,— — W ,7 ,7, 7‘ if , 4. — — 7 ,l I IIIIIII 2833 II m—‘o I I I EFFECTS OF GOAL SETTING AND FINANCIAL INCENTIVES 0N CLERICAL PERFORMANCE Thesis for the Degree of M. A. MICHIGAN STATE UNIVERSITY SYDNEY WILLIAM WHITE 1976 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 31293 10386 6905 ABSTRACT EFFECTS OF GOAL SETTING AND FINANCIAL INCENTIVES 0N CLERICAL PERFORMANCE By Sydney William White The present study tested the effects of goal setting and financial Iincentives on the productivity of individuals employed on a clerical type of task. The objectives were to test (a) whether performance of clerical work is improved by prescribing a goal and/or offering financial incentives, (b) whether the effect on performance of goal setting and financial incentives is additive or interactive, (c) whether performance is affected by the order of presentation of the two incentives in what manner. Forty female undergraduate students were hired through the univer- sity placement service at regular clerical rates. They were individually instructed on an inventory pricing and valuation task and the quantity and quality of their work was recorded during a 15 minute practice period. Each subject was then randomly assigned for a l l/2 hour work period to an experimental condition (N = 8), given one of the five experimental manipulations:G>no goal setting - no financial incentives; IFQOal setting - no financial incentive;GNo goal setting - financial incentivengoal setting followed by financial incentiveigfinancial incentive followed by goal setting. Sydney William White Differences in performance among the experimental conditions were not significant. Results were compared with earlier studies and pos- sible factors contributing to different findings were suggested. Short- comings in research design are discussed and changes which might improve future studies are proposed. Approved by Thesis Committee: ,x} // . / , / Dr. Carl Frost, Chairman ; ,I:?»~w,‘**“/ ,/ Dr. Neal Schmitt Dr. John Wakeley EFFECTS OF GOAL SETTING AND FINANCIAL INCENTIVES ON CLERICAL PERFORMANCE By Sydney William White A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1976 ACKNOWLEDGMENTS I wish to acknowledge the contributions of my committee, which resulted in better planning and execution of this research. Dr. Carl Frost, my chairman, gave unwavering support despite the delays and setbacks involved. Dr. John Wakeley made many comments and suggestions which helped me to think more clearly and work more effectively. Dr. Neal Schmitt gave guidance in correcting methodological errors and finding the right path through the data processing labyrinth. My special thanks go to Dr. Raymond Frankmann for his sound advice on several occasions. The final tribute goes to my wife who not only typed and retyped at great length the various drafts, but, more importantly, was an unfailing source of support and love at each step of the way. ii TABLE OF CONTENTS Page List of Tables ......................... iv INTRODUCTION .......................... 1 Goal Setting Theory .................... l Relation to Financial Incentives ............. 5 Applicability to Industry ................. 7 Plan of Study ....................... l2 Hypotheses ........................ 13 METHOD ............................. 16 Subjects ......................... 16 Task ........................... l6 Experimental Procedure .................. l6 Conditions ......................... l8 RESULTS ............................ 20 DISCUSSION ........................... 22 Comparison of Results With Earlier Research ........ 22 Shortcomings and Possible Improvements .......... 24 APPENDICES ........................... 28 REFERENCES ........................... 37 iii LIST OF TABLES Table Page l Group Means and Standard Deviations for Performance During the Practice and Work Periods ........ 2l iv INTRODUCTION Goal Setting Theory The use of the term "goal setting" as a popular term in performance studies began in the mid-sixties, but the process itself originated much earlier. The expression "level of aspiration“ had been used as early as the l930's (cf. Festinger, l942) to mean the future level of perfor- mance an individual will try to attain. Fryer, (1964) found that sub- jects required to set "levels of aspiration" for themselves before beginning a difficult task performed better than subjects not required to do so. Mace, (T935) reported success in the use of what he called an "intended level of achievement" by setting goals for his subjects in terms of the number of correct answers on an arithmetic task. Locke, (l966) set standards for his subjects of above average performance on each task and described these standards as "levels of intended achieve— ment.“ L0cke and Bryan, (l966) referred to performance standards as goals and in the following year used the expression "goal setting" to describe the process of establishing standards, (Locke, 1967, Bryan and Locke, l967). Of more interest than the derivation of goal setting as a term are the explanations advanced for its results. Mace suggested that subjects who knew their scores set standards for themselves even when not instructed to do so. These standards might be a goal of constant improvement or a specific score higher than the last performance. This "intended level of achievement" was further developed by Locke, (1966) l as resulting from the subject's “need for achievement," (McClelland, 1961) which is described as “not part of the individual's conscious desires." Orne, (1962) suggests that the improvement through goal setting results from higher "demand characteristics" of a type which are inherent in most experimental situations because of the implicit demand to improve performance. Latham and Baldes, (l975) found that the performance of logging crews improved immediately upon the assign- ment of a specific hard goal and suggested that The setting of a goal that is both specific and challenging leads to an increase in performance because it makes clear to the individual what he is supposed to do. This in turn may provide the worker with a sense of recognition. achievement and commitment in that he can compare how well he is doing now vs. how well he has done in the past and in some instances, how well he is doing in comparison to others. Thus, the worker is not only incited to expend greater effort, but he may devise better or more creative tactics for attaining the goals. These explanations for goal setting may be valid but do not form a complete theoretical framework. Since goal setting results in behavioral change it might be well to look at areas of psychology most concerned with such changes for an applicable theory. One of these areas is learning, which has been the fbcus of much experimental analysis over a period of years. Perhaps the best known learning theory is that of operant conditioning, which attempts to explain behavior by factors that can be observed, measured and repro— duced. Operant conditioning deals with the large proportion of behavior which is emitted or operant (i.e., operating on the environment). It began with Thorndike's classic experiment with the hungry cat which "learned" to move the latch on its box to get the food outside, (Thorndike, 1911). This led him to postulate the importance of reward in increasing the probability that a behavior would take place. His Law of Effect stated that any act producing satisfaction became more likely to be repeated. Thorndike's principles were further developed by Skinner, (1938, 1953) as operant conditioning which included the fbllowing definitions and principles: 1. Positive reinforcement means the increase in frequency of a behavior which is followed by a reward (positive reinforcer). Negative reinforcement means the increase in behavior which is followed by removal of a punishment. 2. Positive reinfbrcers include primary or unconditioned reinfor- cers (e.g., food and water) and secondary or conditioned reinforcers (e.g., praise and recognition). 3. Conditioned reinforcers begin as neutral events but become conditioned through repeated pairing with a primary reinforcer (e.g., after several pairings of food with praise, the praise alone can serve as a conditioned reinforcer). 4. Generalized reinforcers (e.g., money) develop from conditioned reinforcers by being associated with a number of other reinforcing events. Primary reinforcers, such as food and water will reinforce behavior without the organism having any previous experience with them, (Reynolds, 1975). Secondary reinforcers acquire power through their temporal association with primary reinforcers over the lifetime of the organism. As they develop, behavior is determined increasingly by the organism's previous experience with these or similar reinforcing conditions. Operant conditioning theory leads naturally to an incentive theory of motivation which postulates that behavior is largely activated by anticipation of reinforcing consequences, (Bandura, 1969). The accomplish- ment of specific goals has become associated in the individual's reinfor- cement history beginning at early stages of development with primary reinforcers and later with social (secondary) reinforcers such as praise, approval and love. As the behavior of goal attainment acquires secondary reinforcing properties, the extrinsic incentives set by parents and teachers may be replaced by a self-set and self-monitored reinforcement system. The individual will also strive to avoid failure, which is associ- ated with the removal of positive reinforcers and with positive punish- ment such as physical and mental duress, deprivation, or perhaps criticism, isolation and other social sanctions. Behavioral theory is consistent with the finding that goal levels are positively correlated with performance levels, (cf. Locke, 1966(a)). In behavioral terms the higher goals may also lead to higher performance because of the individual's past association of failure to meet require- ments with withdrawal of positive reinforcement or even punishment. Knowledge of results (KR) has also been found to have a positive effect on both learning and performance, (Ammons, 1956, Vroom, 1964). Locke and Bryan (1969) found that KR had no effect on performance when goal setting was controlled. However, Cummings, Schwab and Rosen, (1971) found that KR influenced goals set for subsequent performance, and also that correct KR increases goal levels significantly above those set with no KR. They also found that the greater the previous performance, the higher the subsequent goal. These findings are also consistent with the theory that the individual tends to strive for the higher goals which are associated with greater rewards in his reinforcement history. Latham and Kinne, (1974) furnished workers with tables based on a large number of previous workers' production results to aid in setting production goals. They suggested this was valuable because it helped to convince the worker that the goal was attainable and not based either on his own unrealistic expectations or on unreasonable demands of his supervisor. In reinforcement theory terms, information about other workers' production levels produces behavior effects through association with earlier incidents in the individual's reinforcement history. If these incidents resulted in positive reinforcement through the individual's success in meeting or exceeding other workers' production, then they will tend to produce that successful behavior again. On the other hand, if the association of goals assigned by supervisors is with incidents in which the failure to meet goals has resulted in some fOrm of punish- ment, supervisory goal setting will tend to result in avoidance behavior such as absenteeism or quitting the job. Relation to Financial Incentives The present study arose from consideration of two earlier studies which attempted to measure how financial incentives affected performance on tasks for which goals were set. The first of these consisted of five experiments conducted by Locke, Bryan and Kendall, (1968). They based their study on the assumption that goals were the most immediate deter- minants of individual behavior and hypothesized that financial incentives would affect task perfbrmance only by affecting the individual's goals or intentions. Their findings supported this hypothesis by indicating that the same goal level produced the same performance level regardless of whether financial incentives were offered for performance or not. When financial incentives seemed to affect behavior, these behavioral differences could be correlated with equivalent differences in behavioral intentions. When goal or intention differences were controlled or par- tialled out, there was no relationship between the financial incentive condition and behavior. The second study, by Pritchard and Curts, (1973) begins by pointing out that the Locke et a1. findings are in direct contrast with the financial incentive approach to motivation and with expectancy-valence models (cf. Porter and Lawler, 1968). They then draw attention to two possible methodological shortcomings in the Locke et a1. experiments. First, the financial incentives offered were quite small, (ranging from 4/lO¢ to St per item), and may have been "washed out" by other types of rewards such as pleasing the experimenter or feelings of achievement. The second problem is that in some of the experimental conditions, sub- jects were told of the incentive before being asked to set their goals. This raises the possibilitythat because of the prior effect of the in- centive, the actual goal setting would have had little or no direct influence on performance. Pritchard and Curts designed their study to overcome these pos- sible weaknesses. To test whether incentive size was important they offered a 50¢ incentive to one group and a $3 incentive to another. To rule out the possibility that incentives influence subjects in their goal setting and thus their perfbrmance, Pritchard and Curts offered incentives only after the goal had been set. Their results did not confirm the Locke, et al. hypothesis that incentives have no effect on performance outside of their effect on goal setting. They did confirm that goal setting enhances performance, but found that incentives without goal setting can also have positive effects on performance. Moreover, they found that the financial incentive effects are additive and operate independently of the goal setting effects. Applicability to Industry Both the studies referred to above are concerned with the inte- gration of goal setting and financial incentives in industry. Locke et al., suggest that although financial incentives do not appear to influence job performance directly, they may do so in other ways. Incentives may influence the nature of intentions; where the pay off is large the worker is more likely to set a harder goal than where it is small. Incentives may persuade workers to accept assigned goals in industrial situations where such cooperation would not otherwise be forthcoming. Incentives may affect the worker's degree of commitment to his goal, thus increasing his persistence and staying power in the face of difficulties and frus- tration. Locke et al. concluded that incentive programs will not affect behavior automatically; the effect will be a function of the degree to which the individual values money compared with other incentives and the perceived instrumentality of a given behavior in attaining this value. Pritchard and Curts conclude that goal setting procedures should not replace financial incentives. Organizations should use both factors together for maximum positive effect. Before organizational applications are considered, however, at least two questions should be answered. Were the results obtained by Pritchard and Curts different from those of Locke et al. because of the two methodological changes described above, or because of other dif- ferences in experimental methods and conditions? And were the conditions under which the experiments took place similar enough to actual field conditions that practical organizational applications can be considered on the basis of these results? In addition to the changes in the size and order of presentation of financial incentives, there were a number of other differences between the experimental methods and conditions in the two studies. Using Experiment 2 of Locke et al., which had the largest financial incentive, for comparison, the following tabulation contrasts their approach with that of Pritchard and Curts: Locke et al. (Experiment 2)_ Pritchard and Curts SUBJECTS Paid, male, undergraduate students Male and female undergraduate students receiving experimental credit fer participation plus any money earned during the experiment. TASK Individual assembly of a toy from Sorting index cards into stacks boxes of like parts supplied according to information on the card. Errors were eliminated by punching the cards in advance in unique patterns for each possible data configuration and providing corresponding patterns of metal spikes on the sorting board. PROCEDURE AND CONDITIONS Study introduced as experiment in Experiment explained as a study of development of “manual dexterity' information sorting techniques. and 'manipulative abilities.‘ Every attempt was made to avoid making the task appear to be related to valued abilities such as crea- tivity and intelligence. Subject given model assembly and Subject asked to sort 24 practice asked to make two practice assemb- cards, his questions were answered lies. His time was recorded. and he was then given 10 minutes sorting practice to establish a Locke et al._(Experiment 2) Half the subjects were told they would be paid at the rate of 12¢ per assembly for the next 50 minutes. Using his fastest time fer the two practice assemblies, subject was told how many he could make in 50 minutes at that pace. He was then asked to set a goal for number of assemblies and the money he would try to make. The other half of the subjects were told they would earn $3 for participation regardless of performance. They were then asked to set goals in the same manner as the first group. After 25 minutes subjects were given a two minute rest while completed work was counted and subjects set new goals fer the second 25 minutes. Pritchard and Curts baseline performance, then given a two minute rest. Subjects were divided into 5 groups: (1) goal setting-no incentive (2) goal setting-50¢ incentive (3) goal setting-$3 incentive (4; no goal setting-no incentive 5 no goal setting-piece rate incentives Subjects in all goal setting groups were first asked to set a goal of 30% improvement over baseline perb fbrmance. All agreed to do so, although they were not told they would be on incentive. The 50¢ and $3 incentive groups were told they would receive these sums respectively for meeting or exce- eding their goals. Group 5 was told they would receive 3¢ for each 2 cards sorted on a piece rate basis. All groups were then given a second ten minute sorting period. Some of the above differences make it difficult to evaluate the significance of the disparity in results between the two studies. 1) Giving the students experimental credit in the Pritchard and Curts study is a potentially powerful reinforcer which has no counter- part in the Locke et al. experiment. 2) While the Locke et al. study stressed dexterity and manipula- tive ability, and did in fact make use of these skills, the Pritchard and Curts study explicitly avoided any suggestion of ability being required. When intelligence and ability are ruled out for a task, it is probable that the effect of nonmonetary incentives such as need for achievement and desire to please the experimenter will be reduced. 3) The elimination of the error factor from the Pritchard and TO Curts experiment would make it easier for subjects to take advantage of a large monetary incentive to increase their production. This factor also makes the results less realistic in terms of the real world where errors virtually always affect performance. 4) Locke et al. gave their subjects only 2 practice assemblies, told them their rate of production and then asked them to set their own goals; Pritchard and Curts gave their subjects 10 minutes of practice, but then set their goals fbr them without giving them any information about their performance. But knowledge of results has been found to influence positively the level of goals set for subsequent task per- formance (Cummings, Schwab and Rosen, 1971). Thus it is possible that goal setting may have more positive effects on performance in the Locke et a1. experiment than for Pritchard and Curts. To what extent this might modify the results attributed to financial incentives in the latter experiment cannot be measured. Nor can the differences due to experi- menter goal setting vs. self goal setting, although one could speculate that students receiving experimental credit might tend to set higher goals than those who do not. 5) Pritchard and Curts do not comment on the difference in kind between their all-or-none incentive and the Locke et al. piece rate incentive, but imply from their comparisons between the 50¢ and $3 condition that size of incentive alone was significant. But expectancy theory (Porter and Lawler, 1968) predicts that the individual's belief concerning the likelihood that a particular act will be fbllowed by a particular outcome is an important determinant of his effort and sub- sequently his performance. One can assume, therefore, a stronger effect on performance when subjects have high expectations than those with low ll expectations of reaching the task goal. In view of the number and importance of the differences between the two studies it is difficult to reach any firm conclusions regarding their relative merits. I The question whether conditions in the two studies resemble real organizational conditions closely enough to permit consideration of practical applications is easier to answer. Incentive payment systems have many and complex effects on employee behavior. Whyte (1955) flatly predicts that when traditional piece rates are introduced, workers will restrict output well below capacity by pretending to work fast when observed, while actually working slowly. The result will be some piece rates that are too hard, while others are too easy, thus causing endless disturbance in intergroup relations. Marriott, (1971) conc1uded that while in theory incentive systems should increase productivity and benefit workers, employers and customers, actual results have often fallen far short of these goals when production standards could not be attained or maintained because of factors outside the control of the workers, or when output has been deliberately restricted. He has com- piled from a number of sources, including studies by the International Labor Office, a list of 31 basic requirements of sound incentive systems, covering such areas as industrial relations, standards and incentives, quality and supervision. These findings point to the need to consider carefully a number of situational factors before practical applications can be considered, even on a small scale. Since the studies of Locke et al. and Pritchard and Curts dealt with highly artificial conditions, particularly with regard to tasks and methods of payment, it seems doubtful that practical 12 applications can be considered as the result of this research to date. Plan of Stugy The present study tested the effects of goal setting and financial incentives using a clerical task and working conditions which resemble closely those to be found in industrial organizations. This was done by simulating a task on which part-time temporary employees are often employed in industry. By hiring them through the university placement service at the going rates for this kind of work in the community, their conditions of employment resembled more that of the real world and avoided the confounding motivational effects of using volunteers who are given experimental credit for their services. Motivation patterns should tend to follow those of temporary clerks hired for similar jobs in industry. Subjects were instructed by the experimenter as they might be by a supervisor and the task was one'that is performed in many industrial and commercial organizations. Because this kind of clerical work is almost always performed by females, no males were used in the subject sample. At the same time it was realized there are severe restrictions in simulating a real work setting. There was no attempt to pretend that the task was bona fide clerical work. The subjects were told that this was an experiment undertaken in order to determine the effect of different conditions on work performance. The object was to simulate actual work tasks and conditions as much as possible without losing the degree of control of variables which an experimental situation affords. In order to eliminate group incentive effects, subjects were tested one at a time. To reduce subject-experimenter interaction to a minimum, 13 the experimenter left the room as soon as the subject had demonstrated she could perform the task correctly and remained in a separate room fer the duration of the experiment except when administering further job instructions and supplying additional forms. With regard to quality of work, it was felt essential to measure performance by standards similar to those used in industry. Industrial work goals are usually specified in terms of acceptable units produced and incentive bonuses or piece rates are paid only on production which meets quality control standards. For that reason the subjects were instructed that only correct items would be counted as part of their completed work. With regard to goal setting and financial incentives, the study attempted to answer the following questions: 1) Will goal setting improve perfbrmance on a typical clerical task? 2) Does a financial incentive improve performance in the absence of any goals set by the experimenter and when no standards or knowledge of results can be used by the subject to set her own goal? 3) Are the effects of goal setting and financial incentives additive? 4) When both goal setting and financial incentives are used, does the order of presentation of the incentive affect the performance and if so, which is the best order for optimum performance? Hypotheses The hypotheses investigated in the present study are listed below: Hypothesis 1: Goal setting without financial incentives will improve 14 performance. Bryan and Locke, (1967) fOund that setting specific goals fbr a simple addition task resulted in higher production and Locke and Bryan, (1967) found that specific "hard" goals for a range of tasks led to better task performance than when subjects were given only the general instruction to "do your best." Pritchard and Curts, (1973) confirmed these results on a card sorting task. Moreover, reaching a goal is a conditioned reinforcer in itself, (Kazdin, 1975) and should therefore be associated with an increase in the frequency of the work behavior which it follows. Hypothesis 2: Financial incentives for increased production with- out goal setting will improve performance. Expectancy theory, (Vroom, 1964, Porter and Lawler, 1968) predicts that the motivation to perform a task will vary with the individuals' perception of the valence of outcomes and the instrumentality of performance for attainment of these outcomes. Although the subjects were not given specific goals they were assigned general objective of performing better than another group doing the same task without any financial incentive. Money is a power- ful generalized reinforcer, (Skinner, 1953) and should be effective, even when incentive amounts are not known, with individuals whose need for it is relatively high. Hypothesis 3:} When goal setting is followed by an offer of a financial incentive, performance will be improved more than by either goal setting or financial incentive alone, but less than the total improvement for the two variables when they are offered separately. Operant conditioning theory holds that both money and goal attainment are conditioned reinforcers and their effects should therefore tend to be additive. However, since the total potential improvement is limited 15 by each subject's capacity to increase working speed while controlling errors, the performance improvement from the combined incentives will be less than the total of the increases when the two incentives are offered separately. I Hypothesis 4: Financial incentives offered before goal setting commitments are made will result in a greater increase in performance than when they are offered after goals have been set. In this regard, Pritchard and Curts' argument is persuasive that a subject who knows of a financial incentive before the goal is set may be positively influ- enced by it in his goal setting behavior and, subsequently, his performance. METHOD Subjects Subjects for the study were female undergraduate students recruited through the placement center at Michigan State University. They were paid $4.50 for participation at the end of the experiment for approximately two hours work. In addition, subjects working under financial incentive conditions were eligible to receive bonuses as explained below. Ig§H_ The task was similar to a typical job for which temporary part-time clerical workers are hired by business organizations. It consisted of pricing and valuing inventory items on simulated inventory sheets such as those which might be used in taking inventory of stationery and office supplies. Item prices were obtained by the subjects from a separate price list, entered on the inventory sheet opposite the pr0per item and multiplied by the quantityin the "inventory count" column to give the item value when was then entered on the sheet. Experimental Procedure To ensure that each subject received the same initial briefing, she first read a set of instructions. These described the work setting, the ferms to be used and the first two parts of the task, consisting of the learning and practice phases (Appendix A). She was then asked if she understood the instructions and if she had any questions. There were very few questions; when those came they were answered by referring back 16 17 to the instructions or by stating that the question should be answered by the demonstration of the use of the forms which was to follow or by further instructions she would receive at a later stage. When all questions had been dealt with, the subject was given the inventory price list (for sample sheet of price list, see Appendix B) and a sample completed inventory sheet (Appendix C). By reference to the written instructions and the price list, she was shown how the sample sheet had been completed. Any questions were answered and she was given an uncompleted inventory sheet and told to price and value the items on it to demonstrate that she knew how to do the task correctly. At this stage some subjects asked questions about completing the work on the first sheet. These were answered, the completed work was checked by the experimenter and any errors were shown to the subject. Each subject was asked if she believed she could now complete the next stage of timed practice without further help and all of them indicated that they could. For the practice phase each subject was given a small supply of inventory forms (Appendix D), and asked to practice the task in order to familiarize herself with it and improve her rate of job performance. She was asked to do her best with regard to speed, but to remember that only correct items would be counted as part of her completed work. She was told that the experimenter would not be in the room during most of the remaining work time. She was told not to waste time over figures or items she could not understand, but instead go on to the next item. The experimenter then left the subject alone until just before the end of the practice period. Each subject was timed for a 15 minute practice period, following 18 which her work was checked and the gross total, number of errors and net total of completed items were compiled. Subjects were then randomly assigned to one of five conditions for the work phase, as explained below: Condition 1: No Goal Setting:No Financial Incentive Subjects were not told the practice results. They were given written instructions (Appendix El) directing them to continue working at the same task for a further period of 1 1/2 hours. They were asked to do their best with regard to both quality and quantity of work performed. Condition 2: No Goal Setting;Financial Incentive Subjects were not told the practice results. They were given written instructions (Appendix E2) telling them they could earn a bonus during the fbllowing work phase of 1 1/2 hours if they completed more items than the average number completed by another group doing the same work without bonus. For each item in excess of the other group's average they would receive a bonus of 25% of the clerical cost saved. However, the other group's average could not be compiled until all work was completed and therefore the bonus payments could not be computed or made until the end of the experiment. They were asked to do their best with regard to both quality and quantity of work performed. Condition 3: Goal SettinggNo Financial Incentive Subjects were told their practice results. They were asked through written instructions (Appendix E3) and verbally, to commit themselves to a goal of improving their rate of performance by 25% during the following work period of 1 1/2 hours. All of them agreed to this commit- ment which was written down by the experimenter and left with them. 19 Condition 4: Goal Settingyfollowed by Financial Incentive Subjects were told their practice results, then given written instructions (Appendix E4) asking them for the same goal commitment as in Condition 3. Similarly, their commitments were written down and left with them. When goal setting was completed this group was given a secOnd written instruction (Appendix E5) with the same information con- cerning financial incentives as that given to subjects in Condition 2. Condition 5: Financial Incentive Followed by Goal Setting Subjects were not told their practice results initially. They were given a written instruction (Appendix E6) containing the same infbrmation as that furnished to subjects under Condition 2. When subjects agreed they understood the financial incentive, they were told their practice results, then given a second written instruction (Appendix E7) asking them for the same goal commitment as in Condition 3. Again their commitments were written down and left with them. All subjects were timed for the work period of l l/2 hours. They were asked for any comments and suggestions they might have to offer, then paid and allowed to leave. RESULTS Performances during the work periods were compiled for all sub- jects and used to calculate the means and standard deviations for each experimental condition presented in Table 1. Gross figures are the mean total scores for each condition of items completed._ Error figures are the means of incorrect item scores in each condition. Net means reflect the scores for correct performance which are used for further analysis. It is worth noting that the error rate as a percentage of gross mean scores, which is an indication of work quality, remained virtually unchanged from the practice period to the work period. Equivalence of net scores between experimental conditions before introduction of incentive conditions was tested by a one-way analysis of variance among the practice scores and found to be nonsignificant, §_(4, 35) = .73, E.< .58. Net work scores were tested by analysis of covariance, using the net practice scores as covariates. The incentive conditions showed increases from the no-incentive condition but failed to reach signifi- cance, E (4, 34) = 2.57, P_< .055. 20 21 .xwm An cowgoo xgoz goo: pr F ozp Low mogoom Pmcpmmgo as» mcwow>wo za oocwmuao mew: mppamo; oovgoo xcoz ozu .uowgoo mopuougo opacme mp on» yo mupomog on» car: :omwgoosoo oumuwpwomm ope om.o~ Pw.~m om.NN am.o~ ao.o~ om.mp coapaw>ma ataucaom mN.NN mm.¢m e~.¢m mm.o~ ma.e~ mm.F~ omz _m.N om.N om.~ em.P F¢._ N¢.~ mcottm mm.m~ mm.~m oo.~w mm.- Am.me om.¢~ amato memos ozogw m m m w m c nowtma xtoz mm.NF om.mp m~.op _~.om mm.~_ o~.m_ eowpa_>ma egaocapm N¢.No mm._~ Nz.om m~.~m m~.oo m~.mo pmz mm._ m~.N oo.~ mm._ mm._ m~.~ meets“ mm.¢o oo.¢~ N_.No m_.mm om._o oo.oo ameto memos oaogw m m m w m c cowgoa oomuooga m a m N 2 .oz Q=OEo 4_p=ou=_ —ooo am _m_ocmcwm o>wucooc~ o>wucoocu o>pucoo=H oozo_—oz ma oozoppou pmpococwu oz parocmcwu meocmcwu oz m>_p=mo=H u=_uumm -mcwuumm -QCPppmm -mcwopmm pmwocmcmu .mou pmoo poow oz ”moo oz mooFLoa a”to: one mowuooga oz» mcwgoa oucmELomLoa Low mcowump>oo ogoucmpm one memo: quota .p opnmp DISCUSSION Comparison of Results with Earlier Research The results do not confirm any of the findings reported by Locke et al. or Pritchard and Curts. Part of the explanation for the discre- pant findings stems from differences in the type of data used for analysis. Both of the earlier studies used the improvements in perfor- mance from the first to the second trial (gain scores) as the basic data. Use of gain scores for analysis of variance in this study would have yielded significant results, E (4, 35) = 2.65, E,< .05. But some authorities hold that analysis of gain scores is a poorer method than analysis of covariance. Cronhach and Furby (1970) prefer analysis of covariance, stating that there appears to be no need to use measures of change as dependent variables and no virtue in using them. Assuming that errors of measurement of posttest scores are random, these scores constitute an entirely suitable dependent variable. Kirk (1968) and Campbell and Stanley (1966) state that gain scores yield less precise results than the use of analysis of covariance on pretest and posttest scores. Confirmation of results from earlier research was also made more difficult because of the smaller number of subjects employed in the present study. Both Locke et al. and Pritchard and Curts used approxi- mately twice the number of subjects in each condition, thus giving them more powerful tests. The effort to ensure that financial incentives were completely 22 23 separated from goal setting probably also weakened their effect on performance. Subjects were not given any performance level at which bonus payments would start nor any means of calculating or even estima- ting the extra amounts they might earn. This approach contrasts with the piece work and specific goal-linked bonuses offered by Locke et al. and Pritchard and Curts which permitted subjects to anticipate the rewards available for improved performance. The goal setting incentive was also weakened by being completely separated from the financial incentive in order to measure the effect of the order of presentation when both incentives were offered to sub- jects. In the Locke et al. experiment, subjects who were offered a financial incentive had to choose their production goals and their money goals at one time. In the Pritchard and Curts experiment, the subjects were told they could earn the financial incentive only by reaching the minimum goal set for them. This close linkage of incen- tives could be expected to contribute to better performance. Another factor tending to reduce incentive performance in the present study was the arrangement to pay a regular hourly rate to all subjects before they were introduced to any incentives. This was part of the simulation of industrial employment conditions, but it may also have reduced the drive for results demonstrated under the conditions in earlier experiments in which piece-rates or all-or-nothing bonuses constituted the sole financial rewards. The information given subjects in the financial incentive condi- tions that any bonus earnings would not be paid until a later date may also have weakened performance. The purpose of informing them was to ensure that no one worked under the mistaken impression that a bonus 24 might be received at the end of the work trial. But the effect of weakening anticipation of reward would tend to reduce incentive by comparison with the Locke et al. and Pritchard and Curts experiments in which bonuses were paid immediately following the final trial. Shortcomings and Possible Improvements The main purpose of this research was to explore the effects of goal setting and financial incentives when used in combination and to do so under conditions which would give a better indication of their applicability to clerical work in industry. The results indicate that perhaps too much was attempted. In order to control and measure vari- ables, some conditions quite unlike those in industry were imposed, while the simulation of industrial conditions tended to confound the measurement of some variables. This is perhaps best exemplified by the effort to separate goal setting and related factors such as knowledge of results and incentive work standards, from financial incentives. To do so effectively requires the establishment of highly artificial work conditions and the resulting performance has probably little if any applicability to industrial work. Similarly the use of hourly rates of pay to eliminate the artificial course credit incentive and to simulate industrial conditions precludes the proper measurement of performance-contingent financial incentives. An improved approach would require that any questions raised by previous studies be investigated first under purely experimental condi- tions. The outcome could then serve to confirm or reject previous experimental results concerning the nature and effect of goal setting and financial incentives. The question of industrial application of 25 research findings could be investigated later by field studies in which the effect of experimental conditions on subjects' motivation and per- fbrmance could be eliminated. This may be quite important; a number of subjects in the present study spoke of their interest in participa- ting in an experiment and it seems probable that desire to please the experimenter is a powerful incentive in contributing to improvement from pretest to posttest measurements. There is probably no way to eliminate this experimenter effect completely without going to a field study carried out under normal working conditions in which the experi- menter would have no contact with the subjects. There are, however, some possible improvements in goal setting which should be considered and perhaps tested before a field study is attempted. In the present research, although the goal setting level used was based on a pilot study, it is evident from the results that a higher level might have furnished a better incentive. All subjects had a mean increase of 23.4% from practice trial to the work trial and the ten best performers had a mean increase of 48.2%. It might also be better to devise a means by which the subjects can set their own goals instead of accepting an assigned goal. During a pilot test, subjects did not seem able or willing to set their own goals, but this reluctance might be overcome by offering them a choice of several specific goal levels. If available levels ranged from the pretest average performance to the highest performance achieved, the theory that "hard" goals produce greater improvement than "easy" goals could also be tested (Locke and Bryan, 1969). There is also some evidence (Latham and Yukl, 1975), that participative goal setting conditions, in which workers and supervisors discuss and agree on goals, result in 26 improved productivity over assigned or "do your best" conditions. The effect seems to result in part from higher goals being set and in part from greater acceptance of the higher goals by workers. However, the admittedly tentative nature of the explanations for the results of this field study points to the necessity for further experimental investigation of goal setting. With regard to financial incentives, the weaknesses discussed earlier can be largely overcome by recognizing that the method by which bonuses are calculated should be clearly explained to the subjects, even though this may allow them to set personal goals. In industry, work standards and knowledge of results are virtually always available to employees working under financial incentive conditions. They are able to set their own goals and link them directly to desired levels of earnings. Although not significant, the results from this study suggest that the effects of financial incentives may be additive to those 0f goal setting, but that the reverse may not be true. It may be that goal setting followed by financial incentives is more effective because it confbrms to the subjects' own learning patterns. It seems probable that their reinforcement history would contain many experiences in which parents and other authority figures had set goals first and offered financial incentives only after goals were assigned. In relatively few instances would the financial incentive be offered before the goal was set and the individual may not be conditioned to respond to this pattern. Further investigation needs to be undertaken to test for significance of the order of presentation of incentives. The measurement of financial incentives effects could also be 27 facilitated by using several levels of incentive bonuses and observing whether performance tends to vary with the anticipated level of earnings. The use of different bonus levels in conjunction with different goal levels would also yield data which could help establish the optimum combination of the two incentives from the standpoint of performance improvement. APPENDICES APPENDIX A You have been hired to perform a simple clerical job under experi- mental conditions. The instructions which follow give an outline of the setting, the forms to be used and the various phases of the task itself. Please follow them as closely as possible. Setting You will work by yourself in a separate room for approximately two hours. You should leave the room only for personal needs or to consult the experimenter, who will be in a room nearby. You will be furnished with a pen, scratch paper and the forms required for your work. To reduce distraction from hallway traffic it is suggested you leave the room door closed or slightly ajar. 5953; You will use two forms to complete your task: 1. Inventory Price List - comprised of 22 pages of inventory items showing for each item the stock number, description, unit of issue (box-bx; pad-pd., etc.) and price per unit. 2. Inventory Sheets - comprised of 22 sheets with columns for stock number, description, unit of issue, inventory count, price and value. Clerical Task Each inventory sheet has a quantity in the inventory count col- umn for each item. Your first step will be to find the price on the 28 29 inventory price list for an item and enter it on the inventory sheet. Next, you must multiply the quantity times the price to obtain the inventory value, which should then be entered in the value column. A sample sheet, showing in red the figures to be inserted in the clerical operation, has been completed for your guidance. The task is comprised of three phases: 1. Learning Phase: After reading the instructions, you will have the opportunity to ask questions concerning any part of them which you do not fully under- stand. You will then be asked to price and value the items on a speci- men inventory sheet to demonstrate that you can perform the task correctly. You should show this work to the experimenter after comple— ting it as accurately as possible. II. Practice Phase; When you have completed phase I satisfactorily, you will be given a supply of sheets and asked to practice the task for 15 minutes in order to familiarize yourself with it and to improve your rate of performance. You will be asked to do your best with regard to both speed and accuracy of your work. In this phase and phase III which follows, only correct items will be counted as part of your completed work. APPENDIX B INVENTORY PRICE LIST OFFICE AND PAPER SUPPLIES Ihock IUnncfl IhunBOI Doucdpflon Issue Pflco APPOINTMENT HOOKS 160 0016 Ave! noot 00' A! GLANCE A A 6 1/0 BLK EA 2.25 160 0023 APP! noon oAv A1 GLANCE A l 6 1/9 BPOIN (A 2.25 180 0029 0991 900K 00' A1 GLANCE a I 6 7/0 GREEN ' £0 2.08 160 0030 Au»! noon DAV A1 bLAuCE A x 6 7/8 950 EA 2.25 160 00.? APP! A00. uowtw A1 GLANCE 9 A 1 EA 1.25 160 0056 A991 HOOK NAIL SCHOOL CALENDAR JULY 10 JUNE BLUE EA 1.36 160 0069 Ave! 000A UK A1 GLANCE A x 6 7/3 Asst COLORS N076 EA .92 160 0070 Aunt noon UK AT GLANCE a l 6 7/» EA 1.16 160 can. Ave! noon wn/oL 6 I/n I 3 1/6 EA 1.03 160 0095 Ave! noon wA/oL 6 l/nxa 1/6 REFILL EA 1.03 o . ‘5" 'H"‘ 160 0112 ASH tuAvs FEAN BAG BLACK 6 [N 01A EA .83 160 0126 ASN 1~Avs ALAN hAu uLuE 6 1w 01A EA .03 160 0160 Asa IOAVS HEAN nan MHOIN 6 1w 01A EA .53 1A0 015. As" luAvs nEAV HAO GOLD 6 IN 01A EA .03 [60 0168 As" TUAVS BE‘V Nlh GHEEN 6 IN 010 El .83 160 01H? 05H thAvs HFAN «A0 RLD A IN 01A EA .03 ruum_._.m-q-m . HAUGES 160 0195 uAOuE CAHO HOLDEH CLfAu PLST 2 A 3 w/PIN 100/3: ax "IOGF 'NSEH'S 160 0210 AAuut stswi 2 A 3 an! usu SPAutAN SO/PAD 90 .19 160 OPP6 nAJnF NAME SELF-STICK 3 II? I Z [/2 SPAHTY IOOPKG PG 2.50 ”‘05 9A9. D 160 023% «an VAPOR I LB 3 1/246 116 PHN K9011 ZSOIHU “DU '80 ’ .72 160 OPS? ~8H ”491° 2 L* h IIHAT 1/6 HRN x9011 PSO/HO HDU 80 .72 160 026% ~&- wart” 3 La 6 J/«IH 3/. Raw wari 250/50 HDU so .72 |bo o2~o as- VAP[« 5 La 3 l/«All RPM KNA'T PRO/RU "DU 80 .7? 150 020. ‘ min eAutw 10 Lb A 1/(111 3/6 run KHAFT 280/H0 no” 80 .12 160 030» "an PAPOH 20 LB n 1/6115 1/6 nun KRAFT 250/HU HDU BD .12 160 032/ an eww “tum. PAUHU) A wen. we was loo/cs cs lo.h9 160 031" HIM BAPLN Mllleh wountn 0 [/2116 1/2 N06 100/05 (5 6.89 160 0350 «an unfit-I IAlle. HAHN“) 10 112116 NOS Ion/cs CS 6.36 no 0).... «.11. “ten «HIM. 9011.110 12 10:19 N06 SO/Cs CS 6.211 .— 5-1— mwns tau-mm “—9 -l==:===3 I” 1.0 037E huh” UUflfll“ I 116 I l/lh N010 IIGLH A99! 1600 9C< Bl .Jb -Illl=ll====l 14-0 03w “‘4 Bums“! 1 3/6 A 111'. A4012 l/61n Mm: 050 PCS an .36 n—_ 160 0600 -A-‘. Emma.» ( I 1110. N11... 1161.6. ‘91-] 1'50 PCS an .36 ,—- -- 111' LISllkh (UNTIVUTD 0N NEAT PAGE '- _ 111. P06! fl-l 30 APPENDIX C , SAMPLE COMPLETED INVENTORY SHEET OPFTCE AND PAPER surmes Spock M00! Duct-lotion APPOINTMENT BOOKS 160 0016 APPT NOOI DAV AT GLANCE A A 6 7/8 HLI 160 0020 AV»! noon 0Av.A1 GLANCE a A 6 7/9 BPOIN 160 0029 APP! noon 0Av AT bLANCE A A 6 7/0 CATCH 160 0030 AA»! 900A 0Av AT “LANCE A A 6 7/0 0:0 160 0062 A991 900A uowiw A1 GLANCE 9 A 7 160 0050 A991 noon NATL SCHOOL CALENDAR JULY 70 JUNE BLUE 160 0069 ADPT 000A wx AT GLANCE H A 6 TIA ASST COLORS worn 160 0070 Ath noon AA A! GLANCE A l A 7/0 160 0086 val HOOK wA/CL h [In A 3 1/A 160 000A APP! AOOA uA/CL a l/uAJ 1/6 REFTLL - ASA THAYS 160 0112 ASh TUAVS n£A~ hAG ALAcA A IN 01A 160 012A Asa TwAvs MEAN hAG ALnL A IN 01A 160 0160 Asn TDAVS H1A~ hAh HkOdN 6 IN 01A 160 0156 Ash TUATS REAV HA0 GOLD 6 1w 01A 160 Olbh As" TuAvs atAv HAh nutEu A IN uxA 160 01"? As" TpArs uLAv hAb ago A 1A 01A BADGES 160 0196 HADur CARD HOLDEF CLEAw PLST 2 A 3 w/PIN 100/8A «A00! INSEHTS 1A0 0210 nAuoL IASEHT 2 A 3 an! wsu SPARTAN so/PAo 160 0226 “AHGF NAME SELT-STTCA 3 1/2 A 2 1/2 SPAUTV Ioovxn «A05 eApfa 160 021A AAn VAth 1 LB A 1/2An 1/6 PHN KRAFT 250/HD HDU 160 02a; ~An VAPTD 2 L4 ~ 1/HA7 1/. new keArt zsolwn ~00 160 026A ~Ah VArtu 3 LB h J/«Au 316 Run near! 250/30 "DU 160 02~0 6AM v.01. 5 Lu 5 1/.A11 awn KHA‘T PHO/HD now 160 02u~ ~A~ vvau 10 Ln A [/2111 1/6 #9” KRAFT Zso/Hn up” 160 03on «an PAPOH 20 L» n 1/AAIH 1/6 nun AuAfT 250/H0 HUN 160 032/ was 9A9!" “Altlhh Pnuhtn A 1/2A16 1/2 ~03 loo/Cs 160 031a kAu PAwa IAILlhh wAuufH Q 1/2A16 1/2 N06 TOO/Ls 160 0350 ~Av vAP1~ “lllluh PAHHTH 10 112-16 ~05 TOO/CS 160 0).. Han VAvtw ‘Altlhh ”Arvin 12 112119 wan 50/Cs "Anus ”Uflnfiu 160 037v ~~uw awn-ta 1 1/A A 111» wogu IIALH App: 1500 pcc 160 039/ “0%” "unfit“ 1 J/A A Illa N016 ll~ln 09w! 950 PCS |.n 060' ~Aw 91~rn~ / A 1/I~ N016 1/6Lu APHA 250 9C5 -- 111“ llfillnu (unll~usn 0w NTAT PAbf -- 1176 31 thflmof 1000. EA EA EA EA EA EA EA EA EA EA EA EA pD PG BO 50 00 an HO HO CS CS C8 C5 BA HR IAK c~1° °°SH Cwfiflfl‘ VALUE ”‘3 I 2'25 203 I34 72:7 103 73 '23 331 1? 70 AI"? 3H {3 41 33' , APPENDIX D INVENTORY FORM OFFICE AND PAPER SUPPLIES Inna lhunbor Dnncdpflon CALENDAR RASFS 160 0720 CALENDAR HASE-OESK 9L5 A A 7 MIST GREEN 160 0762 CALENnAR AASE-OLSA PLS a A 7 GRAV 160 0790 CALENDAR RASE-OESA PLS n 1/A A 7 3/6 GRAY 160 070A CALENOAA uALL 0A1Lv wLocA 20 1/2 A 27 1 NOIPAGE CALENDAR AFFILLS 160 0812 CALENDAR RLVTLL A A use N0 91 313 1/2 3 160 0626 CALENnAu AtFILL 3 A 3 3/A uro No 51 919 1/2 160 00.0 CALENDAR REFILL 3 3;. A 6 31A are no 55 739 /160 0656 CALENnAn UEFTLL 3 1/2 A A Arc N0 717 1/2 1&0 0360 CALENnAu AEFILL A A 6 3/6 MFG no A6 1/2 160 0082 CALENOAR REFILL 5 A 6 FOR BASE REG N0 658 1/2 .460 0896 CALENDAR REFILL 6 1/. A 10 MFG No ul 610 1/2 CARDS-NAME Ibfl 0°30 TAHLE CARD MARE OR PLACE 2 1/8 A 5 TOO/PKG CARDS-POSTAL 150 0910 CARD POSTAL PLAIN J [/2 I 5 1/2 STAMPED b CENTS CHALK Ibo 0°38 CHILI 26 ASST COLORS 3 [/61315 STKS IAAPCS/hl 1.0 0952 CHALK UNITE “01 OF 16b 1 llélJ/R STKS DUSTLESS CLARRs-vAocw 160 0°66 CLANPs TIN aloe elurn rva CAeAcliv 3/8 IN 160 0040 CCA-Ps (IN .101 PTNCH TYPE CAPACITY 9/16 IN 160 o¢96 CLANDS 11M .106 RINC~ Ives CAPACITY 15/16 1N CLEANERS 160 100” CLEANTR “Ann AATFHLTSS CLFAD LOTIHN I? U! RTL 160 1022 CLEAusw Mann .AtknLrss LnTInN IIPGAL RTL w/nl 160 10;. CLEAHE“ TNI-CflLuDuTNENt GIN paw nrc MACH 1ND soL 160 1050 CLFANT» lvvr.A|1rw A 1/2111 3 s»15/pxo 160 1066 CLEAN"J 1791.31119 rLuin H nl/NIL A/DAquR 160 107A CLEAnsu 1191.A11:u uuubu ) 1/A A | J/A PC 160 1001 CLEANEd 3n «1LT TthHUEAl A01 0/6 1/6 A J 1/0 PAD 166 1007 c11AAru Atunt It" noun a dL/CTN 1176 32 0.1161 In“ 1.... «rte-11'.1 EA EA EA EA PK EA BX 8A EA EA HT dT 6L PG 01 BA CN CT ‘—HW-D—h.'- ““11" ~01: 0.0611 - .04 ..—-. —.—-..-——.--—.-—-—-—~ ”-1—..- ”-6-. —- --—-—--—-—-—— . -m-Ae---- *o—..O-—..—. ---. -——— “an- o~a———- APPENDIX E] 111. Work Phase Following completion of the practice phase you will be asked to continue working at the task for a further period of l l/Z hours. You will be asked to do your best with regard to quality and quantity of work performed. At the end of the work period you will be paid and permitted to leave. APPENDIX E2 III. Work Phase Following completion of the practice phase you will be told to continue working at the task for a further period of l l/2 hours. You will receive additional bonus earnings during the following work phase if you can complete more items than the average number completed by another group which is doing the same work without bonus. For each item you complete in excess of the other group's average, you will receive a bonus of 25% of the clerical cost saved per item. Since the performance of the other group cannot be obtained until all subjects have completed their work, the bonus payments cannot be calculated until all the work is finished. At the end of l l/2 hours work you will be told what your produc- tion has been in terms of completed items. You will be paid for your time and permitted to leave. When the work of the other group is completed, their average production rate will be calculated. If your 33 34 production exceeds this average you will receive payment of the bonus you have earned. APPENDIX E3 III. Work Phase Following completion of the practice phase you will be told how many items you have completed correctly. You will then be asked to commit yourself to increasing your output rate by 25% during the work phase which follows and your commitment will be recorded. At the end of the work period, you will be told how you have performed. You will then be paid and permitted to leave. APPENDIX E4 III. Work Phase Following completion of the practice phase, you will be told how many items you have completed correctly. You will then be asked to commit yourself to increasing your output rate by 25% during the work phase which follows, and the commitment rate will be recorded. APPENDIX E5 III. Work Phase Next, it will be explained that you will receive additional bonus earnings during the following work phase if you can complete more items than the average number completed by another group which is doing the same work without bonus. For each item you can complete in excess of the other group's average you will receive a bonus of 25% of the clerical cost saved per item. 35 Since the performance of the other group cannot be obtained until all subjects have completed their work, the bonus payment cannot be calculated until all the work is finished. At the end of l l/2 hours work you will be told what your produc— tion has been in terms of completed items. You will be paid for your time and permitted to leave. When the work of the other group is com- pleted, their average production rate will be calculated. If your production exceeds this average you will receive payment of the bonus you have earned. APPENDIX E6 III. Work Phase Following completion of the practice phase you will be told to continue working at the task for a further period of l l/2 hours. You will receive additional bonus earnings during the following work phase if you can complete more items than the average number completed by another group which is doing the same work without bonus. For each item you complete in excess of the other group's average, you will receive a bonus of 25% of the clerical cost saved per item. Since the performance of the other group cannot be obtained until all subjects have completed their work, the bonus payments cannot be calculated until all the work is finished. APPENDIX E7 III. Work Phase You will then be told how many items you have completed correctly and asked to commit yourself to increasing your output rate by 25% 36 during the work phase which follows and your commitment will be recorded. At the end of l l/2 hours work you will be told what your produc- tion has been in terms of completed items. You will be paid for your time and permitted to leave. When the work of the other group is completed, their average production rate will be calculated. If your production exceeds this average you will receive payment of the bonus you have earned. REFERENCES REFERENCES Ammons, R. B. Effects of Knowledge of perfbrmance: A survey and tenta- tive theoretical formulation. Journal of General Psychology, 1956, §fl, 279-299. Bandura, A. Principles of behavior modifications. New York: Holt, Rinehart and Winston, l969. Bryan, J. F. & Locke, E. A. Goal setting as a means of increasing motivation. Journal of Applied Psychology, 1967, 51, 274-277. Campbell, D. T. and Stanley, J. C. Experimental and quasi-experimental designs for research. Chicago: Rand, McNally, l966. Cronbach, L. J. & Furby, L. How we should measure "change"--or should we? Psychologjcal Bulletin, l970, 25, 68-80. Cummings, L. L., Schwab, D. P. & Rosen, M. Performance and knowledge of results as determinants of goal setting, Journal of‘Applied Psychology, l97l, 65, 526-530. Festinger, L. A. A theoretical interpretation of shifts of level of aspiration. Psycholpgjcal Review, l942, 42, 235-250. Fryer, F. W. An evaluation of aspiration as a trainingrprocedure. Englewood Cliffs,’N.J.: ‘Prentice-Hall, 1964. Kazdin, A. E. Behavior modification in applied settings. Homewood, Ill.: The Dorsey Press, 1975. Kirk, R. E. Experimental Design: Procedures fbr the behavioral sciences. Belmont, Cal: Brooks/Cole, 1968. Latham, G.P. & Blades, J.J. The "practical significance" of Locke's theory of goal setting. Journal Of Applied Psychology, l975, 60, l22-124. Latham, G.P. & Kinne, S.B. Improving job performance through training in goal setting. Journal of Applied Psychology, l974, 59, l87-l9l. Latham, G.P. & Yukl, G.A. Assigned versus participative goal setting with educated and uneducated woods workers. Journal of Applied Psychology. l975, 69, 299-302. 37 38 Locke, E.A. A closer look at level of aspiration as a training procedure: A reanalysis of Fryer's data. Journal of Applied Psychology, l966, .99, 4l7-420. (a). Locke, E. A. Motivational effects of knowledge of results: knowledge or goal setting? Journal of Applied Psychology, l967, 91, 324-329. Locke, E. A. The relationship of intentions to level of performance. Journal of Applied Psychology, l966, 99, 60-66 (b). Locke, E. A. & Bryan, J. F. Cognitive aspects of psycho-motor perfor- mance: The effects of performance goals on level of performance. Journal of Applied Psychology, l966, 99, 286-29l. Locke, E. A. & Bryan, J. F. Knowledge of score and goal levels as determinants of work rate. Journal of Applied Psychologx, l969. _5_3_9 59-650 . Locke, E. A. & Bryan, J. F. Perfbrmance goals as determinants of level gf pegformance and boredom. Journal of Applied Psychology. l967, . 0-l30. Locke, E. A., Bryan, J. F. 6 Kendall, L. M. Goals and intentions as mediators of the effects of monetary inventives on behavior. Journal of Applied Psycholggy, 1968, pg, l04-l2l. Mace, C. A. Incentives: Some experimental studies. Industrial Health Research Board (Great Britain), 1935, Report No. 72. Marriott, R. Incentive payment systems. London: Staples Press, l97l. McClelland, D.C. The achieving_§ociety. New York: Van Nostrand, l96l. Orne, M. T. On the social psychology of the psychological experiment with particular reference to demand characteristics. American Psychologist, l962, 11, 776-783. Porter, L. V. 8 Lawler, E. E. Managerial attitudes and performance. Homewood, Ill.: Irwin-Dorsey, l968. Pritchard, R.D. & Curts, M.I. The influence of goal setting and fin- ancial incentives on task performance. Organizational Behavior and Human Performance, l973, 19, l75-l83. Reynodls, G. S. A primer of operant conditioning. Glenview, Ill.: Scott, Foresman, T975. Skinner, B. F. The behavior of organisms. New York: Appleton- Century, l938. Skinner, B. F. Science and human behavior. New York: Macmillan, l953. Thorndike, E. L. Animal intelligence. New York: Macmillan, l9ll. 39 Vroom, V. H. Work and motivation. New York: Wiley, 1964. Whyte, w. F. Money and motivation. New York: Harper and Row, l955. HICHIGRN STRTE UNIV. LIBRQRIES ilHllllllllllllli Ill IIIIIIUHIIINI |||1 ill! lllllillilllll Illll 31293103866905