham, and Mitchell (1979), found it intwo studies of clerical personnel, but only forthose who set goals participatively; Hall andHall (1976) found it for the class perfor-mance of second through fourth grade stu-dents in high-support schools; and Ivancev-ich and McMahon (1977a, 1977b, 1977c)found it for skilled technicians who hadhigher order (growth) need strength, werewhite, and had higher levels of education.

Negative results were obtained by For-ward and Zander (1971) with United Fundcampaign workers, Hall and Foster (1977)with participants in a simulated manage-ment game, and Steers (1975) with first-level supervisors.

All the correlational studies are, of course,open to multiple causal interpretations. Forexample, Dossett et al. (1979) implied thattheir results may be an artifact of ability,since ability was considered when settinggoals in the participative groups but not inthe assigned groups. In fact, none of the cor-relational studies had controls for ability.Also, many relied on self-ratings of goal dif-ficulty or performance. The Yukl and La-tham (1978) study found that only objectivegoal level, not subjective goal difficulty, wasrelated to typing performance. None of thecorrelational studies measured the individ-ual's personal goal level, a measure thatMento et al. (1980) found to be the singlebest motivational predictor of performance.Their measures of subjective goal difficultydid not explain any variance in performanceover and above that explained by objectiveand personal goal levels.

Goal Specificity

Specific hard goals versus "do best" goalsor no goals. Previous research found thatspecific, challenging (difficult) goals led tohigher output than vague goals such as "doyour best" (Locke, 1968). Subsequent re-search has strongly supported these results,although in a number of studies, no distinc-tion was made between groups told to dotheir best and those assigned no specificgoals. The latter were typically labeled nogoal groups. We have not found any differ-ences in the results obtained by studies inwhich no goals are assigned and those inwhich subjects are explicitly told to do theirbest. No goal subjects, it appears, typicallytry to do as well as they can on the assignedtask.

Twenty-four field experiments all foundthat individuals given specific, challenginggoals either outperformed those trying to dotheir best or surpassed their own previousperformance when they were not trying forspecific goals: Bandura and Simon (1977)


with dieting; Dockstader (Note 3) with keypunching; Dossett et al. (1979) in two stud-ies, one using a clerical test and the otherperformance evaluations for clerical work-ers; Ivancevich (1977) with maintenancetechnicians; Ivancevich (1974) in two plantswith marketing and production workers (forone or more performance criteria); Ivancev-ich (1976) with sales personnel; Kim andHamner (1976) with telephone service jobs;Kolb and Boyatzis (1970) with personalitychange in a T-group; Latham and Baldes(1975) with truck loading; Latham andKinne (1974) with logging; and Latham andYukl (1975a) with woods workers who par-ticipated in goal setting; Latham and Yukl(1976) with typing; Latham, Mitchell, andDossett (1978) with engineering and scien-tific work; Migliore (1977) with canning(press department) and ship loading (twostudies); Nemeroff and Cosentino (1979)with performance appraisal activities; Um-stot, Bell, and Mitchell (1976) with codingland parcels; Wexley and Nemeroff (1975)with managerial training; and White, Mitch-ell, and Bell (1977) with card sorting. Thestudies by Adam (1975) with die casters,Feeney with customer service workers ("AtEmery Air Freight," 1973), and Komaki,Barwick, and Scott (1978) with pastry work-ers are also included in this group. Althoughthese investigations claimed that they weredoing behavior modification, the major tech-nique actually used was goal setting plusfeedback regarding goal attainment (Locke,1977).

A negative result was obtained by Lathamand Yukl (1975a) with one sample of woodsworkers. Either individual differences orlack of organizational support may havebeen responsible for this failure. (Ivancevich,1974, also cited differences in organizationalsupport as the reason for obtaining betterresults in one of his plants than the other.)

The generally positive results of the fieldstudies were supported by the results of 20laboratory studies: Chung and Vickery (1976;their KR condition included implicit goalsetting) with a clerical task; Frost and Ma-honey (1976) with a reading task (but onlyfor subjects given frequent feedback) andwith a puzzle task; Hannan (1975) with acoding task; Kaplan and Rothkopf (1974)

and LaPorte and Nath (1976) with proselearning; Latham and Saari (1979a) withbrainstorming; Latham and Saari (1979b)with brainstorming again, but only for sub-jects who set goals participatively (thoughthis may have been an artifact since the au-thors reported that the assigned goal subjectsmay not have understood the instructionsclearly); Locke and Bryan (1969a) with adriving task; Locke et al. (1978) with per-ceptual speed (comparing the hard-goal vs.do-best groups only); Mossholder (1980)using two assembly tasks; Organ (1977) withanagrams; Pritchard and Curtis (1973) withcard sorting; Reynolds, Standiford, and An-derson (1979) with learning prose; Rosswork(1977) with a sentence construction taskused with sixth graders; Rothkopf and Bil-lington (1975) and Rothkopf and Kaplan(1972), again with learning prose; Strang,Lawrence, and Fowler (1978) with arith-metic computation (but only for hard-goalsubjects who had feedback); and Terborgand Miller (1978) with tinker-toy assembly.

A negative result was obtained by Organ(1977) on a proofreading task. Evidently thegoals set were moderate rather than hard,since they were set at the median scores forpretest subjects and were surpassed by sub-jects in all conditions. Moderate goals arenot predicted to lead to higher performancethan do-best goals. Locke et al. (1978), forexample, found that although hard-goal sub-jects exceeded the performance of do-bestsubjects, moderate-goal subjects did not.

Seven correlational field studies also sup-ported or partially supported the superiorityof specific hard goals over do-best goals orno goals: Blumenfeld and Leidy (1969) withsoft drink servicemen; Brass and Oldham(1976) and Oldham (1976) with foremen;Burke and Wilcox (1969) with telephoneoperators; Ronan, Latham, and Kinne (1973)with pulpwood producers; Steers (1975)with supervisors (but only those high on needfor achievement); and Terborg (1976) withstudents studying programmed texts.

Clear versus unclear goals or intentions.Relatively few studies have been concernedwith the effect of goal clarity on perfor-mance. Two experimental studies (Kaplan& Rothkopf, 1974; Rothkopf & Kaplan,1972) found that specific prose-learning


goals led to more learning than generallystated goals. Carroll and Tosi (1970) foundthat goal clarity correlated with increasedeffort only for managers who were matureand decisive and who had low job interestand low support from their managers. Ivan-cevich and McMahon (1977a, 1977b, 1977c)found that goal clarity correlated with per-formance mainly for technicians who wereblack, less educated, and high on higher or-der need strength. These correlational stud-ies seem to provide no consistent pattern,which is not surprising in view of the prob-lems inherent in concurrent, self-report de-signs.

The borderline and negative results ofHall and Hall (1976) and Hall and Foster(1977) with respect to goal difficulty andperformance may have been because theirgoals did not consist of clear objectives butof the self-rated strength of the subjects' in-tentions to perform well.

The findings of these studies involvingvague intentions can be contrasted with theorganizational studies by H. Miller, Kater-berg, and Hulin (1979), Mobley, Homer,and Hollingsworth (1978), and Mobley,Hand, Baker, and Meglino (1979). Theyfound significant longitudinal correlationsbetween the specific intention to remain inor leave the organization and the corre-sponding action.

Conclusions

Overall, 48 studies partly or wholly sup-ported the hypothesis that hard goals leadto better performance than medium or easygoals; 9 studies failed to support it. Fifty-onestudies partially or wholly supported theview that specific hard goals lead to betterperformance than do-your-best or no goals;2 studies did not support it. Combining thesetwo sets of studies, we found that 99 out of110 studies found that specific, hard goalsproduced better performance than medium,easy, do-your-best, or no goals. This repre-sents a success rate of 90%.

Most of these studies (at least the exper-imental ones) were well designed; they in-cluded control groups, random assignment,negligible attrition, controls for ability, ob-jective performance measures, and a great

variety of tasks and situations. Thus, con-siderable confidence can be placed in themin terms of both internal and external val-idity.

Mechanisms for Goal-Setting Effects

Given that goal setting works, it is relevantto ask how it affects task performance. Weview goal setting primarily as a motivationalmechanism (although cognitive elements arenecessarily involved). The concept of moti-vation is used to explain the direction, am-plitude (effort), and duration (persistence)of action. Not surprisingly, all three are af-fected by goal setting. One additional, in-direct mechanism is also described.

Direction

Most fundamentally, goals direct atten-tion and action. Perhaps the most obviousdemonstration of this mechanism is thestudy by Locke and Bryan (1969a) in whichdrivers were given feedback regarding fivedifferent dimensions of driving performancebut were assigned goals with respect to onlyone dimension. The dimension for which agoal was assigned showed significantly moreimprovement than the remaining dimen-sions. Similarly, Locke et al. (1970) foundthat subjects modified their speed of reaction(to make it faster or slower) on a simplereaction-time task in the direction of theiroverall objective. Reynolds et al. (1979)found that subjects spent more time readingprose passages that were relevant to theirgoals (consisting of questions inserted in thetext) than to reading parts that were notrelevant. Terborg (1976) found that subjectswith specific goals spent a greater percentageof the time looking at the text material tobe learned than did subjects with nonspecificgoals or no goals. (Terborg labeled this mea-sure effort in his study.) Rothkopf and Bil-lington (1979) found that subjects with spe-cific learning goals, as compared withsubjects with no specific learning goals (do-your-best instructions), spent an equal orgreater amount of time inspecting passageswith goal-relevant material and significantlyless time looking at incidental passages.


Effort

Since different goals may require differentamounts of effort, effort is mobilized simul-taneously with direction in proportion to theperceived requirements of the goal or task.Thus, as Kahneman (1973) and Shapira(Note 1) have argued, more effort is ex-pended on hard tasks (which are accepted)than on easy tasks. Sales (1970) found thathigher work loads produce higher subjectiveeffort, faster heart rates, and higher outputper unit time than lower work loads. Lathamand Locke (1975) and Bassett (1979) foundthat people work faster under shorter thanunder longer time limits. In summary, highergoals produce higher performance than lowergoals or no goals because people simply workharder for the former (Locke, 1968; Ter-borg, 1976; Terborg & Miller, 1978; for ear-lier documentation see Locke & Bryan,1966).

This hypothesis of a positive linear rela-tion between motivation or effort and per-formance (also stated in Locke, 1968, andYates & Kulick 1977), contradicts theYerkes-Dodson inverted-U "law," which as-serts that performance is maximal at mod-erate levels of motivation. Although it is truethat with any given subject, performanceeventually will level off as the limit of ca-pacity or ability is reached (Bavelas & Lee,1978; Kahneman, 1973), this is a separateissue from that of motivation. Of course,subjects may abandon their goals if theybecome too difficult, but the hypothesizedfunction assumes goal commitment. Perfor-mance may also drop if subjects becomehighly anxious, especially on a complex orunderlearned task. But a state of high anx-iety should not be labeled high motivationin the positive sense because it represents astate of conflict rather than of single-mindedgoal pursuit.

Persistence

Persistence is nothing more than directedeffort extended over time; thus, it is a com-bination of the previous two mechanisms.Most laboratory experiments on goal settinghave not been designed to allow for the mea-surement of persistence effects, since time

limits typically have been imposed; fieldstudies to date have measured only the endresults of goal setting rather than how theywere obtained. LaPorte and Nath (1976)allowed some subjects unlimited time to reada prose passage. Those asked to read thepassage to get 90% of 20 postreading ques-tions correct spent more time on the passagethan subjects asked to get 25% of the post-reading questions correct. Rothkopf and Bil-lington (1979) found that more time wasspent on goal-relevant than on incidentalpassages. More studies of this type would behighly desirable.

Strategy Development

Whereas the first three mechanisms arerelatively direct in their effects, this lastmechanism is indirect. It involves developingstrategies or action plans for attaining one'sgoals. Although strategy development ismotivated by goals, the mechanism itself iscognitive in essence; it involves skill devel-opment or creative problem solving.

Bandura and Simon (1977), for example,found that dieting subjects with specific quo-tas for number of mouthfuls eaten changedtheir eating patterns (e.g., by eating morelow-calorie foods that did not count in theirquotas). They also engaged in more planning(e.g., by saving part of their quota for a din-ner out). Latham and Baldes (1975) ob-served that some of the truck drivers as-signed specific hard goals with respect totruck weight recommended minor modifi-cations of their trucks to help them increasethe accuracy of their judgments of weight.

In Terborg's (1976) study, the subjectswho set specific goals were more likely toemploy relevant learning strategies (e.g.,writing notes in the margins) than those whodid not set goals. A unique aspect of Ter-borg's (1976) design was that he was ableto obtain separate measures of direction ofeffort (which he called "effort") and of strat-egy use (which he called "direction"). Hefound that when these mechanisms were par-tialed out, there was no relation betweengoals and task performance. This supportsthe argument that these are some of themechanisms by which goals affect perfor-mance.


In a similar vein, Kolb and Boyatzis(1970) found that behavior change in a T-group was greatest for participants who de-veloped plans for evaluating their perfor-mance in relation to their goals. Such plansevidently were developed only for behaviordimensions that the subjects were trying tochange.

Bavelas and Lee (1978) made detailedanalyses in three experiments to determinethe strategies subjects used to attain hardgoals. They found that subjects would fre-quently redefine the task in a way that wouldpermit them to give "looser" or lower qualityanswers. For example, subjects asked to listvery large numbers of "white, hard, edibleobjects" were more likely to list objects thatwere white but not very hard or hard but notvery edible than were subjects given easiergoals. Similarly, with appropriate training,subjects given hard addition goals wouldmore often estimate rather than calculatetheir answers as compared to subjects witheasy goals.

Subjects given hard goals in Rosswork's(1977) study simply wrote shorter sentencesto meet their quota, which was expressed interms of total sentences written. The subjectsin Sales's (1970) study who were given ahigh work load made more errors, presum-ably by lowering their standards, than thosegiven a low work load. Christensen-Szalan-ski (1980) found that subjects who weregiven a short time limit in problem solvingused less complex and less adequate strate-gies than subjects given a longer time limit.Strategy development is especially impor-tant in complex tasks. If the requisite strat-egies are not developed, the increased mo-tivation provided by the goals will not betranslated into effective performance.

We now examine the influence of feed-back, money, and participation on the ef-fectiveness of goal setting.

Knowledge of Results (Feedback)

In early goal-setting studies, attemptswere made to separate the effects of feed-back (i.e., knowledge of results [KR]) fromthe effects of goal setting to determinewhether KR directly influenced performanceor whether its effects were mediated by goal-

KR No KR

1

3

2

4

Specifichard goal

No specificgoal or do-best goal

Figure 1. Model for analyzing goal-KR studies.(KR = knowledge of results.)

setting activity (Locke, 1967; Locke &Bryan, 1968, 1969a, 1969b; Locke, Car-tledge, & Koeppel, 1968). In the most care-fully controlled of these studies, all subjectswith specific goals also received knowledgeof their performance in relation to theirgoals; individuals in the KR conditions re-ceived knowledge of their actual scores pre-sented in such a way as to preclude their usein setting goals. Such knowledge of scoresdid not lead to better performance than noknowledge of scores. The evidence fromthese and related studies indicated thatknowledge of scores was not sufficient to im-prove task performance. However, sincegroups with goals and no KR were not in-cluded, these studies did not test the possi-bility that KR may be a necessary conditionfor goals to affect performance. Few studiesrelevant to this hypothesis had been con-ducted at the time of the Latham and Yukl(1975b) review.

A number of such studies have since beencompleted in both the laboratory and thefield. Figure 1 illustrates the conditions ofinterest. Cell 1 represents specific, hard goalscombined with KR; Cell 2, specific, hardgoals without KR; Cell 3, KR with no spe-cific goals (or do-best goals that are equiv-alent to no assigned goals); and Cell 4, nei-ther specific goals nor KR.

The studies reviewed here included atleast three of the four cells in Figure 1. Table1 summarizes the results of these compari-sons.

Two types of studies are evident in Table1. The first set consists of comparisons be-tween Cells 1, 3, and 4. Consistent withLocke's (1968) mediating hypothesis, thesestudies indicate that although KR alone isnot sufficient to improve performance (3 =


Table 1Studies Comparing the Effects of Goals and KR on Performance

Study

Comparison performed

I vs. 2 1 vs. 3 2 vs. 4 3 vs. 4

Bandura & Simon (1977)Dockstader (Note 3)Latham, Mitchell, & Dossett (1978)Nemeroff & Cosentino (1979)"At Emery Air Freight" (1973)Komaki, Barwick, & Scott (1978)Becker (1978)"Strang, Lawrence, & Fowler (1978)°

1 >1 >1 >1 >

2222

1 >1 >1 >1 >

3333

22222

= 4= 4= 4= 4<4 b

3333

= 4= 4= 4= 4

/Vo(e. KR = knowledge of results. 1 = specific, hard goals combined with KR; 2 = specific, hard goals withoutKR; 3 = KR with no specific goals (or do-best goals); 4 = neither specific goals nor KR." Included both hard and easy goal plus KR conditions. The performance of easy-goal subjects was no better thanthat in the control condition, b Results differed, depending on performance criterion utilized.

4), KR plus goals results in performance in-creases (1 > 3).

In a study of overweight clients in a weightclinic, participants who kept daily recordsof all the food they consumed but did notset goals to reduce food intake did not altertheir eating habits and performed no differ-ently than a control group who kept no rec-ords and set no specific goals (Bandura &Simon, 1977). However, participants whoset goals based on their daily records sig-nificantly decreased food consumption com-pared with the KR-only group.

Dockstader (Note 3) found no apparenteffect of KR alone on the performance ofkey punch operators, but those provided withKR and a performance standard signifi-cantly exceeded their own previous perfor-mance and that of the KR-only group.

Latham et al. (1978) found no differencesbetween engineers and scientists with do-best goals who were provided with feedbackconcerning their performance on certain ap-praisal criteria and those who received nofeedback; however, the subjects who set orwere assigned specific, hard goals in responseto the feedback performed significantly bet-ter than those in the do-best and controlgroups.

Nemeroff and Cosentino (1979) foundthat supervisors who were provided withfeedback concerning their behavior duringperformance appraisal sessions but who didnot use the KR to set specific goals did not

improve subsequent performance. Those su-pervisors who set specific goals in responseto the feedback performed significantly bet-ter on the 12 behaviors for which they setgoals and conducted significantly more suc-cessful appraisal interviews.

This first set of studies demonstrates thatKR without goals is not sufficient to improveperformance (3 = 4), but given KR, goalsare sufficient for performance to be im-proved (1 > 3). Thus, goals seem necessaryfor KR to improve performance.

The second set of studies consists of com-parisons between Cells 1, 2, and 4. In whatwas called a "positive reinforcement" pro-gram ("At Emery Air Freight," 1973), em-ployees in the customer service departmentand on the shipping docks were given agroup-performance goal, progress towardthe goal was posted, and each employee alsokept a personal record of performance. Per-formance levels increased markedly, butwhen KR was removed and self-reports werenot kept, employee performance returned tobaseline levels "or was almost as bad" ("AtEmery Air Freight," 1973, p. 45), eventhough the performance target remained ineffect (1 > 2, 2 = 4).

In another behavior modification program(actually a goals and KR study; see Locke,1980), Komaki, Barwick, and Scott (1978)examined safe behavior in the making andwrapping of pastry products. The authorsintroduced a specific, hard safety goal and


displayed performance results on a graph inview of all the workers. Substantial perfor-mance improvements occurred, but when theKR was eliminated in a reversal phase, per-formance returned to baseline levels.

In a study of residential electricity use,Becker (1978) manipulated specific goalsand KR. Families included in his study rep-resent Cells 1, 2, and 4 of Figure 1; he alsoincluded easy-goal groups with and withoutKR. The only families whose conservationperformance improved significantly frombaseline levels were those with hard goalsplus KR. All other groups performed no bet-ter than a control group. Strang et al. (1978)conducted a laboratory study utilizing a de-sign similar to Becker's (Cells 1, 2, and 4plus the same two easy-goal conditions asabove). Subjects worked on an arithmeticcomputation task. The performance of sub-jects with hard goals and feedback was sig-nificantly better than that of the goals-onlysubjects (1 > 2). Using time to finish as acriterion, there were no differences betweenthe performance of the goals-only subjectsand that of control group subjects (2 = 4).In terms of number of errors, however, thecontrol group's performance was signifi-cantly better than that of the goals-onlygroup (4 > 2), suggesting that goals withoutKR may even inhibit accurate performance.

The results of this second group of studiesindicate that goals without KR are not suf-ficient to improve performance (2 = 4), butgiven goals, KR is sufficient to effect per-formance improvement (1 > 2). Thus, KRseems necessary for goals to affect perfor-mance.

Although her study is not included in Ta-ble 1 because she used a correlational anal-ysis, Erez (1977) was the first to suggest thatKR is a necessary condition for the goal-performance relation. In her laboratorystudy, subjects worked on a number com-parison task. At the end of one performancetrial, they set goals for a second trial. Halfof the subjects were provided with KR at theend of the first trial and half were not. Erezused a multiple regression analysis to iden-tify the unique contribution of the Goal XKR interaction. The regression equation in-cluded Stage 1 performance, the two maineffects variables (goals, KR), and the

Goal X KR interaction. When all four vari-ables were placed in the regression simul-taneously, the interaction effect wassignificant, but beta weights for goals andKR were not significantly different fromzero. The goal-performance correlation inthe KR group was .60 and in the no-KRgroup, .01. These findings led Erez to con-clude that KR is necessary for goals to affectperformance.

Kim and Hamner's (1976) study of goalsand feedback was not included in this anal-ysis because they acknowledged that theirgoals-only group actually may have receivedinformal feedback. Thus, their study onlyincludes two cells: Cell 1, with differentgroups having different amounts and typesof feedback, and Cell 4, which comprised the"before" scores of the various groups. In thisstudy, as in the one by Frost and Mahoney(1976, Task A), providing explicit or fre-quent feedback clearly facilitated perfor-mance.

Integrating the two sets of studies pointsto one unequivocal conclusion: neither KRalone nor goals alone is sufficient to improveperformance. Both are necessary. This viewof goals and feedback as reciprocally depen-dent seems more useful and more accuratethan Locke's (1968) earlier position, whichviewed goals as mediating the effects of feed-back on performance. Together, goals andfeedback appear sufficient to improve taskperformance (given the obvious contextualvariables such as adequate ability and lackof external blocks to performance). Thestudies demonstrate that action is regulatedby both cognition (knowledge) and motiva-tion.

Table 1 demonstrates that not a singlestudy was designed to allow all of the fourpossible comparisons. In other words, nostudy involved a complete 2 X 2 design withKR/no-KR and specific, hard goals/"do-best" goals, or no goals as the variables. Eventhe studies reported did not always involvetotal control over the variables; for example,spontaneous goal setting among KR-onlysubjects was not always prevented. Such acomplete, controlled study is now being con-ducted by two of the present authors. It ispredicted that Cell 1 (see Figure 1) will showbetter performance than the remaining cells,


which should not differ among themselves.This would parallel the results of Becker(1978) and Strang et al. (1978) using KR/no-KR and hard/easy goal conditions.

Other issues remain to be explored re-garding the role of KR. For example, Cum-mings, Schwab, and Rosen (1971) foundthat providing KR can lead to the setting ofhigher goals than not providing KR; this in-dicates that subjects may underestimatetheir capacity without correct informationabout their previous performance. Relatedto this, Greller (1980) found that supervisorsincorrectly estimated the importance of var-ious sources of feedback to subordinates.These issues deserve further study.

One issue that does not seem to deservefurther study is that of feedback as a rein-forcer. The findings and arguments of An-nett (1969), Bandura (1977), and Locke(1977, 1980) speak convincingly against thethesis that feedback conditions behavior. Itseems more useful and valid to treat feed-back or KR as information, the effect ofwhich depends on how it is processed (e.g.,see Locke, Cartledge, & Koeppel, 1968).

A recent article (Ilgen, Fisher, & Taylor,1979) specifies several dimensions alongwhich KR can vary: amount, type, fre-quency, specificity, timing, source, sign, andrecency. Experimental studies of these di-mensions could reveal the most effectiveform in which to provide KR in conjunctionwith goals. Unfortunately, the studies to datehave not been systematic enough to allowany conclusions about these dimensions.

Our major conclusion, that both goals andKR are necessary to improve performance,provides a clear prescription for task man-agement. Not only should specific, hardgoals be established, but KR should be pro-vided to show performance in relation tothese goals. The "At Emery Air Freight"(1973), Komaki et al. (1978), Latham andKinne (1974), and Latham and Baldes(1975) studies emphasize how inexpensivesuch goals-plus-KR programs can be in fieldsettings relative to their benefits.

Monetary Rewards

It is known that money can be a powerfulmotivator of performance. Locke, Feren,

McCaleb, Shaw, and Denny (1980), for ex-ample, found that individual money incen-tives increased worker performance by amedian of 30%. Locke (1968) argued thatgoal setting may be one mechanism by whichmoney affects task performance.

There are several possible ways that thismight occur. First, money could affect thelevel at which goals are set or the level atwhich intentions are established. In five stud-ies, Locke, Bryan, and Kendall (1968) foundthat in some cases, money did affect goal orintention level. Furthermore, in line with themediating hypothesis, goals and intentionsaffected performance even when the effectsof incentives were partialed out, whereas in-centives were unrelated to performance whengoal and intention level were controlled.

Generally these results have not been rep-licated. For example, Pritchard and Curtis(1973) found that although there was nodifference in the performance effects of noincentive versus a small incentive, subjectswho were offered high incentives performedbetter on a sorting task than those offeredsmall or no incentives even when goal levelwas controlled. Similarly, Terborg (1976)found that partialing out the effects of self-set goals in a programmed learning taskfailed to vitiate the difference between con-tingent and noncontingent pay on perfor-mance. Terborg and Miller (1978) foundsimilar results using a toy assembly task,assigned goals, and piece-rate versus hourlypay. Latham et al. (1978) found a significantmain effect for an anticipated monetary bo-nus independent of a significant goal-leveleffect on the job performance of engineersand scientists. In all four of these studies,goals and money had independent effects onperformance. This was also the case in Lon-don and Oldham's (1976) study, althoughtheir incentive effects were not easily inter-pretable. Chung and Vickery (1976) alsofound independent effects for money andgoals (their KR condition included a goal-setting treatment).

A second possibility is that money mightinduce more spontaneous goal setting thanwould occur without incentives. In supportof this hypothesis, Saari and Latham (Note4) found that the introduction of an incentivesystem led mountain beaver trappers to set


specific goals for themselves. However, in thelaboratory studies by Terborg (1976) andTerborg and Miller (1978), incentive pay didnot lead to more specific goal setting thanhourly pay.

A third possibility, which was stressed byLocke (1968), is that rather than increasingthe likelihood of spontaneous goal setting orincreasing the level at which goals are set(an hypothesis that has not yet been fullytested), incentives affect the individual's de-gree of goal commitment. In other words,offering money may arouse the willingnessto expend more effort to attain a given ob-jective than not offering money. In terms ofexpectancy theory, money rewards endowgoal success with a higher valence or valuethan no money. This is our interpretation ofthe results obtained by Latham et al. (1978),London and Oldham (1976), Pritchard andCurtis (1973), Terborg (1976), and Terborgand Miller (1978).

Attempts to measure this commitmenteffect through self reports have not been suc-cessful (e.g., Latham et al., 1978; Pritchard& Curtis, 1973). The whole issue of whygoal commitment measures have not beenrelated to performance in goal-setting re-search will be discussed at length in a latersection of this article.

The effectiveness of money in mobilizingeffort undoubtedly depends on the amountof money offered. Pritchard and Curtis(1973) found an incentive effect only whenthey offered $3 compared with 50$ or nomoney at all for 10 minutes of work. Simi-larly, Rosswork (1977) found a substantialgoal effect but no incentive effect whenschool children were offered up to 6$ foreach sentence composed during two 5-min-ute periods.

The findings indicate that money can af-fect task performance independently of goallevel. The most plausible mechanism for thiseffect appears to be goal commitment, withthe degree of increased commitment de-pending on the amount of the incentive of-fered. Although direct questions regardingcommitment used in several studies do notsupport this interpretation, the fault may liein poor experimental design, poor measures,or poor introspection by subjects (issues wediscuss later). Incentives may also increase

the likelihood of spontaneous goal setting orof setting high goals, but there has not yetbeen enough research to provide support forthese mechanisms.

Participation and Supportiveness

Participation has long been recommendedby social scientists as a means of obtainingemployee commitment to organizationalgoals and of reducing resistance to change.Nevertheless, an extensive review of the par-ticipation in decision-making literature byLocke and Schweiger (1979) found no con-sistent difference in the effectiveness of top-down ("autocratic") decision making anddecisions made with subordinate participa-tion. We specifically review those studiesthat involved participation in goal setting.

Carroll and Tosi (1970) included a mea-sure of perceived participation in goal settingin a questionnaire administered at a manu-facturing firm that had a Management byObjectives program. The results indicatedthat participation did not correlate signifi-cantly with employee perceptions of goalattainment or employee perception of in-creases in effort.

Negative results were also obtained in afield experiment by Ivancevich (1976). Thisstudy compared participative and assignedgoal setting for sales personnel. Goals wereset for each of four quantitative performancecriteria. Although both goal-setting groupsshowed performance increases, no signifi-cant differences in performance were foundbetween the participative and assigned goalconditions.

In a second study, Ivancevich (1977) ob-tained mixed results with maintenance de-partment technicians. Four performancevariables were measured. With regard toservice complaints and costs, the assignedgoal-setting group showed more improve-ment than the participative group; however,for safety the participative goal group per-formed better than the assigned group.There was no significant difference betweenthe two groups in absenteeism.

A possible drawback of these studies isthat goal difficulty levels were not assessedfor the different goal groups. Conceivably,goal difficulty could have been confounded


with the assigned versus participative ma-nipulations.

The following studies all included mea-surements of goal difficulty. In a field ex-periment involving logging crews, Lathamand Yukl (1975a) found that participativegoal setting resulted in higher performancethan assigned goal setting for uneducated(less than 9 years of education) loggers inthe South. The superiority of participativegoal setting may have been due in part tothe higher goals that were set in the partic-ipative rather than the assigned condition.

In a second field experiment, Latham andYukl (1976) found no significant differencesin the performance of typists with partici-pative and assigned goals. Consistent withthese results, there was no difference in thedifficulty levels of the goals in each condi-tion. Both groups, however, improved theirperformance significantly after specific goalshad been set.

Latham et al. (1978) found that engineersand scientists in a participative goal condi-tion set more difficult goals than their peerswho had assigned goals. However, the per-ceptions of goal difficulty did not differ, andno significant differences in goal acceptancewere found between the two goal conditions.The participative and assigned groups didnot differ significantly in performance, al-though only the participative group signifi-cantly outperformed the control group.

These three studies indicate that partici-pation in goal setting may affect perfor-mance through its influence on goal diffi-culty. Thus, if goal difficulty is held constant,participation should not affect performance.Participation may affect performance onlyif it leads to higher goals being set than isthe case when a supervisor assigns them uni-laterally.

Latham and Saari (1979a) systematicallytested this hypothesis in a laboratory studyusing a brainstorming task. Goal difficultylevels were held constant across the partic-ipative and assigned goal conditions. As pre-dicted, no significant differences in perfor-mance were found between the two goalsetting groups. Moreover, no difference ona measure of goal acceptance was found.

Dossett et al. (1979) replicated this find-ing in two field experiments involving testing

and performance appraisal. In the first ex-periment employees who participated in set-ting their goals on a test attained the sameperformance level as individuals who wereassigned goals of the same difficulty level.This same finding was obtained in a secondstudy, which involved setting goals on a per-formance appraisal form.

Hannan (1975), using a simulated creditapplication evaluation task, also found thatassigned and participatively set goals led tothe same level of performance when goallevel was controlled. (There was a smallGoal X Participation interaction, however.)

Likert (Note 5) has pointed out that whenassigned goal setting is effective, it may bebecause the supervisors who assign the goalsbehave in a supportive manner. Latham andSaari (1979b) tested this assumption in asecond laboratory study using a brainstorm-ing task. Goal difficulty again was held con-stant between the participative and assignedgoal groups. However, the supportiveness ofthe experimenter was varied. The results in-dicated that a supportive supervisory styleled to higher goals being set than a nonsup-portive style. It was also found that it tooksignificantly longer to set goals in the par-ticipative goal conditions than in the as-signed conditions because the subjects askedmore questions regarding what answers wereacceptable. Latham and Saari (1979b) con-cluded that the importance of participationin goal setting may be that it not only leadsto the setting of high goals but it can alsolead to increased understanding of how toattain them—two variables that can have adirect impact on performance.

Although few consistent differences intask performance appear between assignedand participatively set goal groups, severaltentative conclusions regarding the influenceof participation can be drawn. There appearto be two possible mechanisms by which par-ticipation could affect task motivation. First,it can lead to the setting of higher goals thanwould be the case without participation, al-though theoretically, assigned goals can beset at any level the supervisor or experi-menter chooses. Second, participation could,in some cases, lead to greater goal accep-tance or commitment than assigned goals.The first effect has been found twice (La-


tham et al., 1978; Latham & Yukl, 1975a).(We discuss the second effect in the sectionon goal acceptance.)

It may be that supportiveness, as discussedin studies by Latham and Saari (1979b),Hall and Hall (1976), and Ivancevich (1974,who called it "reinforcement"), is more cru-cial than participation in achieving goal ac-ceptance. Participation itself, of course, mayentail supportiveness. Other factors, such asthe power of the supervisors and the rewardsand punishments given for goal attainmentand nonattainment, also may be important,but these have not been systematically in-vestigated.

Further, it is possible that the motiva-tional effects of participation are not as im-portant in gaining performance improve-ment as are its cognitive effects. Locke etal. (1980) found that the single most suc-cessful field experiment on participation todate stressed the cognitive benefits; partici-pation was used to get good ideas from work-ers as to how to improve performance effi-ciency (Bragg & Andrews, 1973). Thepotential cognitive benefits of participationare discussed in some detail in Locke andSchweiger (1979) and were implied in theLatham and Saari (1979b) study.

Individual Differences

Thus far we have been discussing goal set-ting as though it affected every individualin the same manner. To date, individual dif-ferences have received minimal attention inthe goal-setting literature, although severalvariables have been examined in one or morestudies.

Demographic Variables

Of the few goal-setting studies that haveinvestigated demographic variables, mosthave dealt with the effects of education,race, and job tenure.

Education. In a study involving elec-tronics technicians, Ivancevich and Mc-Mahon (1977b) found that perceived goalchallenge was significantly related to per-formance only for educated technicians (12years or more of education). In contrast,perceived goal clarity and goal feedback

were significantly related to performanceonly for less educated technicians (fewerthan 12 years of education).

In their field experiment with loggers,Latham and Yukl (1975a) compared as-signed, participative, and do-best goal-set-ting conditions for educated white (12-16years of education) and uneducated black(0-9 years of education) logging crews. Par-ticipative goal setting significantly affectedthe performance of the uneducated crewsbut did not affect the performance of theeducated crews. The goal-setting programmay not have been administered effectivelyin the latter sample, however; in addition,education was confounded with race.

These findings were not replicated in La-tham and Yukl's (1976) field experiment in-volving female typists. In that study edu-cation did not moderate the effects of eitherparticipative or assigned goal setting. Sim-ilarly, Steers (1975) found no moderatingeffect of education on goal setting in a studyof 113 female supervisors.

Although Latham et al. (1978) did notexamine education as a moderator variable,we mention the study here because of theeducation level of the subjects: Goal settinghad a significant effect on the performanceof engineers and scientists with master's anddoctoral degrees.

We must conclude that there is no con-sistent evidence for the effect of educationas a moderator of goal setting, nor is thereany convincing theoretical reason why thereshould be. Goal setting appears to be effec-tive for individuals of all educational levels,ranging from elementary school children(Masters et al., 1977) to loggers with a meaneducation of 7.2 years (Latham & Yukl,1975a) to engineers and scientists (Lathamet al., 1978) with advanced degrees.

Race, As already noted, Latham andYukl (1975a) found that less educated blackloggers who participated in setting theirgoals were more productive and attainedtheir goals more frequently than crews whowere assigned goals by their supervisors ortold to do their best. However, for the moreeducated white loggers there were no sig-nificant differences among the goal-settingconditions.

A study by Ivancevich and McMahon


(1977a) of technicians supported these find-ings. Perceived participation in goal settingwas related to several measures of perfor-mance for black technicians but not forwhites. Goal clarity and feedback were alsorelated to performance for blacks only,whereas goal challenge was related to per-formance for the whites only. Perhaps goalclarity, feedback, and participation affectedthe performance of blacks because, as Ivan-cevich and McMahon (1977a) stated,It has been found that blacks have a higher need forsecurity in performing their jobs.. . . Oneway to derivemore security in a goal setting program is to have goalclarity, receive feedback, and participate in the process,(p. 298)

Clearly more studies are needed before thisinterpretation can be verified. If it is valid,then the racial factor would be reducible toa personality attribute that presumably wouldcut across racial lines.

Job tenure. Five studies have examinedtenure as a moderator variable in the goal-setting process. Three of them (Ivancevich& McMahon, 1977a; Latham & Yukl,1976; Steers, 1975) found no moderatingeffect. Two studies by Dachler and Mobley(1973), found no significant relation betweenstated goals and productivity for short-ten-ure employees (less than 1-2 years), but asignificant relation between these measuresfor long-tenure employees (1-2 or moreyears). Their explanation for this differencewas that longer tenure employees have moreaccurate perceptions of their chances ofreaching various levels of performanceand of performance-outcome contingencies.Nevertheless, it is not clear why it wouldtake 1 or more years for these perceptionsto become accurate. In sum, the evidence todate does not show much promise with re-spect to job tenure as a moderator.

Age. In the study by Ivancevich andMcMahon (1977b) on technicians, age wasnot related to goal setting or performance.To our knowledge no other studies have in-vestigated the moderating effects of age.However, as previously noted, goal settinghas been shown to be effective for children(e.g., Masters et al, 1977; Rosswork, 1977)as well as adults.

Sex. No study has systematically ex-amined sex differences as a moderator of

goal setting, though goal setting has beenshown to significantly increase the perfor-mance of both males (e.g., Ivancevich &McMahon, 1977b; Latham & Yukl, 1975a)and females (Latham & Yukl, 1976; Steers,1975).

Personality Variables

Need for achievement. Steers (1975), inhis study of female supervisors, found thatperformance was related to feedback andgoal specificity only for high-need-achieve-ment individuals. Participation in goal set-ting, on the other hand, was related to per-formance only among low-need-achievementsupervisors. These findings indicate that highneed achievers perform best when they areassigned specific goals and receive feedbackon their progress toward these goals. Con-versely, low need achievers (who are perhapsless confident) perform best when they areallowed to participate in the setting of theirgoals.

In his study using anagrams, Sales (1970)varied the work load given to subjects. Over-all, productivity for subjects high in need forachievement was not higher than that forsubjects low in need for achievement. How-ever, an interaction occurred between workload and need for achievement. Sales re-ported a positive linear relation betweenneed for achievement and productivity in theunderload condition and a curvilinear (in-verted-U) relation between need for achieve-ment and productivity in the overload con-dition. Since high need achievers prefer goalsof moderate difficulty, they presumably con-sidered the overload condition too challeng-ing for their liking.

In a laboratory experiment, Singh (1972)found that students with high need forachievement set higher goals for themselvesover repeated trials of a mathematical cler-ical task than did low need achievers. Yukland Latham (1978) obtained comparableresults in their study involving typists. Highneed achievers who were allowed to partic-ipate in the goal-setting process set moredifficult goals than did low-need-achieve-ment typists, though they did not performany better than low need achievers.

In the two experiments involving word


processing operators, Dossett et al. (1979)found no moderating effects of need forachievement on performance appraisal mea-sures or on performance on a selection testmeasuring mathematical ability. Goal dif-ficulty was not examined in these studiesbecause it was held constant across goal-set-ting conditions, Overall, the results again areinconsistent and unreliable,

Need for independence. An earlier studyby French, Kay, and Meyer (1966) foundthat employees with a high need for inde-pendence had greater goal acceptance whenparticipation in goal setting was increasedthan when participation was reduced or notchanged. Goal acceptance was not affectedby changes in participation for employeeswith a low need for independence.

The moderating effect of need for inde-pendence has not been found by other re-searchers. For example, Searfoss and Mon-czka (1973) found no moderating effect ofneed for independence on the relationshipbetween perceived participation on the partof managers in setting specific budgetarygoals and subsequent motivation to achievethose goals. Similarly, in their study withtypists, Latham and Yukl (1976) found thatneed for independence did not moderate theeffects of either participative or assignedgoal setting on performance. Dossett et al.(1979) also found no moderating effects ofneed for independence on the performanceof word processing operators.

Higher order need strength. Higher or-der need strength is defined as the degree towhich a person desires enriched work (va-riety, autonomy, task identity, and feedback;see Hackman & Lawler, 1971). To ourknowledge, only one study has examined thisneed as a possible moderator of goal setting.

In the study by Ivancevich and McMahon(1977c) involving technicians, initial anal-yses revealed no consistent relationships be-tween various goal attributes and perfor-mance measures. However, when higherorder need strength was used as a moderator,goal clarity, feedback, and challenge wererelated to effort (toward quantity and qual-ity) and attendance for technicians with highhigher order need strength. Conversely, fortechnicians with low higher order needstrength, goal acceptance was related to ef-

fort (toward quality) and attendance. Noobvious interpretation can be made of thisfinding.

Self-esteem. In the study involving typ-ists (Latham & Yukl, 1976), self-esteem didnot moderate the effects of participative andassigned goal setting on performance. How-ever, it was found that self-esteem and goalinstrumentality interacted in their effects onperformance (Yukl & Latham, 1978). In-strumentality was defined as "the extent towhich desirable outcomes (e.g., job security,pay, promotion) are perceived to be contin-gent upon goal attainment" (Yukl & La-tham, 1978, p. 312). Specifically, when goalinstrumentality was low (goal attainmentnot perceived as linked to important out-comes), typists with high self-esteem showedgreater performance improvement than in-dividuals with low self-esteem. There was noself-esteem effect when instrumentality washigh. When self-esteem was low, typists whoperceived high goal instrumentality showedgreater performance improvement than thosewith low goal instrumentality; when self-es-teem was high, there was no instrumentalityeffect. The integrating principle here maybe that people with high self-esteem willwork hard without practical rewards (forpride?), whereas people with low self-esteemwill not.

Carroll and Tosi (1970) found in a cor-relational study that individuals with highself-assurance increased effort in the face ofincreasingly difficult goals, whereas thosewith low self-assurance worked less hard asgoals became harder. It is likely that differ-ent self-perceptions regarding ability under-lie the self-assurance measure.

Dossett et al. (1979) found that word pro-cessing operators with high self-esteem whowere given performance feedback attainedtheir goals significantly more often than in-dividuals with low self-esteem. These resultsare consistent with those of Schrauger andRosenberg (1970), who found that shifts inperformance following feedback depend onthe self-esteem of the individual. Specifi-cally, high self-esteem people improved theirperformance more than low self-esteem peo-ple following positive feedback; the per-formance of low self-esteem individualsdecreased more than high self-esteem


individuals following negative feedback.Thus, high self-esteem individuals are influ-enced more by positives, whereas low self-esteem people are influenced more by neg-atives.

These results are congruent with Kor-man's (1970) thesis, which asserts thatindividuals are motivated to behave in amanner which is congruent with their self-concept. Thus, people respond more to feed-back that agrees with their self-concept,whether it is positive or negative, than theydo to feedback that is inconsistent with theirself-concept.

Internal versus external control. In thestudy of typists (Latham & Yukl, 1976),belief in internal versus external control wasfound to have no moderating effect on per-formance. Dossett et al. (1979) also foundno moderating effects for locus of control onjob performance appraisal measures or ontest performance for word processors. How-ever, Latham and Yukl (1976) found thattypists with participatively set goals whowere "internals" set more difficult goals than"externals."

Conclusions

The only consistent thing about the studiesof individual differences in goal setting istheir inconsistency. A number of reasons forthis can be offered.

First, the studies were not specifically de-signed to look for individual difference ef-fects. The very fact that most studies as-signed goals to the subjects means that anyindividual differences that did exist wereprobably masked by the demand character-istics of the design. When goals are assigned,subjects typically respond to situational de-mands rather than act in accordance withtheir own styles and preferences. The bestdesign for revealing individual differenceswould be one in which there is free (or aconsiderable amount of) goal choice ratherthan assigned goals. Note that the person-ality variables in the goal-setting studies re-viewed previously were most likely to emergein the participative conditions (where thesubject has some input into the decision) orin the self-set goal conditions.

Second, most of the individual difference

variables included in the studies were notbased on any clear theoretical rationale;thus, even when differences were found, theywere hard to explain. Perhaps the most the-oretically plausible of the variables discussedearlier is that of need for achievement. Needfor achievement theory (e.g., McClelland& Winter, 1971) would predict, for example,that people high in need for achievementwould (a) choose moderate goals; and (b)work hardest when probabilities of successwere moderate, when task performance wasin their control, when there was performancefeedback, and when intrinsic rather thanextrinsic rewards were emphasized. Al-though there is some support for these pre-dictions in the need for achievement litera-ture, goal-setting studies have not beendesigned to test them.

The results for self-esteem are also in-triguing. This variable seems worthy of fur-ther study, since it is logical to expect thatone's self-concept would affect the goals onechooses. Self-esteem, of course, must becarefully separated from ability.

Third, there are difficulties with regardto the measures used for assessing person-ality variables. For example, the personalitymeasures used were not consistent acrossstudies. Steers (1975) used the Gough-Heil-brun Adjective Check List (Gough & Heil-brun, 1965) to measure need for achieve-ment, whereas Latham and Yukl (1976)modified a questionnaire developed by Her-mans (1970). Therefore it cannot be deter-mined whether the different results obtainedin these two studies were due to differencesin the measures or in the population. Fur-ther, the reliability and validity of person-ality measures are often inadequate or notreported. In addition, some personality mea-sures were administered after the experi-mental manipulations had taken place. Thisprocedure can result in a confounding of re-sponses to the personality measures with theexperimental treatment.

Fourth, there may be confounding of in-dividual differences in some studies. To drawfirm conclusions regarding an individual dif-ference variable, it must be independent ofother individual difference variables of in-terest. Researchers often do not report theintercorrelations of individual differences,


yet they draw conclusions on various indi-vidual difference variables obtained from thesame sample.

Fifth, many studies report that an indi-vidual difference variable correlates withperformance for people who score high onthat variable but not for those who score low.However, generally no test of significancebetween the two correlations is reported. Toestablish a moderating effect, a test of sig-nificant differences between correlationcoefficients should be made (Zedeck, 1971).

Future research must overcome these dif-ficulties before any clear conclusions can bedrawn regarding the role of individual dif-ferences in goal setting.

Goal Acceptance, Commitment,and Choice

Goal acceptance and commitment aresimilar though distinguishable concepts. Goalcommitment implies a determination to tryfor a goal (or to keep trying for a goal), butthe source of the goal is not specified. Itcould be an assigned goal or a participativelyset goal or a goal that one set on one's own.Goal acceptance implies that one has agreedto commit oneself to a goal assigned or sug-gested by another person. Both acceptanceand commitment presumably can exist invarying degrees. Since most studies haveused assigned goals, the two concepts canoften be used interchangeably.

Most recent studies of goal setting haveused goals as an independent variable. How-ever, since it is assumed that assigned goalsmust be accepted before they will affect taskperformance, it is also relevant to examinethe determinants of goal commitment or ac-ceptance. Generally, attempts to measuredegree of goal commitment in a manner thatwill differentiate between experimentaltreatments and/or relate to task perfor-mance have failed. None of the experimentalconditions in the studies by Latham andSaari (1979a, 1979b), Latham et al. (1978),Yukl and Latham (1978), or Dossett et al.'sStudy 1 (1979) affected self-report measuresof goal acceptance. Dossett et al.'s (1979)Study 2 found an initial difference, with as-signed goals showing greater acceptancethan participatively set goals, a prediction

contrary to expectations. However, this dif-ference disappeared by the end of the ex-periment. Frost and Mahoney (1976), Lon-don & Oldham (1976), Mento et al. (1980,two studies), Oldham (1975), and Yukl &Latham (1978) found no relationship be-tween measures of goal acceptance and per-formance. Organ (1977) found that goalacceptance correlated with performancewithin some of his assigned goal subgroups,but the pattern of correlations was uninter-pretable theoretically.

There are several possible reasons forthese negative results. First, the measures ofgoal acceptance (which consisted typicallyof direct, face-valid questions such as, "Howcommitted are you to attaining the goal?")may not have been valid. Some evidence thatthe measures of goal acceptance may be atfault was obtained in a study by Hannan(1975) in the credit application evaluationtask noted earlier. He measured goal accep-tance not by a rating scale but by the degreeof difference between the subject's external(i.e., assigned or participatively agreed upon)goal and his or her personal goal (as deter-mined from a questionnaire given after ex-ternal goals were set). Hannan found thatparticipation did lead to greater goal accep-tance (though it had no main effect on per-formance) than assigning goals and that theeffects of participation became progressivelystronger as the difficulty of the external goalincreased. The goal acceptance measure wasrelated to one measure of performance. Han-nan also found that personal goals predictedperformance better than assigned goals, asdid Mento et al. (1980). These findings sug-gest that indirect measures of goal accep-tance may be more valid than direct mea-sures.

Second, in most of the studies where ac-ceptance was measured, nearly all subjectsshowed complete or substantial goal com-mitment; thus the range of scores was quitelimited. Small differences on the scales typ-ically used may not reflect genuine differ-ences in psychological states,

Third, due to limitations in introspectiveability, most (untrained) subjects may notbe able to discriminate small differences inpsychological commitment (see Nisbett &Wilson, 1977; but see also Lieberman, 1979,


for a more sanguine view of the usefulnessof introspection). Recall that in the studiesby Latham et al. (1978) and Pritchard andCurtis (1973) described earlier, there ap-peared to be significant commitment effectsfor monetary incentives based on actual per-formance, but these were not reflected in thedirect goal commitment questions.

The solution to the last two problems maybe to modify the design of the typical goal-setting experiment. Designs that encouragea wide range of goal commitment, such asthose with a choice of various possible goals,with commitment to each being measuredafter choice, may reduce the introspectiveburden and increase the variance of the an-swers on the commitment scale. Within-sub-ject designs, which involve assigning differ-ent goals (under different conditions) to thesame subjects at different times, might alsomake the commitment responses more ac-curate by providing a clearer frame of ref-erence for the subject. In addition, when asubject is less than fully committed to agiven goal, it is important to determine whatother goals he or she is committed to. Forexample, a subject who is not fully commit-ted to a moderately difficult goal could betrying for a harder goal, an easier goal, orno specific goal. Each alternative choicewould have different implications for per-formance.

Different degrees of goal commitmentmight be induced by varying types or degreesof social influence (e.g., approval, disap-proval). Such influences undoubtedly haveprofound effects on goal choice and com-mitment among certain individuals, but adetailed discussion of the social—psycholog-ical literature is outside the scope of this re-view.

Goal acceptance or commitment can beconsidered a form of choice, (i.e., the choicebetween accepting or rejecting a goal thatwas assigned or set participatively). In thissense these studies tie in with the more tra-ditional studies of what is called "level ofaspiration," which allowed subjects to freelychoose their own goals after each of a seriesof trials on a task (e.g., see Frank, 1941;Hilgard, 1958). The factors that affect goalacceptance and goal choice are basically thesame. They fit easily into two major cate-

gories, which are the main components ofexpectancy theory (Vroom, 1964).

Expectations of Success

Other things being equal, individuals aremore likely to accept or choose a given goalwhen they have high rather than low expec-tations of reaching it (Mento et al., 1980).Such expectations evidently stem from self-perceptions about ability on the task in ques-tion (Mento et al., 1980). Presumably theseperceptions are inferences from past perfor-mance. Past performance has consistentlybeen found to predict future goals (Cum-mings et al., 1971; Lopes, 1976; Wilsted& Hand, 1974; Ash worth & Mobley, Note2). Individuals are more likely to becomemore confident and to set higher goals aftersuccess and to become less confident and toset lower goals after failure (Lewin, 1958),although failure may lead to higher goals inpressure situations (Forward & Zander,1971; Zander, Forward, & Albert, 1969) oreven due to self-induced pressure (Hilgard,1958). Generalized self-confidence may alsoaffect goal acceptance and choice.

Values

When the perceived value of attaining ortrying for a goal is higher, the goal is morelikely to be accepted than when the perceivedvalue is low (Mento et al., 1980). The valuedoutcomes involved may range from intrinsicrewards like the pleasure of achievement toextrinsic rewards following performance,such as money, recognition, and promotion.Instrumentality in expectancy theory is thebelief that goal acceptance or goal attain-ment will lead to value attainment. Theo-retically, goal choice and goal acceptanceshould be predictable from the expectancies,values, and instrumentalities the subjectholds with regard to the various choices(Dachler & Mobley, 1973).

This is clearly a maximization-of-satisfac-tion model, which is not without its critics(e.g., Locke, 1975). Nevertheless, treatingexpectancy theory concepts as factors thatpredict an individual's goal choices does sug-gest a way of integrating the expectancy andgoal-setting literatures (Dachler & Mobley,1973; Mento et al., 1980).


Although external factors such as rewardsand pressures presumably affect the individ-ual through their effects on expectancies,instrumentalities, and values, it is worth em-phasizing pressures because they have playeda major role in most of the goal-setting stud-ies. For example, the typical laboratorygoal-setting study simply involves asking thesubject to try to reach a certain goal. Thesubject typically complies because of thedemand characteristics of the experiment(probably reducible to beliefs regarding thevalue of extra credit and the desire to helpthe experimenter).

Similarly, in field settings subjects are typ-ically asked to try for goals by their super-visor. The supervisor, of course, is in a po-sition to reward or punish the employee;furthermore, employees know they are beingpaid to do what the organization asks themto do. Ronan, Latham, and Kinne (1973)found that goal setting among woods work-ers was only effective when the supervisorstayed on the job with the employees. Themere presence of the supervisor could beconsidered a form of pressure in this context.In the studies by Forward and Zander(1971) and Zander et al. (1969), competitiveor community pressures led to setting goalsthat were unrealistically high.

Although pressure is something that socialscientists generally have been against, Halland Lawler (1971) argued that if used ap-propriately (e.g., by combining it with re-sponsibility), it can facilitate both high com-mitment and high performance. Pressure, ofcourse, also can be self-imposed as in thecase of the Type A personality who appearsto be a compulsive goal achiever (Friedman& Rosenman, 1974).

Summary, Conclusions, and Directions forFuture Research

Based on the findings to date, the follow-ing conclusions about goal setting seem war-ranted:

1. The beneficial effect of goal setting ontask performance is one of the most robustand replicable findings in the psychologicalliterature. Ninety percent of the studiesshowed positive or partially positive effects.

Furthermore, these effects are found just asreliably in field settings as in the laboratory.

2. There are at least four mechanisms bywhich goals affect task performance: by di-recting attention and action, mobilizing en-ergy expenditure or effort, prolonging effortover time (persistence), and motivating theindividual to develop relevant strategies forgoal attainment. The latter two mechanismsare most in need of further study.

3. Goals are most likely to affect perfor-mance under the following conditions:

Range and type of goals. Individualswith specific and hard or challenging goalsoutperform individuals with specific easygoals, do-best goals, or no assigned goals.People with specific moderate goals showperformance levels between those of peoplewith easy and hard goals but may not per-form better than individuals with do-bestgoals. A common problem with easy-goalsubjects is that their goals are so easy thatonce they are reached, they set new, highergoals to have something to do, which meansthat they are no longer genuine easy-goalsubjects. Perhaps easy-goal subjects shouldbe told not to try to exceed their goals or notto set new goals when the easy goals arereached.

The wider the range of goal difficulty, themore likely goal setting is to affect perfor-mance (cf., Frost & Mahoney, 1976, withLocke et al., 1978). It is probable that longertime spans will progressively increase thedifference between subjects with hard goalsand those without hard goals.

One aspect of goal setting that has notreceived much attention to date is the use-fulness of setting intermediate goals orsubgoals as an aid to attaining longer termor end goals. Locke and Bryan (1967) foundthat on a 2-hour addition task, setting 15-minute subgoals led to slightly poorer per-formance than setting just end goals. Ban-dura and Simon (1977), however, found thatsetting weekly goals for weight loss only ledto weight loss when daily goals (or multiplegoals within days) were set as well. Thereis probably an optimal time span for the set-ting of goals depending on both the individ-ual and the task situation. Subgoals couldconceivably facilitate performance by oper-ating as a feedback device; they might also


serve to maintain effort over long time spans.On the negative side, they might limit per-formance if the subgoals were treated asperformance ceilings. More studies areclearly needed on this topic.

Goal specificity. Goals seem to regulateperformance most predictably when they areexpressed in specific quantitative terms (oras specific intentions to take a certain action,such as quitting a job) rather than as vagueintentions to "try hard" or as subjective es-timates of task or goal difficulty.

Ability. Individuals must have the abilityto attain or at least approach their goals.(In complex tasks they must choose appro-priate strategies, as noted previously.) Ex-erting more effort will not improve task per-formance if improvement is totally beyondthe individual's capacity. Goal-setting stud-ies should carefully control for ability (suchas by a work sample pretest) to isolate thevariance in performance due to goals fromthat due to ability. If ability is not controlled,it becomes error variance when testing fora motivation effect. The most practical wayto set goals may be to base them on eachindividual's ability on the task in questionas measured by a preexperimental worksample. This usually insures ready goal ac-ceptance and makes it easy to control forability when comparing different goals.

Knowledge of results (feedback). Knowl-edge of performance in relation to the goalappears to be necessary if goals are to im-prove performance, just as goals are neces-sary if feedback is to improve performance.Feedback is probably most helpful as an ad-junct to goal setting when the task is dividedinto trials and feedback is provided aftereach one, although the ideal frequency is notknown. Feedforward, telling the subjectshow fast they will need to work on a futuretrial as compared with their speed on an im-mediately preceding trial may be a partialsubstitute in some cases (e.g., see Mento etal., 1980, Study 1). Knowledge and feed-back, of course, may have purely cognitive(learning) effects on performance (see Lockeet al., 1968, for a discussion of this issue),but these are not the concern of this review.Clearly more research is needed on feed-back, especially research based on the issues

raised by Ilgen et al. (1979), such as timing,frequency, source, interpretation, and so on.

Monetary rewards. Money may be aneffective method of improving performancein relation to a given goal (presumablythrough increased commitment), but theamounts involved must be large rather thansmall (e.g., $3 rather than 3£ in a typicallaboratory experiment).

Further research on money and goal set-ting could be tied into Deci's work on in-trinsic and extrinsic motivation. Deci andPorac (1978) suggested that money rewardsthat encourage the attainment of compe-tence on a task (reaching a challenginggoal?) may enhance rather than decreaseinterest in the task.

Participation and supportiveness. Thereis no consistent evidence that participationin setting goals leads to greater goal com-mitment or better task performance thanassigned goals when goal level is controlled,though it sometimes leads to setting highergoals than the supervisor would have as-signed. One study found that participationfacilitated the acceptance of hard goals(Hannan, 1975).

Supportiveness may be more importantthan participation, although this conceptneeds to be defined more clearly. Lathamand Saari (1979b) defined it as friendliness,listening to subjects' opinions about the goal,encouraging questions, and asking ratherthan telling the subject what to do. Moreexploration of the nature and effects of sup-portiveness in goal setting is clearly war-ranted.

Individual differences. No reliable in-dividual difference factors (other than abil-ity) have emerged in the goal-setting liter-ature, probably because most of the studieshave used assigned goals. Thus, situationalconstraints have prevented personal stylesand preferences from affecting performance.In free-choice situations individual person-ality traits may play a more substantial role.Subjects high in need for achievement shouldprefer to set moderate goals, whereas thoselow in this motive should be more likely toset easy or very hard goals. Individuals withhigh self-esteem should be more likely toaccept and try for challenging goals than


those with low self-esteem. However, it is notclear whether a generalized self-esteem mea-sure would show as great an effect as a moretask-specific measure of perceived compe-tence. Mento et al. (1980; based on Motow-idlo, 1976) found that self-perception ofability added unique variance to perfor-mance even when expectancy, valence, andgoal level were controlled.

Goal acceptance and choice. A basic as-sumption of goal setting research is that theindividual accepts (is actually trying for) thegoal that was assigned or was set. Personalgoals usually predict performance betterthan related measures such as assigned (orobjective) goal difficulty or subjective goaldifficulty. Direct measures of goal accep-tance have been found to be generally un-related to either experimental treatments ortask performance. For example, rewardssuch as money may affect performance, withgoal difficulty controlled, even though goalacceptance questions do not indicate in-creased commitment. Indirect measures, suchas the difference between the personal andthe assigned goal, show more promise. How-ever, better experimental designs (e.g.,within-subject designs and designs allowingfree choice of goals) may show effects evenusing direct questions.

Goal choice and acceptance are influencedby numerous factors, including pressure, allof which may work through influencing theindividual's expectancies, values, and per-ceived instrumentalities. Support on the partof higher management for goal-setting pro-grams in organizations seems critical fortheir success, as is the case for most socialscience interventions (e.g., see Hinrichs,1978; Ivancevich, 1974; Woodward, Koss,& Hatry, Note 6). In an organizational con-text support may include insuring or secur-ing the commitment of middle and lowermanagers. It is likely that the degree of con-tinuing support for goal-setting programswill determine the duration of their effects.The Latham and Baldes (1975) study withtruck drivers has continued to be successfulfor the past 7 years (reported in Latham& Locke, 1979, Figure 1, Footnote b).

Other issues. Not mentioned in the abovediscussion was how the type of task affects

goal-setting effectiveness. Obviously, indi-viduals must have some control over taskpace, quality, method, and so on for goalsetting, or any other motivational technique,to affect performance. We do not agree withthose who claim that goal setting might workonly on certain types of tasks. However, itwill undoubtedly be the case that the fourmechanisms noted earlier are differentiallyimportant in different tasks. For example,where more effort leads to immediate re-sults, goals may work as long as they leadthe subject to work harder. On the otherhand, where the task is complex, hard goalsmay only improve performance if they leadto effective strategies.

Regarding the relation of goals to re-wards, an intriguing finding by Masters etal. (1977) was that children who were toldto evaluate their performance after each trialblock while speaking into a tape recorder(e.g., "I did very good [sic];" "I didn't dovery good [j/c]") all reached assymptote onthe task regardless of their assigned goals.Self-reward ultimately vitiated what hadbeen highly significant goal effects. Thisfinding is clearly worthy of future study.

Competition in relation to goal setting alsorequires further study. Both Latham andBaldes (1975) and Komaki et al. (1978)found that goal setting plus feedback led tospontaneous competition among subjects.White et al. (1977) found that telling sub-jects that their performance would be com-pared to that of others ("evaluation appre-hension," in their terminology) had apowerful effect on task performance inde-pendent of a separate goal manipulation.However, spontaneous goal setting withinthe evaluation apprehension condition wasnot measured. It is likely that competitioncould lead people to set higher goals thanthey would otherwise (other people's perfor-mances become the goals) and/or lead togreater goal commitment (Locke, 1968).

Another issue that has not been investi-gated is whether hard goals combined withhigh pressure might lead to a conflict situ-ation and therefore high anxiety. It has beenshown that anxiety disrupts performance oncomplex tasks when it leads subjects toworry rather than concentrate on the task


(Wine, 1971). As noted earlier, conflictsmay also occur among different goals, al-though this has not been studied. Conflictingpressures in goal setting may vitiate theusual goal-performance relationship (For-ward & Zander, 1971). Nor has the issueof individual versus group goal setting re-ceived much attention. (Group goals are dis-cussed in Zander, 1971.)

A final note is in order with respect to thepractical significance of the technique ofgoal setting. In a review of all available ex-perimental field studies of goal setting, Lockeet al. (1980) found that the median improve-ment in performance (e.g., productivity,quality) that resulted from goal setting was16%. In one company the use of goal settingon just one job saved a company $250,000(Latham & Baldes, 1975). Combined withthe use of monetary incentives, Locke et al.(1980) found that goal setting improved per-formance by a median of more than 40%—a finding of great practical significance.

A model for the use of goal setting in fieldsettings has been developed by Latham andLocke (1979). White and Locke (in press)have documented the high frequency withwhich goals actually regulate productivityin business settings. Locke (1978) has ar-gued that goal setting is recognized explicitlyor implicitly in virtually every theory of andapproach to work motivation.

Reference Notes

1. Shapira, Z. Goal difficulty and goal selling as de-terminants of task motivation. Unpublished manu-script, Hebrew University, 1977.

2. Ashworth, D. N., & Mobley, W. H. Relationshipsamong organizational entry performance goals, sub-sequent goals, and performance in a military setting(Tech. Rep. TR-6). Columbia: Center for Manage-ment and Organizational Research, University ofSouth Carolina, July 1978.

3. Dockstader, S. L. Performance standards and im-plicit goal setting: Field testing Locke's assumption.Paper presented at the meeting of the American Psy-chological Association, San Francisco, August 1977.

4. Saari, L. M., & Latham, G. P. Hypotheses on rein-forcing properties of incentives contingent upon per-formance. Unpublished manuscript, University ofWashington, 1980.

5. Likert, R. Personal communication, August 1977.6. Woodward, J. P., Koss, M. P., & Hatry, H. P. Per-

formance targeting in local government.' An exam-ination of current usage, impacts, and implemen-tation factors. Washington, D.C.: Urban Institute,1978.

References

Adam, E. E. Behavior modification in quality control.Academy of Management Journal, 1975, 18, 662-679.

Andrews, F. M., & Farris, G. F. Time pressure andperformance of scientists and engineers: A five-yearpanel study. Organizational Behavior and HumanPerformance, 1972, S, 185-200.

Annett, J. Feedback and human behaviour. Baltimore,Md.: Penguin Books, 1969.

At Emery Air Freight: Positive reinforcement boostsperformance. Organizational Dynamics, 1973, 7(3),41-50.

Bandura, A. Social learning theory. Englewood Cliffs,N.J.: Prentice-Hall, 1977.

Bandura, A., & Simon, K. M. The role of proximalintentions in self-regulation of refractory behavior.Cognitive Therapy and Research, 1977, ;, 177-193.

Bassett, G. A. A study of the effects of task goal andschedule choice on work performance. OrganizationalBehavior and Human Performance, 1979, 24, 202-227.

Bavelas, J. B. Systems analysis of dyadic interaction:Prediction from individual parameters. BehavioralScience, 1978,23, 177-186.

Bavelas, J. B., & Lee, E. S. Effects of goal level onperformance: A trade-off of quantity and quality.Canadian Journal of Psychology, 1978, 32(4), 219-240.

Becker, L. J. Joint effect of feedback and goal settingon performance: A field study of residential energyconservation. Journal of Applied Psychology, 1978,63, 428-433.

Blumenfeld, W. S., & Leidy, T. R. Effectiveness of goalsetting as a management device: Research note. Psy-chological Reports, 1969, 24, 752.

Bragg, J. E., & Andrews, I. R. Participative decision-making: An experimental study in a hospital. Journalof Applied Behavioral Science, 1973, 9, 727-735.

Brass, D. J., & Oldham, G. R. Validating an in-baskettest using an alternative set of leadership scoring di-mensions. Journal of Applied Psychology, 1976, 61,652-657.

Burke, R. J., & Wilcox, D. S. Characteristics of effec-tive employee performance review and developmentinterviews. Personnel Psychology, 1969, 22, 291-305.

Campbell, D. J., & Ilgen, D. R. Additive effects of taskdifficulty and goal setting on subsequent task perfor-mance. Journal of Applied Psychology, 1976, 61,319-324.

Carroll, S. J., Jr., & Tosi, H. L. Goal characteristicsand personality factors in a management-by-objec-tives program. Administrative Science Quarterly,1970, 75, 295-305.

Christensen-Szalanski, J. J. J. A further examinationof the selection of problem-solving strategies: The ef-


fects of deadlines and analytic aptitudes. Organiza-tional Behavior and Human Performance, 1980, 25,107-122.

Chung, K. H., & Vickery, W. D. Relative effectivenessand joint effects of three selected reinforcements ina repetitive task situation. Organizational Behaviorand Human Performance, 1976, 16, 114-142.

Cummings, L. L., Schwab, D. P., & Rosen, M. Per-formance and knowledge of results as determinantsof goal setting. Journal of Applied Psychology, 1971,55, 526-530.

Dachler, H. P., & Mobley, W. H. Construct validationof an instrumentality-expectancy-task-goal model ofwork motivation: Some theoretical boundary condi-tions. Journal of Applied Psychology, 1973, 58, 397-418. (Monograph)

Deci, E. L., & Porac, J. Cognitive evaluation theory andthe study of human motivation. In M. R. Lepper &D. Greene (Eds.), The hidden costs of reward. Hills-dale, N.J.: Erlbaum, 1978.

Dossett, D. L., Latham, G. P., & Mitchell, T. R. Theeffects of assigned versus participatively set goals,KR, and individual differences when goal difficultyis held constant. Journal of Applied Psychology,1979, 64, 291-298.

Erez, M. Feedback: A necessary condition for the goalsetting-performance relationship. Journal of AppliedPsychology, 1977, 62, 624-627.

Fishbein, M., & Ajzen, I. Belief, attitude, intention, andbehavior: An introduction to theory and research,Reading, Mass: Addison-Wesley, 1975.

Forward, J., & Zander, A. Choice of unattainable groupgoals and effects on performance. OrganizationalBehavior and Human Performance, 1971, 6, 184-199.

Frank, J. D. Recent studies of the level of aspiration.Psychological Bulletin, 1941, 38, 218-226.

French, J. R. P., Kay, E., & Meyer, H. H. Participationand the appraisal system. Human Relations, 1966,19, 3-20.

Friedman, M., & Rosenman, R. H. Type A behaviorand your heart. New York: Knopf, 1974.

Frost, P. J., & Mahoney, T. A. Goal setting and thetask process: I. An interactive influence on individualperformance. Organizational Behavior and HumanPerformance, 1976, 17, 328-350.

Gough, H. G., & Heilbrun, A. B. The Adjective Check-list manual. Palo Alto, Calif.: Consulting Psycholo-gists Press, 1965.

Greller, M. M. Evaluation of feedback sources as a func-tion of role and organizational level. Journal of Ap-plied Psychology, 1980, 65, 24-27.

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

Hall, D. T., & Foster, L. W. A psychological successcycle and goal setting: Goals, performance, and at-titudes. Academy of Management Journal, 1977, 20,282-290.

Hall, D. T., & Hall, F. S. The relationship betweengoals, performance, success, self-image, and involve-ment under different organizational climates. Journalof Vocational Behavior, 1976, 9, 267-278.

Hall, D. T., & Lawler, E. E. Job pressures and research

performance. American Scientist, 1971, 59(1), 64-73.

Hamner, W. C, & Harnett, D. L. Goal-setting, per-formance and satisfaction in an interdependent task.Organizational Behavior and Human Performance,1974,72, 217-230.

Hannan, R. L. The effects of participation in goal set-ting on goal acceptance and performance: A labo-ratory experiment. Unpublished doctoral dissertation,University of Maryland, 1975.

Hermans, H. J. M. A questionnaire measure of achieve-ment motivation. Journal of Applied Psychology,1970, 54, 353-363.

Hilgard, E. R. Success in relation to level of aspiration.In C. L. Stacey & M. F. DeMartino (Eds.), Under-standing human motivation. Cleveland, Ohio: How-ard Allen, 1958.

Hinrichs, J. R. Practical management for productivity.New York: Van Nostrand Reinhold, 1978.

Ilgen, D. R., Fisher, C. D., & Taylor, M. S. Conse-quences of individual feedback on behavior in orga-nizations. Journal of Applied Psychology, 1979, 64,349-371.

Ivancevich, J. M. Changes in performance in a man-agement by objectives program. Administrative Sci-ence Quarterly, 1974, 19, 563-574.

Ivancevich, J. M. Effects of goal setting on performanceand job satisfaction. Journal of Applied Psychology,1976, 61, 605-612.

Ivancevich, J. M. Different goal setting treatments andtheir effects on performance and job satisfaction.Academy of Management Journal, 1977, 20, 406-419.

Ivancevich, J. M., & McMahon, J. T. Black-white dif-ferences in a goal-setting program. OrganizationalBehavior and Human Performance, 1977, 20, 287-300. (a)

Ivancevich, J. M., & McMahon, J. T. Education as amoderator of goal setting effectiveness. Journal ofVocational Behavior, 1977, 11, 83-94. (b)

Ivancevich, J. M., & McMahon, J. T. A study of task-goal attributes, higher order need strength, and per-formance. Academy of Management Journal, 1977,20, 552-563. (c)

Kahneman, D. Attention and effort. Englewood Cliffs,N.J.: Prentice-Hall, 1973.

Kaplan, R., & Rothkopf, E. Z. Instructional objectivesas directions to learners: Effect of passage length andamount of objective-relevant content. Journal of Ed-ucational Psychology, 1974, 66, 448-456.

Kim, J. S., & Hamner, W. C. Effect of performancefeedback and goal setting on productivity and satis-faction in an organizational setting. Journal of Ap-plied Psychology, 1976, 61, 48-57.

Kolb, D. A., & Boyatzis, R. E. Goal-setting and self-directed behavior change. Human Relations, 1970,23, 439-457.

Komaki, J., Barwick, K. D., & Scott, L. R. A behavioralapproach to occupational safety: Pinpointing and rein-forcing safe performance in a food manufacturingplant. Journal of Applied Psychology, 1978, 64, 434-445.

Korman, A. K. Toward a hypothesis of work behavior.Journal of Applied Psychology, 1970, 54, 31-41.


LaPorte, R. E., & Nath, R. Role of performance goalsin prose learning. Journal of Educational Psychol-ogy, 1976,65, 260-264.

Latham, G. P., & Baldes, J. J. The "practical signifi- •cance" of Locke's theory of goal setting. Journal ofApplied Psychology, 1975,<50, 122-124.

Latham, G. P., & Kinne, S. B., III. Improving job per-formance through training in goal setting. Journal ofApplied Psychology, 1914,59, 187-191.

Latham, G. P., & Locke, E. A. Increasing productivitywith decreasing time limits: A field replication of Par-kinson's law. Journal of Applied Psychology, 1975,60, 524-526.

Latham, G. P., & Locke, E. A. Goal-setting: A moti-vational technique that works. Organizational Dy-namics, 1979,5(2), 68-80.

Latham, G. P., Mitchell, T. R., & Dossett, D. L. Im-portance of participative goal setting and anticipatedrewards on goal difficulty and job performance. Jour-nal of Applied Psychology, 1978,65, 163-171.

Latham, G. P., & Saari, L. M. The effects of holdinggoal difficulty constant on assigned and participa-tively set goals. Academy of Management Journal,1979, 22, 163-168. (a)

Latham, G. P., & Saari, L. M. Importance of supportiverelationships in goal setting. Journal of Applied Psy-chology, 1979, 64, 151-156. (b)

Latham, G. P., & Yukl, G. A. Assigned versus partic-ipative goal setting with educated and uneducatedwoods workers. Journal of Applied Psychology, 1975,60, 299-302. (a)

Latham, G. P., & Yukl, G. A. A review of research onthe application of goal setting in organizations. Acad-emy of Management Journal, 1975, 18, 824-845.(b)

Latham, G. P., & Yukl, G. A. Effects of assigned andparticipative goal setting on performance and job sat-isfaction. Journal of Applied Psychology, 1976, 61,166-171.

Lewin, K. Psychology of success and failure. In C. L.Stacey & M. F. DeMartino (Eds.), Understandinghuman motivation. Cleveland, Ohio: Howard Allen,1958.

Lieberman, D. A. Behaviorism and the mind: A (lim-ited) call for a return to introspection. American Psy-chologist, 1979, 34, 319-333.

Locke, E. A. Motivational effects of knowledge of re-sults: Knowledge or goal setting? Journal of AppliedPsychology, 1967, 51, 324-329.

Locke, E. A. Toward a theory of task motivation andincentives. Organizational Behavior and Human Per-formance, 1968, 3, 157-189.

Locke, E. A. Purpose without consciousness: A contra-diction. Psychological Reports, 1969, 25, 991-1009.

Locke, E. A. Critical analysis of the concept of causalityin behavioristic psychology. Psychological Reports,1912,31, 175-197.

Locke, E. A. Personnel attitudes and motivation. AnnualReview of Psychology, 1975, 26, 457-480.

Locke, E. A. The myths of behavior mod in organiza-tions. Academy of Management Review, 1977, 2,543-553.

Locke, E. A. The ubiquity of the technique of goal set-ting in theories of and approaches to employee mo-

tivation. Academy of Management Review, 1978, 3,594-601.

Locke, E. A. Latham versus Komaki: A tale of twoparadigms. Journal of Applied Psychology, 1980, 65,16-23.

Locke, E. A., & Bryan, J. F. Cognitive aspects of psy-chomotor performance: The effects of performancegoals on level of performance. Journal of AppliedPsychology, 1966, 50, 286-291.

Locke, E. A., & Bryan, J. F. Performance goals as de-terminants of level of performance and boredom.Journal of Applied Psychology, 1967,57, 120-130.

Locke, E. A., & Bryan, J. F. Goal-setting as a deter-minant of the effect of knowledge of score on perfor-mance. American Journal of Psychology, 1968, 81,398-406.

Locke, E. A., & Bryan, J. F. The directing function ofgoals in task performance. Organizational Behaviorand Human Performance, 1969, 4, 35-42. (a)

Locke, E. A., & Bryan, J. F. Knowledge of score andgoal level as determinants of work rate. Journal ofApplied Psychology, 1969, 53, 59-65. (b)

Locke, E. A., Bryan, J. F., & Kendall, L. M. Goals andintentions as mediators of the effects of monetary in-centives on behavior. Journal of Applied Psychology,1968, 52, 104-121.

Locke, E. A., Cartledge, N., & Knerr, C. S. Studies ofthe relationship between satisfaction, goal setting, andperformance. Organizational Behavior and HumanPerformance, 1970, 5, 135-158.

Locke, E. A., Cartledge, N., & Koeppel, J. Motivationaleffects of knowledge of results: A goal-setting phe-nomenon? Psychological Bulletin, 1968, 70, 474-485.

Locke, E. A., Feren, D. B., McCaleb, V. M., Shaw,K. N., & Denny, A. T. The relative effectiveness offour methods of motivating employee performance.In K. Duncan, M. Gruneberg, & D. Wallis (Eds.),Changes in working life. New York: Wiley, 1980.

Locke, E. A., Mento, A. J., & Katcher, B. L. The in-teraction of ability and motivation in performance:An exploration of the meaning of moderators. Per-sonnel Psychology, 1978, 31, 269-280.

Locke, E. A., & Schweiger, D. M. Participation in de-cision-making: One more look. In B. M. Staw (Ed.),Research in organizational behavior (Vol. 1). Green-wich, Conn.: JAI Press, 1979.

London, M., & Oldham, G. R. Effects of varying goaltypes and incentive systems on performance and sat-isfaction. Academy of Management Journal, 1976,19, 537-546.

Lopes, L. L. Individual strategies in goal-setting. Or-ganizational Behavior and Human Performance,1976, 15, 268-277.

Masters, J. C., Furman, W., & Barden, R. C. Effectsof achievement standards, tangible rewards, and self-dispensed achievement evaluations on children's taskmastery. Child Development, 1977, 48, 217-224.

McClelland, D. C., & Winter, D. G. Motivating eco-nomic achievement. New York: Free Press, 1971.

Meichenbaum, D. Cognitive-behavior modification. NewYork: Plenum Press, 1977.

Mento, A. J., Cartledge, N. D., & Locke, E. A. Mary-land vs. Michigan vs. Minnesota: Another look at therelationship of expectancy and goal difficulty to task


performance. Organizational Behavior and HumanPerformance, 1980, 25, 419-440.

Migliore, R. H. MBO: Blue collar to top executive.Washington, D.C.: Bureau of National Affairs, 1977.

Miller, G. A., Galanter, E., & Pribram, K. H. Plansand the structure of behavior. New York: Holt, 1960.

Miller, H. E., Katerberg, R., & Hulin, C. L. Evaluationof the Mobley, Homer, and Hollingsworth model ofemployee turnover. Journal of Applied Psychology,1979, 64, 509-517.

Mobley, W. H., Homer, S. O., & Hollingsworth, A. T.An evaluation of precursors of hospital employee turn-over. Journal of Applied Psychology, 1978, 63, 408-414.

Mobley, W. H., Hand, H. H., Baker, R. L., & Meglino,B. M. Conceptual and empirical analysis of militaryrecruit training attrition. Journal of Applied Psy-chology, 1919,64, 10-18.

Mossholder, K. W. Effects of externally mediated goalsetting on intrinsic motivation: A laboratory experi-ment. Journal of Applied Psychology, 1980,65, 202-210.

Motowidlo, S. J. A laboratory study of the effects ofsituational characteristics and individual differenceson task success and motivation to perform a numer-ical task. Unpublished doctoral dissertation, Univer-sity of Minnesota, 1976.

Motowidlo, S., Loehr, V., & Dunnette, M. D. A labo-ratory study of the effects of goal specificity on therelationship between probability of success and per-formance. Journal of Applied Psychology, 1978, 63,172-179.

Nemeroff, W. F., & Cosentino, J. Utilizing feedbackand goal setting to increase performance appraisalinterviewer skills of managers. Academy of Manage-ment Journal, 1979, 22, 566-576.

Ness, R. G., & Patton, R. W. The effect of beliefs onmaximum weight-lifting performance. CognitiveTherapy and Research, 1979, 3, 205-211.

Nisbett, R. E., & Wilson, T. D. Telling more than wecan know: Verbal reports on mental processes. Psy-chological Review, 1977, 84, 231-259.

Odiorne, G. S. MBO: A backward glance. BusinessHorizons, October 1978, pp. 14-24.

Oldham, G. R. The impact of supervisory characteristicson goal acceptance. Academy of Management Jour-nal, 1975, 18, 461-475.

Oldham, G. R. The motivational strategies used by su-pervisors: Relationships to effectiveness indicators.Organizational Behavior and Human Performance,1976, 15, 66-86.

Organ, D. W. Intentional vs arousal effects of goal-set-ting. Organizational Behavior and Human Perfor-mance, 1977, 18, 378-389.

Pritchard, R. D., & Curtis, M. I. The influence of goalsetting and financial incentives on task performance.Organizational Behavior and Human Performance,1973, 10, 175-183.

Rand, A. Introduction to objectivist epistemology. NewYork: The Objectivist, 1967.

Reynolds, R. E., Standiford, S. N., & Anderson, R. C.Distribution of reading time when questions are askedabout a restricted category of text information. Jour-nal of Educational Psychology, 1979, 77, 183-190.

Ronan, W. W., Latham, G. P., & Kinne, S. B., III.Effects of goal setting and supervision on worker be-havior in an industrial situation. Journal of AppliedPsychology, 1973, 58, 302-307.

Rosswork, S. G. Goal setting: The effects on an aca-demic task with varying magnitudes of incentive.Journal of Educational Psychology, 1977, 69, 710-715.

Rothkopf, E. Z., & Billington, M. J. A two-factor modelof the effect of goal-descriptive directions on learningfrom text. Journal of Educational Psychology, 1975,67, 692-704.

Rothkopf, E. Z., & Billington, M. J. Goal-guided learn-ing from text: Inferring a descriptive processing modelfrom inspection times and eye movements. Journalof Educational Psychology, 1979, 71, 310-327.

Rothkopf, E. Z., & Kaplan, R. Exploration of the effectof density and specificity of instructional objectiveson learning from text. Journal of Educational Psy-chology, 1972, 63, 295-302.

Ryan, T. A, Intentional behavior: An approach to hu-man motivation. New York: Ronald Press, 1970.

Sales, S, M. Some effects of role overload and role un-derload. Organizational Behavior and Human Per-formance, 1970, 5, 592-608.

Schrauger, J. S., & Rosenberg, S. F. Self-esteem andthe effects of success and failure feedback on perfor-mance. Journal of Personality, 1970, 38, 404-417.

Searfoss, D. G., & Monczka, R. M. Perceived partici-pation in the budget process and motivation to achievethe budget. Academy of Management Journal, 1973,16, 541-554.

Singh, J. P. Some personality moderators of the effectsof repeated success and failure on task-related vari-ables. Unpublished doctoral dissertation, Universityof Akron, 1972.

Steers, R. M. Task-goal attributes, n achievement, andsupervisory performance. Organizational Behaviorand Human Performance, 1975, 13, 392-403.

Steers, R. M., & Porter, L. W. The role of task-goalattributes in employee performance. PsychologicalBulletin, 1974, 81, 434-452.

Strang, H. R., Lawrence, E. C. & Fowler, P. C. Effectsof assigned goal level and knowledge of results onarithmetic computation: A laboratory study. Journalof Applied Psychology, 1978, 63, 446-450.

Taylor, F. W. The principles of scientific management.New York: Norton, 1967. (Originally published,1911.)

Terborg, J. R. The motivational components of goal set-ting. Journal of Applied Psychology, 1976, 61, 613-621.

Terborg, J. R., & Miller, H. E. Motivation, behavior,and performance: A closer examination of goal settingand monetary incentives. Journal of Applied Psy-chology, 1978, 63, 29-39.

Umstot, D. D., Bell, C. H., Jr., & Mitchell, T. R. Effectsof job enrichment and task goals on satisfaction andproductivity: Implications for job design. Journal ofApplied Psychology, 1976, 61, 379-394.

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

Wexley, K. N., & Nemeroff, W. F. Effectiveness ofpositive reinforcement and goal setting as methods of


management development. Journal of Applied Psy-chology, 1975, 60, 446-450.

White, F. M., & Locke, E. A. Perceived determinantsof high and low productivity in three occupationalgroups: A critical incident study. Journal of Man-agement Studies, in press.

White, S. E., Mitchell, T. R., & Bell, C. H., Jr. Goalsetting, evaluation apprehension, and social cues asdeterminants of job performance and job satisfactionin a simulated organization. Journal of Applied Psy-chology, 1977, 62, 665-673.

Wilsted, W. D., & Hand, H. H. Determinants of as-piration levels in a simulated goal setting environmentof the firm. Academy of Management Journal, 1974,17, 172-177.

Wine, J. Test anxiety and direction of attention. Psy-chological Bulletin, 1971, 76, 92-104.

Yales, J. F., & Kulick, R. M. Effort control and judg-

ments. Organizational Behavior and Human Perfor-mance, 1977, 20, 54-65.

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

Zander, A. Motives and goals in groups. New York:Academic Press, 1971.

Zander, A., Forward, J., & Albert, R. Adaptation ofboard members to repeated failure or success by theirorganization. Organizational Behavior and HumanPerformance, 1969, 4, 56-76.

Zedeck, S. Problems with the use of "moderator" vari-ables. Psychological Bulletin, 1971, 76, 295-310.

Received November 29, 1980

goal setting and task performance: 1969-1980 · goal intensity has not been studied as such,...

Documents