EGO NEED GRATIFICATION. EXTRA-WORK SOCIALIZATION, AND ATTITUDES TOWARD THE JOB Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY ROBERT ALLEN RUH 1970 Th‘ ‘79:. 0-169 Date "' a.“ *1 I l " T. , LIBR A" 6"“ 1 W11 {LIMIT TRY/W W L j i 'mfl‘t . ”pawn—vim- VI "U. This is to certify that the thesis entitled EGO NEED GRATIFICATION, . . ~1ze EXTRA—WORK‘SOCIALIZATION, ‘ } AND ATTITUDES TOWARD THE JOB * presented by Robert A. Ruh has been accepted towards fulfillment of the requirements for» .. __PL_D_-_ degree in_ESM;leogy{ a July 20, 19 LIBRARY BINDERS I Inna-o u..-...-... 7 “Wm-WW. . f . u u“: 1'. L‘a-l Jag?" K W .. J:l,/ ( ' ‘ ' ' ‘ I“ L; Mgfiwkrms LET" ] war-rm ABSTRACT EGO NEED GRATIPICATION, EXTRA-WORK SOCIALIZATION, Am ATTITUDES roman THE JOB By Robert Allen Ruh This study investigated the following hypotheses derived from the theoretical positions of the "human relations" theorists and their critics: l. 3. S. The extent of ego need gratification provided by the job is positively related to enployee attitudes toward the job. The extent of ego need gratification provided by the job is nore positively related to job attitudes for those enployees subject to extra-work socialization which more strongly enphasizes the inportance of such gratification than for those individuals subject to socialization which enphasizes such gratification less strongly. Extra~work socialization fostering internalization of middle class work norns is positively related to the inportance of ego need gratification from the job. The relationship between the extent of ego need gratifi- cation provided by the job and enployee attitudes toward the job is more positive for those individuals for who. such gratification is lore inportant than for those in- dividuals for when such gratification is less inportant. The extent of ego need gratification provided by the job is positively related to the importance of such gratification. The neasures of each of the variables investigated were based on the questionnaire responses of 2755 enployees of six industrial organizations. Level of education, level of father's education, father's occupation, size of con-unity raised in, size of present Robert Allen Ruh community, and sex were used to measure extra~work socialization. Hierarchical level, "collar color," and a self report index were used to measure the extent of ego need gratification provided by the job. Likert scales designed to measure job satisfaction, identification, ego involvement, and commit-cut served as measures of job attitudes. The data were analyzed with analysis of-variance techniques. The results strongly supported the hypothesized positive relation- ships between ego need gratification, employee attitudes toward the job, and the importance of ego need gratification. All 15 of these relationships were in the predicted direction at less than the .001 level of significance. The results were rather inconsistent, however, in relation to hypotheses 2, 3, and 4. In general, the results obtained employing education, father's education, and sex as measures of extra-work socialization supported hypotheses 2 and 3, although not always at conventional levels of statistical significance. The results obtained for the size of community raised in and size of present com-unity measures, however, were consistently opposite in direction to those predicted by these hypotheses. Finally, no support was obtained for the hypothesized interaction between the extent of ego need gratifica- tion provided by the job and the importance of ego need gratification. The results of this study, therefore, provided more support for hypotheses derived from the "human relations" theoretical position than for hypotheses derived tron criticisms of that position. The discussion of these results, however, emphasized possible restrictions on the generality of these results due to luaitations of design and measurement and nonrandom subject sanpling. mo nan cmrulcarmu, mas-m socuuzsnou, m ATTITUDES means In Jos By Robert Allen Ruh A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1970 "1 .._, . .-‘p,.:-~A. b ’- (ix/i) VJ 7""! ACKNOWLEDGMENTS As is usually the case, this dissertation reflects the contri- butions of many people in addition to its author. Richard Ruch, Maurie Hertenberger, and John Rrause were largely responsible for gaining the support and cooperation of the Midwest Scanlon Associates and the member companies which participated in the survey. Responsi- bility for the rather formidable and often tedious tasks of data collection and coding was shared by James Morrison, Douglas Little, Steve Reinen, Dr. Carl Frost, and Dr. John flukeley. The specific advice and recommendations and the general intellectual stimulation provided by the members of my dissertation committee, Drs. Eugene Jacobson, Lawrence Hbssz', and Terrence Allen, were, of course, invaluable. It is customary to reserve "special recognition" for the chairman of one's dissertation committee. This custom is particularly appro- priate in my case. The effect of Dr. Carl Frost's guidance and inspi- ration on the quality of this dissertation is only one example of his immeasurable contribution to my graduate training. The ultimate credit for the completion of this dissertation belongs to my wife, Pam. Her patience and understanding were indispensable, not only for the completion of this dissertation, but throughout my graduate training. ii TABLE OF CONTENTS LIST OF TABLm .0.0..00.0.00...OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO. v mm 1: Imam!“ COOOOOOIOOOOO0.0000......00.00.000.000... 1 Theory OCOOOCOOOOCOOOOO0.000000000000000...OOOOOOOOOOOOOOOOOO 1 Human Relations Theorists .............................. 1 Critics of the Human Relations Theorists ............... 7 Arm1.. Rebutt.l .0OOOOOOOCOOOOOOOOOO00......00.0.0.0..10 Re.E‘rCh OOOOOOOOOOOOOOOOOOOOOOOO0..0......OOOOOCOOOOOOOOOOOOII Leadership Style and Participative Decision Making .....12 Job Complexity ....31 Critical Summary of Previous Research ..................42 Hypotheses .............................................46 mm 11: mm OOOOOCOOOOOCOOOOOOO0......00.0.0000000000000049 Data Collection Procedures ..................................49 Research Sites ..............................................50 Subject Sample ..............................................51 Definition and Measurement of Variables .....................52 Data Coding .55 Data Analysis ...............................................56 own 111: REst .0O...0.0000000000000000.0.0000000000000000058 Hypothesis 1 ................................................58 Hypothesis 2 ................................................58 Hypothesis 3 ................................................69 Hypothesis 4 ................................................7O Hypothesis 5 ................................................79 0mm Iv: Dlsmslm ooo..o......o......................o......81 smry Of Re.u1t. eseeessseeeeoeeeeseeeeseases sssss seeeeoeee81 mt.t1°u eseeeeeeeeeeeeoeeeseeseeseseseoeosseeoeeoeeeeseee85 IfltOrpTOtOtionl .“ conc1u.1°n. ...o....o....o..........ooo..89 BIBLIM .0.0.0.0...OOOOOCOOOOOOOO...00.00.000.000.00.00.000.092 iii APPENDIX A Analysis of Variance Sui-series for Interactions between Ego Heed Gratification and Extra-Work Socialization . . . . . . . . . 101 mmn I Mean Job Attitudes for Ego Heed Gratification by htr.‘w°rk soc1.1tz‘t1°n sseseeoessesseoeeessseeesseeeseeeeesllo APP-IX C Su-aries of Simple Effects Analyses of Variance for Ego Heed Gratification on Job Attitudes within Levels Of htr.-worksoc1‘11'.t1°n 0.0.0....OOOOOOOOOOOOOOOOOOOOOOO.128 APPENDIX D Questionnaire Items used to Measure Extra-Work Socialization, Ego Heed Gratification, the Impor- tance of Ego Heed Gratification, and Attitudes tw.rd the Jab OOOOOOOOOOOOOOOOOOOO0.0.0000...O..0000.000.00.132 APPENDIX E Correlations between Measures of Ego Heed Gratification, Extra-work SOCECIEZOEEOB, Cu Jab Attitllde. - o e e o o o o o e e o o o e o o 136 iv Table Table Table Table Table Table Table Table Table Table Table Table Table 10 ll 12 13 LIST OF TABLES Summary of Analysis of variance for the Relationships between Hierarchical Level (A)8MJ°bAtt1tude. 00.00.000.000.00.000.00.000000059 Summary of Analyses of Variance for the Relationships between Self Report (A) and Job Attitude, OOOOOOOOOO...OOOOOOOOOOOOOIOOOO..0059 Summary of Analyses of Variance for the Relationships between Collar Color (A) and Jab Attitude. OIOOOOOOOOOCOOOOOOOOOOOO00.0.00000060 Mban Job Attitudes by Hierarchical Level.............60 Mean JOb Attitudes by Self Report ...................60 Mean JOb Attitudes by Collar Color ..................61 F and 2,Values for Interactions between Extra-work Socialization and Hierarghigal u!e1 .0OOOOOOOOOOOOOOOOOOOOOOO0....0.0.0.0000000000061 P and p,Values for Interactions between Extra-work Socialization and Self Report ............62 P and 2,Values for Interactions between Extra-Work Socialization and Collar gglo ...........63 Mean Differences in Job Attitudes between High and Low Categories of W within Levels of Extra-work Socialization ...........64 Mean Differences in Job Attitudes between High and Low Categories of e within Levels of Extra-work Socialization ...........65 Mean Differences in Job Attitudes between White and Blue Collar Groups within Levels of Extra-work Socialization..........................66 Summary of Analyses of Variance for the Rela- tionships between Extra-work Socialization and the Importance of Ego Heed Gratification ........71 V Table Table Table' Table Table Table Table Table Table Table Table Teble Table Table Table Table Table Table 14 15 17 18 19 2O 21 22 23 24 25 26 27 28 29 30 31 Mhan Importance of Ego Heed Gratification for Levels of Extra-work Socialization (EWS)........71 Summary of Interactions between Hierarchical Level (A) and the Importance of Ego Heed Gr‘tiftc‘tion (3)0000...0.000.000.0000...000.00.000.73 Summary of Interactions between Self Report (A) and the Importance of Ego Heed Gratification (8)....73 Summary of Interactions between Collar Color (A) and the Importance of Ego Heed Gratification (3)....74 Mean §g§igfigg§igg,for Importance by Hierarchical “VGIOOOOO0.0000000000000COOOOOOOOO0.0.0.0000000000074 than Identifiggtign for Importance by Hierarchical “veIOOOO.OO...0..OOOOOOOOOOOOOOOO00.000.000.00.0.0.74 Mean Lgyglgglggg for Importance by Hierarchical “velOOOOOOOO0.0000000000000000000...00.0.000000000075 Mean W for Importance by Hierarchical “veIOOOOOOOOOO0.0......00.000.000.000...00.00.00.0075 Mean Satisfactiog for Importance by Self Report.....75 Mean Identificatiog for Importance by Self Report...76 Mean W for Importance by Self Report.. .. . .76 than M for Importance by Self Report. . . . . . .76 than Satisfactiog for Importance by Collar calorOOOOOOOOOOOOCOOOOOOOOOOOOCOOOOOOOOOOOOOOO0.0.0.77 Mean Identifigation for Importance by Collar calor...OOOOOOOOOOOOOOOOOCOOOOOOOOOOOOOOOOOOOO0.0.0.77 than Lnolvemegt for Importance by Collar calorIOOCOOOOOOOCOOOOOOOOOOOOOOOO0.00.00.00.0000000077 Hhan‘ggggiglgng,for Importance by Collar colorOOCOCOOOOOOOOO00......OOOOOOOOOOOOOOOOO0.0.0.0.78 Summary of Simple Effects Analyses for Self Report (A) on Identification within Levels of Importance of Ego Heed Gratification (El-Low, Rz-lhdium, 33-High Importance)......................78 Summary of Simple Effects Analyses for Collar Color (A) on Identification*within Levels of Importance of Ego Heed Gratification (Bl-Low, Hz-lhdium, 33-High Importance)......................78 vi Table Table Table Table Table Table Table Table Table Table Table Table Table Table 32 33 36 37 38 39 4O 41 42 43 44 45 Summary of Analyses of Variance for Hierarchical Level (A), Self Report (B) and Collar Color (C) on the Importance of Ego Heed Gratification............................80 Mean Importance of Ego Heed Gratification by Hierarchical Level, Self Report, and coll‘r color...0....0.00.00.00.00...00.00.000.000..0080 Analysis of Variance Summaries for Inter- actions between Education (A) and Hier- ‘rcnical “val (3)00000000OOOOOOOCOOOOOOOOOO...0.0.0101 Analysis of Variance Summaries for Inter- actions between Pather's Occupation (A) and Hierarchical Level (E)..........................101 Analysis of Variance Sun-arias for Inter- actions between Pather's Education (A) and nier‘rcutc.luve1(3)000000000000000000000C0.0.0.0.102 Analysis of Variance Summaries for Inter- actions between Community Raised (A) and Hierarchical Level (D)..............................102 Analysis of Variance Summaries for Inter- actions between Present Community (A) and nier.rCh‘-C‘l uvel (3)0000000000000000.0000000000000103 Analysis of Variance Su-aries for Inter- actions between Sex (A) and Hierarchical “v.1(3)00000000000OOOOOOOOOOOOOOOOOOOOOOO0.0.00000103 Analysis of Variance Su-aaries for Inter- actions between Education (A) and Collar c010: (3)00...OOOOOOOOOOO0.0000000000000000000000.0.1“ Analysis of Variance Summaries for Inter- actions between Pather's Education (A) .nd call‘r col” (3)0000000000OOOOOOOOOOOOOOOOOOOOOOIM Analysis of Variance Summaries for Inter- actions between Pather's Occupation (A) and 0011.: col" (3)00000000OOOOOOOOOOOOOOOOOOOOOO0.0.0.105 ‘Analysis of Variance Summaries for Inter- actions between Community Raised (A) and calm cal“ (B)...........................C.......0105 Analysis of Variance Summaries for Inter- actions between Present Community (A) and cal-1‘: color (3)00000000000000000000000......0.0.0.0106 Analysis of Variance Summaries for Inter- actions between Sex (A) and Collar Color (D)........lO6 vii Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table 46 47 48 49 SO 51 52 S3 55 56 S7 58 59 60 61 62 Analysis of variance Summaries for Inter- actiona between Education (A) and Self Report (B)...107 Analysis of Variance Summaries for Inter- actions between Pather's Education (A) .“ self Rewrt (n)...OOOOOOOOOOOOOOOOOOO00.0.0.0...107 Analysis of Variance Summaries for Inter- actions between Pather's Occupation (A) .“ self Report (a)...OOOOOOOOOOOOOOOO00.0.0.0...00.108 Analysis of Variance Summaries for Inter- actions between Community Raised (A) and salt Rem: (”)OOOOOOOOOOOOIOOOOOO.COO...00.0.0.0...108 Analysis of Variance Summaries for Inter- actions between Present Community (A) ‘u self Rem: “)00000000000000000000.00.00.000000109 Analysis of Variance Summaries for Inter- actions between Sex (A) and Self Report (B).........109 Mhan Jab Attitudes for Education by mer‘rCh1C‘l “VGIOOOOOOOOOOOOOOOO0.00...0.00.00.00.110 Mean Job Attitudes for Father's Education by Hierarchical Level.....................lll than Job Attitudes for Father's Occupation by Hierarchical Level....................112 than Job Attitudes for Con-unity hi..dbynier‘rCh1c.l “"100000000000000.000000000113 Mban Job Attitudes for Present Community by Hierarchical Level.....................ll4 than Job Attitudes for Sex by Bier.rChic‘1 “VGIOOOOOOOOOCIOOOOOOO000.00.00.000000115 Mean Job Attitudes for Education by coll.r ColorOOOOOOCOOOOOOOOOOOOO00a...0000.00.00.116 than Job Attitudes for Father's Bduc.t1°n by Cal“: colorOOO.......COC0.0.0.00000000117 Mean Job Attitudes for Father's menp‘tionby C0111]? calorsssseeaessssseeeeseseesesells than Job Attitudes for “unity n1.ed by c011.r COIMOOCCCCCCOOCOCO0.0.00.00.00.000119 than Job Attitudes for Present cmnit’ by colur CalorOOOOOOOOOOOOOOO0.00.00.00.0120 viii Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table 63 65 66 67 68 69 7O 71 72 73 74 75 76 77 78 Mean Job Attitudes for Sex by Collar Color..........121 Mean Job Attitudes for Education by self Report.....OOOOOOO0.0.0.000....00.000.000.00000122 Mean Job Attitudes for Father's Education by self RaportOOOOOOOOOCOOOO0.00.00.00.000123 than Job Attitudes for Father's Occupation by Self Report...........................124 Mean Job Attitudes for Community m1.ed by self Report.....OOOOOOCCOOOOOOOOOOOOO0.0.0125 Mean Job Attitudes for Present Community by self Report.....OOOO...OOOOOOOOOOOOOOOO0.0.00.000126 Mean Job Attitudes for Sex by Self Report...........127 Simple Effects Analyses of Variance for Collar Color (A) on Satisfaction within Levels (l-Low, 2.kd1u. 3-n18h) Of Ed“.t1°n (’) O O O O O O O O O O O O O O O O 0 O 128 Simple Collar Levels Simple Collar Levels Simple Collar Levels Simple Collar Levels Simple Collar Levels Simple Collar Effects Analyses of variance for Color (A) on Identification'within 0f EduC‘tion (3)000...OOOOOOOOOOOOOOO...0.0.0128 Effects Analyses of Variance for Color (A) on Involvement within Of muc.ti°n (3)0000000000000000000.0.00.0000128 Effects Analyses of Variance for Color (A) on Commitment within 0f Educ.t1°n (3)0000000000000000000.0.0.00000129 Effects Analyses of variance for Color (A) on Satisfaction‘within Of he.ent mutt, (”)000000000000.0.0.0...129 Effects Analyses of variance for Color (A) on Commitment within of Present Community (B).....................129 Effects Analyses of variance for Color (A) on Identification within Sexes (ll-female, 32-male)..........................129 Simple Collar Effects Analyses of Variance for Color (A) on Involvement within Sexes (El-female, Ez-male)..........................l30 Simple Collar Levels Effects Analyses of Variance for Color (A) on Involvement within of Father's Education (D)....................l30 ix Table Table Table Table Table Table Table 79 8O 81 82 83 85 Simple Effects Analyses of Variance for Collar Color (A) on Satisfaction*within Level of Father's Education (B).....................130 Simple Effects Analyses of variance for Self Report (A) on Identification‘within Levels of Community Raised (B)......................130 Simple Effects Analyses of Variance for Self Report (A) on Commitment within Levels of Community Raised (D)......................l3l Simple Effects Analyses of Variance for Self Report (A) on Identification within Levels of Present Contunity (B).....................131 Correlations Among Hhaaures of Job Attitudes........138 Correlations Among thasures of Ego Heed Gratification.OOOOOOOOOOOOOOOO0.000000...00.0.0.0...138 Correlations Among Measures of Extra-Work SOCializ‘tionOOO.O...00......OOOOOOOOOOOOOOOOO0.0.0.139 CHAPTER I INTRODUCTION Theory The theme of the following review of the theory relevant to the effects of the ego need gratification and extra-work socialization processes on employees' responses to the job is that these variables and the relationships between them have played, and continue to play, roles of central importance in the theoretical literature of organization- al psychology. The role of these variables in the theories of the most prominent*writers identified with the "human relations school of thought" will be presented first, and then the treatment of these variables by the most articulate critics of the human relations theorists will be discussed. Finally, the most elaborate "rebuttal" by the human relations theorists to their critics' discussion of these variables will be presented. The discussion of the human relations theorists will be structured around the potential sources of on-the-job ego need gratification differentially emphasized by these theorists. at o e st A proposition underlying the theories of the most prominent writers identified with the "human relations" school of thought in organizational psychology (Argyris, 1964; Herzberg, 1966; Likert, 1961; thGregor, 1960) is that the extent of ego need gratification provided by the job is a prime determinant of various employee reactions to the job such as job satisfaction, ego involvement, commitment, and identi- fication with the organization. This proposition, for example, is the basis for the rather polemdc attacks of such writers on the traditional organization theories (Mooney, 1939; Taylor, 1911; Urwick, 1939; Weber, 1947) and the modern industrial organization practices based on those theories. This proposition is also the basis for the "human relations" theorists rather "evangelistic" advocacy of participative management, considerate supervision, job enlargement, and job enrichment. The ego needs most frequently mentioned include the needs for achievement, autonomy, self-esteem, recognition, status, and self-actualization. These needs are said to be gratified on the job through the provision of challenging, difficult, meaningful work, considerate supervision, participation in decisions concerning work-related goals and methods, and opportunities for personal growth and development, and to be frus- trated by fractionated, routine work, inconsiderate supervision, lack of control over goals and methods, and personal stagnation. Although they all agree that job dissatisfaction, apathy, alienation, and lack of positive motivation in relation to organization goals mainly result from.on-the-job ego need deprivations, and that increasing the extent of ego need gratification provided by the job would increase job satisfaction, involvement, commitment and identification with the organi- zation, the different human relations theorists focus their attention on somewhat different potential sources for such gratification. McGregor (1960) and Likert (1961), while they are definitely concerned with job content, give somewhat greater eqhasis to leader- ship style and participation in decision making as potential sources of on the job ego need gratification. According to McGregor, for example, the main deficiency of traditional organization theory and the modern industrial practices based on that theory is the exclusive emphasis placed on hierarchical authority as a means of social influence by that theory and practice. McGregor begins his criticism of traditional or- ganizational theory by stating that the "single assumption.which per- vades conventional organization theory is that authority is the central, indispensable means of managerial control" (McGregor, 1960). He then goes on to argue that more "democratic" forms of influence may be generally more effective in modern organizations because hierarchical authority does not provide as adequate an opportunity for the grati- fication of ego needs as the more "democratic" forms of influence do. thre specifically, according to McGregor: "The philosophy of management by direction and control--regardless of whether it is hard or soft--is inadequate to motivate because the human needs on which this approach relies are relatively unimportant motivators of behavior in our society today. Direction and control are of limited value in motivating people whose important needs are social and egoistic. People deprived of opportunities to satisfy at work the needs that are now important to them, behave exactly as we might predict-~with indolence, passivity, unwillingness to accept responsibility, resistance to change, willingness to follow the demagogue, unreason- able demands for economic benefits." (McGregor, 1960, p. 42) Similarly, thGregor's views concerning the potential gains from increasing the opportunities for ego need gratification on the job are clearly evident in his alternative to the assumptions about human moti- vation and behavior implied by traditional organization theory and practice. thGregor labeled his alternative assumptions "Theory Y": 1. The expenditure of psychical and mental effort in‘work is as natural as play or rest. The average human being does not inherently dislike work. Depending upon controllable conditions, work may be a source of satisfaction (and will be voluntarily performed) or a source of punishment (and will be avoided if possible). 2. External control and the threat of punishment are not the only means for bringing about effort toward organizational objectives. Hanrwill exercise self-direction and self- control in the service of objectives to which he is committed. 3. Commitmegt to objettivea is a functign of the rtgtrds atso- tiatad with their achiegemegt. The most significant of such rewards, e.g., the satisfaction of ego and self-actualization needs, can be direct products of effort directed toward organizational objectives. kW only to attept hat to gtek tetpongibility. Avoidance of responsibility, lack of ambition, and emphasis on security are generally consequences of experience, not inherent human characteristics. 5. e i o e at e de e o i - t o e a ea t i t e sol tio 0 or a i a- tio o 1: ti e te pggglatiog. 6. ggdet the toggitiogs of modetn industtial lite, the igtelleg- ttal pottnttalitiet ot the gvetage hm; Bill! ate only pattially gttlized (McGregor, 1960; pp. 47-48). Obviously, Theory Y relies heavily on the importance of ego need grati- fication as a determinant of employee reactions to the job. In a shmilar vein, the underlying basis for Likert's advocacy of his "participative-group" form of organization is the supposed superiority of this form of organization over management systems based more on tra- ditional theory in terms of the opportunities provided by each for the gratification of ego needs. According to Likert, "...subordinates react favorably to experiences which they feel are supportive and contribute to their sense of importance and personal worth. Simdlarly, persons react unfavorably to experiences which are threatening and decrease or minimize their sense of dignity and personal worth....Each of us wants appreciation, recog- nition, influence, a feeling of accomplishment, and a feeling that the people who are important to us believe in us and respect us" (Likert, 1961; p. 102). The importance attached to the extent of ego need gratification pro- vided by the job in Likert's theory can be readily seen in the "princi- ple of supportive relationships," which underlies his entire theory. "The leadership and other processes of the organization must be such as to ensure a maximum probability that in all interactions and all relationships with the organiza- tion each member will, in the light of his background, values, and expectations, view the experience as support- ive and one which builds and maintains his sense of personal worth and importance" (Likert, 1961; p. 103). In elaborating on the meaning of "supportive relationships" Likert states that each member of the organization "should see his role as difficult, important, and meaningful. This is necessary if the indi- vidual is to achieve and maintain a sense of personal worth and impor- tance. When jobs do not meet this specification they should be reorganized so that they do" (p. 103). Herzberg's (1966) interest in ego need gratification is exclusively focused on the extent of such gratification provided by the actual con- tent of the job itself, as opposed to the interpersonal relations and leadership practices emphasized by McGregor (1960) and Likert (1961). The essence of Herzberg's theory is that only those job-related factors which produce ego need gratification--opportunities for achievement, recognition, the use of valued skills and abilities, responsibility, and advancement-~can lead to positive job satisfaction and motivation. Similarly, only factors "surrounding" the job such as company policies and administration, supervision, salary, interpersonal relations, and working conditions, which provide gratification for "lower-level" or "non-ego" needs, according to Herzberg, can lead to job dissatisfaction. Herzberg then goes on to recommend "job enrichment," which basically consists of increasing the extent of ego need gratification provided by the actual content of the job itself, as the only way to effectively increase job satisfaction and motivation. Argyris (1964) is concerned with the effects of leadership practices, interpersonal relations, managerial control systems, and job content on the extent of ego need gratification provided by the job and the consequences of this variable in terms of employee reactions to the job. He is more concerned with the effects of job content and control systems at the lower organization levels and with the effects of leadership practices and interpersonal relations at the higher levels of the organization, since he feels that these variables are differentially important at these levels. The key variable in Argyris' most recent statement of his theory (1964) is "psychological success,‘ which he treats as synonymous with self-esteem. Argyris contends that the following factors contribute to psychological success and self-esteem'within organizations: 1. The individual is able to define his own goals. 2. These goals are related to his central needs or values. 3. Be is able to define the paths to these goals. 4. The achievement of these goals represents a realistic level of aspiration for the individual. A goal is "realistic" when it represents a challenge or a risk that requires hitherto unused, untested effort to overcome which the individual can make available to himself. A goal that is too difficult or one that is below his present level of achievement will not lead to success--indeed it may lead to failure and a lowering of his self-esteem. According to Argyris, the conditions present in most formal organi- zations make high degrees of psychological success very difficult to obtain for most members. At the lower levels of the organization this is due to the highly fractionated and routine nature of the work itself and to managerial control systema, while at the higher levels of the organization, authoritarian leadership practices and interpersonal relations hinder the achievement of psychological success and self- esteem. The frustration of desires for psychological success, accord- ing to Argyris, leads to frustration, failure, short time perspective, and conflict, and these conditions in turn lead to absenteeism, turnover, aggression, unionization, and apathy in relation to organization goals. 0 e elatio s eoris s The basic contention of the human relations theorists concerning the effects of ego need gratification on employee reactions to the work situation has, of course, not gone without criticism in the litera- ture. Strauss (1963) and flalin and Blood (1968) present the most elaborate and cogent criticisma relevant to the present discussion. The basic theae’of both of these articles is that as a result of extra-work socialization processes, many, and perhaps most, members of contemporary industrial organizations, especially the blue collar, rank and file workers in urban, highly industrialized communities, do not particularly desire ego need gratification from their jobs, but rather view work merely as a means to procure the financial resources necessary to pursue their real interests and satisfy their important needs 2£I.£h£.1223 According to these authors, employees subject to extra~work sociali- zation processes which deemphasize the importance and desirability of ego need gratification and emphasize the basic instrumentality of work in the satisfaction of needs off the job are not particularly frustrated by the absence of ego need gratification on the job. Similarly, according to this view, employees subject to such socialization processes can hardly be expected to respond favorably to the organiza- tional changes advocated by the human relations theorists on the basis of their overgeneralized conception of the effects of ego need gratifi- cation on employee reactions to the work situation. Strauss‘ article is more general than Hulin and Blood's in that he addresses himself to the interaction between extra work socialization and the extent of ego need gratification provided by all the potential sources of such gratification discussed by the human relations theorists. That is, Strauss contends that extra-work socialization should moderate the relationships between the extent of ego need gratification provided by the leadership practices, participative decision making, and inter- personal relations emphasized by McGregor (1960), Likert (1961), and Argyris (1964). He also, however, contends that extra-work socialization should moderate the relationship between job size or complexity, which Herzberg (1966) and Argyris (1964) emphasize as a source of ego need gratification, and employee reactions to the work situation. The specific relationship which Strauss discusses most systematically in this connection is an hypothesized interaction between the need for achievement and par- ticipation in decision making, such that the extent of participation in decisions should be more positively related to motivation for those individuals with relatively strong needs for achievement than for those individuals with relatively weak needs for achievement. Hulin and Blood (1968), on the other hand, focus exclusively on the interaction between extra-work socialization and job size or com- plexity as determinants of employee reactions to the work situation. These authors present a highly critical review of those studies re- porting results frequently cited as confirming the human relations theorists' advocacy of job enlargement. They then devote considerable attention to the few studies with results indicating the importance of extradwork socialization in structuring workers' responses to their jobs (Blood and Rulin, 1967; Katzell et. al., 1961; Turner and Lawrence, 1965; Whyte, 1955). Hulin and Blood then presented a model which was intended to integrate the results of both the studies supporting the "job enlargement thesis" and those studies indicating the importance of extra-work socialization process. The key variable in Hulin and Blood's (1968) model is "alienation from middle class norms" concerning the work situation. The middle class norms involved in this discussion are those commonly referred to collectively as the "Protestant Ethic" (Weber, 1930), i.e., beliefs concerning the goodness of work and the spiritual obligation to demon- strate one's heavenly predestination through hard work and success on earth. Essentially, Hulin and Blood contend that the extravwork sociali- zation process involved in being raised and living in an urban, blue collar, highly industrialized environment lead to alienation from middle class norms concerning the work situation, whereas the extra-work socialization processes involved in other types of environments, rural, non-industrialized communities, for example, foster integration with middle class work norms. The degree of alienation from, or integration with, such norms, according to Hulin and Blood, moderates the relation- ship between job size or complexity and worker responses to the job, such that the relationship between job complexity and job satisfaction is positive for those individuals subject to extra-work socialization 10 processes which foster integration with middle class work norms and negative for those individuals subject to extra-work socialization processes which foster alienation from such norms. Aggytit' gebuttal The most elaborate rebuttal by the human-relations theorists to criticisms similar to those discussed above is presented by Argyris (1964). Argyris' arguments in this connection are rather involved and somewhat contradictory to the main themes of his theory, but the following points are highly relevant to the present discussion. First of all, Argyris admits that there are employees who do not desire ego need gratification from their jobs and that these employees will not be par- ticularly frustrated by the absence of such gratification. He implies that such employees are definitely in the minority, however, and the "real" source of their lack of desire for ego need gratification from the job is the past and present deprivation in relation to these desires which they have experienced and are experiencing on the job, rather than the extra-work socialization process discussed by Hulin and Blood (1968) and Strauss (1963). In addition, Argyris contends that the lack of desire for on the job ego need gratification expressed by such employees does not re- present their "true" feelings in this respect. According to Argyris, such employees have merely repressed their desires for ego need gratifi- cation because of their realistic perceptions concerning the opportuni- ties for such gratification in their present and likely future situations. n Such employees will, however, according to this reasoning respond "favorably when realistic opportunities for ego need gratification on the job are 11 provided for them. The important point for the present discussion is that Argyris contends that the determinants of expressed desires for ego need gratification on the job are the past and present extent of such gratification rather than extra-work socialization processes. Reteatch The research results relevant to the effects of ego need gratifi- cation and extra-work socialization processes on employee responses to the job have been rather extensive, but, of course, far from conclusive with respect to the theoretical positions discussed above. No claim is made that the following review of that literature is in any way exhaustive. The objective of this review is to provide a somewhat detailed picture of the kind of research.which has been conducted in this area, the general results of that research, and the directions for future research which seem.most fruitful to the present author. In order to provide some structure for this rather voluminous body of literature, the review will be organized in the following manner. The research on leadership style and participative decision making as potential sources of ego need gratification on the job will be discussed first. Then those studies investigating the relationships between job size or complexity and employee responses will be presented. Within each of these broad areas, those studies which may be viewed as generally supporting the human relations theory position will be discussed first, and then those studies which obtained results contradictory to this hy- pothesis and those studies which yielded results indicating the impor- tance of extra-work socialization processes will be presented. 12 a a ati e 1 10 1 Over the past thirty years a considerable amount of research re- lating leadership practices and participative decision making to various measures of performance and satisfaction has accumulated in the litera- ture of organizational psychology. Those studies will be presented first which obtained results generally supporting the human relations hypothesis that more "democratic" or "participative" leadership prac- tices result in higher productivity and satisfaction. Those studies which obtained results basically contradictory to this hypothesis and those studies indicating the moderating effect of extra-work sociali- zation process on the relationships between leadership practices and employee responses will then be discussed. Only a few of the studies which are representative of the different methodologies employed in this research will be presented in any detail. Several laboratory experiments (Day and Bamblin, 1964; Lewin, Lippit, and White, 1939; Lewin, 1947; Hissumi, 1959), small-scale field experi- ments (Coch and French, 1948; Lawrence and Smith, 1955; Lawler and Hickman, 1969, Levine and Butler, 1952; Strauss, 1955), field surveys (Back-an, et. al., 1968; Backman, et. al., 1966; Bowers, 1964; Indik, et. al., 1961; Katz, et. al., 1950; Kata, et. al., 1951; Smith and Ari, 1964; Smith and Tannenbaum, 1963; Tannenbaum and Smith, 1964; Yuchtman, 1968), and large-scale field experiments (Marrow, et. al., 1967; Seashore and Bowers, 1963) have obtained results which, while far from unambiguous, have generally supported the human relations hypothesis concerning the effects of participative decision making on employee responses. 13 The Day and Ramblin (1964) study illustrates the type of laboratory research which may be cited as generally supporting the human relations hypothesis. This study investigated the effects of "closeness of super- vision" in a laboratory simulation of an assembly line. Their subjects were freshman and sophomore female college students, and they manipulated closeness of supervision by varying the detailed nature of the instruc- tions given the subjects by their supervisors, who were experimental "stooges." Highly statistically significant results in the predicted direc- tion were obtained. The average productivity in the "close" super- vision condition was 252 less than that in the "general" supervision condition. In addition, aggressive feelings, measured by post experi- mental questionnaires, toward both the coworkers and the supervisors were higher under close supervision than under general supervision. Day and Hamblin interpreted all of these results in terms of the ex- pression of aggression as a result of the frustration of ego needs produced by close supervision. They gave credit to Gouldner (1950) for this interpretation, but the influence of Lewin (1939) is also apparent. Coch and French's (1948) small-scale field experiment is frequently cited in support of the human relations theorists' advocacy of partici- pative management by those writers favorable to this point of view. This study was intended to be a direct extension of Lewin's early studies on the effectiveness of democratic leadership (1939) and group decision (1947) as compared to authoritarian leadership and lecture presentations. The study was conducted at the main plant of the Harwood manufacturing company. The plant, which produces pajamas, is located in Marion, Virginia, 14 and at that time, employed about 500 women and 100 men, recruited from the rural mountain areas surrounding the town. A chronic problem facing the management at Harwood, several members of which were former students of Lewin, had been the "resistancd‘of employees to various changes in job methods and content necessitated by product style changes. This resistance had been expressed in the form of grievances about new rates, high turnover, low efficiency restriction of output and verbal aggression against management follow- ing such changes. The experiment was designed to test the effectiveness of "participation" in overcoming such resistance to change. The design included four groups of from seven to eighteen employees each. In the "no participation" group, the change was implemented as usual. The employees were simply informed of the changes in methods, rates, etc. involved. In the "participation through representation" group a meeting was held during which the need for change was presented "as dramatically as possible." After agreement was reached that the change was necessary and desirable, this group then chose several operators to help determine the new methods and piece rates and to help train the other operators in the new methods. The procedure was much the same for the two "total participation" groups. Since these groups were smaller, however, all the operators helped to determine the new methods and rates. Immediately after the change all four groups displayed the usual decline in productivity. within a few days, however, striking differ- ences appeared. The no participation group remained at its immediate post-change low efficiency level for thirty two days subsequent to the change. There was marked conflict with the methods engineer and hostility 15 expressed against the supervisor. Several grievances were filed, and seventeen percent of the group quit. After thirty-two days this group was broken up, and the individuals were reassigned to jobs scattered throughout the plant. The representative group, on the other hand, "showed an unusually good relearning curve." After fourteen days the group's production.was slightly above standard; they cooperated with the engineer and their supervisor, and there were no grievances filed and no turnover for the first forty days. The total participation groups recovered faster and to a higher level of productivity than the representative group. They achieved better than standard production on the second day after the change, and progressed to a sustained level of about fourteen percent above standard. There were no expressions of aggression and no turnover in these groups in the first forty days. Finally, two and a half months after the no participation group had been dispersed they were reassembled and transferred to a new job using the total participation procedure. In contrast to their earlier behavior, the group quickly recovered from an initial decline in performance, and progressed to fourteen percent above standard within eighteen days. There was no aggression or turnover in the group for nineteen days after the change. The Hatz et. al. (1950) study provides a good illustration of the type of survey research which may be cited as generally supporting the human relations theory position on leadership practices and partici- pative decision making. This study was the first project carried out in the Human Relations Program.at the Survey Research Center. The objectives of the study were to discover: (1) employee attitudes related to productivity, (2) supervisory beliefs and practices related 16 to productivity, and (3) the interrelationships of the various dimensions of morale and their determinants. Only data bearing on the first two objectives were reported in this publication. The study was carried out at the home office of the Prudential Insurance Company. Twelve section (work group) pairs which handled the same type of work.with the same work organization but differed in their productivity were compared. All of the 419 non-supervisory and 73 supervisory employees were inter- viewed. The supervisors of the high producing sections reported spending more time in supervision ptt_ggf-overseeing and planning the work of their staff--than the low-producing supervisors did. The high-producing supervisors also were coded as less "production oriented" and more "employee oriented" and as employing less "close" and more "general" supervision than the low-producing supervisors. "Close" supervision was defined as "the degree to which the supervisor checks up on his employees, frequently, gives them detailed and frequent instructions, and in general, limits the employees' freedom to the work in their own way." Finally, the high-producing supervisors were coded (blind) by their interviewers as more democratic and less authoritarian, and they reported that they were supervised less closely by their supervisors and that they were more satisfied with the degree of their authority and responsibility than the low-producing supervisors. The great majority of the survey research supporting the human relations position on leadership practices and participative decision making has been conducted by the Survey Research Center at the University of Michigan. Since Likert naturally identifies strongly with this research, he, more than the other human relations theorists, relies l7 heavily on such research in the development and presentation of his theory. Also, since Likert is "closer to the data," than these other theorists, his theory has been modified on the basis of empirical research results more than the theories of‘HcGregor (1960), Herzberg (1966) and Argyris (1964) have been. There have been two large-scale field experiments published which have been cited as generally supporting the human relations theorists' contention concerning the relative effectiveness of participative management over more authoritarian leadership practices (Harrow, Bowers, and Seashore, 1967; Seashore and Bowers, 1963). Both studies were con- ducted by the Survey Research Center at the university of Michigan and employed Likert's (1961) theory as the conceptual framework for the design of the research. The Seashore and Bowers (1963) study involved an attempt to bring three departments of a manufacturing company's main plant closer to Likert's (1961) "participative group" form of organization. An addi- tional two departments served as "controls." Changes were attempted through supervisory seminars and discussion sessions, individual coun- seling sessions, and meetings of employees conducted by line supervisors. Questionnaires were administered in 1958 before introduction of the change program, in 1959 during the course of the program, and finally in 1961. The change programwwas aimed at increasing the following independent variables: the emphasis on groups as opposed to individual supervision, the rate of interaction and influence among group members, the degree of participation in decision making and control activities in lower echelons, and the degree of supervisory supportiveness. An analysis of relative changes, based on questionnaire data, indicated that seven 18 of the eleven variables changed in the predicted direction. In general, the results for changes in satisfaction, machine efficiency, and absen- teeism were positive for the experimental groups and negative or neg- ligible for the control departments. The authors cautiously interpreted these results as supporting Likert's (1961) theory. Smith and Jones (1968) analyzed data from this study not included in the Seashore and Bowers (1963) report. They investigated Likert's theory of an interaction influence system1which intervenes between the independent and dependent variables mentioned above. The aspect of that theory relevant to our present purpose states that increases in the independent variables mentioned above result in, among other things, increased total control and increased rank-and-file control relative to higher-level control, which results in higher motivation and greater consensus and uniformity, which in turn result in increased performance and satisfaction. Questionnaire measures of "general control" plus an index of specific control over pay raises, division of labor, work methods and standards, were analyzed to test this theory. The results provided some support for the total control hypotheses, but no support at all was obtained for the relative control hypothesis. For both the general and specific measures, total control in- creased in the experimental departments and decreased in the control departments. The relative changes were significant only for the in- dex of specific influences, however. According to both the general and specific measures, positive slope decreased in both the experimental and control departments, although the decreases were significant only for the control departments. An analysis of the changes in control at each hierarchical level revealed that the increases in total control 19 for the experimental department were primarily the result of increases in control attributed to the middle levels of the organization. The increases in performance and satisfaction, therefore, were not mediated through increased rank-and-file control. The authors interpret these results as supporting the notion that participative management may be a means of increasing management's influence (March and Simon, 1958; Tannenbaum, 1968) as well as, if not more than, that of the rank-and- file employees. The Marrow et. a1. (1967) monograph also describes a change pro- gram designed to bring a manufacturing company closer to Likert's (1961) participative management model of organization. The change target, the Weldon Manufacturing Company, was acquired by the Harwood owners in 1962. The companies were in similar product lines, but their markets overlapped only slightly. Performance measures were obtained weekly throughout the change program, which lasted about two years. Questionnaire measures were also obtained from random samples of employ- ees in both Harwood and Weldon in 1962, in Weldon in 1963, and again in both Harwood and Weldon in 1964. Pre-post change comparisons were made, with Harwood serving as a "control." The change program*was rather all-encompassing. Many "technical" changes were made in addition to the more "social" ones. On the technical side, new machinery was added, individual jobs were re-engineered, and the work flow was reorganized. 0n the social side, all the managers, down to first level supervision, were involved in one form or another of "sensitivity training," joint problemesolving meetings were held between foremen.and workers, and there was a general emphasis on in- creasing the involvement of lower echelons in decisions affecting their 20 jobs. Other changes included an intensive training program for the operators, the hiring of an additional engineer and a personnel manager, an "earnings development" program, and "tougher" policies towards absenteeism and chronic low producers. There can be little doubt that Weldon's performance improved greatly from 1962 to 1964. Some of the highlights of that improve- ment included: 32% increase in return on capital invested, 81 decrease in make-up pay, 251 increase in production efficiency, 6% decrease in monthly turnover rates, and a 32 decrease in daily absenteeism rates. It is impossible, of course, to determine exactly which changes in which combinations resulted in which improvements. Bowers and Seashore, however, on the basis of an analysis of weekly performance records and the chronology of the various change elements, concluded that they could confidently trace short-run improvements in operator performance to the earnings development program, the weeding out of low producers, the training of supervisors in interpersonal relations, and the joint problem solving between workers and foremen. The measured changes in attitudes were much less pronounced than the changes in performance. Attitudes toward the company, compensation, and fellow employees improved, but only slightly from 1962 to 1964. In addition, attitudes toward the company and compensation improved more at Harwood than at Weldon during this period. Measures of the workers' perceptions of supervisory behavior were particularly difficult to interpret because by 1964 the assistant supervisors had largely taken over the role of first-level supervisors. The data did indicate, however, that the workers viewed the supervisors in 1964 as supervising less closely, and less concerned exclusively with production. Similarly, 21 there were no significant changes in control, measured by Tannenbaum's (1968) procedure, at Weldon but there were significant increases at Harwood. Finally, all these "negative" results were in striking con- trast to the views of management. Self, peer, and superior retroactive change reports, designed to assess the impact of the sensitivity train- ing procedures, produced highly significant changes, as did management ratings on Likert's (1967) 43-factor "Profile of Organizational and Performance Characteristics." There are also many published studies which have obtained results contradictory to the human relations hypothesis concerning leadership practices and employee responses. The following laboratory (Anderson and Piedler, 1964; Bennett, 1955; Piedler, Bass, and Piedler, 1961; Piedler, Muewese and Oonk, 1961; Muewese and Piedler, 1965; Shaw and Blum, 1966; Shaw, 1955) and field survey studies (Cleven and Piedler, 1956; Fiedler, 1955; Fiedler, 1954; Fleishman, 1953; Fleishman, 1957; Pleishman, Harris, and Burtt, 1955; Fleishman and Peters, 1962; Godfrey, Piedler, and Hall, 1959; Halpin and Winer, 1957; Hemphill, 1955; Hunt, 1967; Hutchins and Fiedler, 1960; Parker, 1963; Spitzer and McNamara, 1964) have been cited as conflicting with the human relations hypothesis. The type of laboratory research which has produced results conflict- ing with the human relations hypothesis is illustrated by an experiment by Shaw (1955). This experiment investigated the problem solving per- formance and satisfaction of four-man groups in different "Bavelas-type“ communication nets under democratic and authoritarian leadership. The subjects were male undergraduate college students. As predicted, the problems were solved faster and with fewer errors under authoritarian leadership, but satisfaction was higher under democratic leadership. 22 Shaw used the concepts "saturation" and "independence” to predict and interpret his results. Independence refers to the degree of freedom with which a member may operate, and saturation refers to the communi- cation requirements imposed on a group member. Independence is said to correlate positively with efficiency and morale, but after a certain optimal point, saturation tends to counteract these favorable effects. In addition, morale is influenced more by independence than by satura- tion, while performance is influenced more by saturation than indepen- dence. These concepts are related to leadership style and its effects on performance and satisfaction in the following manner. Authoritarian leadership should decrease independence for most group members and hence decrease morale, and should decrease saturation for all group members and hence improve performance. Democratic leadership should increase independence for all group members and hence increase morale, and should increase saturation for all group members and hence decrease performance. The survey research which may be cited as contradicting the human relations hypothesis concerning leadership practices has come from two main sources: (1) the work of Fred Fiedler and his associates (summarized and integrated in Fiedler, 1967), and (2) the work stemming from the Ohio State leadership studies (summarized in Harman, 1966). In their summaries of this work, both Piedler (1967) and Herman (1966) conclude that the human relations hypothesis is grossly overgeneralized. Piedler (1967), on the basis of a large number of field surveys and laboratory experiments conducted by him and his associates, con- tends that the human relations hypothesis only holds true under a very limited set of circumstances. Essentially, Fiedler contends that 23 the relative effectiveness of task centered, directive, and controlling leadership as opposed to relationship oriented, nondirective, permissive leadership depends upon a complex interaction between the leader's position power, the task structure, and the relationship between the leader and the rest of the group. The nature of this interaction, according to Piedler, is such that nondirective, permissive leadership is more effective than directive, controlling leadership only when the leader's position power, his relationship with the group, and the task structure combine to produce a "moderately favorable" situation for the leader. In contrast, when the leader's position power, his relationship with his group, and the task structure combine to produce either very favorable or very unfavorable situations for him, directive, controlling leadership is more effective than permissive leadership. Equally contradictory to the human relations hypothesis, Korman (1966) concluded that the only generalizations which could be drawn from the studies relating "Consideration" and "Initiating Structure," as measured by the Ohio State Leader Behavior Description Questionnaire and Leadership Opinion Questionnaire, to various measures of performance and satisfaction was that the relationships obtained were generally weak and highly inconsistent. "Initiating Structure" and "Consideration" were defined in the following manner (Pleishman and Peters, 1962): Lpitiatigg Strutture (S): Reflects the extent to which an individual is likely to define and structure his role and those of his subordinates toward goal attainment. A high score on this dimension characterized individuals who play a more active role in directing group activities through planning, communicating information, scheduling, trying out new ideas, etc. Considetgtion (C): Reflects the extent to which an individual is likely to have job relationships characterized by mutual trust, respect for subordinates ideas, and consideration of their feelings. A high score is indicative of a climate of good rapport and two-way communication. 24 A low score indicates the supervisor is likely to be more impersonal in his relations with group members. Based on a review of the studies investigating the relationships between these variables and performance and satisfaction, Korman (1966) came to the following conclusion, "What seems most apparent, however, is that there is very little evidence that leadership behavioral and/or attitudinal variation, as defined by scores on the Leadership Behavior and Leadership Opinion Questionnaire, are predictive of later effective- ness and/or satisfaction criteria." (p. 354) Finally, the results of the first large-scale field experiment (Morse and Reimer, 1956) designed to test the human relations hypothesis of a positive relationship between participative decision making and productivity and satisfaction could be interpreted as contradicting this hypothesis, at least in relation to productivity. The site for this experiment was a large department of an industrial organization engaged in routine clerical work. The rank-and-file decision making power was increased in two divisions of this department and decreased in two comparable divisions. The experiment lasted for a year and a half. A before questionnaire was administered; six months were spent training supervisors; the experi- mental conditions were in effect for a year, and then an after question- naire was administered. The results supported the hypotheses concerning satisfaction rather strongly. There were significant before-after in- creases for the "autonomy" program.and decreases for the "hierarchical" program for "self-actualization," satisfaction with supervision, and satisfaction with the company, and nonsignificant changes in the expected directions for job satisfaction. In addition, on several open-ended 25 questions concerning the program itself the clerks in the autonomy program.typically: "...wanted their program to last indefinitely, did not like the other program, felt that the clerks were one of the groups gaining the most from the pro- gram, and described both positive and negative changes in interpersonal relations among the girls. The clerks in the hierarchically-controlled program, on the other hand, most frequently: wanted their program to end immediately, liked the other program and felt that the company gained the most from their program. Not one single person in the hierarchically-controlled program mentioned an improvement in interpersonal relations as a result of this program." The hypothesis concerning productivity, however, was not con- firmed. Both programs resulted in a significant increase in produc- tivity, and the increase in the hierarchical program was significantly greater than that in the autonomy program. As Morse and Reimer point out, however, the productivity measures were far from perfect and sort of "stacked against" the autonomy program. The volume of work was not under the control of the units under study. The only way to increase productivity, therefore, was to decrease the number of employees required to do the unit's work. In the hierarchical program this pre- sented no great problem; the management simply cut their staff. It was obviously not so easy, however, for the clerks in the autonomy program to decide to eliminate some of their peers' jobs. These clerks did reduce their numbers, but they accomplished this through not replacing employees who left and by attempting to find new jobs for employees who wanted to change jobs. It is certainly to be expected, however, that for a given unit of time, the hierarchical programmwould be able to eliminate more employees than the autonomy program. Finally, Likert (1967) has suggested that the differential attitude changes would have 26 resulted in the eventual superiority of the autonomy program had the experiment lasted for a longer period of time. Several studies, including laboratory experiments (Aronoff and Messe', 1969; Haythorn, et. al., 1956) small scale field experiments (French, Israel, and As, 1960), field surveys (Aronoff, 1967; vroom, 1959; 1960), and large scale field experiments (Tannenbaum and Allport, 1956) have also produced results which may be interpreted as indicating the importance of extra-work socialization as a moderator of the re- lationships between leadership practices and employee responses. Haythorn et. a1. (1956), for example, varied the authoritarianism, as measured by the California F-scale, of leaders and followers within a 2 x 2 factorial design, such that there were four four-man problem solving groups within each combination of leader authoritarian (high - low) and follower authoritarianism (high - low). The task consisted of a group discussion of a "human relations" film and the composition of a case related to the film. The data collected included observer and subject ratings of various leader and follower behaviors and attitudes. Several hypotheses, derived mainly from.the Adorno, et. a1. (1950) theory of the authoritarian personality, were confirmed. High-F leaders, for example, were rated as being less equalitarian, less concerned with group approval, more autocratic and less sensitive to others than low-F leaders. Similarly, leaders with high-F followers were rated as more autocratic than leaders with low-P followers, and high-P followers were more satisfied with their appointed leaders and less critical of their groups than low-F followers. On the other hand, contrary to pre- dictions, homogeneous groups were significantly less satisfied with their leaders than heterogeneous groups, and there were no unambiguous, consistent 27 differences between homogeneous and heterogeneous groups in terms of productivity and morale. Despite these somewhat mixed results, the authors interpret their findings as demonstrating the usefulness of conceptualizing leadership as an interaction between the needs and personality structures of both leader and followers. A small scale field experiment*which is frequently cited as indi- cating the importance of extra-work socialization as a determinant of employee responses to different leadership practices was conducted by French, Israel, and As (1960) in a Norwegian shoe factory. The purpose of this study was to replicate the original Coch and French (1948) study, in a different culture using more careful methods and a more precise theory of participation. The general hypothesis predicted positive relationships between participation and productivity, management-worker relations, and job satisfaction. The effects of several conditioning variables were also discussed. Briefly, the authors hypothesized that the above relationships would vary with the perceived legitimacy of par- ticipation, the resistance to the participation process, the importance of the area of participation, and the relevance of the decisions to the dependent variables. The following rationale was given for the general hypotheses. Increases in participation should increase productivity because the decisions involved should be better and because workers should be more motivated to implement them. Participation should relate positively to management-worker relations because the mutual influence involved should lead to greater understanding between the parties and because the implication that the workers are intelligent, competent, and worthy should increase their perception of being valued. Finally, 28 participation should be positively related to job satisfaction because workvrs' john should be improved in ways that are relevant to their needs and because participation should directly satisfy various ego needs. Nine groups of four workers each were employed in the experiment, which involved variations in the participation allowed to the groups in decisions concerning seasonal changes in production. Two of the ' which con- experimental groups were allowed "moderate participation,‘ sisted of participation in decisions about the allocation of articles, length of the training, division of labor, and job assignment. The other three experimental groups were given "weak participation," which involved making decisions about the allocation of articles only. The four control groups did not participate in any of these decisions. Questionnaire data revealed that these decisions were of intermediate importance to the workers. The results revealed no post-change significant differences be- tween the experimental and control groups in productivity. The authors attribute this result to the less than overwhelming importance of the decisions, the low relevance of the decisions to productivity, and to strong group norms restricting productivity. Slight and generally non- significant differences in the predicted direction were obtained for the measures of labor-management relations and job satisfaction, and some support was obtained for the postulated conditioning effects of legitimacy and resistance. Especially when the effects of the condi- tioning variables are considered, the authors interpret their results as being generally consistent with those of the Coch and French (1948) study. The literature, however, has emphasized the fact that the Coch and French (1948) results were not replicated and that this was probably 29 due to the cultural differences between the Norwegian and Virginia workers involved in the studies. Vroom's (1959; 1960) field survey of the interaction between employee needs and personality structure and participation undoubtedly has provided the most clear-cut evidence of the importance of individual differences in determining employee responses to leadership practices. Vroom tested the hypothesis that participation would be more positively related to performance and attitudes for persons low in authoritarianism and for persons with strong independence needs than for persons high in authoritarianism and for persons with weak independence needs. The data were collected from the supervisory personnel of the two largest "plants" of a delivery company. The need for independence was measured with a 16 item questionnaire developed by Tannenbaum and Allport (1956), and authoritarianism was measured with the F-scale. Participa- tion and job attitudes were measured with questionnaire items, and superior ratings were used to measure performance. The total sample was divided into three approximately equal groups and participation was correlated with attitudes and performance for the total sample and within each group. The results strongly supported the hypotheses. Par- ticipation was correlated positively and significantly with both attitudes (.36) and performance (.20) for the total sample. Both correlations, however, were higher for supervisors high in need for independence (.55 and .25) than for supervisors low in independence (.13 and .01). In addition, both correlations were lower for high authoritarian supervisors (.03 and .06) than for low authoritarian supervisors (.53 and .27). Vroom's theoretical treatment of his results was also more adequate than that involved in most of the studies in this area. Given the 30 assumptions that: (1) persons derive satisfaction from successfully carrying out decisions in which they have participated, (2) the more an individual has influenced a joint decision, the more satisfaction he obtains from its successful execution; and (3) the motives for in- dependence and power-equality are satisfied by successful execution of joint decisions, Vroommwas able to predict and interpret his results in terms of a general motive, expectancy, incentive theory of motivation (Atkinson, 1964). Finally, Tannenbaum and Allport (1956) used data collected in the Morse and Reimer (1956) study to test the hypothesis that personality characteristics and organizational structure, in this case the "hier- archical" and "autonomy" change programs, should interact to determine work-related attitudes. Essentially, their hypothesis, derived from Allport's (1954) event-structure theory, was that individuals with personalities "suited" to the different experimental programs would have more favorable attitudes towards these programs than individuals with less "suited" personalities. The specific conceptual and operational definitions of personality "trend structures" and their "suitedness" to the experimental programs were derived from Allport's theory, which is rather esoteric and beyond the scope of a few paragraphs. In more familiar language, however, an individual's personality is "suited" to a given social structure to the extent that that structure provides him with the opportunity to achieve his important goals or satisfy his dominant motivea--trends or trend structures in event-structure termin- ology. Twenty-six such trends were identified as being relevant to the experimental programs. The "potency" (importance) of the trends for each of the subjects was measured by Likert-type questionnaire 31 items, and three judges estimated the extent to which the different programs would provide "closure" (realization) for each of the trends. The attitudes measured included desired length of the program, satis- faction with the way the program operates, and two measures of degree of liking for the program. Although the differences were not extremely large, nor always statistically significant, especially when each program*was considered by itself, the results generally confirmed the hypothesis. When the data for the two programs were combined, three of the four attitude measures yielded the predicted differences significant at the .05 level, and the differences for the fourth measure, "satisfaction with the way the program operates,‘ were significant at the .10 level. Job Eggplexity A considerable portion of the research efforts of industrial- social psychologists has also been devoted to investigations of the relationships between job size or complexity as a source of ego need gratification and employee responses to the job. Such investigations have basically been of two varieties: (1) those specifically designed to test Herzberg's (1959; 1966) theory of job satisfaction and motiva- tion, and (2) those designed to explore the relationships between job complexity and worker responses from a "non-Herzberg" theoretical orientation. Experimental and survey methodologies have been employed in both types of studies, and, of course, both positive and negative results have been obtained. We will first discuss those studies specifically designed to test Herzberg's theory and then present those investigations of the relationships between job complexity and worker 32 responses not specifically designed to test that theory. Within each of these sections, those studies which produced generally "positive" results will be discussed first, and then those studies yielding "negative" results will be presented. Finally, those studies indicating the importance of extra-work socialization as a moderator of the rela- tionship between job size or complexity and worker responses to the job will be discussed in detail. The original study (Herzberg, et. al., 1959) which led to Herzberg's theory of job satisfaction and motivation has been frequently cited in support of the human-relations hypothesis that the extent of ego need gratification provided by the job is an important determinant of employee reactions to the job. This study, which was designed to determine the sources and consequences of on the job need satisfactions, employed a subject sample of 200 engineers and accountants. A rather unique "critical incidents-type" survey design was employed. Mbre specifically, the subjects were first requested to recall a time when they had felt exceptionally good about their jobs. They were then further questioned concerning the reasons for their feelings of satisfaction and the effects of their satisfactions on their performance, personal relationships and well-being. An analogous procedure was then followed for negative incidents, that is, incidents in which their feelings about their jobs were exceptionally negative. The results obtained with this procedure, which have formed the basis for Herzberg's two-factor theory, indicated that the job characteristics "intrinsically" related to the nature of the work itself were more frequently mentioned as sources of satisfaction than as sources of dissatisfaction and that the job characteristics related to the context or environment surrounding 33 the job were more frequently mentioned as sources of dissatisfaction than as sources of satisfaction. The intrinsic job factors most frequently mentioned included achievement, recognition for achievement, the work itself ( a sense of performing interesting and important work), job responsibility, and advancement. Those factors most frequently mentioned as sources of dissatisfaction included company policy and administration, supervision, working conditions, interpersonal relations, and salary. Herzberg has labeled those factors which were most frequently mentioned as sources of satisfaction "satisfiers" and "motivators" and those factors most frequently mentioned as sources of dissatisfaction "dissatisfiers" and "hygiene" factors. Without providing adequate references, Herzberg (1966) cites the following studies which employed the same critical-incidents methodology but used widely varying subject samples as replications of his original (1959) study (Herzberg, 1966; pp. 122-123): étthors Sample Descriptiog §_ Schwartz, M. et. a1. male supervisors in lower half of 111 management echelons Walt, D. Professional women doing analytical 50 work in economics, languages, mathe- matics and engineering Clegg, D. County agriculture administrators 58 Myers, M. Scientists 50 Engineers . 55 Manufacturing supervisors 50 Male technicians 75 Female hourly assemblers 52 Herzberg, F. Finnish foremen 139 Gendel, H. Housekeeping workers 119 Perczel, J. Engineers 78 34 In summarizing these studies, Herzberg (1966) presents the follow- ing conclusions, " ...of the 41 significant differences reported for the six motivator factors, gxtty_ggt_was in the predicted direction. For the 57 significant hygiene factors, 54 were in the predicted direction. In sum, then the predictions from the theory were wrong in less than 3 per cent of the cases." (p. 125) Herzberg (1966) also cites several studies as producing results supporting his theory with methodologies somewhat different from that employed in the original (1959) study (Frantz, 1962; Friedlander and Walton, 1964; Gibson, 1961; Hahn, 1959; Schwartz, 1959), and he cites several more studies as supporting his theory with more conventional job attitude survey methods (Fine and Dickman, 1962; Friedlander, 1964; Saleh, 1964; Sanvold, 1962; Wernimont and Dunnettee, 1964). Herzberg either ignored, discredited, or reinterpreted the results of those studies which have been cited as contradicting his theory. The only such studies he even mentioned were those of Friedlander (1965), Ewen (1964), and Wernimont and Dunnette (1964), and, in each case, he first depricated the methods employed and then reinterpreted the results as supporting rather than contradicting his theory. Herzberg (1966) summarizes his review of the studies testing his theory with methods other than the critical incidents technique with the following conclusion: "The results of the many studies reviewed indicate a striking consistency. The original findings have been ex- tended to a series of diverse populations working in many different types of organizations and by diverse data collection and analytic methods.’ (p. 161) 35 The authors of a somewhat more objective and inclusive review of the studies designed to test Herzberg's theory (House and Wigdor, 1967), not surprisingly, came to conclusions quite different from those pre- sented by Herzberg. House and Wigdor (1967) reviewed all the studies discussed by Herzberg plus the several studies with results conflicting with his theory which were not included in his review (Burke, 1966; Centers and Bugental, 1966; Dunnette, et. al., 1967; Eran, 1966; Ewen, et. al., 1966; Friedlander, 1963; 1966a; 1966b; Gordon, 1965; Green, 1966; Malinovsky and Barry, 1965; Ott, 1965; Singh and Baumgartel, 1966). These authors concluded, first of all, that the direct replications Herzberg cited as supporting his theory actually could be interpreted as contradicting that theory. They based this conclusion on a reanalysis of the data of these studies which resulted in the following rank order of "dissatisfying factors" (the ranking was based on the number of people across all the studies employing the critical incidents method who mentioned each factor as a source of dissatisfaction): Company policy and administration supervision achievement recognition working conditions work itself relations with superior advancement responsibility According to House and Wigdor (1967), "Our secondary analysis of the data presented by Herzberg (1966) in his most recent book yields conclu- sions contradictory to the proposition of the two-factor theory that satisfiers and dissatisfiers are unidimensional and independent. Although many of the intrinsic aspects of the job are shown to be more frequently identified by respondents as satisfiers, achievement and recognition 36 are also shown to be very frequently identified as dissatisfiers. In fact, achievement and recognition are more frequently identified as dissatisfiers than working conditions and relations with the superior. (p. 385) Similarly, based on their review of the studies not using the critical incidents technique, House and Wigdor (1967) offer the follow- ing conclusions: 1. A given factor can cause job satisfaction for one person and job dissatisfaction for another person, and vice versa... Variables that partially determine whether a given variable will be a source of satisfaction or dissatisfaction on the job were shown to be: Job or octunatigntl level: Centers and Bugental (1966), Myers (1964), Rosen (1963), Friedlander (1966b), Dunnette (1965) Age at tennongentt: Singh and Baumgartel (1966), Ssleh (1964), Friedlander (1966b), Wernimont (1966) Set of tennondents: Centers and Bugental (1966), Gibson (1961), Myers (1964) Fogntl Education: Singh and Baumgartel (1966) Culture: Turner and Lawrence (1965) Time-d io a iab e: Ewen (1964), Wernimont (1966) Resngngent's standing in his gtoup: Eran (1966) A given factor can cause job satisfaction and dissatisfaction in the same sample (Dunnette, 1965; Ewen, 1964; Gordon, 1965; Burke, 1966; Ewen, Smith, Hulin and Locke, 1966; Friedlander, 1963; Wernimont, 1966; Halpern, 1966; Ott, 1965, Hinricks and Mischkind, 1967; Green, 1966, Malinovsky and Barry, 1965). Intrinsic job factors are more important to both satisfying and dissatisfying job events (Dunnette, et. al, 1967; Wernimont, 1966, Ewen, et. al., 1966; Graen, 1966; Friedlander, 1964). These conclusions lead us to agree with the criticismnad- vanced by Dunnette, Campbell, and Hakel (1967) that the 37 two-factor theory is an oversimplication of the relation- ships between motivation and satisfaction, and the sources of job satisfaction and dissatisfaction. Several studies have been published since House and Wigdor's (1967) review which have resulted in conclusions similar to those cited above (Graen and Hulin, 1968; Hinton, 1968; Hulin and Smith, 1967; Kosmo, 1969). warren (1958), MacKinney, Wernimont, and Galitz (1962), and Hulin and Blood (1968) have reviewed the literature on the relationships between job size or complexity and worker responses which does not deal directly with Herzberg's theory. The two latter reviews are particularly critical of the human relations proposition that job com- plexity as a source of ego need gratification is positively related to job satisfaction and other "favorable" worker responses. Both of these reviews emphasize the importance of individual differences which are completely ignored by the human relations theorists in this connection. According to MacKinney et. a1.(l962), for example: The most compelling argument against specialization as a major cause of job dissatisfaction lies in the fact of individual differences. This is the central fact of life in the behavioral sciences, and yet the would-be reformers apparently believe that all people must react in exactly the same*way to the same job. The observer says to himself, "That job would drive me nuts in half an hour." From this he somehow concludes that it must drive everyone else nuts as well." (p. 17) Since the Hulin and Blood (1968) article is the most recent and most comprehensive review of this literature, it will receive the most detailed attention. These authors begin by reviewing those studies typically cited as supporting the human relations hypothesis concerning job size or complexity (Argyris, 1959; Bigsnne and Stewart, 1963; Elliott, 1963; Kornhauser, 1965; Marks, 1954; Walker, 1950; Walker and Guest, 1952; 38 Walker and Marriott, 1951; Wyatt, Fraser, and Stock, 1929b; Wyatt, Langdon, and Stock, 1937). Hulin and Blood present two main conclu- sions concerning these studies and the interpretations typically drawn from them: (1) Most of the studies were highly deficient in relation to the standard cannons of research methodology, and (2) the results hbtained were highly dependent upon the subject samples employed and therefore cannot be generalized to broader subject samples. Hulin and Blood's treatment of the Kornhauser (1965) study illustrates their approach in relation to this literature. Kornhauser (1965) reported a positive relationship between job level and an index of mental health obtained from extensive interviews with 665 blue collar workers in the Detroit area. On the basis of these data, Rornhauser concluded that job simplification and routinization is a cause of poor mental health. Hulin and Blood criticize this study on several grounds. First of all, they point out the possible bias involved in the interviews. They argue, for example, that the workers in repetitive jobs could have felt hesitant to admit that they were not dissatisfied with their work because they could have felt that their middle-class interviewers dis- approved of such responses. Similarly, they point out that the inter- viewers or coders, if they knew the hypothesis of the study, could have systematically interpreted the responses in a manner most favorable to the hypothesis. In addition, Hulin and Blood criticize Rornhauser for generalizing from an urban blue-collar sample to all production workers and for ignoring differences in the workers' personal backgrounds even though his data indicated that the relationships between background variables and mental health were at least as strong as that between job level and mental health. 39 Finally, Hulin and Blood contend that Rornhauser's index of mental health was based upon middle-class value judgments concerning "healthy" views about life in general and work in particular, and that his data, therefore, indicated merely that there are differences by job level among urban workers in the extent to which they adhere to such middle- class value systems. Hulin and Blood took exception especially in this connection to the tendency of Kornhauser's mental health index to emphasize striving for personal betterment more than a "realistic evaluation of the situation" as indicating positive mental health. They took issue, for example, with Kornhauser's scoring of agreement with items such as "there's such a thing as beating your brains against the wall. Some things you just can't change; might as well adjust yourself to them." as indicating poor mental health. Hulin and Blood offer the following conclusion based on their review of the studies typically cited as supporting the human relations hypothesis that job complexity leads to positive worker responses: "Though the human relations approach has gained widespread popular support, the data are unconvincing. These supportive data present us with severe restrictions either because of methodological problems or because of the nature of the samples" (p. 47). These authors also briefly discuss a few studies which may be cited as contradicting the human relations hypothesis concerning the effects of job complexity (Kennedy and O'Neill, 1958; Killbridge, 1960; Turner and Miclette, 1962). The Killbridge (1960) study found that the majority of 202 assembly line workers employed by a radio and television set factory in Chicago stated that they would prefer a smaller job. The Kennedy and O'Neill (1958) study obtained results indicating that assembly 40 operators performing highly routine and repetitive tasks held work~ related attitudes which were no more negative than those held by utility men.who were performing a much more varied set of tasks. Turner and Miclette (1962), finally, reported results indicating that the majority of 115 female assembly workers in an electronics plant were basically satisfied with work that was extremely repetitive and routine. Hulin and Blood (1968) devote considerable attention in their review to those studies which produced results indicating the impor- tance of extra-work socialization in structuring workers' responses to their jobs (Blood and Hulin, 1967; Katzell, Barrett, and Parker, 1961; Turner and Lawrence, 1965; Whyte, 1955). Only the Blood and Hulin (1967) and Turner and Lawrence (1965) studies specifically inves- tigated the interaction between extra-work socialization and job com- plexity as determinants of worker responses to the job. The Katzell et. al., (1961) and Whyte (1955) studies merely produced results in- dicating the importance of extra-work socialization as a determinant of worker responses in general. More specifically, Hulin and Blood cite Whyte's (1955) descrip- tions of "rate busters" and "quota restricters" as indicating the im- portance of workers' cultural backgrounds as determinants of responses to the job. Whyte reported that workers classified as "rate buster," i.e., those workers who produced above the group norm, were also predomi- nantly from rural or small-town backgrounds, had fathers who had been entrepeneurs or farmers, were Protestant Republicans, and had tended to look toward their parents for authority sanctions rather than toward their peer group for approval. The "quote restricters," i.e., those workers who produced below the group norm, on the other hand, were more 41 likely to have come from large cities and working class families, to have been Catholic Democrats and to have belonged to boys gangs as youths. Similarly, Katzell, et. a1. (1961) found, among other things, strong relationships between both satisfaction and productivity and community characteristics in a sample of warehouse workers drawn from a number of locations, such that the most satisfied and productive ‘workers were from plants located in small towns with non-unionized work forces. Turner and Lawrence (1965) directly investigated the interaction between extra-work socialization and job complexity as determinants of employee responses to the job. These authors initially hypothesized positive relationships between such task attributes as complexity, responsibility, variety, authority, etc. and satisfaction and attendance. Their sample included 470 blue collar workers from 47 different jobs in eleven different industries. The initial hypothesis concerning job level and attendance was confirmed but the hypothesized positive rela- tionship between job level and satisfaction was not supported. After several additional analyses, Turner and Lawrence found that the workers from plants located in small, rural towns were responding quite differently than those from more urban, industrialized settings. The workers from plants located in large cities, however, demonstrated no relationships between task attributes and attendance, and reversed the relationship between task attributes and satisfaction predicted by the human relations theorists. That is, workers from plants located in large cities were lgtt_satisfied with "larger" jobs than'with "smaller" jobs. Similarly, Blood and Hulin (1967) reanalyzed data gathered from 1,300 blue-collar workers employed in 21 different plants and found 42 support for their predictions that workers from plants located in large, industrialized communities with large slum areas, etc. would respond "negatively" to large jobs, while workers from small communities, with low standards of living, and few slums‘would respond in the opposite manner. The correlation between job level and satisfaction in the largest, most industrialized community was -.50, whereas the correlation between these same variables was .40 in the smallest, least industrialized community. These correlations are reported in the Hulin and Blood (1968) review article, however, not in the original Blood and Hulin (1967) re- search report. The results reported in the original report were not nearly so clear-cut and consistent as these correlations might lead one to think. Actually, 168 correlations coefficients were reported in the original article, a very small percentage of which were significantly in the predicted direction. Critical Summary of Previous Research In summary, there is considerable indirect support for the human relations hypothesis that the extent of ego need gratification provided by the job is positively related to favorable employee responses to the job. This support is indirect in that it consists of reported positive relationships between stimulus variables such as leadership practices and job complexity, which are presumed to provide ego need gratification, and employee responses. The extent of ego need gratification provided by the job has rarely been actually measured directly and related to either employee responses or leadership practices and job complexity. There are also, however, many indirect research results conflicting with the human relations hypothesis. 43 In addition to the ambiguities mentioned above, the previous re- search designed to test the human relations hypothesis has employed a rather narrow range of employee responses as dependent variables. While the human relations theorists discuss a broad range of employee attitudes and behavior in relation to this hypothesis, the research designed to test this hypothesis has focused almost exclusively on productivity and satisfaction as dependent variables. This state of affairs is particu- larly surprising in that the human relations theorists emphasize employee attitudes such as ego involvement, commitment, and identification much more than job satisfaction in relation to this hypothesis. There are also a few studies indicating the role of individual differences and extra-work socialization processes as moderators of the relationships between the extent of ego need gratification provided by the job and employee responses to the job. Hulin and Blood (1968) contend that the majority of studies supporting the human relations hypothesis employed subjects which had internalized middle class work norms. If this contention is accepted, then the theoretical position of the critics of the human relations theorists would seem to explain all of these data much more adequately than the human relations hypothesis. The data simply aren't available, however, to confirm or reject this contention with much confidence. Similarly, even the few studies ex- plicitly cited as supporting the critic's position (Blood and Hulin, 1967; Katzell, et. al., 1961; Turner and Lawrence, 1965; Whyte, 1955) are far from unambiguous in this respect. The only studies which specifi- cally investigated the interaction between the extent of ego need grati- fication and extra-work socialization as determinants of employee re- sponses (Blood and Hulin, 1967; Turner and Lawrence, 1965) employed 44 highly indirect and very possibly confounded measures of extra-work socialization. In both of these studies, plant site was used as the index of extra-work socialization. Workers from plants located in large cities were considered to be subject to socialization processes fostering alienation from middle class work norms, and workers from plants located in rural and small town communities were considered to be subject to socialization processes fostering internalization of middle class work norms. It is obvious that plant site is a rather indirect index of extra-work socialization and that it is very likely thoroughly confounded with, among other things, plant size. Turner and Lawrence (1965) made some crude attempts to measure extra-work socialization more directly, but they were unsuccessful in this respect. Since the data used by Blood and Hulin were collected for other purposes, however, they were unable even to attempt more direct measurement of this variable. Even the most ardent supporters of this research (Hulin and Blood, 1968) were not unaware of this problem. Although they defended their indirect measurement in the original research report, Hulin and Blood (1968) explicitly mentioned the need for more direct measurement of extra-work socialization in their review article. Even if these studies had employed more direct measures, there would still be a need for replication of their findings. As indicated earlier, the final data analyses reported as supporting the human rela- tions critics' position with respect to ego need gratification and extra-work socialization in both studies were actually only small pro- portions of the total set of analyses performed on the same set of data and were not performed until several other analyses failed to 45 produce significant results. In addition, Blood and Hulin (1967), although they failed to mention this point in their review article, did not find support for their hypotheses in their original sample which included both white and blue collar employees. Even when they restricted their analysis to blue collar workers, only 12 out of 84 correlation coefficients were significant at the .05 level, and only four of these were significant at the .01 level. Such results would seem to indicate the need for additional investigations of the effects of extra-work socialization on the relationships between ego need gratification and employee responses to the job. Another problem‘with these studies--a problem shared with the research supporting the human relations hypothesis-~is that the extent of ego need gratification provided by the job also was not measured directly. "Job level" was used to index this variable in the Hulin and Blood study, but no details were provided concerning exactly how "job level" was measured. The Turner and Lawrence study did employ somewhat more direct measurement of this variable in the form of "expert" ratings of various job characteristics such as the amount of autonomy, authority, responsibility, opportunities for interaction, etc. Still, it would be desirable to obtain measures of the employees' perceptions concerning the extent of ego need gratification provided by their jobs. Finally, there are no published studies which have directly tested Argyris' hypothesis that the extent of past and present ego need grati- fication rather than extra-work soxislization determines employee desires for ego need gratification. There are a few articles which have reported relationships between "job level" and employee desires for ego need 46 gratification (Friedlander, 1965; Porter, 1963), but, unfortunately, these studies did not obtain measures of extra-work socialization. It would seem, therefore, that additional investigations of the relationships between the extent of ego need gratification provided by the job, extra-work socialization processes, employee desires for ego need gratification on the job, and employee responses to the job would be worthwhile. Men. In order to help clarify some of the ambiguities mentioned above the present study was designed to test the following hypotheses. gyngthesis 1. The extent of egg need gratification nrovided by the job is ngsitively related to employee attitudes toward the job. This hypothesis is derived directly from the human relations theories. There are two main differences between the present study and most of the previous research designed to test similar hypotheses: (l) The present study employed more direct measurement of the extent of ego need gratification provided by the job, and (2) more of an attempt was made in the present study to measure the variety of job attitudes discussed by the human relations theorists. flynothesis 2. There is an interaction between the extent of ego need gratifi- cation provided by the job and extra~work socialization, such that the exttnt of ego need gtatification nrovided by the job is more nositively telated to attitudes for those individuals snbject to extra-work socitlization 47 r e se whi mo e tro 1 e hasize the i orta ce 0 uch rati i- nation tntn for thone indiyiduala subjent to nocialization.which em- nnasizet sucn gtatification less strongly. This hypothesis is derived from the theoretical arguments and data presented by the critics of the human relations theorists. In addition to the differences mentioned above, the main difference between the present study and the previous research designed to test similar hypotheses is that the present study employed a wider variety of more direct measures of extra-work socialization. This hypothesis is, however, somewhat at variance with the theoretical position of Hulin and Blood (1968) in that these authors predict a negative rela- tionship between ego need gratification and job attitudes for those individuals subject to extra-work socialization fostering alienation from middle class work norms. The present hypothesis merely predicts that the relationship between ego need gratification and job attitudes will be less positive for such individuals than for individuals sub- ject to socialization which fosters more internalization of such norms. flynothenis 3. ere is s ositi e e atio hi betwee extra ork socia i ati fostering internnlization of middle class yotk norms and the inportanne attacned to ego need gtatification on the job. flynnthesis 4. There is an interaction between the importance attached to ego need gratification on the job and the extent of ego need gratification actually provided by the job, such that the gelationshin between the exte o e o eed ratificatio rovided b the o and e lo ee attitudes 48 towa d t e ob is more 0 itive for those 1 dividua fo whom such ggatitication is more innnrtant than tor those individuals for whom nntn gtatification is less inpnttant. These hypotheses represent attempts to investigate the psychological processes underlying the effects of extra work socialization on the relationship between ego need gratification and attitudes toward the job. flyflthes is 5 . Enere is a nositiye relationshin between the extent of ego need ati i atio rovided b the ob and the i rtance attached to such gtatificatinn. This hypothesis is derived from Argyris' rebuttal to the critics of the human relations theorists in relation to the determinants of employee desires for ego need gratification on the job. That is, for the purposes of the present study, it is hypothesized that the deter- minants of employee desires for ego need gratification on the job lie in both off the job socialization and past and present on the job ego need gratification. CHAPTER II 'METHODS Data Collectinn Procedure; The measurement of all the variables in this study was based on responses to a job attitude questionnaire administered to all the employees of six medium sized manufacturing organizations. The spe- cific items designed to measure each variable are presented in Appendix D. This survey was conducted at the request of the Midwest Scanlon Associates, an organization of companies employing the Scanlon Plan, for a first attempt at systematic investigation into the dynamics of the operation of the Scanlon Plan. Representatives from the Midwest Scanlon Associates, who were also managers in three of the six organi- zations where the questionnaire was administered, were involved in all phases of the planning of the research in which the data examined in this study were collected. These representatives were influential in gaining the cooperation of the six organizations which directly par- ticipated in this research. Research teams of from three to five professors and graduate students from the Department of Psychology at Michigan State University visited each organization in the late summer of 1968 in order to ex- plain the purpose and nature of the research and to administer the questionnaires. The president of each organization had sent a letter to each employee at his home explaining the general purpose and nature 49 50 of the research and asking for cooperation before the arrival of the research teams. At each organization, the research teams also held meetings with various management and supervisory groups, the Scanlon Plan Screening Committee, and departmental groups of rank-and-file employees in order to explain the purpose and nature of the research and to answer any questions. The research teams then distributed the questionnaires, which were returned the following day or night. The questionnaires were handed to the employees in envelopes addressed to the Department of Psychology, Michigan State University. Stamped envelopes addressed in this manner were provided for those employees who were absent when the questionnaire was administered. The employees were not asked to sign their names, but enough detailed "background" questions were on the questionnaire that identi- fication would clearly have been no major problem. The research teams mentioned this fact in their meetings with the employees. It was em- phasized, however, that the individual questionnaires would be kept at Michigan State University and their confidentiality was guaranteed. It was specifically emphasized that no one employed by any of the or- ganizations involved in the research would see, or be informed about, any individual questionnaire response. General feedback of the research results was promised for the Scanlon Associates Fall Conference, however, and each company was promised more individualized feedback following this conference. geseartn §ittt The six organizations involved in this study all had the Scanlon Plan and were members of the Midwest Scanlon Associates. These 51 organizations ranged in size from approximately 3,000 to approximately 150 employees. The products involved included casual shoes, rear view mirrors, display items, office furniture, beauty salon furniture, and boat propellers. Hone of the companies could be considered highly automated. Five of the companies were located in the mid-Michigan area, and one company had plants located in Illinois and Canada. Employees from twenty-two geographically separate plants participated in the study. Two companies had one plant; three companies had two plants, and one company had fourteen plants. The plants ranged in size from approximately 10 to approximately 500 employees. new As mentioned above, the questionnaires were distributed to all the employees of these six organizations. A total of 4162 question- naires were given out and 2755 usable questionnaires were returned, providing a return rate of .66. The median educational level for this sample of 2755 employees was 12th grade. Thirty-seven per cent of the sample did not graduate from high school, and 19 per cent had some formal education beyond high school. The median family income for the sample was $7,501-9,500. Forty per cent of the sample had a total family income less than $7,501, and thirty per cent had an income greater than $9,500. Twenty-seven per cent of the respondents' fathers were unskilled workers. Fifty per cent of the sample had fathers who were semiskilled or skilled workers, and twenty-three per cent of the respondents' fathers were in some type of white collar occupation. The median educational level of the fathers of this sample was eighth grade. Twenty per cent of the sample had fathers with less than an 52 eighth grade education, and forty per cent of the respondents' fathers completed more than eight years of formal education. Forty-six per cent of the subjects were raised on farms. Forty-six per cent of the respondents were raised in communities ranging in size from less than 2,000 to 100,000 citizens, and nine per cent of the subjects were raised in cities with populations greater than 100,000. Fifty-one per cent of the subjects were males and 49 per cent were females. ngigitiog agd Hbasurement of Variables The above hypotheses involve four main variables: extra-work socialization, the extent of ego need gratification provided by the job, the importance of ego need gratification, and employee attitudes toward the job. The definition and measurement of each of these variables are discussed below. gxtga-Epgg §ocia1iggtiog As mentioned above, this variable is of central concern to the critics of the human relations theorists (Bulin and Blood, 1968; and Strauss, 1963). Although these authors are not terribly precise in this respect, it is clear that they are referring to conditions and experiences to which the employee is subject off the job. It also seems clear that these conditions and experiences are highly related to variables typically used to index social status or social class. For the purposes of the present study, extra-work socialization is defined as "experiences or conditions to which the employee has been subject off the job and which can theoretically be expected to foster varying degrees of internalization of middle class norms and values concerning work." Hare specifically, the following six variables 53 were used to index the extent of extra-work socialization fostering internalization of middle class work norms: Education Father's education Father's occupation Type of community raised in (urban vs. rural) Type of community of present residence (urban vs. rural) Sex UIJ-‘qu-e I. 0.. 0‘ o E 0 Need ati icatio o ided b the ob This variable is of central importance to the human relations theorists. For the present study, this variable was defined as "the total ego need gratification, from all the sources discussed by the human relations theorists, provided by the job situation." More specifically, the following three variables were used to index the extent of ego need gratification provided by the job. 1. Hierarchical level (position in the hierarchy of authority) 2. Collar Color (White collar - Blue collar) 3. Employee self reports of the extent of challenge, recog- nition, responsibility, freedom to make decisions, chances for promotion, opportunities to learn new skills, and feelings of importance and accomplishment provided by the job. The I ta e of E o e rs ificatio o the o Hypotheses 3 and 4 represent attempts to investigate the psychologi- cal processes involved in the effects of extra work socialization on the relationships between the extent of ego need gratification provided by the job and employee attitudes toward the job. In investigating these processes, the present study focused directly on employee desires for ego need gratification on the job. Related alternatives, of course, would include the broader personality traits and need structures, e.g., authoritarianism, needs for esteem and security, etc., or the more 54 general norms and values, e.g., the "Protestant Ethic," discussed in the literature. Quite simply, the importance of ego need gratification on the job was defined as "the extent to which the employee, considers it important that his job be challenging, and important, and provide recognition, responsibility, a sense of accomplishment, freedom to make decisions, chances for promotion, and opportunities to learn new skills." tti ude Towa d e ob As mentioned above, an attempt was made in the present study to investigate the variety of job attitudes discussed by the human relation theorists. The present study, therefore, included measures of ego involvement, commitment, and identification as well as job satisfaction. While these variables are rarely given precise definitions by the human relations theorists, the following definitions were employed in the present study: 1. Ego involgemcgg: The degree to which an individual is psychologically identified with his work, or the importance of work in his total self image. (Lodahl and Kejner, 1965) 2. Cmitmengz The extent to which the individual is "motivated to perform well because of subjective rewards or feelings that he expects to receive or experience as a result of performing well." (Lawler and Hall, 1970) 3. Identifiggtion: The extent of perceived communality of interests or goals between the individual and the organi- zation. (Patchen, 1965) 4. Job gatisgaggiog: The degree of overall positive affect which the individual feels for his job situation. The correlations between the different measures of extra-work socialization, ego need gratification, and job attitudes are presented in Appendix E. 55 W After the data were collected, the joint frequency distributions for each combination of the different measures of ego need gratifica- tion and extra-work socialization, excluding sex and collar color, were examined, and trichotomies were formed for each of these variables by combining categories according to the following criteria: (1) as few cells as possible should include less than 25 subjects; (2) the combinations of categories should not be entirely arbitrary, but should "make psychological sense;" and (3) the shape of the original distribution should not be grossly distorted. This process resulted in the following categories: 1. Education: High - some college and beyond Medium - high school graduate Low - less than high school graduate Father's education: High - high school education and beyond Medium I some high school Low - eighth grade or less Father's occupation: High - scientist, professional, businessman, executive Medium.- sales, office, supervisor, subprofessional Low - blue collar worker Community raised: High - farm - 10,000 Medium - 10,000 - 100,000 Low - larger than 100,000 Present community: High - farm - 10,000 Medium - 10,000 - 100,000 Low - larger than 100,000 56 6. Hierarchical level: High I supervisors above the first level Medium I first level supervisors Low I rank and file 7. Self report index: High I 3.6 - 5.0 Medium I 2.6 - 3.5 Low I 1.0 - 2.5 For the analyses in‘which the importance of ego need gratification served as an independent variable, it was trichotomized into approxi- mately the top and bottom 30% and saddle 40% of the distribution. Qgta Analysis The data were analyzed with analysis of variance techniques because of the interactive nature of hypotheses 2 and 4. The unequal cell frequencies which were involved in each of the analyses were assumed to reflect population frequency differences. The specific analyses used to test each hypothesis are presented below. Hypothesis 1 predicts a positive relationship between the extent of ego need gratification provided by the job and employee attitudes toward the job. Twelve one-way analyses of variance for unequal cell frequencies were performed to test this hypothesis: one analysis for each combination of the three measures of ego need gratification and the four measures of employee attitudes toward the job. Hypothesis 2 predicts that the extent of ego need gratification provided by the job is more positively related to job attitudes for those individuals subject to extra-work socialization processes which more strongly emphasize the importance of such gratification. Seventy- two least squares two-way analyses of variance were performed to test 57 this hypothesis: one analysis for each combination of the six measures of extra-work socialization, the three measures of ego need gratifica- tion, and the four measures of job attitudes. The interaction terms from these analyses provide the appropriate tests of this hypothesis. Hypothesis 3 predicts a positive relationship between extra-work socialization and the importance of ego need gratification. Six one-way analyses of variance were performed to test this hypothesis: one analysis for each of the six measures of extra-work socialization. Hypothesis 4 predicts that the relationship between the extent of ego need gratification provided by the job and employee attitudes toward the job is more positive for those individuals for whom ego need gratification is relatively more important than for those individuals for whom such gratification is relatively less important. Twelve least squares two~way analyses of variance were performed to test this hypothesis: one analysis for each combination of the three ‘meaaures of ego need gratification and the four measures of attitudes toward the job. The interaction terms from these analyses provide the appropriate tests of this hypothesis. Hypothesis 5 predicts a positive relationship between the extent of ego need gratification provided by the job and the importance of such gratification. Three onedway analyses of variance for unequal cell frequencies were performed to test this hypothesis: none analysis for each of the three measures of ego need gratification. CHAPTER III RESULTS H ot esis l The summaries of the analyses of variance for each combination of the three measures of ego need gratification and the four measures of job attitudes are presented in Tables l-3. The mean job attitude scores for the different levels of each of the measures of ego need gratification are presented in Tables 4-6. The results presented in these tables strongly support hypothesis 1. All twelve of the analyses of variance produced results significant at less than the .001 level. In addition, inspection of the means pre- sented in Tables 4-6 reveals that all the relationships between the measures of ego need gratification and job attitudes are positive and linear. Hypothesis 2 Tables 7-12 summarize the results relevant to the hypothesized interaction between ego need gratification and extra~work socialization. Tables 7-9 present the P and p_values for the interaction terms of the seventy-two analyses of variance (six measures of extra-work socializa- tion X three measures of ego need gratification x four measures of job attitudes). ‘Hore complete analysis of variance summaries for these analyses are presented in Appendix A. 58 59 Table 1. Summary of Analyses of Variance for the Relationships between Hierarchical Level (A) and Job Attitudes. a[ob Attitude Source d_f_ Q E Satisfaction A 2 75.50 72.64* Error 2562 1.04 Identification A 2 20.45 60.52* Error 2566 .34 Involvement A 2 120.85 242.19* Error 2542 .50 Commitment A 2 68.12 154.61* Error 2566 .44 *p < .001 Table 2. Summary of Analyses between Self Report (A) and Job Attitudes. W Satisfaction Identification Involvement Commitment §22££2 of Variance for the Relationship g; 2 2492 2495 2483 2495 HS 91.00 1.01 17.04 .32 45.86 .55 33.65 .46 E 90.09* 52.52* 84.22* 73.60* ”(.001 Table 3. Summary of Analysis of Variance for the Relationships between Collar Color (A) and Job Attitudes. a[ob Attitude Source g; y; g Satisfaction A 1 129.79 122.38* Error 2746 1.06 Identification A 1 24.19 68.97* Error 2750 .35 Involvement A 1 127.32 233.91* Error 2722 .54 Commitment A 1 64.43 135.77* Error 2750 .48 *p‘<.001 Table 4. Mean Job Attitudes by Hierarchical Level. 0 de Satisfaction Identification Involvement Commitment Table 5. Mean Job Attitudes by Self Report. ob A de Satisfaction Identification Involvement Coumitment e ort m 3.48 3.32 3.14 3.48 flieragghical Legel 3.88 3.56 3.38 61 Table 6. Mean Job Attitudes by Collar Color. Collar Qotgt pr attitude glue White Satisfaction 3.31 3.78 Identification 3.12 3.58 Involvement 3.00 3.33 Commitment 3.83 4.03 Table 7. F and 2_Va1ues for Interactions between Extra-Herk Socialization and flietttchital Legel. lob Attitudes Education §atilfiattiog Identititation Inntvement W §_ 1.678 .927 2.032 .105 p_ .152 .447 .087 .981 Father's Education 2_ 1.224 .255 1.616 .269 p, .299 .907 .168 .898 Father's Occupation §_ .468 .193 .571 1.176 2, .760 .942 .683 .320 Community Raised §_ .160 .487 .894 .525 2, .958 .745 .467 .718 Community Present I. 1.995 .853 1.079 .817 p_ .093 .492 .365 .514 Sex 2_ .887 2.440 2.356 .856 g, .412 .087 .095 .425 Table 8. Education I. 2. Father's Education I. 2. Father's Occupation E. 2. Community Raised F .2 Co-unity Present F .2 Sex Fir“ §atittattion .400 .808 .659 .621 .820 .512 2.230 .063 .980 .417 2.789 .062 62 F and p_Values for Interactions between Extra-Work Socialization and Selt Repgt . Job Attttudeg Identifitatiop 1.277 .277 O 742 .563 O 752 .556 3.605 .006 3.695 .005 .776 .460 w .461 .765 .962 .427 .370 .830 2.324 .054 .662 .618 2.872 .057 W 1.316 .261 .951 .433 1.273 .278 2.580 O 036 2.051 .085 2.073 .126 63 Table 9. F and 2,Values for Interaction between Extra-work Socialization and Cottar 0010;. MM Education Satitfacttog Ldentifitation Iggolvemegt Qommitment §_ 23.683 22.231 26.335 6.928 '2 .0005 .0005 .0005 .001 Father's Education 2_ 6.424 1.521 6.277 2.295 2_ .002 .219 .002 .101 Father's Occupation §_ 1.298 2.517 2.866 1.170 2, .273 .081 .057 .310 Community Raised §_ 1.599 1.377 1.910 1.611 2_ .202 .252 .148 .200 Community Present 2, 6.669 1.027 2.911 4.266 2, .001 .358 .055 .014 Sex 2, .011 12.132 12.734 .386 2, .882 .001 .0005 .542 Table 10. Education E d. |§ If Father's Education Father's Occupation 153E Community Raised m 122. 12:. Community Present Satitgacti03 1.04 .76 .56 .69 1.00 .78 .79 .79 .65 .73 1.03 .32 .85 64 Job Attttudes 151th if itatiog W 1.11 .89 .99 .67 .78 .58 .99 .89 1.07 .88 .98 .62 1.02 .76 O 98 O 84 .92 .67 .96 .74 1.01 .70 O 85 O 51 1.00 .70 .85 .55 .85 .78 .65 .49 .94 .72 Hban Differences in Job Attitudes between High and Low Categories of fiterarthical level within Levels of Extra- HOrk Socialization. ggggitment .45 .46 .38 .31 .47 .42 .23 .44 .39 .42 .26 65 Table 11. Mean Differences in Job Attitudes between High and Low Categories of Self geport within Levels of Extra-Work Socialization. Job Attitudes Education Sa tisfatt iog Idegt if icat iog Invo 1veme2t Comiggnt Higt .81 .63 .48 .48 121. .74 .57 .47 .30 M .84 .64 .52 .37 Father's Education 2132 .95 .58 .59 .43 _H_¢_e_<_1_. .69 .41 .54 .29 tgg_ .76 .61 .60 .35 Father's Occupation 12.811 .64 .48 .48 .33 __41. .95 .65 .54 .39 152; .83 .59 .50 .37 Community Raised gggg .85 .62 .50 .37 £22, .45 .18 .27 .17 £22, .88 .58 .57 .39 Community Present M .82 .63 .49 .38 l_ile_c_1_. .60 .30 .38 . 19 Leg, .86 .24 .39 .40 Sex Zenale .70 .54 .44 .31 Male .94 .57 .54 .42 Table 12. Education 1 WE Father's Education 156E rather's Occupation 2.1111 Eli Community Raised Lian £51- Les. Conmunity Present am M- 1.421 Sex £29.15 gale Satisfactiog 1.05 .54 .21 .78 .62 .41 .61 .64 .46 .41 .61 .26 .95 .47 Idegtificatiog 84 :43 .26 .55 .62 .47 .63 .55 .44 .42 .53 .55 .46 .58 .50 .74 .29 .16 .52 .46 .28 .48 .45 .30 .08 .42 .20 .29 .26 .56 .18 .39 Mean Differences in Job Attitudes between White and Blue Collar Groups within Levels of Extra-Work Socialization. Jot Atttttdes Involvement 92m; .41 .22 .15 .30 .18 .23 .33 .22 .17 .27 .28 .19 .12 .44 .19 .23 67 Tables 10-12 present the differences in mean job attitude scores between the high and low levels of ego need gratification within each level of extra-work socialization. More complete matrices of means for these analyses are presented in Appendix B. Ten, or approximately 141, of the total 72 analyses yielded re- sults significant at less than the .01 level in the predicted direction. The P values for 7 of these 10 analyses were significant at less than the .001 level. lbre specifically, results of the following analyses were in the predicted direction at less than the .01 level of signifi- cance: education by collar color on satisfaction, education by collar color on identification, education by collar color on involvement, education by collar color on commitment, father's education by collar color on satisfaction, father's education by collar color on involve- ment, community raised by self report on identification, present community by self report on identification, sex by collar color on identification, sex by collar color on involvement. The following three analyses yielded results significant at less than the .05 level in the direction opposite to that predicted: communi- ty raised by self report on commitment, present community by collar color on satisfaction, present community by collar color on commitment. Overall, 44 of the 72 analyses yielded results in the predicted direction. That is, for 611 of the analyses relevant to hypothesis 2, the mean differences in attitudes between the high and low levels of ego need gratification were higher for the levels of extra-work socialization expected to foster "high" internalization of middle class norms than for the levels of extra-work socialization expected to foster "low" internalization of such norms. Twenty-eight of the 68 72 analyses yielded results opposite to those predicted. That is, for 391 of the analyses relevant to hypothesis 2, the mean differences in job attitudes between the high and low levels of ego need gratifi- cation were nggt_for the levels of extra-work socialization expected to foster "high" internalization of middle class work norms than for the levels of extra-work socialization expected to foster "low" levels of internalization of such norms. In general, then, across the different measures of ego need gratification, extradwork socialization and job attitudes, the re- sults provided marginal support for hypothesis 2. A more detailed examination of tables 7-12, however, indicates that the different measures of ego need gratification and extradwork socialization yielded differential results in relation to this hypothesis. For example, only 10 of the 24 analyses employing the self report measure of ego need gratification yielded results in the predicted direction, while 16 of the analyses employing collar color and 18 of the analyses em- ploying hierarchical level as measures of ego need gratification yielded such results. The six measures of extra-work socialization also produced different results in relation to this hypothesis. There were 12 analyses for each of these measures. Of these, only 3 of the analyses employing present community, 5 of the analyses employing community raised, and 6 of the analyses employing father's occupation as the measures of extra-work socialization yielded results in the pre- dicted direction. On the other hand, 9 of the analyses employing edu- cation, 9 of the analyses employing father's education, and all 12 of the analyses employing sex as the measure of extra-work socialization yielded results in the predicted direction. 69 1e e t A as In order to clarify the interpretation of the relationships between ego need gratification and job attitudes reported in relation to hypothesis 1, shmple effects analyses of variance were performed for those analyses which yielded significant interactions between extra- work socialization and ego need gratification. More specifically, the simple effects of ego need gratification on job attitudes were analyzed within each level of extra-work socialization for the 13 combi- nations of ego need gratification, extra-work socialization, and job attitude which produced significant interactions. Summaries of these analyses are presented in Appendix C. All but 3 of these 37 analyses yielded results in the direction predicted by hypothesis 1 at less than the .01 level of significance. That is, the relationships be- tween ego need gratification and job attitudes were significant and positive in 34 of the 37 simple effects analyses. The relationships between ego need gratification and job attitudes were also positive for the three analyses which were not statistically significant: self report on identification for the "medium level" of community raised 0-1.56, de2, 2414, 2).05) self report on coursitment for the "medium level" of co-annity raised (PI1.35, de2, 2426, 2).05); self report on identification for the "low level" of present community (PI1.01, de2, 2412, 2).”) . W The results relevant to hypothesis 3 are presented in Tables 13-14. Summaries of the analyses of variance for the six measures of extradwork socialization on the importance of ego need gratification are presented 70 in Table 13, and the mean importance scores by level of extra-work socialization are presented in Table 14. As indicated in Table 13, the relationships between each measure of extra-work socialization and the importance of ego need gratification.were significant at less than the .0005 level. In addition, inspection of Table 14 reveals that the relationships between the importance of ego need gratification and education, father's education, father's occupation, and sex are in the direction predicted by hypothesis 3. That is, for these measures of extra-work socialization the mean importance of ego need gratifica- tion'was higher for the levels of extra-work socialization expected to foster "high" internalization of middle class norms than for the levels of extra-work socialization expected to foster "low" levels of internalization of such norms. In contrast, the community raised and present community measures yielded relationships with the importance of ego need gratification in the opposite direction. That is, the size of community raised in and the size of community of present (as of 1968) residence were positively related to the importance of ego need gratifi- cation. gypothesit 4 The results relevant to hypothesis 4 are presented in Tables 15-29. Summaries of the interactions between the three measures of ego need gratification and the importance of ego need gratification on satisfac- tion, identification, involvement, and commitment are presented in Thbles 15-17. The matrices of means for these analyses are presented in Tables 18-29. 71 Table 13. Summary of Analyses of Variance for the Relationships between Extra-Work Socialization and the Importance of Ego Heed Gratification. §ogrce t1; Q 1 Education 2 9.768 16.024* Error 2550 .609 Father's Education 2 6.789 11.331* Error 2293 .599 Father's Occupation 2 4.866 8.103* Error 2427 .600 Community Raised 2 4.748 7.712* Error 2552 .616 Present Community 2 8.245 13.412* Error 2549 .615 Sex 1 30.058 49.483* Error 2662 .607 ' *p‘(.001 Table 14. Mean Importance of Ego Heed Gratification for Levels of Extra-Work Socialization (EWS). mu 915.129.1213. 140' Medium High Education 3.52 3.57 3.76 Father's Education 3.54 3.69 3.70 Father's Occupation 3.56 3.68 3.73 Community Raised 3.66 3.73 3.56 Present Community 3 . 75 3 . 73 3 . 55 Sex (low-female, 3.48 3.69 high-male) 72 Examination of the results reported in these tables reveals little support for hypothesis 4 across the three different measures of ego need gratification. Only two of the interaction term F values reported in Tables 15-17 reach the .05 level of significance (self report by importance on identification, collar color by importance on identification). Furthermore, inspection of Table 23 reveals that one of these significant interactions (self report by importance on identification) is opposite to the direction predicted by hypothesis 4. That is, the mean difference in identification between the low and high levels of self report is greater for the lgg_level of importance (.76) than for the high level of importance (.40). In addition, inspection of Tables 18-29 reveals that all 8 of the analyses for the hierarchical level and self report measures of ego need gratifi- cation yielded results opposite in direction to those predicted by hypothesis 4. Only the 4 analyses in'which collar color served as the measure of ego need gratification yielded results in the direction predicted by hypothesis 4, while 3 of the 4 analyses employing hier- archical level and all 4 of the analyses employing the self report measure of ego need gratification produced results opposite to those hypothesized. §tgple Eftettt Simple effects analyses of variance were performed for collar color on identification and self report on identification within each level of the importance of ego need gratification. The results of these analyses are summarized in Tables 30 and 31. Table 15. Summary of Interactions between Hierarchical Level (A) and the Importance of Ego Heed Gratification (B). Job Attttgde Satisfaction Identification Involvement Commitment Sogtte A x 8 Error A x 3 Error A x B Error A x B Error 9; 4 2481 4 2462 4 2484 4 2484 Q E. .339 .331 1.022 .704 1.456 .484 .407 .9519 .427 .333 1.039 .320 Table 16. Summary of Interactions between Self Report (A) and the Importance of Ego Heed Gratification (B). 122_A5tit222. Satisfaction Identification Involvement Commitment 22253 A x B Error A x 8 Error A x B Error A x B Error g; 4 2451 2441 2453 2453 11 2. 1.882 1.875 1.004 1.303 2.431* .536 .821 1.817 .452 .174 .545 .319 *p(.05 Table 17. Sumary of Interactions between Collar Color (A) and the Importance of Ego Heed Gratification (B). o i u e Sottte g; Lg 1; Satisfaction A x B 2 .676 .649 Error 2658 1.041 Identification A x B 2 1.582 3.010* Error 2636 .526 Involvement A x B 2 .466 1.015 Error 2661 .459 Comitment A x B 2 .282 .854 Error 2661 .331 *p ( .05 Table 18. Mean Sattgtattion for Importance by Hierarchical Level. Importance 1a 3.4- 192 3.27 3.92 m. 3.37 3.97 High 3.49 3.95 Total 3.37 3.95 iea c Total fiigh-Low 3.35 .76 3.48 .85 3.60 .73 3.48 Table 19. Mean Idegtifttation for Importance by Hierarchical Level. attrartgital Level Importance Lg; M. 2131; Total gigh-Low a 3.13 3.87 4.03 3.22 .90 m. 3.13 3.90 4.28 3.27 1.15 M 3.27 3.94 4.12 3.41 .85 Total 3.17 3.91 4.15 3.30 75 Table 20. Mean W for Importance by Hierarchical Level. flietarthical Legal Importance 129. an. m ____.Toca1 mm 1422 2.96 3.53 3.71 3.03 .75 221- 3.01 3.61 3.84 3.12 .83 m 3.14 3.65 3.75 3.25 .61 Total 3.03 3.61 3.77 3.13 Table 21. Mean Cmmaitment for Importance by Hierarchical Level. gietatthical Level Importance E! 34. 3.1.811 Total 1 h- M 3.83 4.25 4.22 3.87 .39 l_ie_d_. 3.85 4.15 4.30 3.90 .45 M 3.93 4.19 4.22 3.98 .29 Total 3.86 4.18 4.25 3.92 Table 22. Mean Satittattiog for Importance by Self-Report. e -Re t Importance l_.gw_ mg. m Total i -Low 1.5 2.94 3.40 4.00 3.35 1.06 1114- 3.15 3.49 3.81 3.47 .66 m 3.09 3.58 3.87 3.60 .78 Tottl 3.06 3.48 3.87 3.47 Table 23. Importance 0t Table 24. Importance EEEE Table 25. Importance 1:. Le: use. nun 1.9.1 124. M 1.2.1 76 Mean Idgntttitatton for Importance by Self-Report. mm .ng 624- m L_ota1 mL-ww 2.90 3.26 3.66 3.21 .76 2.98 3.32 3.53 3.28 .55 3.15 3.44 3.55 3.43 .40 2.99 3.33 3.57 3.30 Mean Wt for Importance by Self-Report. Selt-3e2gtt 1a. su- m 12.—tel nus-:14: 2.79 3.07 3.43 3.04 .64 2.93 3.13 3.35 3.12 .42 2.99 3.25 3.38 3.25 .39 2.89 3.14 3.38 3.14 Mean W for Importance by Self-Report. Se - Les. nes- m T___1°t8 mm 3.71 3.92 4.12 3.89 .41 3.77 3.89 4.07 3.90 .30 3.79 3.96 4.12 3.99 .33 3.75 3.92 4.10 3.92 77 Table 26. Mean Satigftttton for Importance by Collar Color. Collat Color Importance Whtte 2192 Total flite-Blue kg! 3.61 3.22 3.32 .39 M. 3.81 3.31 3.47 .50 M 3.88 3.42 3.59 .46 Total 3.79 3.31 3.46 Table 27. Mean Identifitatiog for Importance by Collar Color. gum Importance mm 3.1112 mm W m 3.44 3.10 3.19 .34 Ldt 3.61 3.09 3.26 .52 5122 3.70 3.24 3.41 .46 Total 3.60 3.13 3.28 Table 28. Mean M for Importance by Collar Color. Col a o 0 Importance ice use 1.9.22.1. W _1.ng_ 3.20 2.94 3.01 .26 _M_e_cl. 3.33 2.99 3.10 .34 High 3.46 3.10 3.24 .30 Tota 3.34 3.00 3.11 78 Table 29. Mean Cglgitmegt for Importance by Collar Color. tolls; Color Importance ite glue Total White-Blue ng_ 3.96 3.82 3.86 .14 522. 4.04 3.82 3.89 .22 High 4.11 3.90 3.98 .21 Total 4.05 3.84 3.91 Table 30. Summary of Simple Effects Analyses for Self-Report (A) on Identification within Levels of the Importance of Ego Heed Gratification (Bl-Low, BZIMedium, B3IHigh Importance). §ourte g; E g A/Bl 2 18.602 34.704* A/Bz 2 16.364 30.529* A/B3 2 6.278 11.713* Error 2441 .536 *p¢L.001 Table 31. Summary of Simple Effects Analyses for Collar Color (A) on Identification‘within Levels of the Importance of Ego Need Gratification (B1ILow, BzIMedium, B3IHigh Importance). oce u 1s 2 A/B1 1 17.044 32.403* Ale 1 62.517 118.854* A/B3 1 37.622 71.525* Error 2636 .526 *p<(.001 79 Inspection of Tables 30 and 31 reveals that the self report and collar color measures of ego need gratification are related to identi- fication at less than the .001 lvel of significance within each level of the importance of ego need gratification. Inspection of Tables 23 and 27 also reveals that in each case the relationships are positive and linear. Hy22thesis 5 The results relevant to the hypothesized relationship between the extent of ego need gratification and the importance of ego need gratification are presented in Tables 32-33. Summaries of the analyses of variance for hierarchical level, self report, and collar color on the importance of ego need gratifi- cation are presented in Table 32. The means for these analyses are presented in Table 33. All three of the F values reported in Table 32 are significant at less than the .001 level. Inspection of the means reported in Table 33 also reveals that in each case the relationships are positive and linear. 80 Table 32. Summary of Analyses of Variance for Hierarchical Level (A), Self Report (B) and Collar Color (C) on the Importance of Ego Heed Gratification. Sogrce g; Q g A 2 5.923 9.693* Error 2490 .611 B 2 41.239 70.928* Error 2459 .581 C 1 14.366 23.406* Error 2665 .614 *p‘(.001 Table 33. Mean Importance of Ego Need Gratification by Hierarchical Level, Self Report and Collar Color. Hieratthical Level Self gepott gollat golot High 3.79 3.95 3.70 Medium 3.72 3.53 Low 3.52 3.44 3.54 CHAPTER IV DISCUSSION Sggggry of Resglts The results of this study consistently supported the human relations hypothesis of a positive relationship between the extent of ego need gratification and employee attitudes toward the job situation. The same results were obtained whether hierarchical level, collar color, or employee perceptions were used to index the extent of ego need gratification: the greater the ego need gratification provided by the job, the greater the reported satisfaction, identifi- cation, involvement, and commitment. All twelve of these relation- ships reached high levels of statistical significance (p (.001). Due to the methodological problems discussed below, these results cannot be considered conclusive, but within the limitations of the present study, the human relations hypothesis did receive substantial support. The support obtained for the hypothesized interaction between the extent of ego need gratification provided by the job and extra- work socialization, on the other hand, was quite marginal. The results of only 141 of the total 72 analyses relevant to this hypo- thesis were statistically significant in the predicted direction. As indicated in the results section, however, there was variability in the extent of support provided for this hypothesis by the different 81 82 measures of ego need gratification and extra-work socialization. In general, the results obtained employing education, father's education, and sex as measures of extra-work socialization supported the hypothesized interaction, while the results obtained employing father's occupation, community raised, and present community did not support this hypothesis. There were twelve interactions for each measure of extra-work socialization. Of these 12 interactions, 9 employing education were in the predicted direction, 4 at less than the .0005 level of significance, and 3 were in the opposite direction, none at less than the .05 level of significance. Of the 12 interactions employing father's education, 9 were in the predicted direction, 2 at less than the .01 level of significance, and 3 were in the opposite direction, none at the conventionally minimal level of significance. All 12 of the interactions employing sex were in the predicted direction, 2 at less than the .001 level of significance. On the other hand, there were no significant interactions employing father's occupation as the measure of extra-work socialization, and half of the results obtained with this measure were opposite to the hypothesized direction. Simi- larly, only 5 of the twelve interactions for community raised and only 3 of the 12 for present community were in the predicted direc- tion. Community raised produced one interaction in the predicted direction at less than the .01 level of significance and one in the opposite direction at less than the .05 level of significance. One of the interactions for present community was in the predicted direction at less than the .01 level of significance, and two of the interactions for this variable were statistically significant in the opposite direction. 83 The results obtained for the collar color and hierarchical level measures of ego need gratification generally supported the hypothesized interaction between the extent of ego need gratification and extra-work socialization, while the results obtained for the self report index of ego need gratification did not support this hypothesis. Excluding the analyses which employed community raised, present community, and father's occupation as the measures of extra-work socialization, there were twelve analyses for each measure of ego need gratification. All twelve of these analyses which employed collar color as the measure of ego need gratification produced interactions in the predicted direction, and half of these were significant at less than the .001 level. Similarly, although none of the analyses for hierarchical level produced interactions which reached statistical significance, ll of the 12 were in the pre- dicted direction. In contrast, none of the 12 analyses for the self report measure of ego need gratification produced significant inter- actions, and only 7 of the interactions obtained with this measure were in the predicted direction. Although direct comparison is impossible since all the measures differed greatly, it seems interesting that the measures of extra- work socialization employed in the present study (community raised and present community) which appear most parallel to the measures of this variable employed in the research (Blood and Hulin, 1967; Turner and Lawrence, 1965) cited by the human relations critics (Hulin and Blood, 1968) as supporting their hypothesis produced the least support for this hypothesis in the present study. These results are especially surprising since the measures employed in the present study would appear to be more direct indices of the variable of interest than the measures 84 previously employed (plant site). The results of the simple effects analyses provided no support for the theoretical position of the critics of the human relations theorists, although they were not inconsistent with the hypotheses tested in the present study. Both Strauss (1963) and Hulin and Blood (1968) argue that ego need gratification is negatively related to job attitudes for those employees who have been "alienated" from middle class work norms. The present study found no evidence to support this hypothesis, and considerable evidence which may be interpreted as contradicting it. All 72 of the mean differences in job attitudes between the high and low levels of ego need gratification within the levels of extra- work socialization expected to foster the least internalization of mdddle class work norms‘were in the direction predicted by the human relations hypothesis. Similarly, the results of 34 of the 37 simple effects analyses for the relationships between ego need gratification and job attitudes within levels of extra-work socialization were in the direction predicted by the human relations theorists at less than the .01 level of significance. The results relevant to the hypothesized role of the importance of ego need gratification as an intervening variable "producing" the interaction between extra-work socialization and the extent of ego need gratification provided by the job were rather inconsistent. The hypothesized positive relationship between the extent of ego need gratification and the importance of ego need gratification was supported for all three measures of ego need gratification. In 85 addition, the results strongly supported the hypothesized positive relationship between extra-work socialization and the importance of ego need gratification for 4 of the six measures of extra-work sociali- zation (education, father's education, father's occupation, and sex). Contrary to predictions based on the results of the Blood and Hulin (1967) and Turner and Lawrence (1965) studies, however, the size of community raised in and the size of present community were positively related to the importance of ego need gratification. Again, the measures of extra-work socialization.which most closely paralleled the measures used by Blood and Hulin (1968) and Turner and Lawrence (1965) provided the least support for the results of these studies. Finally, no support was obtained for the hypothesized interaction between the importance of ego need gratification and the extent of ego need gratification provided by the job. These results are hardly surprising in light of the rather marginal support obtained for the interaction between extra-work socialization and the extent of ego need gratification provided by the job. L131 tat iogs As is typically the case in organizational psychology, the most serious deficiencies of the present study involve measurement and re- search design. The measurement of ego need gratification and extra- work socialization, although generally superior to the previous research in this area, were particularly deficient in relation to the hypothe- tically ideal situation. The "ideal" measurement of the extent of ego need gratification would have involved the systematic observation of job behavior and the organizational context in which this behavior 86 occurs. Time and resource limitations typically preclude the ideal, however, and hierarchical level, employee self reports, and collar color, were used to index the extent of ego need gratification provided by the job. While there is little doubt that these measures are generally valid indices of the extent of ego need gratification pro- vided by the job, they are far from models of precision in this re- spect. For example, while white collar jobs and jobs higher in the hierarchy of authority do provide more ego need gratification than blue collar jobs and jobs lower in the hierarchy on the average, there very likely is a great deal of overlap in these distributions, and there certainly is a great deal of variance of ego need gratification within such broad groupings of jobs. Similarly, job status and hierarchical level reflect differences other than the extent of ego need gratifi- cation provided by the jobs. Among other things, salary and working conditions can be expected to be related to these variables. Employee self reports, of course, are subject to all of the limitations of re- sponse bias and outright distortion involved in any self report measure. The measurement of extra-work socialization employed in the present study was also far from perfect. In this case, the ideal measurement would involve direct observation of the socialization process of interest, e.g. child rearing practices and early life peer and refer- ence group normative influence. In the absence of such difficult to obtain data, employee reports of educational level, father's education, father's occupation, type of community raised in, type of present community, and sex were used to index the extent of extra-work sociali- zation fostering integration with middle class work norms. Although it would seem that these measures would provide more direct indices 87 of the socialization processes of interest than plant site, the index used by Blood and Hulin (1967) and Turner and Lawrence (1965), they are certainly subject to the same limitations of imprecision dis- cussed in relation to the measures of ego need gratification. Another possible limitation of the measurement of extra-work socialization could involve the non-random subject sampling employed in the present study. Even though 2755 employees from 6 organizations and 22 geographically separated plants participated in this study, these subjects can hardly be considered a representative sample of industrial employees. To begin‘with, all of the organizations in- volved in this study were operating under the Scanlon Plan and were committed to the philosophy of participative management at the policy level. Secondly, all of the plants were located in the Midwest, and only 2 of these plants were located in large, industrialized cities. Both of these factors could restrict the range of extra-work sociali- zation, such that the extremely "alienated" tail of the distribution may have been excluded from the present sample. The fact that all of the companies operated under the Scanlon Plan could result in the above bias through self and organizational selec- tion and retention processes. That is, it may be that highly "alienated" employees simply do not choose to join, are not selected by, or do not for any appreciable length of time remain in Scanlon Plan companies. Such processes would certainly seem plausible. Similarly, the present sample is obviously biased geographically. When one thinks of the typical modern American industrial employee, one does not think of someone working in a small town in the Midwest. The fact that "type of community" seemed to "work" in the Blood and Hulin (1967) and Turner 88 and Lawrence (1965) studies but did not "work" in the present study would tend to reinforce this suspicion of bias if one assumes, as seems likely, that there was greater variance in the geographical location of the plants involved in these studies than in the present study. In addition to the problems of imprecise measurement and possibly biased subject sampling, the most serious limitation of the present study lies in the cross sectional design employed. Conclusive answers to the theoretical questions concerning the effects of ego need gra- tification, and the interaction between this variable and extra-work socialization on employee responses to the job can only be provided by tightly controlled experimental or longitudinal designs. The practical obstacles to such designs are, of course, formidable. Such obstacles must be overcome, however, if we are to obtain conclusive answers to research problems of the type involved in the present study. In addition to the sampling problem discussed above, for ex- ample, the failure of the results of the present study to provide much support for the theoretical position of the human relations critics could have been due to the fact that even.within the different "levels" of extra-work socialization, those individuals who were more integrated with middle class work norms‘were more likely to choose and/or be selected into jobs which provide high levels of ego need gratification than those individuals less integrated with middle class work norms. Certainly the questions of causal direction involved in theoretical positions of both the human relations theorists and their critics can only be conclusively answered through experimental or longitudinal re- search designs. 89 Lgtetptetatiogt agd Cogtlgstons The limitations of measurement and design discussed above must be recognized in any interpretation of the results of this study. However, the results obtained cannot be discounted or ignored because of these limitations. The extent of ego need gratification provided by the job, as indexed by hierarchical level, collar color, and employee self reports, was consistently related to employee attitudes in the manner predicted by the human relations hypothesis. In con- trast, little support was obtained for the interaction between extra- work socialization and ego need gratification predicted by the critics of the human relations theorists. Similarly, the results of the simple effects analyses, concerning'which the human relations theorists and their critics make opposite predictions, strongly supported the human relations theorists prediction and directly contradicted the prediction of their critics. For the measures and subject sample employed in the present study, therefore, the human relations hypothesis received much stronger support than the theoretical position of the critics of that hypothesis. The limitations discussed above necessitate caution in generaliz- ing these results for different measures and different subject samples. Considerable generality is implied, however, by the fact that the results were based on mltiple measures of each variable from over 2700 employees in 22 separate plants of 6 different organizations. The problem of restriction of the generality of the present re- sults due to the non-random selection of organizations raises some interesting theoretical questions. The discussion of possible bias resulting from the selection of organizations employing the Scanlon 90 Plan essentially suggests that the human relations hypothesis may be, in part, a "self fulfilling prophesy," in that organizations which are committed to this hypothesis as a matter of policy may attract, select, and retain employees who are more likely to confirm the hypothesis. The effects of management assumptions about the motivation of its employees on the behavior of these employees are given a great deal of attention by the human relations theorists, especially Argyris (1964) and McGregor (1960). This attention, however, is focused on the effects of such assumptions on the extent of ego need gratification provided for employees. It is suggested here that per- haps such assumptions also help to determine which type of people become and remain members of a given organization, and therefore how the employees of a given organization will respond to the ego need gratification which is provided. These processes need not be mutually exclusive, of course, but rather should probably be mutually reinforc- ing and cumulative in their effects. iMsngement assumptions about employee motivation might also affect employee reactions to the extent of ego need gratification provided by the job through the effects such assumptions cauld have on the "socialization" to which an employee is subject within the organization. It seems reasonable that the socialization to which an employee is subject within the organization could influence his behavior on the job just as much, if not more than, the socialization to which he is subject outside of the organization. "Intra-organizational" socialization would include formal orientation and training programs, as well as informal peer and reference group influence processes. 91 The main point of this discussion is that while it seems theore- tically reasonable to expect individual differences in responses to the extent of ego need gratification provided by the job, it seems fruitful to view such differences as subject to change, and to look for the sources of variance in such differences in the conditions which the employee experiences within the organization, as well as the sociali- zation to which he is subject outside the organization. The results of the present study, while far from conclusive, are certainly consis- tent with such a perspective. BIBLIOGRAPHY BIBLIOGRAPHY Adorno, T.W., Frankel-Brunswick, Else, Levinson, D. 6. Sanford, R. The Authgtitarian personality. New York: Harper, 1950. Allport, F. T_h_eories of pettepttop apt! the topcept ot pttpttute. New York: Wiley, 1954. Anderson, L. and Fiedler, F. The effect of participatory and supervisory leadership on group creativity. Jpprnal of Applied Pspthotogp, 1964, 48, 227-236. Aronoff, J. Pspchologipal Eeds apd cultural systegp. New York: Van Nos trum, 1967 . Aronoff, J. 6: these, L. Experiments on the effects of personality processes on small group structure. Unpublished manuscript, Michigan State University, 1969. Argyris, C. I te ati the i ivid l a t e or a i ation. New York: Wiley, 1964. Atkinson, J. Ap introduttiop to motivatiop. New York: Van Nostrum, 1964 . Backman, J., Bowers, D., 6 Marcus, P. Bases of supervisory power: a comparative study in five organizational settings. Paper delivered at A.P.A. Convention, 1965. Backman, J., Smith, C., 6- Slesinger, J. Control, performance, and satisfaction: an analysis of structural and individual effects. legal 0: P9150211“! and §oc1a1I Ptpthptpgy, 1966, 4, 127-136. B'avelas, A. 6: Strauss, C. Group dynamics and intergroup relations. In W. Bennie, K. Benne, R. Chin (Eds.) The Planning 0; thapge. New York: Holt, 1961, pp. 587-591. Bennett, E.B. Discussion, decision, commitment, and consensus in group decision. up Relatiopt, 1955, L, 251-274. Biganne, J., 6: Stewart, P. Job enlargement: A case study. Research Series, No. 25, 1963, State University of Iowa, Bureau of Labor and Management . 92 93 Alienation, environmental characteristics , Blood, M. G: Hulin, C. and worker responses. {091921 at Appligd Pspcpologp, 1967, 21 , 284-290. Bowers, D. Organizational control in an insurance company. Sptiomettx, 1964, 27, 230-244. Bowers, D. G: Seashore, S. Predicting organizational effectiveness with a four-factor theory of leadership. Admipistrative Sptepttftt mattertp, 1966, LUZ): 238-263. Are Herzberg's motivators and hygienes undimensional? Burke, R. Jyrgl of Apthed Psxphologg, 1966, Q, 317-21. Centers, R. 6: Bugental, D.E. Intrinsic and extrinsic job motivations among different segments of the working pepulation. o of Applied Pszthologz, 1966, _6_Q, 193-97. Cleven, W. 6: Fiedler, F. Interpersonal perceptions of open-hearth foreman and steel production. legal ot Applted Pspchologz, 1956, Q, 312-314. Coch, L. 6 French, J. Overcoming resistance to change. m 3.22116“, 1946, _1_, 512-532. Day, R. 6: Hamblin, R. Some effects of close and punitive styles of supervision. Amettcap Mpg; ot §otio1pgp, 1964, 22, 499-510. Dunnette, M.D., Campbell, J.P., 6 Hakel, M.D. Factors contributing to job satisfaction and job dissatisfaction in six occupational groups. MW. 1967. 2, 143-74. Elliott, J. Increasing office productivity through job enlargement. 11‘s tap tide 0; tbs ofttte gmgep's job. (No. 134 Office Management Series) New York: American lbnagement Association, 1953, 3-15. Et‘n, M. Relationships between self-perceived personality traits and job attitudes in middle management. Joprppl of Applied_ mum. 1966. 9.2. 424-30. A study of the a'eln, R.B. Some determinants of job satisfaction: generality of Herzberg's Theory. MIMI ot Applieg Pppttptogz, 1964, it, 161-63. E'Ien, 11.3., Smith, P.C., Hulin, c.1.., and Locke, 3.6. An empirical test of the Herzberg Two-Factor Theory. legal of Applied Ppptpologz, 1966, 29,, 544-50. l”‘IItz, R. lbtivationsl factors in re-habilitation. Unpublished Ph.D. thesis, Western Reserve University, 1962. 94 Fiedler, F. A theorp of leaderthip effectivepess. New York: IEGraw-Hill , 1967 . Fiedler, F. Assumed similarity measures as predictors of team effectiveness. Japtppl of Abnotgpl app Social Pspthologp. 1954, 22, 381-388. Fiedler, F. The influence of leader-keyman relations on combat crew effectiveness. ou of Ab o l and Social Pa h lo , 1955, 2;, 227-235. Fiedler, F., Bass, A., and Fiedler, J. The leader's perception of co-workers, group climate, and group creativity: a cross- validation. urbane, Illinois; Group Effectiveness Research Laboratory, University of Illinois, 1961. (Mimeograph) Fiedler, F., Mbuwese, and Oonk, S. Performance of laboratory tasks requiring group creativity. Acta prppologict, 1961, _1_§_, 100-119. Fine, S. and Dickman, R. Satisfaction and productivity. Paper presented at the American Psychological Association Convention, St. Louis, Missouri, September, 1962. Fleishman, F. A leader behavior description for industry. In R. Stogdill 6.A. Coons (Eds.) Leader fighavior, its description and meapuremept. Columbus, Ohio: Bureau of Business Research, 1957. Fleishman, E. The measurement of leadership attitudes in industry. Jogtnal pf Applied Pszthologg, 1953, 21, 153-158. Fleishman, E., Harris, E. 6 Burt, H. Leadeptplp t tpperyisiOp ip lpgppttz, Columbus, Ohio: Ohio State university, Bureau of Educational Research, Monograph No. 33, 1955. Fleishman, F. 6 Peters, D. Interpersonal values, leadership attitudes and managerial success. gppppnngl_£pyphplpgg, 1962, l2, 127-143. French, 3., Israel, 6.As. D. An experiment on participation in a Norwegian factory. flpppp gelatipps, 1960, l§(l), 3-19. Friedlander, F. Job characteristics as satisfiers and dissatisfiers. gpptppl of Applied Psptpologg, 1964, g&, 388-92. Friedlander, F. Relationships between the importance and the satis- faction of various environmental factors. Joutppl at Applied M281: 1955. 9.2. 160'154- Friedlander, F. Underlying sources of job satisfaction. Journal of Applied Pspthologp, 1963, 21, 246-50. 95 Friedlander, F. and Walton, E. Positive and negative motivations toward work. Administrative §ciepte Qpapterlp, 1964, 9, 194-207. Friedlander, F. IMotivation to work and organizational performance. Joprnal of Applied Ppptpologp, 1966, 29, 143-52. (a) Friedlander, F. Importance of work versus non-work among socially and occupationally stratified groups. gpppnal ct Applied Psychology, 1966, 99, 437-41. (b) Gibson, J. W. Sources of job satisfaction and job dissatisfaction as interpreted from analysis of write-in responses. Unpublished Ph.D. thesis, Western Reserve University, 1961. Godfrey, E., Fiedler, F., and Hall, D. BosrdpI mappgerg, and poppanx sptcess. Danville, Illinois: Interstate Press, 1959. Gordon, G.G. The relationship of satisfiers and dissatisfiers to productivity, turn-over, and morale. Paper presented at the American Psychological Association Convention, Chicago, Illinois, September, 1965. Gouldner, A. Patternp of indpsttial bureaucraty. New York: Free Press, 1954. Graen, G. Motivator and hygiene dimensions for research and develop- ment engineers. Journal of Applied Pschologp, 1966, 99, 563-66. Graen, G. 6 Hulin, C. Addendum to "an empirical investigation of two implications of the two-factor theory of job satisfaction." Joptppl of Applied Ptpphology, 1968, 92, 341-342. Hahn, C. Dimensions of job satisfaction and career motivation. Un- published manuscript, 1959. Cited in House Wigdor, Herzberg's Theory of job satisfaction and motivation. Persepnel Psychologp, 1967, 29, 369-389. Halpern, G. Relative contributions of motivator and hygiene factors to overall job satisfaction. gaprppl of Applied Payphologp, 1966, 29, 198-200. Halpin, A. 6 Winer, B. A factorial study of the Leader Behavior Description. In R. Stogdill 6.A. Coons (Eds.) Leader Ephavior: its desttlptiop apd ppapptpppn . Columbus, Ohio: Bureau of Columbus, Ohio: Bureau of Business Research, 1957. Hamlin, R.M. and Name, R.S. Self-actualization in choice scores of improved schizophrenics. gbprnal of Qlipltal Psptpologp, 1962, 29, 51-54. Haythorn, W., Haefner, D., Couch, A., 6 Carter, L. The effects of varying combinations of authoritarian leaders and followers. Japtppl of Abpptmal egg §otla1 Pszthologz, 1956, 2;, 210-219. 96 Haywood, H.C. and Dobbs, V. Motivation and anxiety in high school bOYS- Jasrnal_2f_£stsenslitx. 1964. 32. 371-79- Hemphill, J. Leadership behavior associated with the administrative reputation of college departments. Jpppppl_pfi_£gppptlpngl W. 1955. i. 385-401- Herzberg, P. War a the tu e of man. Cleveland: The World Publishing Co., 1966. Herzberg, F., Mausner, B. and Snyderman, B. The motivatipn to wort. New York: John Wiley, 1959. Hinrichs, J.R. and Mischkind, L.A. Empirical and theoretical limita- tions of the two-factor hypothesis of job satisfaction. Journal of Applied Pszchologp, 1967, él, 191-200. Hinton, B. An empirical investigation of the Herzberg Methodology and the two-factor theory. Organizational Behavior and Human Perfotmance, 1968, 9, 286-309. House, R. and Wigdor, L. Herzberg's Dual-factor theory of job satis- faction and motivation: a review of the evidence and a criticism. PertOpnel Paxthologp, 1967, 29, 369-389. Hulin, C. and Blood, M. Job enlargement, individual differences and worker responses. Ptgtgologlcal pulletip, 1968, 29(1), 41-55. Hulin, C. and Smith, P. An empirical investigation of two implications of the two-factor theory of job satisfaction. leptppl of Applied Pszchologp, 1967, 22, 396-402. Hunt, J. A test of the leadership contingency model in three organiza- tions. Urbana, Illinois: Group Effectiveness Research Laboratory, University of Illinois, 1967. (Mimeograph) Hutchins, E., and Fiedler, F. Task-oriented and quasi-therapeutic role functions of the leader in small military groups. Sotiomettg, 1960. 22. 393-406. Indik, B., Georgopoulons, B., and Seashore, S. Superior-subordinate relationships and performance. Persoppel Psychologp, 1961, l9, 357-374. Katz, D., Maccoby, N., Gurin, G., and Floor, L. Productivity super- vision and morale among railroad workers. Survey Research Center, University of Michigan, 1951. Katz, 0., Maccoby, N., and Morse, N. Productivity, supervision and morale in an office situation. Survey Research Center, University of Michigan, 1950. 97 Katzell, R., Barrett, R., and Parker, T. Job satisfaction, job per- formance, and situational characteristics. 0 o A ed W: 1961: fie 65'72° Kennedy, J.E., and O'Neill, H.E. JOb content and worker's opinions. l o A e Ps he 0 , 1958, 92, 372-375. Rilbridge, MQD. Do workers prefer larger jobs? Peppegnel Psychology, 1960, 91, 45-48. Norman, A. "Consideration," "initiating structure," and organizational criteria--a review. PetsoApel Paycgology, 1966, l9, 349-361. Kornhauser, A.W. Mental health of the ipdusttlal worter; A Detroit Stpdy. New York: Wiley, 1965. Kosmo, R. and Orlando, B. Single continuum job satisfaction vs. duality: an empirical test. Persoppel Psythology, 1969, 22, 327-334. lawler, E., and Hackman, R. Impact of employee participation in the development of pay incentive plans: A field experiment. qutppl of A ied olo , 1969, 99, 467-471. Lawler, E. and Hall, D. The relationships of job characteristics to job involvement, satisfaction and intrinsic motivation. Journal of Applied Payphology, 1970, (in press). lawrence, L. and Smith, P. Group decision and employee participation. Journal of Applied Psychology, 1955, 99, 334-337. Levine, J. and Butler. Lecture vs. Group decision in changing behavior. Jpptppl ot Applled Psychology, 1952, 99, 29-33. Lewin, R. Group decision and social change. In 6.8. Swansan, T. Newcomb, and E. Hartley (Eds.) Readipgs lp Socipl Ppychology, rev. ed., New York: Holt, 1947, 459-473. Lewin, R., Lippitt, R., and White, B. Patterns of aggressive behavior in experimentally created social climates. 9ourpa1 of Social Payppology, 1939, 29, 271-299. Likert, R. Leg EEEGIR! pf mappgemept. New York: McGraw-Hill, 1961. Likert, R. fie 9mg otgaplpatiop. New York: McGraw-Hill, 1967. LOdahl, T. and Kejner, M. The definition and measurement of job in- volvement. 923221 of Applled Psychology, 1965, 59, 24-33. .Uieein, A. Participative decision making: almodel, literature critique and prescriptions for research. 9tganizptioppl fighayiot and 39322 Pertormapte, 1968, 9, 68-106. 98 Lewin, A., and Craig, J. The influence of performance on managerial style: an experimental object-lesson in the ambiguity of corre- lational data. Orgapigatioppl gpayioy EH! 3922 Petfotmappe, 1968, .3, 440-458. Lewin, A., Hrapchak, W. 6 Havanagh, M. Experimental investigation of consideration and initiating structure. Administrative Science Mrterly, 1969, 29(2), 238-263. MacRinney, A.C., Wernimont, P.F., 6 Galitz, W.O. Has specialization reduced job satisfaction? Personnel Psychology, 1962, 99(1), 8-17. Malinovsky, M.R. and Barry, J.R. Determinants of work attitude. Joggel of Applied Psyphology, 1965, 49, 446-451. March, J. 6 Simon, H. Organizations. New York: John Wiley, 1958. Marks, A.R.N. An investigation of modifications of job design in an industrial situation and their effects on some measures of economic productivity. Unpublished doctoral dissertation, University of California, 1954. (Summarized in L.E. Davis 6 R.R. Center, Job Design Research. Journal of Industrial lipgineeripg, 1956, 1(6), 275-282 .) Marrow, A., Bowers, D., 6 Seashore, S. Management by participation. New York: Harper 6 Row, 1967. PbGregor, D. The human side of enterptise. New York: McGraw-Hill, 1960. Mauwese, W. 6 Fiedler, F. Leadership and group creativity under varying conditions of stress. Urbana, Illinois: Group Effectiveness Research Laboratory, University of Illinois, 1965. (Mimeograph) Missumi, J. Experimental studies on group dynamics in Japan. Psychologia, 1959, 2, 229-235. Morse, N. 6 Reimer, E. The experimental change of a major organizational variable. Journal of Abnormal and Social Psychology, 1956, 92, 120-129. Neully, S. 6 Blood, M. Leadership performance of nursing supervision at two organizational levels. Journal of Applied Psycholpgy, 1968, 92, 414-422. Ott, D.C. The generality of Herzberg's two-factor theory of motivation. Unpublished Ph.D. thesis, The 0110 State University, 1965. Parker, T. Relationships among measures of supervisory behavior, group behavior and situational characteristics. Pertoppel Psytpology, 1963, lg, 319-334. 99 Patchen, M. Some questionnaire measures of epployee motivation and morale. Monograph No. 41 Survey Research Center, The University of Michigan, 1965. Saleh, S. A study of attitude change in the pre-retirement period. Jogrppl of Applied Ppytgology, 1964, 99, 310-12. Sandvold, K. The effect of effort and task orientation on the motivator orientation and verbal responsivity of chronic schizo- phrenic patients and normals. Unpublished Ph.D. thesis, University of Illinois, 1962. Seashore, S. 6 Bowers, D. Chapging the structure apg fupttiOpipg a a - . Ann Arbor: Institute for social research, 1963. Schwarz, P. Attitudes of Middle Management Personnel. Pittsburgh: American Institute for research, 1959. Shaw, M. .A comparison of two types of leadership in various communica- tion nets. Journal 09 Abpptppl and Sotial Payphology, 1955, 99, 127-134. Singh, T.M. 6 Baumgartel, H. Background factors in airline mechanics' work motivations. gourppl of Applied Psyppology, 1966, 99, 337-339. Smith, C. 6 Jones, C. The role of the interaction-influence system in a planned organizational change. In A. Tannenbaum goptrol ip otganigations. New York: McGraw-Hill, 1968. Smith, C. 6 Tannenbaum, A. Organizational control structure: a comparative analysis. W 1963, _1_6_, 299-316. Spitzer, M. 6 leamsra, W. A managerial selection study. Pprpogpel Psyppology, 1964, ll, 19-40. Strauss, G. Some notes on power-equalization. In H. Leavitt (Ed.) e o is sci e o tio . Englewood Cliffs, N.J.: Prentice-Hall, 1963. ‘Tannenbaum, A. gonttol ip organizatiOps. New Ybrk:‘McGraw-Hill, 1968. frannenbaum, A. 6 Allport, F. Personality structure and group structure: an interpretative study of their relationship through an event- structure hypothesis. Japyppl of Appptmal app §ocial Psyppology, 1956, 99, 272-280. Tannenbaum, A. 6 Smith, C. Effects of member influence in an organiza- tion: phenomenology versus organization structure. qurppl of Abpptppl apg 90pia1 Payphology, 1964, 99, 401-410. 100 Turner, A. 6 Lawrence, P. a a d e er: A e ti- e to . Boston: Harvard University Graduate School of Business Administration, 1965. Turner, A.M. 6 Miclette, A.L. Sources of satisfaction in repetitive work. OptppatiOppl Ppytpology, 1962, 99, 215-231. Vroom, V. Some personality determinants of the effects of participation. 1 o A 1 a o l P 010 1959, 99, 322-327. vroom, V. S so lit dete i s o effec s o rti i atio . Englewood Cliffs, New Jersey: Prentice-Hall, 1960. Walker, C.R. 6 Guest, R.B. The;ppp on the attembly lipe. Cambridge: Harvard University Press, 1952. Walker, C.R. The problem of the repetitive job. Hatvard Business Review, 1950, 29(3), 54-58. Walker, C.R. 6 Marriott, R. A study of attitudes to factory work. 9tpppat10ppl Psychology, 1951, 29, 181-439. Warren, N.D. Job simplification versus job enlargement. Joprppl at ngustrial Epglpperipg, 1958, 9(5), 435-439. Wernimont, P.F. Intrinsic and extrinsic factors in job satisfaction. goprppl of Applied Payphology, 1966, 99, 41-50. Wernimont, P. 6 Dunnette, M. Intrinsic and extrinsic factors in job satisfaction. Paper presented at Mid-Western Psychological Association, St. Louis, 1964. wyatt, 8., Fraser, J.A. 6 Stock, F.G.L. The comparative effects of variety and uniformity in work. Report No. 52, 1929, Industrial Fatigue Research Board. (a) Wyatt, 8., Fraser, J.A. 6 Stock, F.G.L. Fatigue and boredom in repeti- tive work. Report No. 77, 1929, Industrial Fatigue Research Board. (b) Wyatt, 8., Langson, J.N. 6 Stock, F.G.L. Fatigue and boredom in repetitive work. Report No. 77, 1937, Industrial Fatigue Research Board. Whyte, W.F. mppy apd motivatiop; An pglyslp of ipteptiye in industry. New York: Harper, 1955. APPENDICES APPENDIX A Analysis of Variance Summaries for Interactions between Ego Need Gratification and Extra-Work Socialization Table 34. Analysis of Variance Summaries for Interactions between Education (A) and Hierarchical level (B). Qtpepdept yatiable §Optce 99_ 99 §_ Satisfaction A x B 4 1.689 1.678 Error 2445 1.007 Identification A x B 4 .463 .927 Error 2427 .499 Involvement A x B 4 .892 2.032 Error 2451 .439 Commitment A x B 4 .036 .105 Error 2451 .346 Table 35. Analysis of Variance Summaries for Interactions between Father's Occupation (A) and Hierarchical Level (B). Dependept yariable Soyrte 99 m 9 Satisfaction A x B 4 .482 .468 Error 2333 1.032 Identification A x B 4 .095 .193 Error 2318 .491 Involvement A x B 4 .255 .571 Error 2339 .446 Commitment A x B 4 .410 1.176 Error 2339 .349 101 102 Appendix A - continued Table 36. Analysis of Variance Summaries for Interactions between Father's Education (A) and Hierarchical Level (B). Dependept Vaplatle Soprce 95. Q l: Satisfaction A x B 4 1.234 1.224 Error 2204 1.008 Identification A x B 4 .124 .255 Error 2189 .487 Involvement A x B 4 .717 1.616 Error 2209 .444 Commitment A x B 4 .091 .269 Error 2209 .338 Table 37. Analysis of Variance Summaries for Interactions between Community Raised (A) and Hierarchical Level (B). 22239122213212 Same 511 la 1'. Satisfaction A x B 4 .164 .160 Error 2444 1.026 Identification A x B 4 .242 .487 Error 2426 .496 Involvement A x B 4 .392 .894 Error 2449 .439 Commitment A x B 4 .180 .525 Error 2449 .343 103 Appendix A - continued Table 38. Analysis of Variance Summaries for Interactions between Present Community (A) and Hierarchical Level (B) Degendent gar iable Source 9; £8; E Satisfaction A x B 4 2.039 1.995 Error 2447 1.022 Identification A x B 4 0.418 .853 Error 2428 .491 Involvement A x B 4 .469 1.079 Error 2451 .434 Commitment A x B 4 .276 .817 Error 2451 .338 Table 39. Analysis of Variance Summaries for Interactions between Sex (A) and Hierarchical Level (B). ggegdent Variable Soggce d_f_ _M_S_ E Satisfaction A x B 2 .922 .887 Error 2559 1.040 Identification A x B 2 1.201 2.440 Error 2539 .492 Involvement A x B 2 1.051 2.356 Error 2565 .446 Commitment A x B 2 .298 .856 Error 2565 .349 104 Appendix A - continued Table 40. Analysis of Variance Summaries for Interactions between Education (A) and Collar Color (B). £32929; ya; iable Source 1; kg 1 Satisfaction A x B 2 23.759 23.683*** Error 2604 1.003 Identification A x B 2 11.811 22.231*** Error 2584 .531 Involvement A x B 2 12.266 26.335*** Error 2611 .466 Commitment A x B 2 2.468 6.928*** Error 2611 .356 ***2g(.001 Table 41. Analysis of Variance Summaries for Interactionabetween Father‘s Education (A) and Collar Color (B) Dependegt Variable Sogge g Q I: Satisfaction A x B 2 6.523 6.424** Error 2334 1.015 Identification A x B 2 .807 1.521 Error 2318 .530 Involvement A x B 2 3.022 6.277** Error 2340 .481 Commitment A x B 2 .808 2.295 Error 2340 .352 **2 (.01 105 Appendix A - continued Table 42. Analysis of Variance Summaries for Interactions between Father's Occupation (A) and Collar Color (B). Depe gent Vagiab 1e Source g; pg 1 Satisfaction A x B 2 1.360 1.298 Error 2476 1.048 Identification A x B 2 1.331 2.517 Error 2459 .529 Involvement A x B 2 1.391 2.866 Error 2483 .485 Commitment A x B 2 .426 1.170 Error 2483 .364 Table 43. Analysis of Variance Summaries for Interactions between Community Raised (A) and Collar Color (B). Dependth Variable Source 1; _M__S_ Z Satisfaction A x B 2 1.665 1.599 Error 2610 1.041 Identification A x B 2 .737 1.377 Error 2589 .535 Involvement A x B 2 .908 1.910 Error 2616 .475 Commitment A x B 2 .581 1.611 Error 2616 .360 106 Appendix A - continued Table 44. Analysis of Variance Summaries for Interactions between Present Community (A) and Collar Color (B). Qgpegdent Variable Sourge EL; _M_$_ r Satisfaction A x B 2 6.868 6.669*** Error 2611 1.030 Identification A x B 2 .540 1.027 Error 2588 .526 Involvement A x B 2 1.356 2.911 Error 2616 .466 Commitment A x B 2 1.509 4.266** Error 2616 .354 ***p,(.001 *fip4gJ31 Table 45. Analysis of Variance Summaries for Interactions between Sex (A) and Collar Color (B). Dependent Variable Sorrre gr! gr ‘_ Satisfaction A x B 1 .011 .011 Error 2743 1.062 Identification A x B 1 6.413 12.132*** Error 2719 .529 Involvement A x B 1 6.108 12.734*** Error 2750 .480 Commitment A x B 1 .142 .386 Error 2750 .366 ***p_(.001 107 Appendix A - continued Table 46. Analysis of Variance Summaries for Interactions between Education (A) and Self Report (B). Qgpegdegt Var gable Sorrce gr pg 1: Satisfaction A x B 4 .402 .400 Error 2420 1.004 Identification A x B 4 .690 1.277 Error 2411 .540 Involvement A x B 4 .211 .461 Error 2423 .457 Commitment A x B 4 .427 1.318 Error 2423 .324 Table 47. Analysis of Variance Summaries for Interactions between Father's Education (A) and Self Report (B). Degendert Variable Sogrre <_1_f_ _l§_ E Satisfaction A x B 4 .659 .659 Error 2192 .999 Identification A x B 4 .406 .742 Error 2183 .547 Involvement A x B 4 .452 .962 Error 2194 .470 Commitment A x B 4 .302 .951 Error 2194 .317 108 Appendix A - continued Table 48. Analysis of Variance Summaries for Interactions between Father's Occupation (A) and Self Report (B). De ende t Variable m gr _l§_ Z Satisfaction A x B 4 .832 .820 Error 2308 1.015 Identification A x B 4 .404 .752 Error 2300 .537 Involvement A x B 4 .172 .370 Error 2311 .465 Commitment A x B 4 .414 1.273 Error 2311 .325 Table 49. Analysis of Variance Summaries for Interactions between Community Raised (A) and Self Report (B). Beardegt Variable §ourre d_f_ LB; 1 Satisfaction A x B 4 2.231 2.230 Error 2423 1.001 Identification A x B 4 1.920 3.605** Error 2414 .533 Involvement A x B 4 1.054 2.324 Error 2426 .453 Commitment A x B 4 .833 2.580* Error 2426 .323 **£< .01 112905 109 Appendix A - concluded Table 50. Analysis of Variance Summaries for Interactions between Present Community (A) and Self Report (B). Dependent Variable Sourre gr Q g Satisfaction A x B 4 .977 .980 Error 2422 .997 Identification A x B 4 1.937 3.695** Error 2412 .524 Involvement A x B 4 .294 .662 Error 2424 .444 Commitment A x B 4 .649 2.051 Error 2424 .317 *tg .01 Table 51. Analysis of Variance Summaries for Interactions between Sex (A) and Self Report (B). Dggendegt Variable Source gr LS; E Satisfaction A x B 2 2.813 2.789 Error 2488 1.009 Identification A x B 2 .405 .776 Error 2479 .522 Involvement A x B 2 1.294 2.872 Error 2491 .451 Commitment A x B 2 .672 2.073 Error 2491 .324 APPENDIX B Mean Job Attitudes for Extra-Work Socialization by Ego Need Gratification Table 52. Mean Job Attitude for Education by Hierarchical Level. W me W L0" “Edi“ 1118b TONI Low 3.56 3.99 4.12 3.60 Satisfaction Medium, 3.24 3.96 4.10 3.32 High 3.16 3.92 4.20 3.52 Total 3.36 3.95 4.17 3.46 Low 3.22 3.91 4.00 3.27 Identification Medium 3.12 3.88 4.11 3.22 High 3.10 3.91 4.21 3.48 Total 3.16 3.90 4.16 3.29 Low 3.13 3.62 3.71 3.17 Involvement Medium 2.96 3.54 3.63 3.02 High 2.92 3.66 3.81 3.24 Total 3.02 3.61 3.76 3.12 Low 3.91 4.23 4.29 3.93 Comitment Medium 3 . 82 4. 19 4 . 28 3 . 87 High 3.77 4.14 4.22 3.93 Total 3.85 4.18 4.25 3.90 110 111 Appendix B - continued Table 53. Mean Job Attitude for Father's Education by Hierarchical Level. Hierarchical Level Father's Attitude Education Low Medium High Total Low 3.43 3.98 4.12 3.51 Satisfaction Medium 3.24 3.83 4.32 3.37 High 3.17 3.89 4.17 3.36 Total 3.34 3.92 4.17 3.45 Low 3.18 3.94 4.16 3.30 Identification Medium 3.06 3.91 4.13 3.22 High 3.18 3.88 4.17 3.37 Total 3.16 3.92 4.16 3.30 Low 3.06 3.63 3.68 3.14 Involvement Medium 2.92 3.63 3.80 3.05 High 2.95 3.61 3.84 3.12 Total 3.01 3.62 3.76 3.12 Low 3.88 4.22 4.30 3.93 Commitment Medium 3.77 4.19 4.24 3.85 High 3.79 4.10 4.17 3.87 Total 3.84 4.18 4.24 3.90 112 Appendix B - continued Table 54. Mean Job Attitudes for Father's Occupation by Hierarchical Level. grerarghical Level Father's Attitude chgprtion Low Medium High Total Low 3.36 3.98 4.14 3.44 Satisfaction Medium 3.29 3.91 4.29 3.51 High 3.43 3.94 4.12 3.61 Total 3.36 3.96 4.17 3.45 Low 3.14 3.88 4.06 3.24 Identification Medium 3.22 3.92 4.20 3.45 High 3.23 4.02 4.25 3.51 Total 3.16 3.91 4.15 3.30 Low 3.02 3.60 3.69 3.09 Involvement Medium 2.98 3.59 3.82 3.17 High 3.08 3.76 3.82 3.29 Total 3.02 3.62 3.76 3.13 Low 3.85 4.21 4.29 3.90 Commitment Medium 3.79 4.12 4.31 3.91 High 3.88 4.20 4.11 3.96 Total 3.85 4.19 4.25 3.91 113 Appendix B - continued Table 55. Mean Job,Attitudea for Community Raised by Hierarchical Level. e ar 1 1e e1 Community Artitude Raireg Low Medium High Total Low 3.49 3.96 4.14 3.68 Satisfaction Medium 3.55 4.08 4.34 3.70 High 3.34 3.90 4.13 3.42 Total 3.37 3.94 4.17 3.48 Low 3.34 3.99 4.19 3.59 Identification Medium 3.30 3.93 4.31 3.48 High 3.12 3.88 4.08 3.22 Total 3.16 3.90 4.15 3.29 Low 3.20 3.66 3.71 3.36 Involvement Medium 3.15 3.62 3.85 3 28 High 3.00 3.59 3.74 3 08 Total 3.03 3.60 3.75 3 13 Low 3.93 4.15 4.19 4.01 Comitment Medium 3.90 4.20 4.32 3 98 High 3.84 4.18 4.23 3 88 Total 3.85 4.18 4.24 3 90 114 Appendix B - continued Table 56. Mean Job Attitudes for Present Community by Hierarchical Level. Hierrrchical Level Present Attitude Cgrrgnitz Low Medium High Total Low 3.26 4.08 4.29 3.63 Satisfaction Medium 3.65 3.92 4.25 3.76 High 3.32 3.93 4.05 3.39 Total 3.37 3.95 4.16 3.47 Low 3.39 4.10 4.24 3.70 Identification Medium 3.29 3.86 4.14 3.48 High 3.12 3.87 4.12 3.22 Total 3.16 3.90 4.15 3.29 Low 3.10 3.76 3.88 3.38 Involvement Medium 3.23 3.66 3.78 3.36 High 2.99 3.54 3.69 3.05 Total 3.03 3.61 3.75 3.13 Low 3.88 4.23 4.31 4.03 Commitment Medium 3.93 4.13 4 22 3.99 High 3.84 4.19 4 24 3.88 Total 3.86 4.18 4 24' 3 91 115 Appendix B - continued Table 57. Mean Job Attitudes for Sex by Hierarchical Level. ie a i a1 Level Artirgde Sgr. Low Medium High Total Female 3.39 3.96 3.71 3.41 Satisfaction Mlle 3.34 3.95 4.19 3.53 Total 3.37 3.95 4.17 3.47 Female 3.09 3.62 3.74 3.11 Identification Mlle 3.24 3.97 4.18 3.46 Total 3.15 3.90 4.15 3.28 Female 3.00 3.40 3.49 3.02 Involvement Male 3.05 3.65 3.77 3.22 Total 3.02 3.60 3.75 3.12 Female 3.88 4.13 4.27 3.89 Co—itment Male 3.81 4.19 4.24 3.91 Total 3.85 4.18 4.24 3.90 Appendix B - continued 116 Table 58. Mean Job Attitudes for Education by Collar Color. Qollar golor Attitude Satisfaction Identification Involvement Commitment Educatiog Low Medium High Total Low Medium High Total Medium High Total Medium High Total Blue 3.54 3.16 2.88 3.29 White 3.75 3.70 3.93 3.80 Total 3.58 3.31 3.52 3.45 117 Appendix B - continued Table 59. Mean Job Attitudes for Father's Education by Collar Color. Collar Color Father's Attitude Education Blue White Total Low 3.38 3.79 3.50 Satisfaction Medium 3.18 3.80 3.37 High 3.04 3.82 3.35 Total 3.28 3.80 3.44 Low 3.15 3.62 3.29 Identification Medium 3.02 3.64 3.21 High 3.13 3.68 3.35 Total 3.12 3.64 3.29 Low 3.04 3.32 3.12 Involvement Medium 2.91 3.37 3.05 High 2.90 3.42 3.11 Total 2.99 3.36 3.11 Low 3.87 4.05 3.92 Commitment Medium 3.75 4.05 3.84 High 3.74 4.04 3.86 Total 3.82 4.05 3.89 118 Appendix B - continued Table 60. Mean Job Attitudes for Father's Occupation by Collar Color. Co a 0 or Father's Artitpde Occupation Blue White Total Low 3.30 2.76 3.42 Satisfaction Medium 3.24 3.88 3.51 High 3.29 3.90 3.59 Total 3.29 3.80 3.45 Low 3.11 3.55 3.23 Identification Medium 3.20 3.75 3.43 High 3.17 3.80 3.48 Total 3.13 3.63 3.28 Low 3.00 3.30 3.08 Involvement Medium 2.98 3.43 3.16 High 3.02 3.50 3.26 Total 3.00 3.36 3.11 Low 3.82 4.04 3.89 Commitment Medium 3.76 4.09 3.90 High 3.83 4.06 3.95 Total 3.82 4.05 3.90 119 Appendix B - continued Table 61. Mean Job Attitudes for Community Raised by Collar Color. Collar gglor Community Attitude Raised Blue White Total Low 3.34 3.95 3.65 Satisfaction Medium 3.44 4.01 3.68 High 3.29 3.70 3.41 Total 3.31 3.79 3.46 Low 3.29 3.84 3.57 Identification Medium 3.25 3.78 3.47 High 3.09 3.51 3.21 Total 3.12 3.60 3.27 Low 3.13 3.53 3.33 Involvement Medium 3.08 3.50 3.26 High 2.98 3.26 3.06 Total 3.00 3.34 3.11 Low 3.86 4.14 4.00 Commitment Medium 3.86 4.13 3.97 High 3.82 3.99 3.87 Total 3.83 4.04 3.89 120 Appendix B - continued Table 62. Mean Job Attitudes for Present Community by Collar Color. Collar Color Present Altitude W Blue White Total Low 3.12 4.07 3.64 Satisfaction Medium 3.65 3.91 3.75 High 3.28 3.71 3.38 Total 3.30 3.79 3.46 Low 3.38 3.96 3.70 Identification Medium 3.28 3.74 3.46 High 3.08 3.50 3.20 Total 3.13 3.60 3 28 Low 3.08 3.64 3.38 Involvement Medium 3 . 23 3 .49 3 33 High 2 96 3.25 3.04 Total 3 00 3.34 3.11 Low 3.80 4.24 4.04 Commitment Medium 3.92 4.04 3.97 High 3.82 4.01 3.87 Total 3.83 4.04 3.90 121 Appendix B - continued Table 63. Mean Job Attitudes for Sex by Collar Color. Collar Colpr Attitude §gr, Blue White Total Female 3.32 3.78 3.42 Satisfaction Male 3.30 3.77 3.49 Total 3.31 3.78 3.46 Female 3.04 3.32 3.10 Identification Male 3.22 3.72 3.42 Total 3.12 3.59 3.27 Female 2.98 3.16 3.02 Involvement Male 3.02 3.41 3.18 Total 2.99 3.32 3.10 Female 3.85 4.04 3.89 Commitment Male 3.79 4.02 3.88 Tota1 4.03 3.89 122 Appendix B - continued Table 64. Mean Job Attitudes for Education by Self Report. Self Report Attitude Education Low Medium High Low 3.11 3.60 3.95 Satisfaction Medium 2.99 3.33 3.73 High 3.14 3.57 3.95 Total 3.05 3.48 3.86 Low 2.88 3.31 3.52 Identification Medium 2.97 3.23 3.54 High 3.15 3.55 3.78 Total 2.99 3.33 3.57 Low 2.89 3.17 3.41 Involvement Medium 2.83 3.04 3.30 High 2.99 3.29 3.47 Total 2.88 3.14 3.38 Low 3.75 3.94 4.12 Commitment Medium 3.76 3.87 4.06 High 3.69 3.96 4.17 Total 3.74 3.92 4.10 123 Appendix B - continued Table 65. Mean Job Attitudes for Father's Education by Self Report. Self Report Father's Attitude Education Medium High Tota1 Low 3.12 3.53 3.88 3.52 Satisfaction Medium 3.05 3.34 3.74 3.56 High 2.91 3.44 3.86 3.39 Total 3.05 3.47 3.85 3.46 Low 2.99 3.32 3.60 3.30 Identification Medium 2.97 3.24 3.38 3.21 High 3.06 3.44 3.64 3.38 Total 3.01 3.33 3.58 3.31 Low 2.94 3.14 3.54 3.14 Involvement Medium 2.78 3.09 3.32 3.07 High 2.84 3.18 3.43 3.14 Total 2.89 3.14 3.37 3.13 Low 3.78 3.93 4.13 3.94 Commitment Medium 3.72 3.84 4.01 3.85 High 3.66 3.93 4.09 3.89 Total 3.74 3.91 4.10 3.91 124 Appendix B - continued Table 66. Mean Job Attitudes for Father's Occupation by Self Report. Self Repgrt Father's Attitude Qgppprrrpp’ Low Medium High Tota1 Low 3.00 3.46 3.83 3.43 Satisfaction Medium 2.97 3.53 3.92 3.51 High 3.38 3.58 4.02 3.62 Total 3.05 3.48 3.87 3.46 Low 2.94 3.27 3.53 3.25 Identification Medium 3.02 3.52 3.67 3.46 High 3.29 3.53 3 77 3.52 Total 3.00 3.33 3.57 3.31 Low 2.85 3.11 3 35 3.10 Involvement Medium 2.86 3.22 3.40 3.19 High 3.08 3.28 3.56 3.28 Total 2.88 3.14 3.38 3.13 Low 3.73 3.91 4.10 3.91 Commitment Medium 3.66 3.98 4.05 3.94 High 3.85 3.94 4.18 3.96 Total 3.74 3.92 4.10 3.92 125 Appendix B - continued Table 67. Mean JOb Attitudes for Community Raised by Self Report. Attitude Satisfaction Identification Involvement Conani tment Community Raised Low Medium High Total Low Medium High Tota1 Medium High Tota1 Low Medium High Total 2.94 3.15 2.85 2.88 §e1f Report Medium 3.78 3.68 3.42 3.49 High 3.92 3.88 3.86 3.87 3.74 3.56 .51 .42 .35 .38 UUWU .10 .08 .10 .10 ##5## 126 Appendix B - continued Table 68. Mean Job Attitudes for Present Community by Self Report. We Present Attitude Qgggpprrx_ Low Medium High Total Low 3.27 3.65 4.13 3.69 Satisfaction Medium 3.36 3.78 3.96 3.77 High 2.99 3.40 3.81 3.38 Total 3.05 3.49 3.87 3.47 Low 3.60 3.69 3.84 3.71 Identification ZMedium 3.23 3.54 3.53 3 49 High 2.91 3.25 3.54 3.23 Total 2.99 3 33 3.56 3.30 Low 3.19 3.40 3.58 3.41 Involvement Medium 3.12 3.38 3.50 3.37 High 2.83 3.06 3.32 3.06 Total 2.89 3.14 3.38 3.14 Low 3.84 4.04 4.24 4.05 Commitment Medium. 3.86 4.04 4.05 4.01 High 3.72 3.89 4.10 3.89 Total 3.75 3.93 4.10 3 92 Table 69. Mean Job Attitudes for Sex by Self Report. Attitude Satisfaction Identification Involvement Commitment Appendix B - concluded Sex Female Male Total Female Male Total Female Mele Total Female Male Total 127 Low 3.11 2.99 3.06 Self Report Medium 3.44 3.53 3.48 High 3.81 3.93 3.88 Total 3.43 3.52 3.47 APPENDIX C Summaries of Simple Effects Analyses of Variance for Table 70. Source A/Bl A/Bp A/B3 Error Table 71. 529552 A/Bl A/Bz A/B3 Error Table 72. §22£££ A/Bl A/82 A/B3 Error Ego Heed Gratification on Job Attitudes Within Levels of Extra-Work Socialization. Simple Effects Analyses of Variance for Collar Color (A) on Satisfaction within Levels (1-1ow, Z-medium, 3-high) of Education (B). §§. 22. HE. E. .2 6.923 1 6.923 6.903 .01 70.358 1 70.358 70.147 .01 130.622 1 130.622 130.232 .01 2612.408 2604 1.003 Simple Effects Analyses of variance for Collar Color (A) on Identification within Levels of Education (B). £2. 22. BE. 2. .2 10.539 1 10.539 19.847 .01 42.858 1 42.858 80.712 .01 83.648 1 83.648 157.529 .01 1372.874 2584 .531 Simple Effects Analyses of Variance for Collar Color (A) on Involvement within Levels of Education (B). 22. .25. HE E. .2 3.935 1 3.935 8.444 .01 20.398 1 20.398 43.772 .01 65.203 1 65.203 139.920 .01 1216.161 2611 .466 128 Table 73. §2§££2 A/81 A/Bz A/B3 Error Table 74. §ourge A/B1 A/Bz A/B3 Error Table 75. §22222 A/Bl AIB2 A/Bg Error Table 76. 522222 A/Bl A/Bz Error 129 Appendix C - continued Simple Effects Analyses of variance for Collar on Commitment within Levels of Education (B). Color (A) 22, 22. £2. E. .2 3.178 1 3.178 8.928 .01 11.010 1 11.010 30.927 .01 19.627 1 19.627 55.132 .01 930.323 2611 .356 Simple Effects Analyses of Variance for Collar Color (A) on Satisfaction within Levels of Present Community (B). §§. 22. 121 EL .2 35.209 1 35.209 34.183 .01 7.685 1 7.685 7.461 .01 859.938 1 859.938 83.435 .01 2688.887 2611 1.030 Simple Effects Analyses of Variance for Collar Color (A) on Commitment within Levels of Present Community (B). .5: 2:. BE. 2. .2 7.707 1 7.707 21.771 .01 1.674 1 1.674 4.729 .05 15.113 1 15.113 42.693 .01 925.367 2616 .354 Simple Effects Analyses of variance for Collar Color (A) on Identification within Sexes (Bl-female, B2-ma1e). §§. 92. BE. 2. 2. 18.233 1 18.233 34.467 .01 84.674 1 84.674 160.063 .01 1437.242 2719 .529 Table 77. Source A/Bl A/BZ Error Table 78. §oprce A/Bl A/Bz A/B3 Error Table 79. Soprce A/Bl A/Bz A/B3 Error Table 80. §22l CE A/Bl A/Bz A/B3 Error 130 Appendix C - continued Simple Effects Analyses of variance for Collar Color (A) on Involvement within Sexes (Bl-female, Bz-male). .2. 22. 22 Z. .2 7.739 1 7.739 16.123 .01 52.215 1 52.215 108.781 .01 1319.130 2750 .480 Simple Effects Analyses of Variance for Collar Color (A) on Involvement within Levels of Father's Education (B). 22. 22. 22. I. .2 22.941 1 22.941 47.695 .01 17.367 1 17.367 36.106 .01 36.232 1 36.232 75.326 .01 1126.395 2340 .481 Simple Effects Analyses of Variance for Collar Color (A) on Satisfaction within Levels of Father's Education (B). 22 22. 22. z. .2 48.375 1 48.375 46.028 .01 31.378 1 31.378 29.855 .01 80.402 1 80.402 76.500 .01 2369.846 2334 1.01 Simple Effects Analyses of Variance for Self Report (A) on Identification within Levels of Community Raised (B). 22. 22. 121 E. 2. 11.091 2 5.545 10.404 .01 3.119 2 1.559 2.926 >.05 80.869 2 40.434 75.862 .01 1285.516 2414 .533 Table 81. Sourre A/Bl A/Bz A/B3 Error Table 82. Source A/Bl A/Bz A/B3 Error 131 Appendix C - concluded Simple Effects Analyses of Variance for Self Report (A) on Comitment within Levels of Cal-unity Raised (B). 22. 22 :52 I. .2 5.193 2 2.596 8.038 .01 . 874 2 . 437 1 . 354 > .05 28.135 2 14.067 43.553 .01 782.625 2426 .323 Simple Effects Analyses of Variance for Self Report (A) on Identification within Levels of Present Community (B). .42 22. BE. 2. 11 1.067 2 .534 1.018 7.05 5.457 2 2.729 5.207 .05 78.147 2 39.074 74.568 .01 1264.496 2412 .524 APPENDIX D Questionnaire Items Used to Measure Extra-Work Socialization, Ego Heed Gratification, the Importance of Ego Heed Gratification, and Attitudes toward the Job. 1. Extra-work socialization: Employee responses to the following items were used to index this variable: How many years of school have you completed? 6 or less 7 8 9 10 11 12 some college college graduate work The place in which you spent the most time during your early life was a: farm town of less than 2,000 town of 2,000 or‘more, but less than 10,000 city of 10,000 to 100,000 city larger than 100,000 In what type of community are you now living? in the country town of less than 2,000 town of 2,000 or more but less than 10,000 city of 10,000 to 100,000 city larger than 100,000 How many years of school did your father complete? Same alter- natives as question a. 132 133 Appendix D - continued e. What was your father's chief occupation? unskilled worker semi-skilled or skilled worker sales or office work subprofessional (bookkeeper, pharmacist, draftsman, etc.) scientist professional (lawyer, physician, teacher, etc.) businessman executive of large business or industry f. Your sex: 2. The e te t of e 0 me d ati i atio rovided b the ob The following variables were used to index the extent of ego need gratification provided by the job: a. Hierarrpicpl leve : This variable was based on employee's report of supervisor's name and organization charts. b. Collar color: This variable was based on employee's report of job title and company records. c. Self report: This variable was based on the average of responses to the following S-point Likert scale items: 1. To what extent do you feel the work you are doing is important? 2. To what extent are you recognized and appreciated for doing good work? 3. How much responsibility do you have on your job? 4. How good are your chances for promotion in this company? 5. To what extent does your work provide you with a feeling of accomplishuent? ' 6. To what extent are you able to decide how to do your job? 7. To what extent are you able to learn new skills and gain experience on your job? 8. How challenging is your job? 3. ing 4. 134 Appendix D - continued The lgportappe of ego need grarification pp the job. This variable was indexed by averaging the responses to the follow- 5-point Likert scale items: How important to you is: 8. having a challenging job? b. having good chances for promotion? c. being recognized and appreciated for doing good work? d. having responsibility on your job? e. being able to decide how to do your job? f. a feeling that the work you are doing is important? 3. being able to learn new skills and gain experience on your job? h. a sense of achievement in the work you are doing? Job atrltpdes: a. Job sarigfpgrion: This variable was measured by averaging responses to the following 5-point Likert scale items: 1. How much do you look forward to coming to work each day? 2. How much do you actually enjoy performing the day to day activities that make up your job? b. W: This variable was measured by averaging responses to the following 5-point Likert scale items: 1. How often do you discuss your work with people outside the company? 2. my job means a lot more to me than just money. 3. How much are you personally involved in your job? 4. I'm really interested in my work. 5. To me, my work is only a small part of who I am. 6. I lie awake at night thinking about my job. C. d. 9. 10. 135 Appendix D - concluded I would probably keep working even if I didn't need the money. ‘Most things in life are more important than work. The most important things that happen to me involve my work. The major satisfaction in my life comes from my job. Commitmep : This variable was measured by averaging responses to the following S-point Likert scale items: 1. 2. I show up for work a little early to get things ready. I'll stay overtime to finish a job even if I'm not paid for it. I'm really a perfectionist about my work. I feel bad when I make mistakes in my work. How much do you really want to do a good job? How often do you really want to work hard at your job? I avoid taking on extra duties and responsibilities in my work. Ideptificatlop: This variable was measured by averaging responses to the following 5-point Likert scale items: 1. 2. How much conflict do you see between the goals of the company and your own personal goals? How much do you feel your personal satisfactions are related to how well you do your job? How much is the welfare of the company you work in related to your own personal welfare? How much does the company's achievement of its goals help your achievement of your own personal goals? '4‘: APPENDIX E Correlations Among Measures of Ego Heed Gratification, Extra-Work Socialization, and Job Attitudes The correlations among the different measures of ego need grati- fication, extra-work socialization, and job attitudes are presented in Tables 83-85. The correlations between the different measures of job attitudes are moderately high and positive, ranging from .468 to .666. The correlations between four of the six measures of extra-work socialization--education, father's education, father's occupation, and sex--are low and positive, ranging from .053 to .345. The community raised and present community measures, however, are positively corre- lated with each other and negatively correlated with the other four measures of extra-work socialization. Similarly, the correlation between the hierarchical level and collar color measures of ego need gratification is moderately high and positive, but the correlations between these measures and the self report measure of ego need gratifi- cation are positive but quite low. In general, these correlations are consistent with the results obtained for these measures in relation to the hypotheses under investi- gation. More specifically, the results obtained for the measures of extra-work socialization which correlated positively with each other generally supported the hypotheses under investigation, while the results obtained for the community raised and present community measures in each 136 137 Appendix E - continued case were opposite to those predicted. Similarly, while the results for all three of the measures of ego need gratification generally supported the hypotheses under investigation, the self report measure provided less consistent support, especially for hypothesis 2, than the hierarchical level and collar color measures. 138 Appendix E - continued Table 83. Correlations Among Measures of Job Attitudes. Satisfaction Identification Involvement Commitment Satisfaction Identification .468 Involvement .666 .567 Commitment .626 .604 .614 Table 84. Correlations Among Measures of Ego Heed Gratification. Hierarchical Level Self Report Collar Color Hierarchical Level Self Report .097 Collar Color .650 .025 139 Appendix E - concluded NMN.1 ¢¢~.1 onc. cud. nno. awe. ~mo.: -H.o w-.u nn~.u enu.s mo~.s nan. mmN. mum. gum emu-oum commas domususvm couusmsouo coda-use” humans-co Samuelson a.uucuam s.uueusm .moauuwufismoom xuo31suuuu we consume: wooed msomusaowuoo mammal-o0 unusuum vacuum mamas-Ion mouuaosvm a.uoeusm mouusmsooo s.uo£usm coausosmm .nw wanna nICHIan STATE UNIV. LIBRARIES VIHlmlll"WWIWUIVINIWNWWWW1 31293011019514