ABSTRACT EFFECTS OF INDIVIDUAL DIFFERENCES ON THE RELATIONSHIP BETWEEN PARTICIPATION AND JOB ATTITUDES By John Kenneth White This study was designed to investigate the role of personal vaTues and background factors on the relationship between opportunity for participation and job attitudes. Previous studies have investigated the moderating effect of background variables and inferred needs and values from them. The present study directly investigated the moderating effect of l0 values and at the relationship between these values and background factors. Specifically, it was hypothesized that: l. Opportunity for participation correlates positively with the job attitudes involvement, commitment, and identification. 2. The values accomplishment, equality, freedom, ambitious, capable, imaginative, independent, responsible, self- controlled and participation moderate the above relationship such that the relationship is greater for those who attach high importance to each of the values. 3. The values correlate with 15 background factors related to residence, religion, occupation, and education. All the data came from a questionnaire filled out by 2755 employees from 6 Midwest manufacturing concerns, and was analyzed with correlational techniques. John Kenneth White Strong support was found for the first hypothesis but generally no support was found for the other two. The nature of the opportunity for participation measure and the ipsative nature of the values were suggested as two possible reasons for the lack of results. EFFECTS OF INDIVIDUAL DIFFERENCES ON THE RELATIONSHIP BETWEEN PARTICIPATION AND JOB ATTITUDES By John Kenneth White A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1972 ACKNOWLEDGEMENTS Dr. Frank Schmidt who is responsible for any methodological competence that this thesis might demonstrate. Dr. Carl Frost whose help and guidance on this thesis is just a small part of his contribution to my entire education. Dr. Robert Ruh who, in spite of my efforts to the contrary, helped me onto a goal directed path. My wife Sharon for being my wife Sharon. Thank You. 11' TABLE OF CONTENTS PAGE LIST OF TABLES ....................... . . . iv CHAPTER I. INTRODUCTION ...................... l Individual Difference Contingency Theory ........ l Sunmary ....................... 8 Interpretation .................... 10 Plan of Study .................... ll Hypothesis ...................... ll Values and Background Factors ............. 12 II. METHOD .......................... l6 Sample ........................ 16 Data ......................... l6 III. RESULTS AND DISCUSSION .................. 23 Preliminary Analysis ................. 23 Individual Difference Contingency Theory ........ 25 Results ....... . ............... 25 Discussion ...................... 30 Conclusion ...................... 34 Values and Background Factors ............. 34 Results ...................... . 34 Discussion ...................... 37 Conclusion ...................... 39 FOOTNOTES ............................ 41 REFERENCES ............................ 43 iii TABLE LIST OF TABLES Item-Scale Correlations and Scale Reliabilities ...... Inter-Scale Correlations . . . . . . . . . . . . . . . . . Correlations between Attitude Measures and Opportunity for Participation for Different Hierarchial Levels Correlations between Opportunity for Participation and Job attitudes for the Value Subgroups of Workers Correlations between Opportunity for Participation and Job Attitudes for the Value Subgroups of Managers . . . . Correlations between Opportunity for Participation and Values . . . . . . . . . . . . . . . . . . . . . . . Comparison of Correlations of Attitudes and "other" Items with Correlations of Opportunity for Participation and These “other” items . . . . . . . . . . ....... Correlations between Background Items and Values ..... Correlation Matrix of Background Items . . . . . . . . . . iv Page 24 24 25 26 27 29 31 38 LIST OF TABLES Item-Scale Correlations and Scale Reliabilities ...... Inter-Scale Correlations ................. Correlations between Attitude Measures and Opportunity for Participation for Different Hierarchial Levels Correlations between Opportunity for Participation and Job attitudes for the Value Subgroups of Workers Correlations between Opportunity for Participation and Job Attitudes for the Value Subgroups of Managers . . . . Correlations between Opportunity for Participation and Values . . ............ . . . . . . . . . Comparison of Correlations of Attitudes and "other" Items with Correlations of Opportunity for Participation and These "other” items ................. Correlations between Background Items and Values ..... Correlation Matrix of Background Items .......... iv Page 24 24 25 26 27 29 3I CHAPTER I INTRODUCTION Individual Difference Contigency_Theory In response to rather broad acceptance of much of human relations theory, several authors have postulated contigency models - arguing that the human relation predictions of improved attitudes and performance resulting from enriched jobs1 are not universally valid. The variables which have been suggested as prerequisites for these relationships fall into two broad overlapping classes. The first group consists of situational variables, several of which are discussed by Lawrence and Lorsch (T967). The second group, which is the major focus of the present study, consists of individual difference variables. There are many studies that have demonstrated effects of individual differences on job attitudes and performance. However, this review of the literature is limited solely to those studies that focus directly on the interac- tion between individual differences and characteristics of the job situation. In a study of 470 workers from ll industries working on 47 different jobs, Turner and Lawrence (l965) attempted to demonstrate a relationship between job complexity and satisfaction and absenteeism. Using job level and job ratings as measures of job complexity, the hypothesized relation- ship was found for absenteeism but not for satisfaction. After several reanalyses of the data the authors found that both of the hypothesized 1 relationships held for those working in small towns, but for those working in large cities there was no relationship between job complexity and absenteeism, and a negative relationship between job complexity and satisfaction. The authors suggest that the hetrogeneous social cultures in the larger city result in workers from these areas being anomic. In a later study, Blood and Hulin (1967) attempted to demonstrate a negative relationship between the degree to which workers were employed in more urbanized and slum areas, and the degree to which job satisfaction correlated with job complexity. They found no support at all for their original analysis. Upon reanalyzing the data using only the blue collar portion of their sample the authors claim reasonable support for their hypotheses. However, only three of the fourteen response variables dealt with satisfaction with the work itself. These were the relationships between satisfaction with the work itself and skill level, job level, and general job satisfaction, as affected by the demographic variables. However, since job level is very likely negatively correlated with alienation from middle class norms (the authors' interpretation of the urbanization/slum variables) and since it is rather a nebulous concept when the sample is restricted to blue collar workers, clearly the most relevant relationship is that between satisfaction with the work and estimates of skill requirements of the job. However, the correlation between this relationship and six demographic variables ranged from -.02 to -.lO none significant. The authors interpret their urbanization measure as alienation from middle class norms and use the data to refute three other possible interpreta- tions: economic status, differences in needs that result from urban 3 and rural environments as suggested by Katzell, Barett, and Parker (l96l), and anomie-lacking all norms rather than just middle class nonns (Turner & Lawrence, l965). However, in lieu of the fact that the data only marginally supported the effect of any variable, it seems a little premature to positively identify that variable on the basis of that same data.2 In a subsequent article (Hulin & Blood, 1968) the same authors reviewed the literature relevant to individual difference contigency theory and concluded that the available empirical evidence does not justify the broad acceptance that human relations theory has enjoyed. Drawing particularly on Turner and Lawrence (1965) and their own previous research, the authors interpret their measures of the demographic variables as a continuum of alienation/integration and suggest that while human relations predictions of a positive relationship between satisfaction and job complexity are valid for highly integrated workers, the relationship decreases as alienation increases and is in fact highly negative for extremely alienated workers. Several authors in referring to this study emphasize city size as the significant variable. However, the measure that provided the best support was the extent to which the area in which the plant was located was a slum. Therefore it seems equally plausible that the real dimension is the extent to which the area is dominated by lower class non whites.3 In discussing their alienation from middle-class norms continuum the authors write (Blood & Hulin, 1967): At the integrated end of the construct are found workers who have personal involvement with their jobs and aspirations within their occupation. Their goals are the type of upward mobility, social climbing goals generally associated with the American middle class. At the Opposite pole of the construct, workers can be described as involved in their jobs only instrumentally; that is, the job is only a provider of means for pursuing extra- occupational goals. The concern of these workers is not for increased responsibility, higher status, or more autonomy. They want money and they want it in return for a minimal amount of personal involvementx’ This difference between integrated and alienated workers is similar to Dalton's (1947) discrimination between overproducers who are likely to hold middle-class ideals. The construct of alienation being proposed in this study stands in obvious relation to the Protestant ethic proposed and discussed by Weber (1958). It is likely that conditions fos— tering integration with middle-class norms will also foster adhercence to the Protestant ethic since the latter is an aspect of the former (p. 284-285). In their subsequent article Hulin and Blood, (1968) give the following definition of Protestant ethic: "Work hard and you will get ahead. You are responsible for your destiny. Acceptance into the Kingdom of Heaven is dependent on hard work on this mortal earth (p. 386)." They define middle class norms as "positive affect for occupational achievement, a belief in the intrinsic value of hard work, a striving for the attain- ment of responsible positions, and a belief in the work-related aspects of Calvinism and the Protestant ethic (p. 386)," and suggest that this corresponds (inversely) to what Kornhauser (1965) describes as "the very limited self-expectations, the degree of positivity fatalism, and resignation that characterize many of the workers (p. 24l)." The opposite of this is described as (Hulin & Blood, 1968): "every person is responsible for his own situation rather than being influenced by forces beyond his control (p. 383)." MacKinney, Wernimont, and Galitz (1962) reviewed the studies concerning the relationship between job enlargement and job satisfaction and concluded the results were unconvincing. They commented: 5 The most compelling argument against specialization as a major cause of job dissatisfaction lies in the fact of individual differences. This is a 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. This simply is not so! (For that matter, it's highly probably that many of the workers inter- viewed by sympathetic social scientists privately regard their questioners' activities as a pretty terrible way to earn a living too) (p. 17). In a similar vein Strauss (1963) argues that human relations theory depends on need for achievement and participation which may vary considerably in the p0pu1ation, and even if they don't vary too much there certainly is variance in workers' perceptions of the appropriate- ness of the work situation (as opposed to off the job activities) for satisfying these needs. In a large postal delivery company, Vroom (1960) conducted a study to test the hypothesis that there would be a positive relationship between satisfaction/performance and participation, and that this relationship would be greater for those with a high need for indepen- dence, and lower for those high on authoritarianism. Participation and satisfaction were measured by four and three Likert type questionnaire items respectively. Need for independence was measured by a 16 item scale, authoritarianism was measured with the F-scale, and performance was based on supervisory ratings. The results yielded strong support for the hypothesized relationships. The overall correlations between participation and satisfaction and performance were .36 and .20 respectively. The correlations between satisfaction and participation for the high, medium, and low need for independence groups were .55, .31, .13, and .03, .35, and .53 for the high, medium, and low authori- tarianism groups. The comparable measures for performance and partici- pation were .33, .19, .06 and .08, .28, and .28. However, in a repli- cation of the Vroom experiment with a different sample, Tosi (1970) failed to find the same moderating effect for authoritarianism and need for independence. He did confirm the substantial relationship between satisfaction and participation but found no differences when the sample was subgrouped on the same measures as in the original study. However, it should be pointed out that his performance measure was return on investment as opposed to Vroom's supervisory ratings. In an ominously titled Harverd Business Review article ("Behavioral Theory vs. Reality”) Lee (1971) argues that human relations theory will not gain wide scale success until the values and beliefs inherent within it become more generally prevalent in the workforce. These values and beliefs include increases in autonomy, knowledge, and participation, and decreases in authoritarianism and elitism. According to the author, these changes will ccme about, but extremely slowly because they are directly tied to the evolution of the general culture. He argues that where human relations programs have succeeded it is chiefly a result in changes in beliefs and attitudes of those subcultures and is "unrelated to the direct efforts of behavioral science theorists (p. 159)." Argyris (1964) is the only theorist who has responded to the criticisms that human relations hypothesis are contingent upon factors that vary among individuals. He argues that while there are individuals who may not desire intrinsic rewards from the work situation, they are a minority and that they have developed this fatalistic, but realistic, 7 acceptance of work as a result of experiencing jobs with no opportunities for participating. Several authors (Ruh, 1970; Morrison, 1970; Siegel, 1971) have identified Hulin and Blood and Argyris as being in clear opposition to each other: However, if we look at the original Blood and Hulin (1967) data rather than at their subsequent theorizing, the conflict is not so apparent. Because these authors swnned across very disparate items in creating their demographic variables it is difficult to conceptualize what they really were measuring. However, their independent variable that fared best, slum conditions, was mainly a measure of the degree to which the area could be characterized as a racial slum (see footnote 3). While it appears at first glance with the 23 plants that the authors have a "continuous" measure, considering how small the obtained correlations were it seems very likely that the variance could be accounted for by a dichotomy between those who work in racial slums and everybody else. However, minority slum.dwellers are often characterized as having a fatalistic acceptance of life as a result of extremely limited Opportunities. By interpolation, this coincides very closely with Argyris' argument that these people have repressed any desire for satisfying egoistic needs in the work situation as a result of previous experiences with jobs offering little such opportunity. However, it should be pointed out that while Argyris would agree that human relations predictions would not hold (at least as well) for these people (minority slum dwellers), he would certainly not accept the negative relationship suggested by Hulin and Blood. It is harder to explain Vroom's (1960) findings with Argyris' argument. However, none of the obtained or predicted relationships were significantly negative, so the discrepancy is not as great. l V 8 There are two recent Michigan State studies that directly inves- tigated the contingency hypothesis. Ruh (1970) hypothesized that extra- work socialization would moderate the relationship between ego need gratification provided by the job and job attitudes. Job attitudes satisfaction, involvement, commitment, and identification were measured by averaging the responses to 2, 10, 7, and 4 item Likert scales respectively. Ego need gratification provided by the job was measured by hierarchial level, collar color, and by averaging responses to an eight item Likert scale. Extra-work socialization was indexed by education, father's education, father's occupation, sex, size of community raised in, and size of community now living in. The results yielded very marginal support for the hypothesis. Only l0 of the 72 total analysis were significant in the predicted direction. Education tended to give the strongest support. No support was found for a second hypothesis that importance attached to ego need gratification provided by the job, measured on 8 item Likert scale, would moderate the same relationship. Using very similar data, Siegel (1971) investigated the.relation- ship between opportunity for participation and job involvement. The relationship was significantly higher for groups with more education, urban residence, and (somewhat ambiguous) more intrinsically oriented. Summary It appears that the issue at stake is not whether individual differences affect the relationship between characteristics of the job situation and attitudes - a readily acceptable idea - but what differ- ences and why. The studies reviewed here have empirically established 9 some effect for size of city where plant is located (Turner & Lawrence, 1965; Blood & Hulin, 1967), size of city of residence and education (Ruh, 1970; Siegel, 1971), intrinsic/extrinsic orientation (Siegel, 1971), sex (Ruh, 1970), authoritarianism and need for independence (Vroom, 1960), and slum conditions (Blood & Hulin, 1967). Altogether the following variables have been postulated: Needs/traits/values - self actualization, independence, achievement, Fatalism(reversed) - authoritarianism (reversed), intrinsic orientation, religiosity, accomplishment, autonomy, knowledge, elitism (reversed), social mobility, responsibility, status, equality participation, and ego need gratification provided by the job. The degree to which a person believes he can affect his personal outcomes-at the job level as suggested by Argyris (1964), and at the life in general level as in Hulin and Blood's (1968) interpreta- tion of the Kornhauser (1965) findings. (This is a broad category that involves internal versus external control and undoubtedly overlaps with the previous concepts.) Alienation from middle class norms - Hulin and Blood (1968) define middle- Protestant ethic - Background Factors - class norms as a positive affect for achievement, work, responsibility, etc. Alienation from these norms is measured, in part, by per capita motor vehicle deaths, percentage of workers in wholesale, and percentage of change in farm level of living. This has been suggested as working in both direc- tions. Hulin and Blood (1968) suggest that since it embodies many of the values mentioned pre- viously it should increase the relationship. However, Siegel (1971) argues that beliefs in the Protestant ethic will exert a strong direct influ- ence on job attitudes and hence lessen the effects of the job situations. Similarily it relates to fatalism in that the benefits to be derived from hard work do not directly result from the worker's efforts but rather are mediated by an outside Force beyond his control. sex, education, father's occupation and education, hierarchial level, slum conditions, rural/urban, and previous work experience. 10 The only other variable is Strauss' (1963) suggestion of differences in perception of the appropriateness of the work situation for satisfying egoistic needs. However, this is very similar to the (relative) importance of ego need gratification provided by the job as postulated by Ruh (1970). Interpretation It seems that all these variables can be classified into three categories. Then, for enrichment to result in improved job attitudes: l) The workers must have needs, traits, and values compatible with the enrichment process. In the case of participative management, for example, there must be a relatively high need for independence, a high value for participation itself, a low value for elitism, and low authoritarianism. 2) The workers must have some positive affect for the potential "benefits"to be derived from enriched jobs. This would include virtually to all the values and needs listed previously, plus the background factors that would cause or at least predict them. 3) The workers must believe that they have meaningful control over the "benefits" to be derived from the job situation. Although it might appear from an "objective" assessment of the situation that this is the case, previous work experience, early socialization, or a host of other reasons may cause the worker to perceive the situation differently. He must believe that the "benefits" can be derived as a direct result of his efforts. (Fatalism and the Protestant ethic). However, it is very likely that in practice the three go together very closely. It is difficult to imagine, for example, having a nega- tive value for participation and a positive value for equality (a result of participation) or vice versa. Similarly, cognitive dissonance would likely reduce any large discrepancy between a worker's aspirations and his beliefs in what he could actually achieve. Or, as Argyris (1964) would argue, these aspirations would be repressed. 11 Plan of Study The major focus of this study was the contingency effect of several Rokeach values on the relationship between job attitudes and opportunity for participation. It was the availability of the data that limited the present investigation to the use of Rokeach values. However, values are the most frequently suggested class of variables in the literature. Rokeach (1968) defines a value as an "enduring belief that a specific mode of conduct (instrumental value) or an end state of existance (terminal value) is . . . preferable to alternate modes of conduct or end states of existence (p. 160)." However, it seems, at least on a superficial level, that this is not too different from needs (drives) which is the other frequently suggested class of variables. It was thought that rather than looking directly at the contin- gency effects of background variables and inferring values or whatever from them, it would be more appropriate to identify the moderating values directly and then determine their background correlates. At the minimum this approach permits more confidence in the interpretation of the results.4 The fatalism concept introduced previously was not included in the moderating variables. However, differences in perceptions of the situation did likely exert an influence on the measure of Opportunity for participation in that it was a self report description of the opportunities provided by the job. Hypgtheses 1) There is a positive relationship between the opportunity for_partici- pation_provided by the job and the job attidues involvement, commitment, 12 nd identification.5 - A basic tenet of human relations theory. r 2) Values will moderate the above relationship such that the correlation between attitudes and_pppprtunity-for participation provided by the job is greater for those persons who attach high importance to each of the values than for those persons who attach low importance to each of the values. Following directly from the literature discussed previously are the values accomplishment, equality, freedom, ambitious, rgsppnsible, independent, and participation.6 Self-Controlled was investigated because it has connotations of intrisic orientation which in turn relates to fatalism. Also investigated were capable and imaginative. Values and Background Factors As was mentioned previously, background factors were investigated as they relate to values rather than focussing directly on their contin- gency effects. In the studies that were reviewed here there were several theoretical arguments as to why specific background factors should relate to certain values - the most involved being that of Hulin and Blood (1968). However, there is an existing framework of theoretical and empirical developments that seems particularly relevant to this 7. relationship - modernism/traditionalism. This concept incorporates background factors, values (especially accomplishment, achievement, and power equalization), and ability to achieve them (fatalism). Kahl (1968) writes: Traditional values are compulsory in their force, sacred in their tone, and stable in their timelessness. They call for fatalistic acceptance of the world as it is, respect for those in authority, and submergence of the individual in the collectivity. Modern values are rational and secular, permit choice and experiment, glorify efficiency and change, and stress individual responsibility p. 6 . 13 Smith and Inkeles (1966) write: [Modernism is] a complex but coherent set of psychic dispositions manifested in general qualities such as a sense of efficacy, readiness for new experience and interest in planning, linked, in turn, to certain dispositions to act in institutional relations-- as in being an active citizen, valuing science, maintaining one's autonomy in kinship matters, and accepting birth control. As indicated above, we assumed these personal qualities would be the end product of certain early and late socialization experiences such as education, urban experience, and work in modern organiza- tions. [p. 377] These authors (Smith & Inkeles, 1966) have done the most extensive empirical work regarding "attitudinal modernity." Drawing on previous investigations and their own theoretical framework they constructed a large pool of relevant items which were administered to 5500 men in six countries. Using both an internal criterion and an external criterion of education, occupation, and urban experience, they identified 119 purely attitudinal items which they included measure "an underlying dimension of psychological modernity (p. 362)."8 Included in these items are several that are very comparable to the values used in this investigation. These are attitudes toward education, status, responsi- bility, ambition, fatalism, intelligence, industry, and equality. The hypothesized relationships between values and background factors that were investigated here were derived from the early sociali- zation processes identified by these authors and others in this field. Urban experience and education, as mentioned, were two of the relevant background factors. In addition the following factors were investigated: geographic background, section of town, geographic mobility, education, religion, and father's occupation. Specifically it was hypothesized that, compared to those individuals who attach low importance to each 14 of the (same)_values, those individuals who attach high importance to each of the values would: 4a. 4b. 4c. 4d. 4e. 4f. 4h. 4i. 45. 4k. 41. 4m. 4n. 40. Have a smaller proportion of persons Have a greater proportion of persons to rura areas Have a greater proportion of persons Opposed to rural areas. Have a greater proportion of persons town. Have a greater proportion of persons towns. Have a smaller proportion of persons or parochial schools. raised in the American South. raised in urban as opposed currently living in urban as raised in "better" sections of raised in more industrialized who attended military, private, Have had a greater number of books around the house where they grew up. Have more education. Have father's with more education. Have mother's with more education. Have experienced greater geographic mobility. Have fathers with more skilled/ professional occupations. Attend church less frequently. Have been raised in homes where religion was less important. Have been less active in the church during their youth. No attempt was made here to integrate these hypothesized relation- ships with the vast literature relating to the formation of values. However, there are two recent Michigan State studies that have results relating directly to these hypotheses. In addition to the moderation effect discussed previously, Ruh (1970) found that importance of ego need gratification provided by the job (which includes job participation) T "I‘ll‘l'll It‘ll III ['1‘] 11"!!! I -111 III [III I [II I .I‘ I’ll III, {III ll 15 was related positively to education, father's education, father's occupation, size of community where raised, and size of community where currently residing. Morrison (1970) found a positive relation- ship between the personal job goals" advancement and achievement and size of conmunity where currently residing, size of comnunity where raised, father's occupation, education (self, mother and father), and family income, a negative relationship with church activity during youth, and no relationship with frequency of church attendance. The two personal job goals correlated .48 with each other and related positively to the Rokeach values sense of accomplishment, ambitious, imaginative, and independent. With the exception of frequency of church attendance these results are in complete agreement with the present hypothesis. i. 1 I III l l I l ‘1' l ii (I 'll‘ Ill ( CHAPTER II METHOD Sample All of the measures used in this study were obtained from a questionnaire that was given to all the employees in nineteen plants of six manufacturing organizations. All of the companies were located in the Midwest, were operating under the Scanlon Plan, and were members of the Midwest Scanlon Associates. Of 4162 questionnaires distributed, 2755 usable ones were returned. Various subgroups of these 2755 respondents were used in the present study. The median level of education for this sample was twelfth grade. Thirty-seven per cent did not graduate from high school, and 19 per cent had some formal education beyond high school. The mean age was approximately 35 years. The mean tenure was 3.89 years. Fifty-one per cent of the subjects were males. Forty-six per cent of the respondents were raised on farms; 46 per cent were raised in communities ranging in size from less than 2,000 to 100,000 in population; and 9 per cent were raised in cities with populations greater than 100,000. Data The following items from the questionnaire were used to index the attitude, opportunity for participation, and importance of participation measures . 16 Involvement For me, the first few hours at work really fly by. My job means a lot more to me than just money. I'll stay overtime to finish a job, even if I'm not paid for it. I'm really interested in my work. I would probably keep working even if I didn't need the money. The major satisfactions in my life came from my job. The most important things that happen to me involve my work. How much do you actually enjoy perfonning the day to day activities that make up your job? 9. How much do you look forward to coming to work each day? Conmitment l. I feel bad when I make mistakes in my work. 2. How often do you really want to work hard at your job? 3. How much do you really want to do a good job? 4. I'm really a perfectionist about my work. 5. How much do you feel your personal satisfactions are related to how well you do your job? Identification 1. How much is the welfare of your company related to your own personal welfare? ' 2. How much does the company's achievement of its goals help your achievement of your own personal goals? Opportunity for Participation 1. 2. To what extent are you able to decide how you do your job? In general, how much say or influence do you have on what goes on in your work group? In general, how much influence do you have on decisions which affect your job? 18 4. In general, how much say or influence do you have on how you perform your job? 5. My superiors are receptive and listen to my ideas and suggestions. Importance of Participation How important to you is: 1. Everyone in the company, no matter how low his status level having a say in the decisions affecting his job? 2. Being able to decide how to do your job. There were five possible responses to each of the questions ranging from strongly agree/a1ways/extremely/very much, to strongly disagree/never/ not at all/very little. The score for each subject for each variable was obtained by averaging his responses to the appropriate items. Values The values, with the exception of participation mentioned above, were measured ipsatively with the Rokeach Value Survey. Each subject was asked to rank eighteen instrumental and eighteen terminal values according to their importance to him. Six of the values used in this study, self-controlled, responsible, independent, imaginative, capable and ambitious were from the instrumental values; the other three, freedom, equality, and a sense of accomplishment from the terminal values. The values were coded from 1 to 18 (l = most important) according to their position in the ranking. 1. Part of country where raised In what part of the country did you live most of the time before you were 18? a. The Northeast b. The South c. do e. 19 The Middle West The Pacific Coast Outside the United States 2. Urban versus Rural Background The place in which you spent the most time during your early life was a. a. b. c. d. e. Fann Town of less than 2000 Town of 2000 or more, but less than 10,000 City of 10,000 to 100,000 City larger than 100,000 3. Urban versus Rural Residence In what type of community are you now living? a. b. c. . d. e. In the country Town of less than 2000 Town of 2000 or more but less than 10,000 City of 10,000 to 100,000 City larger than 100,000 4. Section of Town In what section Of town did your family live longest while you were growing up: a. b. Lived in one of the most exclusive sections of town. Lived in a good but not the best section Lived in an average section of town Lived in one of the poorer sections of town Lived in a rural area 5. 20 Industrialization of Home Town In your home town what was the main source of income? 6. a. Agriculture, dairy, etc. b. Industry or manufacturing c. Wholesale, retail, or tourist trade d. Petroleum or mining e. Diversified Type of Schooling What kind of school did you attend between the ages of twelve to 18? . . 7. a. Military b. Private or Parochial c. Public d. Vocational and trade e. Did not attend school Books around Home When you were growing up, about how many books were around the house? 8. a. A large library b. Several bookcases full c. One bookcase full d. A few books e. Only a few children's books Education How many years of school have you completed? 6 or less 7 8 9 10 11 12 Some college college graduate work Father's Education (format as above) 21 10. Mother's education (format as above) 11. Geographic Mobility In how many different cities, towns, or townships have you lived? a. b. c. d. e. 1 to 3 4 to 6 7 to 9 10 to 12 13 or more 12. Skill level of Father's Occupation What was your father's chief occupation? a. b. i. Unskilled worker Semi-skilled or skilled worker Sales or office work Supervisory work Sub-professional (bookkeeper, pharmacist, draftsman, etc.) Scientist (geologist, engineer, chemist, etc.) Professional (lawyer, physician, teacher, etc.) Businessman Executive of large business or industry 13. Frequency of Church Attendance Under usual conditions how often do you attend religious services? a. b. C. Twice or more a week Every week Once or twice a month On special occasions (such as Easter) Do not attend services 22 14. Importance of Religion Religion in your home was considered as: a. An essential part of home life b. One of several factors which was important c. A relatively unimportant factor d. Something to be left out of our family life e. One's own business 15. Church activity During Youth As a young man, were you any of the following? a. A church member and active in church activities b. A nominal church member c. A religious rebel d. A non-churchman without any great religious conviction e. None of the above These background items were coded "continuously" as they appear above with the exception of four items that were dichotomized. Item 1, part of country where raised, was dichotomized between the South (b) and all other responses (a, c, d, e,). Item 5, industrialization of home town, was dichotomized between agriculture, diary, etc. (a) and all other responses (b, c, d, e). Item 6, type of schooling was dichotomized between military, private, or parochial (a, b) and all other responses (c, d, e). Item 15, church activity during youth, was dichotomized between active church member (a) and all other responses (b, c, d, e). CHAPTER III RESULTS AND DISCUSSION Preliminary Analysis The items used to index involvement, conmitment, identification, and opportunity for participation were selected on the basis of cluster analysis conducted as part of previous investigations using essentially the same data. The two items used to index importance of participation were selected on the basis of face validity. Inter item correlations and scale reliabilities are presented in Table l. The correlations between the three attituded scales and the opportunity for participation scale are presented in Table 2. Inspection of the inter item correlation matrix and scale reliabilities provided justification for the a priori grouping. In all cases the alpha estimates of internal consistency for each of the scales exceeded the corresponding inter scale correlations. To investigate hypothesis 1 and 2 the sample was divided into first line workers (those with no supervisory responsibilities) and managers (everyone else). To investigate hypothesis 2 both of these groups were then trichotomized into high, medium, and low thirds on each of the nine Rokeach values and importance of participation. 23 24 TABLE 1 ITEM-SCALE CORRELATIONS AND SCALE RELIABILITIES Scale Number Item Total r's Median Item r Name Of Items High Median Low with the other scales Alpha Involvement 9 .81 .66 .31 .43 .87 Commitment 5 .62 .55 .51 .32 ' .69 Identifica- 2 .72 .46 tion Opportunity for Partici- pation 5 .79 .66 .58 .50 .81 Importance of Participa- tion 2 .53 -.03 .41 Note: N = 2030 TABLE 2 INTER-SCALE CORRELATIONS (ALPHA ESTIMATES IN THE DIAGONAL) Opportunity for Involvement Commitment Identification Participation Involvement (.87) Commitment .57 (.69) Identification .58 .46 (.66) Opportunity for . Participation .53 .38 .47 (.81) Note: Sample size ranged from 2732 to 2752. Median N = 2741 25 Individual Difference Contingency Theory Results Hypothesis 1 stated that there is a positive relationship between opportunity for participation and the job attitudes, involvement, commitment, and identification. 'As Table 3 shows the relationship was substantial for all three of the job attitudes both for first line workers and managers. TABLE 3 CORRELATIONS BETWEEN ATTITUDE MEASURES AND OPPORTUNITY FOR PARTICIPATION FOR DIFFERENT HIERARCHIAL LEVELS SAMPLE Attitude Total Workers Managers Involvement .53 .44 .53 (2750) (2183) (384) Commitment .38 .34 .39 (2750) (2183) (384) Identification .47 .40 .39 (2730) (2168) (383) Note: The figures in parentheses indicate the number of observations on which the correlations were computed. Hypothesis 2 stated that values would moderate the relationship between opportunity for participation and the three job attitudes. Table 4 shows these correlations for workers on the high, medium, and low groups for each of the values. Table 5 gives the corresponding results for managers. An inspection of the data indicates no support 26 TABLE 4 CORRELATIONS BETWEEN OPPORTUNITY FOR PARTICIPATION AND JOB ATTITUDES FOR THE VALUE SUBGROUPS OF WORKERS Value Group Rank order of correlations If Monotonic If in predicted in predicted direction, similar Value Hi Med Low Monotonic? direction? finding for managers? Accomplishment Involvement .45 .46 .41 No Commitment .34 .32 .33 NO Identification .43 .41 .36 Yes Equality Involvement .41 .44 .49 Yes No Commitment .27 .33 .39 Yes NO Identification .39 .38 .43 NO Freedom Involvement .40 .50 .43 No Commitment .26 .37 .34 No Identification .38 .48 .34 No Ambitious Involvement .43 .45 .43 ‘ No Commitment .27 .36 .34 No , Identification .40 .42 .38 No Capable Involvement .48 .42 .42 Yes Yes No Commitment .36 .33 .30 Yes Yes Yes Identification .43 .39 .39 No Imaginative Involvement .51 .46 .34 Yes Yes NO Commitment .41 .32 .25 Yes Yes No Identification .44 .40 .35 Yes Yes NO Independent Involvement .41 .48 .44 No Commitment .33 .33 .33 No Identification .38 .43 .40 No Responsible Involvement .45 .42 .46 No Commitment .29 .34 .35 Yes No Identification .39 .38 .44 No Self-Controlled Involvement .46 .44 .43 Yes Yes No Commitment .32 .32 .34 No Identification .39 .41 .41 No Participation Involvement .44 .46 .42 NO Commitment .35 .38 .27 No Identification .41 .45 .33 No Note: Sample size ranged from 483 to 838. Median N = 603. 27 TABLE 5 CORRELATIONS BETWEEN OPPORTUNITY FOR PARTICIPATION AND JOB ATTITUDES FOR THE VALUE SUBGROUPS OF MANAGERS f I Value Group Rank order of If Monotonic If in predicted . correlations in predicted direction, similar Value Hi Med Low monotonic? direction? finding for worker? Accomplishment Involvement .45 .55 .50 No Commitment .41 .33 .36 No Identification .25 .45 .50 Yes No Equality Involvement .63 .48 .45 Yes Yes No Commitment .22 .40 .45 Yes NO Identification .49 .42 .34 Yes Yes No Freedom Involvement .55 .50 .49 Yes Yes No Commitment .40 .47 .24 No Identification .44 .38 .40 No Ambitious Involvement .59 .57 .43 Yes Yes No Commitment .46 .54 .11 NO Identification .38 .42 .44 Yes No Capable Involvement .61 .41 .51 NO Commitment .48 .31 .30 Yes Yes Yes Identification .32 .37 .52 Yes No Imaginative Involvement .44 .57 .54 No Commitment .39 .45 .31 No Identification .34 .40 .47 Yes No Independent Involvement .46 .57 .57 NO Commitment .40 .46 .30 No Identification .41 .37 .47 No Responsible Involvement .43 .54 .62 Yes No Commitment .40 .46 .30 Yes No Identification .42 .26 .52 No Self-Controlled Involvement .53 .48 .58 No Commitment .20 .48 .45 No Identification .49 .41 .33 Yes Yes No Participation Involvement .55 .53 .51 Yes Yes No Commitment .39 .42 .33 No Identification .42 .32 .50 No Note: Sample size ranged from 97 to 149. Median N = 120. 28 for this hypothesis was found. Out of a total of 60 rankings of correlations only 14 rankings were in the predicted direction, 10 were in the opposite direction, and the remainder were curvilinear. The ghgpge_expectation is that 10 would be in the predicted direction. Unlike the Blood and Hulin (1967) study, no improvement was found when only one hierarchial level was investigated. The 14 rankings that were in the predicted direction were divided evenly between managers and workers - and only one of these (capable x identification) was in the predicted direction for both samples. Furthermore, there was little differential support among the values. For only one of the values, imaginative, did the rank order of correlations go in the predicted direction with all three attitude measures — but only for first line workers; for managers two out of three were curvilinear and the third was in the opposite direction. The 14 out of the 60 rankings that were in the predicted direction were distributed over eight of the ten values. Of these eight values, five also had rankings in the Opposite direction. Finally there was no systematic difference in support among the three attitude measures. To determine if their was restriction of range in opportunity for participation in the value subgroups the relationship between opportunity for participation and each of the values was investigated. Table 6 shows these correlations for just line workers and managers. The correlations ranged from -.16 to .07 with a median correlation of -.01. Although several of the correlations did exceed conventional levels of significance it seemed inappropriate to attach any great significance to them because of the inconsistency and low absolute levels. Essentially no relationship was found between opportunity for participation and the values. 29 TABLE 6 CORRELATIONS BETWEEN OPPORTUNITY FOR PARTICIPATION AND VALUES SAMPLE First Line Workers Managers Correlation SignifTEance Correlation Significance Value Level Level Accomplishment -.03 .14 -.15 .003 Equality -.OO .86 +.Ol .84 Freedom +.03 .15 -.ll .03 Ambitious -.O8 .001 -.O9 .09 Capable -.03 .26 -.16 ' .002 Imaginative -.03 .17 -.12 .03 Independent +.00 - .92 +.03 .56 Responsible -.04 .13 +.Ol .78 Self-Controlled +.Ol .59 +.06 .28 Participation +.O7 .OOl +.06 .25 Note: Sample size ranged from 2242 to 2665. Median N = 120. 30 Discussion Method variance. Both the attitude measures and the opportunity for participation measure were assessed by Likert type items. Further- more all of these items had a "right" or socially desirable answer. Therefore it is possible that these variables were in part measuring a common trait - a response bias to answer Likert type items of this form in a consistent manner. One way to check if two measures are assessing the same trait is to compare their correlations with other traits measured by different methods. To do this the correlations of the attitude and opportunity for participation measures with all other itens in the present study were compared. Table 7 shows the correlations of the opportunity for participation and each of the attitude measures with each of these "other" variables whenever one of the correlations 9 To demonstrate that it is a with the other variable exceeded .10. conmon trait would require proving that the correlations are the same - the null hypothesis - and cannot be done. However, as inSpection of Table 7 reveals, the attitude and opportunity for participation measures do correlate with nine other items in a very comparable fashion. The rank order of the correlations for involvement, commitment, and identi- fication with the nine "other items correlated .83, .88, and .87 respectively with the rank order of correlations between opportunity for participation and these same nine items. Therefore, it seems likely that both attitude and opportunity for participation measures were in part measuring a common trait and to that extent their inter correlations were inflated. This effect should be kep in mind when interpreting the strong support found for the first hypothesis. 31 TABLE 7 COMPARISON OF CORRELATIONS OF ATTITUDES AND "OTHER" ITEMS WITH CORRELATIONS OF OPPORTUNITY FOR PARTICIPATION AND THESE “OTHER'ITEMS Opportunity Attitude "Other" for ' Items Participation Involvement Commitment Identification Urban versus rural back- ground +.l7 +.12 +.05 +.ll Urban versus rural resi- dence +.21 +.l8 +.O9 +.l7 Section of town +.13 +.ll +.06 +.lO Skill level of father's occupation +.l4 +.ll +.O4 +.05 Frequency of Church attend- ance -.O8 -.13 -.05 -.06 Church activity during youth -.11 -.15 -.09 -.ll Accomplishment —.13 -.13 -.O7 -.09 Ambitious -.10 -.15 -.O7 -.10 Capable -.11 -.13 -.09 -.10 Rank order correlation between these correlations and those for opportunity for participa- . tion .83 .88 .87 Note: Sample size rangedfrom 2234 to 2520. Median N = 2564 32 However, the fact that the reliability exceeds the interscale correlations indicates that there is some unique trait variance. It is also conceivable that this method variance could account for part of the failure of values to moderate this relationship. If the trait, response bias.were correlated with the values it could have nonuniformly inflated the obtained correlations and thus camoflouged any underlying relationship. Confounding. The main reason for investigating the relationship between opportunity for participation and values was that they posed a potential confound for hypothesis 2. Had opportunity for participation correlated with the values there would have been restriction in range in opportunity for participation when the subjects were subgrouped on the values. However, the essentially zero relationships that were found indicate that this is not the case.10 Vglggg, There are three aspects of the value measure that make it open to criticism. The first is the difficulty of the task of filling out the instrument - ranking the 18 values. Inspection of the data shows that many of the subjects did not comprehend, or at least did not follow, the instructions. Secondly, because the Value Survey appeared at the very end of the lengthy questionnaire the attention of many of the subjects may have been reduced. The fact that 1011 (37%) of the subjects chose not to fill out the instrument or answered incorrectly (or their responses were coded incorrectly) makes the reliability of the values suspect. This could account for part of the failure to get any support for hypothesis 2 and 3. Thirdly, it must be remembered that the Value Survey is completely ipsative. It is 33 quite conceivable, for eXample, that many of the subjects in the low accomplishment group actually attach greater importance to accomplish- ment than some of those in the high accomplishment group. This brings up the question of whether the theory predicts that human relations hypothesis should hold for those who attach a high absolute amount of importance to accomplishment (normative) or whether they should hold for those who attach high importance to accomplishment compared tg_ other life values (ipsative). While no reference to this point was made in the literature reviewed for this study it appears that most of the authors have assumed the normative interpretation. It should be pointed out that one of the values, importance of participation, was not difficult to fill out, appeared earlier in the questionnaire, and was not ipsative. However, it failed to provide any better support for either hypothesis than did the other values. ngplg, There are several peculiarities of the sample that may have influenced the results. a) All of the companies were operating under some form of the Scanlon Plan which should, at least in theory, have put a floor effect on the range of Opportunity for participation provided by the job. b) None of the subjects lived in a extremely large city and therefore differed in that respect from the Blood and Hulin (1967) sample. c) The religious make up of the sample, while perhaps representative of that geographic area, is certainly not representative Of American society and very likely in sharp contrast with the Blood and Hulin (1967) sample. This may have exerted an influence on their values that was not detected by the ipsative instrument. However, the median rankings of the values for the 34 present sample and the Rokeach (1968) national white sample did correlate .90 and .93 for the terminal and instrumental values respectively. Conclusion There is, of course, one other possible interpretation for the lack of moderating effects. That is, the participative management theorists may be correct in deemphasizing the importance of individual differ- ences on the relationships between job and organizational characteris- tics and employee responses. It is difficult to imagine that individual differences in general and values in particular have no systematic effects on employee responses to participation in decision making. However, it may be that such effects are not as great as some of the critics of the participative management position (Hulin & Blood, 1968; MacKinney gt, gl,, 1962; Strauss, 1963) would have us believe - at least not great enought to emerge through the limitations of the present study. Clearly, more research is needed before it can be concluded that values exert any systematic influence on the relationship between participation in decision making and attitudes toward the job. Values and Background Factors Results Hypotheses 4a. through 40. stated that certain background factors correlate with the values. Table 8 shows the obtained correlations along with their hypothesized direction. Of the 150 correlations, 86 35 mNPm mPQEmm .NONN u z :mwvwz .oP.u cusp mmmp men so OP. umwuxm page mmozp mew mcowpmpmccou nmaFLmucz .momm op wowy Eocm ummcmc umpoz Rum moor xmm x0e xmm mum compumcwo empuwumca cw pcwucma mo.- no.- eo.- mo.- mo.+ mo.- - gaze» damage spw>wpaa eucseu No.- Fo.- No.+ mo.o _o.- oo.+ - commeflac co maeaccoaaH mocmu 00.- mo.- mo.- mo.- No.- _o.- - -cappe zucsea co saeazamcu :owpma mp.+ oF.+ mo.+ Fo.- mo.- __.+ + usouo m.cmcmmm mo Fw>mF F_m4m mo.+ eo.+ No.- No.- oo.+ No.+ + spcpwnoe accaecmomw mo.+ No.+ Fo.- mm.. m_.- m_.- + cowpmosnm m.cm;poz mo.+ No.+ Fo.- mm.-. om.- mm.- + :owpmuzum m.cm;pmd mo.+ mo.+ oo. om.u NN.- N_.- + cowgmuswm mp.- mo.+ eo.u Fo.+ No.+ mo.- + mac; eczocw mxoom AzofiungHWFPE Po.- oo.+ fio.- mo.+ No.+ _o.- + \_mw;aocmav m:w_oo;am co mash czop eo.+ eo.+ mo.- _o.- Po.-. mo.+ + ego; mo sewpmNPFMWLHmzvcH A304 F_.+ __.+ No.- No.- oo.- mo.+ + n.0am Loony egos co compaam Azopupmczcv mucmv oP.+ NF.+ No.+ mo.- mo. F_.+ + -wmmc Pegs; msmcm> can»: AEOPuchscv uczocm mp.+ __.+ m0.. Fo.- No. op.+ + -xomn Fees; msmcm> cone: Azopuzpzomv uwmwmc 00.- Fo.+ Fo.+ No.+ oo.u mo.+ + mews: Acpcsou mo “Lea m>wpmcwmmEH mrnmawu msowumae< Ecummcm upmezcu pcmESmwpanuu< uwpuwumcm mEmpH uczocmxumm cowuumcwo mspm> mm34<> oz< mZMHH ozaomoxu0000 000000 $00 00,- P0.+ N0.+ m_.- 00P00_ac co macwmcoaeH mom mo.- mo.+ po.+ w0.- mucmucwupm :uczgu mo zucmscmcm Rom No.- mo.- vo.+ No.+ cowpmazuoo m.ewcpmm mo Fm>mp FFme $00 00.- 00.- N0.+ 00.+ 00w_000E 0200mc0000 mow mo.- No.- No.+ oo.- cowpmosz m.gm;poz mom No.- no.- oo.+ No.- cowpmazuuo m.cm;wmd Rom No.- mo.- ao.+ Fo.- covumusvm Non mo.- 00.. .ao.+ uo.+ mac; ucsocm mxoom fiom eo.+ Fo.- oo.- po.+ AzopuzgmawFwE\—0wcoocmavmcwFoocum mo max» xom oo.- mo.+ mo.+ mo.+ czop use; mo covpmepwwcpmzvcH .wom No.- Fo.+ Po.+ oo.+ Azop u .umm Loony cxop mo :Owwuwm x00 Po.+ 00.- do.+ —o.+ Axopu—mczcvmucmuwmmc Fags; mzmcm> cane: Rom Fo.- oo.- eo.+ mo.- Azopupmcacvuczocmxumn Fags; mzmcm> 200;: mom mo.- 00.- oo.- vo.+ Azopuspzomvummwmc mews; Advance mo peed compuwcwo mampH ucsocmxomm 000000000 cavemacuwpcaa pcwucwa wOPFocpcou mpnwmcoammm ucmncmamvcw -mpmm fleascepcoov w m4mwuum £00050 00.+ 00.+ 00.- mo.- 00.- N0.- 00.+ 00.+ ~0.- 00.+ __.+ 00.+ mucmncmppm - zuczcu mo aucmscmcm 00.+ 00.- 00.- 00.- 0_.- 00.+ 00.+ 00.+ 00.+ 00.+ 00.+ compeazuoo -. . m.mepmm mo pw>m~ prxm NP.- 0?. 00.- 00.- 00.+ 00.+ 00.+ 0_.+ 00.+ m0.+ cowpmozum m.cm;uoz 00.+ 00.+ 00.+ 00.+ 00.+ N0.+ m0.+ mo. 00.+ 000000000 m_co;pmm mm.+ 00.+ _0.- Np.- 00.+ 00.- 00.- N0.+ 000000000 0N.+ F0.+ PF.- 00.+ mo.- 00.- 00.+ mac; wczocm mxoom 00.- 0_.- 00.+ 00.- 00.- 00.- 000_00000 co 0000 mN.+ No.- mm.+ 00.+ m0.+ czop ago: 00 covpmNPP00cpmvucH N0.+ 00.+ 00.+ 00.- mucmuwmmc 000:; msmcm> 000;: F0.+ N0.+ 0_.- prP0noE Owgamcmoww 00.- 00.+ 0300 mo corpumm 00.+ vcsocmxumn 000:; msmcm> 000;: ummwmc 000:: 0000000 mo 9:00 00 mp NF PF 00 m m n 0 m w m N 0 EwuH m2mHH ozsomwxuqm no mep