THE EFFECT 6F MB AND PREMARY MYERACTEQN EXPERIENCES 0N VOLUNTARY ORGANEZATMN MEMBERSHEP: A STUDY OF AUTOMDBELE WORKERS EN FOUR NATIONS Thesis for the. Degree of Ph: D. MICHIGAN STATE UNWERSETY NOREEN LGUSSE DULZ 1973 m {“11 V LIUI 1": Li. v 1 lumen State [,7 Universit 5' M” llllllllllllzllllllllllllllllHill"!!!llllllllllllllll 321 93 10523 2429 “Iv-Hm This is to certify that the thesis entitled The Effect of Job and Primary Interaction Experiences on Voluntary Organization Membership: A Study of Automobile Workers in Four Nations presented by Noreen L. Dulz has been accepted towards fulfillment of the requirements for Ph .D. degree in Sociology Owns 6‘. 946%— Major professor Date May 8, 1973 0-7639 was HDAS & SONS’ 800K BINDERY lNC LIBRARY BINDERS srnlueronr mcmm ACKNOWLEDGEMENT I would very much like to express my appreciation to Dr. William Form for his invaluable assistance in this work. First, he gave me access to his data - far more extensive and significant than any I could have obtained otherwise. Second, he was a most helpful advisor in the planning and writing of this work. Third, he was my most important teacher throughout my graduate education. And, finally, he has become a friend and colleague. For all these gifts, I am grateful and I thank him. The vet zationa has ”9 moat 11k 1m“ beat. . the ”“3313 the high then} azed, Little 18mg, the ac ABSTRACT THE EFFECT OF JOB AND PRIMARY INTERACTION EXPERIENCES ON VOLUNTARY’ORGANIZATION MEMBERSHIP: A STUDY OF AUTOMOBILE WORKERS IN FOUR NATIONS By Nbreen Louise Dulz The vast majority of literature which deals with voluntary organi- zations has focused on the question of which segments of the population are most likely to belong to organisations and which social character- istics best account for varying membership rates. Many studies carry the analysis back a step further to investigate what factors account for the high membership rates of whites, of the upper classes, or the middle aged. Little attention, however, is given to the other side of the issue, the social context that leads to low membership rates of other social categories. In.this study, we test the few suggested explana- tions for the low rate of organisational membership that characterizes blue-collar workers. These explanations describe the blue-collar worker as oriented toward primary relationships, uncomfortable with those of a secondary nature, and largely unable to function well in formal structures. The explanations suggest that these interaction patterns of the worker create barriers which curtail his organisational membership. ‘we test these assertions by ascertaining whether workers who gg.have experience with secondary relationships and formal organisations in some context also belong to theories have a then the man higher members} In a secor ”37 interactic blue-collar we: interaction as “Mined for t} the Motion. The data , “Menu in four nation: us to he“ on mung“, ”I! 98th Of fmu. m ”risen 1‘ baal M” ”one. on the 11'" O. 2 Noreen Louise Duls also belong to voluntary organisations in the commnity. If the theories have accurately identified the factors that limit membership, than the ruoval of these barriers for some ought to result in their higher mbership rates. In a second section of the study we focus on several types of pri- mary interaction and investigate the pattern which characterises the blue-collar worker. Here models which describe primary and secondary interaction as being of either a cumulative or cmansatog nature are examined for their applicability to the interaction of this segment of the population. The data used for this analysis are drawn from a larger study con- ducted by william H. Tom. The sample is taken from automobile workers in four nations of varying levels of industrialisation. The data allow us to focus on an important segnent of workers and, in addition, to investigate the relationships between types of social interaction in each of four nations. The major prediction of the cross-cultural com- parison is based on Alex Inkeles' (1960) mdustrial Han theory. This theory aphasises the uniform and mediate impact of industrialisation on the lives of workers, an impact strong enough to override previous social structures. Based on this approach, we will twpothesise that the carry-over from Job to organisational membership will be found in each of the four nations. In addition, we will predict a similar rela- tionship between primary interaction and membership in each of the four settings. The data, when used to test the first hypothesis, did not offer strong support for the suggested relationship between interaction on the Job and membership in a voluntary organisation. Although, in general, the correlatior ations and were ported. The t: the United Stat masher-ship rate they describe c for the social tionsl task, at. tially llter t}. link between t: explain ”Saul: In regard °’ “1° amulet: between Mr? tion dopendg on action is '1 th that in °rganiz those the arena become 1,33 ho W“ t° be 3 Nbreen Louise Duls the correlations were in the expected direction, they were weak associ- ations and were not strong enough for us to consider the theories sup- ported. The hypotheses were most strongly supported by the data from the United States. This suggests that attempts to explain differential membership rates of blue-collar workers may be culturally bound so that they describe only the experiences of the American.worker. Controlling for the social characteristics of the workers (age, education, occupa- tional task, skill level, and residential background) did not substanp tially alter these findings. we concluded that whereas there may be some link between the two sets of variables, interaction on the Job could not explain organisational membership. In regard to the second hypothesis, the data showed that instead of the cumulative or compensatory theory describing the relationship between primary interaction and organisational membership, the associa- tion depends on the particular arena of primary interaction. 'Hhen inter- action is with the family, a type of interaction quits dissimilar from that in organisations, the compensatory theory holds; interaction in these two arenas tends to be mutually exclusive. As the interaction becomes less home centered and more oriented toward the community, it tends to be accompanied by organisational membership and the cumulative pattern is most applicable. This pattern appears in all four nations, regardless of the level of industrialisation, although the findings, again, are stronger for the United States sample. The relationships found in these data did not substantially change with the introduction of the social characteristics of the worker as control variables. It was further found that among those who report good friends on the Job and also engage in interaction with fellow workers off the 36b, the rate of me: ation is only 1 nor-ken in org: between job ani ‘ue Job can be throagh intera offers acme 811‘ meats that 9h? 18 not as The concl 13339? context 3131311310” 0 w the” appli b Noreen Louise Dulz the rate of membership in an organisation is even higher. This associ- ation is only partially explained by the opportunity to see fellow workers in organisations. The finding points out an even stronger link between.Job and membership than originally identified. Good friends on the Job can be seen as indirectly enabling organisational membership through interaction with fellow workers away from.the work place. This offers some support for the explanations of blue-collar membership but suggests that the relationship between secondary interaction and member- ship is not as simple as it is said to be. The conclusions of the study are, in a final chapter, fit into the larger'context of other factors affecting organisational membership. Explanations of differential membership rates are questioned in regard to their applicability to the blue-collar worker. THE EFFECT OF JOB AND PRIMARY INTERACTION EXPERIENCES ON VOLUNTARY ORGANIZATION MEMBERSHIP: A STUDY OF AUTOMOBILE WRKERS IN FOUR NATIONS By Noreen Louise Duls A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Sociology 1973 I. II. III. V. 0“ TABLE or CONTENTS mmUCTION O O O O O O O O O O O 0 O O O O O O 0 O O MI W WC}! DESIGN 0 O O O O O O O O O O O O 0 Interaction on the Job and Organisational Hunbership The Industrial Setting Correlations and Causality Voluntary Organizations Primary Interaction and Organizational Membership The Four Nations The Research Design m mrom SWDY O O O O O O O O O O O O O. O O 0 DISTRIBUTIONOFTHEVARIABLES.. ... .. . ... Distribution of Job Experiences Personal Interaction Variables Distribution of the Control Variables: Age, Education, Skill Level, Occupation, and Urban Background RELATIONSHIP BETWEEN ORGANIZATIONAL MEMBERSHIP AND SOCIAL INTERACTION EXPERIENCE ON THE JOB . . . Conclusions and Summary of the Zero-Order Relationships Controlling for Social Characteristics of the Worker Conclusions About Hypothesis I RELATIONSHIP BETWEEN ORGANIZATIONAL MEMBERSHIP ANDPRIHARIFORJ‘BOFINTERACTION. . . . . . . . . Patterns of Primary lhteraction: A Continuum of Variables Controlling For Social Characteristics of the Worker ii 13 20 26 33 to L2 117 51 62 70 72 8h Table Of C01 VI. 5"” Table Of Contents (cont'd.) VI. SUHHKRI'AND CONCLUSIONS OF THE STUDY . . . . . . . . Hypothesis One Hypothesis Two The Four Nations Implications of the Study iii 95 95 102 103 went in Perc Social 1 Plant 31 Organiz: man-m Interac' Distrib‘ Distrib Physics; VOlunta Requi re Volunta hate;- VOlunta mutter VOlunta 10. 11. 12. 13. 1b. 15 . 16. 17. LIST OF TABLES Extent of Formal Group Association by Ebttent of Infomal Contacts, in Percentages Social and Economic Characteristics of Four Cities Plant and Labor Force Background Data Organisational Membership Rates in Each of the Four Nations Distribution of Job Variables for Each Nation Interaction Rates in Axelrod's Detroit Area Study, in Percentages Distribution of Primary Interaction Variables for Each Nation Distribution of Control Variables for Each Nation Physical Mobility While on The Job by Percent Who Belong to Voluntary Organisations Required Interaction While on The Job by Percent Who Belong to Voluntary Organisations Number Can Talk To While on the Job by Percent Who Belong to Voluntary Organisations Number of Good Friends on the Job by Percent Who Belong to Voluntary Organisations Ability to Discuss Personal Problems With Fellow Workers by Percent Who Belong to Voluntary Organisations Correlation Matrix of the Job Interaction Index Items to Organisational Membership: The Values of Q Distribution of Job Interaction Index for Each Nation Job Interaction Index by Membership in Voluntary Organisations for Each Nation Job Variables by Organisational Membership for Each Educational Level, for Each Nation: The Value of Q iv List of Tabll 15. 19. 20. 21. 26. 2?, 29, r)! Job 7. for E. Job V Back; Job V for E Job V Task Famil 4 ..3. Neil,“ Orgar Inter Orgar Into} to Vt Inte: ”fifth 3111': Math List of Tables (cont'd.) 18. 19. 21. 22. 23. 2h. 25. 26. 27. 28. 29. 30. 31. 32. Job Variables by Organisational Manbership for Each Age Group, for Each Nation: The Value of Q Job Variables by Organisational Membership for Each Residential Background, for Each Nation: The Value of Q Job Variables by Organisational Membership for Each Skill Level, for Each Nation: The Value of Q Job Variables by Organisational Membership for Each Occupational Task Group, for Each Nation: The Value of Q Family Interaction by Percent Who Belong to Voluntary Organisations Neighborhood Interaction by Percent Who Belong to Voluntary Organisations Interaction With Friends by Percent Who Belong to Voluntary Organisations Interaction With Fellow Workers Off the Job by Percent Who Belong to Voluntary Organisations Interaction With Fellow Workers Off the Job by Organisational Mubership, Controlling for Good Friends of the Job: The Value of Q Sol-nary of Primary Interaction Variables by Organisational Menbership: The Values of Q Correlation Matrix of the Primary Interaction Variables: The Value of Q Interaction With Family by Organisational Membership for Each Control Variable, for Each Nation: The Value of Q Interaction with Neighbors by Organisational Membership for Bob Control Variable, for Each Nation: The Value of Q Interaction with Friends by Organisational Membership for Each Control Variable, for Each Nation: The Value of Q Interaction with Fellow Workers Off the Job by Organisational Membership for Fach Control Variable, for Each Nation: The Value of Q INTRIDUCTION The vast majority of literature which deals with voluntary organi- sations has focused on the question of wish seasents of the population are most likely to belong to organisations and which social character- istics best account for varying membership rates. Maw studies carry the analysis back a step further to investigate what factors account for the hit: m-bership rates of whites, of the upper classes, or the middle aged. Little attention, however, is given to the other side of the issue, the social context that leads to low membership rates of other social categories. In this study, we test the few suggested explana- tims for the low rate of organisational membership that charseterises blue-collar workers. These explanations describe the blue-collar worker as oriented toward primary relationships, uncomfortable with those of a secondary nature, and largely unable to function well in formal structures. The explanations suggest that these interaction patterns of the worker create barriers which curtail his organisational membership. We test these assertions by ascertaining another workers who do have experience with secondary relationships and formal organisations in some context also belong to voluntary organisations in the cosmunity. If the theories have accurately identified the factors that limit mubership, then the 1 reml of these sebonhip rates. To test this benhip, m inns 305, focusing on national setting Nah me dsterai W Winissti Mob mums worship, In I second “‘7 interaction 2 rueval of these barriers for some ought to result in their higher mmsbership rates. To test this relationship between interaction experience and mem- bership, we investigate interaction gained by the worker while on the Job, focusing on required and informal interaction in this formal organ- isationel setting. We measure five aspects of Job interaction and in each case determine the correlation with membership in a cos-unity, vol- unteer organisation. The hypothesis is that those who report these on- the-Jcb interactions will also have a higher rate of organisational membership. In a second section of the study we focus on several types of pri- mary interaction and investigate the pattern which characterises the blue-collar worker. Bore models which describe primary and secondary interaction as being of either a cumulative or M nature are exmsinod for their applicability to the interaction of this segment of the population. Four types of primary interaction - that with family and kin, neidsbors, friends, and fellow workers off the Job - are analysed to find their correlation with involvuent in a secondary organisation in the oo-unity. The hypothesis here asserts the cumula- tive nature of interaction, positing that primary and secondary inter- action accompaw one another, although the applicability of both theories is discussed. The data used for this analysis are drawn frm a larger shady con- ducted by William 1!. Form. The sample is taken from automobile workers in four nations of varying levels of industrialisation. The data allow us to focus on an important segment of workers and, in addition, to investigate the relationships between types of social interaction in each of four nations. In effect, we can ”control for culture. and investigate use: our and {as Isjor P 11c Inkslss' (l utiom and ins n inset strong on this spproac}. Emisstional s In addition, a. union and lambs his study mm “Pom: I'm, ‘0 focusf 3 whether our predicted relationships still exist. The major prediction of the cross~cultural comparison is based on Alex Inkeles' (1960) Industrial Men theory. This theory emphasises the uniform and i-ediate inset of industrialisation on the lives of workers, an impact strong enough to override previous social structures. Based on this approach, we will hypothesise that the carry-over from Job to organisational mubership will be found in each of the four nations. In addition, we will predict a similar relationship between primary inter- action and membership in each of the four settings. This study of social participation of the blue-collar worker makes several important contributions to the work and leisure issue in general. First, by focusing on the link between work experience and leisure time use, me can explain the cutout to which such arenas are linked and the extent to which skills gained on the Job have relevance in non-work situ- ations. By foaming on cognitive and structural factors rather than the usual effectual and motivational variables, we offer valuable information to the issues of the impact of the work place on the behavior of the worker. By focusing on a wide range of formal and informal interaction arenas, our study helps to correct the organisational bias that current- ly exists in our knowledge about participation. It helps to answer some unresolved questions about the linkages between formal and informal inter- action and to test the assertion that they are positively correlated. A major contribution of the study is the understanding of the inter- . action of the traditionally "low participaters' , the blue-collar workers. In addition, as Manner (196M has suggested, it is important to under- stand sub-groups of blue-collar workers and this study offers data on one important sub-group, the sutuobile worker. The social interaction a! the ssh has with who for I larger cal Ch: :- putative n isolated s an occnpa our think: altar-e, h of the automobile worker and of the larger category, blue-collar workers, has largely been ignored. Information such as we will provide has further value for larger inquiries into the linkage of such individuals to the larger enmity and the consequences of mch linkages. Our research also makes valuable contributions because of its con- parative nshsre. The impact of industrialisation can be more easily isolated since other important variablesn-industry, work organisation, and occupation-ohave been controlled and we can new test the validity of our thinking about social participation beyond the boundaries of our own culture. 1h Chapter (be, we describe in greater detail the theoretical back- ground and the gaps in upiricsl research mich led to this study. In addition, we present the research design -- the hypotheses, and the variables and their meaurements used in testing the hypotheses. The second chapter describes the larger four nation study of which this is a part. The first of the data chapters, Chapter Three, presents the distribution of the smuples across each of the variables used in this study, including the control variables that are introduced into the analysis. In Chapter Pour, the data are used to test Hypothesis I, dealing with the relationship between organisational mesabership and interaction experience on the Job. Chapter Five presents the data analysis used to test the relatiomhip between organisational membership and primary forms of interaction, Hypothesis II. The concluding chapter, Chapter Six, reviews this data analysis, summarises the study as a whole, and places the findings in a broader sociological context. CHAPTER ONE m1 AND RESEARCH DESIGN Interaction m The Job And Organisational anbershig Industrial society offers two major arenas for expansion of social interaction beyond the family and neighborhood interaction: coasmsnity voluntary manisations, and 2!; 93 .1213. interactions. In this study we will test the assertion that membership in voluntary organisations can be explained by social interaction rates in other contexts. We will measure interaction as it occurs in the context of the job and deter- mine its relationship to membership in community organisations. In this manner, the study focuses on membership of blue-collar workers while it also offers infonsation about experiences on the job. Blue-collar workers, in contrast to white-collar workers, report a low rate of membership in voluntary organisations . Dhrrsy Rausknecht (1962), for example, found that only twentyhfive percent of his working class snple, including the occupations of skilled, semi-skilled and non- farm labor, belonged to an association. mly nine percent belonged to more than one. Sinilarly, sum and Wright (1911) have found that forty- one percent of the labor class (including skilled, semi-skilled, and unskilled) who earn above 87000 a year belong to an organisation and twenty-eight percent who earn less than this figure belong. Bennett Berg- or (1960) found that thirty percent of his working class sample belonged to an organisation; only eight percent belonged to more than one. The 5 lav lube tilted in its mt u we‘ 6 low nubership rate of this segnent of the population is well substan- tisted in the literature and is seldu disputed. nae explanatims for the nubership rate of blue-collar workers are not as well defined and have not been eupirically tested. We have developed our research design to test explanations which focus on the worker's lack of the necessary interaction experience - experiences dish are often gained throud) the context of one's occupation, which when not present are said to liuit organisational n-bership. The blue-collar worker is described by these theorists as relative- ly I'person centered. so that interaction with others in bureaucratised or fonsl settings is difficult. The worker is said to relate to person- alities and individuals rather than to roles and organised structures. (Killer and lies-an, 1961; nauskneeht, 1961:; Gordon and Anderson, l96h.) Still other discussion have suggested that the blue-collar worker is suspicious and unocufortable in the 'alien' world outside his neighbor- hood. He is lacking the ability to cope well with the broader environ- nent. He desires stability and security in relationships, interacts prinsrily with fadJy and childhood friends and has new friendships that result fru work contacts (Miller and Rises-an, 1961; Cohen and Hodges, 1963: Hausknecht, 1961:; Guest and walker, 1952; Dotson, 1951). Alan Blue (1961;) points out that the social networks of blue-collar workers are typically snell ... honogeneous, leading to a fear and distrust of |'strangers" and an avoidance ef the heterogeneous social contacts typically found in voluntary organisations. The blue-collar worker, these theorists suggest, lacks the interactions that would lead to or- ganisational nenbership. He is not experienced in interaction which takes place in a fornslised structued setting, where relationships are based on re wily, nei I! the fact pmhib: fiction: as {all}, we 1 inter-nation and with no: Ofalnisatio: men would filltionghip label-ship 1 hYEW-hells w ““18 blue-c Since t. '2‘” Done g E” ”389st 'mcesa i1 “hmm the 1d.. of t we r. Th mum-101; t Eh, .nd t: Miriauy hm" “lit “NM“! h|ctly 7 based on roles. Nor is he experienced in interaction beyond that with fanily, neighbors, and childhood friends. If these explanations are comet, and limited interaction does in fact prohibit organisational nenbership, then we can nake several pre- dictions as to relationships that will appear in the data. Host specif- ically, we would expect that those who vary from the pattern of limited interaction and do. have interaction experiences in formalised settings and with non-family numbers, would be likely to belong to voluntary organisations. On the other hand, those without such interaction experi- ences would not be likely to report ueuberehip. The presence of this relationship, a positive correlation between interaction experience and neubership in a voluntary organisation, is the basis of the first hypothesis which is designed to test these enlanations of nenbership ancng blue-collar workers. Since these theories fone the basis of our research design, and since none are clearly detailed, it is important that we note the process they suggest and the assunptions iuplicit in then. The theories imply a process thereby “something” is gained by interaction in one setting and is then applied in another interaction setting; they are based on the idea of a carry-over frou interaction in one arena to interaction in another. The important elements here are that which is ‘_a_i_._ngd_, the assunption that whatever is gained in one setting is applicable else- where, and that it will, in fact, be applied or utilised elsewhere. abseifically, the theories suggest that interaction beyond the usual narrow liuits provides experiences which are carried-over, will facili- tate , and result in, organisational nenbership. hastily what it is that is gained and carried over is rarely speci- “"05 nmdnmfl or Ian scum hrtin Heissm of persona d: activities. that which is level, focusi: Elma such . he ntitudin “Wells m¢flieb 8 fied, although we night assune it to be sous type of interaction Algal; or sons neasure of _t__ru_s_t_ or fauiliaritz with the previously unknown. Hartin Heisener (1971) considers the carry over to be the application of personal discretion or decision asking in regard to the choice of activities. In this study we have not attempted to identify or measure that which is sanded-over, but have instead rained on the behaviorial level, focusing on the association between interaction in two settings. harass such a stance enables us to avoid ease of the difficulties of the attitudinal, enotional, or personal skills analysis, it does not entirely elininate the concern. Our study, because of the theories upon thich it is based, still involves the seeond sleuent of this process, the asmtion that interaction experience in one arena facilitates interaction in another. We have avoided the question of he; it facili- tates nubership, what is carried over, but we still assune that the experiences in one arena apply elsewhere. For our specific study, because we are analysing data about interaction experiences on the Job, we seems that interaction experiences which occur at work are applicable in the setting of voluntary organisations, that a positive relationship indicates that fish experiences have facilitated membership. The causal direction iuplied by these theories is, of course, open to question as is the assumption of a cos-non sleunt. If the expected relationship is found in our data, we will have given support to the explanations of neubsrship rates but we will not have verified the causal priority assigned the experiences of the worker as he interacts outside of f-ily and neighborhood and outside the voluntary organisation. For our specific study, with its focus on the Job setting, we can offer rationale for several causal priorities. He will exanine these nore closely test the and d: the study The tenction 'ie have p: and that i that it no 111111! elem we settiq ‘18“. that to Wm Q 15"! he t1 0): 9 closely when we discuss the job setting as the situation in which we test the explanations of blue-collar nenbership. Our suggestion of a causal direction will, however, be tentative and the central focus of the study is upon the presence of the predicted positive association. The third step in the process inplied by the theories is that in- teraetion in one setting will, in fact, lead to interaction elsewhere. He have pointed out that the theories suggest that something is gained, and that it applies in other settings. We now focus on the assertion that it not only applies elsewhere but 1L1; be applied elsetlhere. ‘lhis third elenent is our prinary research question ~— does interaction in one setting lead to or accompaw interaction in another. He will inves- tigate whether interaction at the Job aecupanies or is positively related to nubership in a voluntary organisation. In su-sry, our study will focus on the actual behavior, the relat- ionships between reported interaction, rather than ask what personal skills the individual gains frcn the interaction. we do assune that interaction in one arena is linked to interaction in another. is we discuss the Job setting we will see that there is theoretical support for this seemed link between Job and off the Job interaction. Our research question is whether interaction in the Job setting, in fact, correlates with interaction in an organisation. we hypothesise that it does. If these expectations are borne out , the explanations of blue- collar nenbership rates will have gained sons upirical support. The Industrial Setting He will test the theories about blue-collar nenbership with a focus on interaction in the industrial work place. The explanations of nonber- hip rate: 9:311 «31 1111 1mm gob also t suggests I now cf ix offer s cc 1120 mm quaint: The I been dflel “‘13-‘18! I ml, so ll” than 15514). In ”I Job t M" 101' r lO ship rates suggest that the worker lacks experience in interacting in fennel settings and beyond the confines of family and neighborhood. Us will investigate mother those who obtain such experiences while on the Job also belong to organisations in the calamity. 111s literature, which suggests a linkage between Job and non-Job interactions, considers the flow of influence to be frou Job to nonc-job. Although we can, and will, offer a counter-rationale, the causal priority assigned to the Job is also inplisd in the theories we are testing as they suggest that non- organisational interaction experiences explain organisational neubership. The suggestion of a relationship between Job and non-Job life has been developed in a large body of research vhich questions whether feelings and attitudes about the work situation carry-over to non-work arenas, so that, for ex-ple, workers who are alienated on the Job are also alienated off the Job. (Seenan, 1967; Friedlander, 1966; Blauner, 19614). In this sane tradition, we investigate the flow of influence free Job to non-Job arenas but focus on correspondence of interaction behavior rather than on attitudes and feelings. Some researchers who have focused on the impact of the work situation on experiences awey from the Job often begin with questions about the over-all leaning of work and the inset of industrialisation on a society. Other theorists have dealt nore directly with the experiences of the ssuent of the population of interest to us, the blue-collar worker. The suggestions of these theorists hsve provided the research questions to be investigated in this study. Harold Hilansky (1961), discussed the importance of the Job to off- Job linkage and suggested that 'the vitality of social participation, priuary and secondary, and the strength of attaclnent to the emity ll and to the under institutional spheres of society are in part a function of cunulative experience in the economic systen..." and 'Partieipation in enmity life is a natural extension of participation in the labor market; orderly and pleasant experiences in the latter provide active and opportunity for the fomr.‘ (p. 522). Hagedorn and Labovits (1968) have further suggested that the process by which work because a shaping factor on eff-Job interaction is one in which the individual acquires responses or values in the work situation which then carry over for application in other interaction arenas. is are investigating this linkage with a focus on social interaction responses gained on the Job. is we have already noted, other sociologists who are interested specifically in blue-collar workers have iuplied this sane relationship between Job experiences and conunity involve-cut. Their attempts to explain wig the “bump rate in voluntary organisations is not as high for blue as white collar workers have focused on the blue-collar workesz lack of the necessary interaction experiences -- experiences which are often gained through the context of one's occupation. The descriptions of the blue-collar worker's characteristic inter- action patterns suggest two dinensions for investigation in our study. no 'person-centered' orientation and inability to relate well to inform- al, bureaucratic structures prompts our decision to investigate $3321.29. interaction on the Job - that interaction dictated by the Job structure, based on forual roles rather than friendship. This type of secondary interaction also characterisss cmnity organisations where secondary- type relationships exist because of the structural constraints of the organisation itself and because of its goals and purposes. The descrip- tion of the worker as suspicious of strangers, likely to interact only w: {a in Itic task. 1'. provides and neig! of indivi tion in v to have p action at. in: which 12 uth f-ily lid neighbors, suggests we also focus on that Job interact- ion which is inforual, possible while on the Job but not essential to the task. The chance to interact on a personal level with fellow workers provides opportunity to gain contacts outside the small circle of fauily and neighborhood and to increase one's interaction skills with a range of individuals. Such a skill is also necessary for successful participa- tion in voluntary organisations where infornal relationships develop. is have predicted that the opportunity for required and infernal inter- action while on the Job provides the worker with experience in interact- ing which then carries over to nubership in voluntary organisations, where the sane skills are essential. Only a few atteqts hsve been made to empirically validate the claiu of a relationship between interaction experiences on the Job and in the co-unity. Martin Heissnsr (1971) conducted a study which this study will closely approxiaaate. He found that responses and experiences gained on the Job carry over to further interaction off the Job. In constructing his explanation, heissner identified the interaction skill gained on the Job and applied in the act of Joining a co-mnity organisation as one of |‘exercising discretion.” These two types of interaction vary together - when a worker has the option to interact at work, he also engages in discretionary activities off the Job, i.e. , he Joins an organisation. Heissner's research also contains indicators of technical constraints on the Job such as ability to move freely, and he finds a positive relation- ship between the absence of spatial constraints and the decision to Join an organisation. Other research has indicated fliat workers involvement in the life of the plant leads to attendance at union neetings. (Dean, 1951a: Lip- act, at. .- interactic closely a the influl linkedto 13 set, at. al., 19563 lore and Danseraau, 1951). this suggests that the interaction of the worker while on the Job does carry over, at least to closely associated arenas of interaction. He will investigate whether the influence of work is as far reaching as orgausations not directly linked to the Job. Correlations and Causality ilthouda a correlation between interaction in voluntary organisa- tions all! on the Job, by itself, does not imply a causal relationship, the theories do and it is incl-taut to ex-ine the rationale behind the assuqtion that organisational mabership is caused or explained by interaction in another arena. In particular, we must examine the slain that interaction on the Job miflit be such a csusal variable. The writings of Vilensky and of Hagedorn and Labovits which describe the shaping impact of experiences on the Job, as well as those of others which discuss the carry-over of feelings and attitudes from the Job, offer Justification of a causal direction from Job to off the Job. Our assertions, however, about interaction experience and behavior have not been as thoroughly dealt with, although Heissnsr's study lends empirical support in this regard. We can offer a few further rationale but it lust be realised that the suggestion of a causal direction is still a highly tentative one. We have reasoned that the Job experiences are more scheduled and regulated than those of secondary organisations, so that expansion of social contacts beyond family and neighborhood is most likely to occur there first. That is, the work situation is seen as the least voluntary and in that lidit oomtitutes the set of experiences most readily evailable to the worker. This rationale is basically the ‘ai use as thl suggestions priorities; mantis he re tired and r em, for s in both the mt b! tor tint, we an ml! dei Wits 0th” “It! in the Selection 1! m1 31th 01 11; same as that utilised by Haissner. The blue-collar theorists , whose suggestions we are testing, do not offer a rationale for their causal priorities; our reasoning helps to Justify their claims of the causes of organisational n-bership when the process is tested in the Job setting. has reasoning that Job experiences are less voluntary and nore struc- turad and regulated overlooks several counter-arguments. It may be the case, for example, that a broader set of experiences shapes interaction in both the Job aid organisations. lh this case, it might be suggested that by testing the theories of blue-collar lubership in the Job set- ting, we are focusing on a range of interaction experiences that are too narrowlydefinsd. Hamstbearinmind that this studydoes mtincor- porste other possible sets of explanatory variables. A further diffi- culty in the suggested causal direction is that it assumes that the Job selection is involuntary; it does not allow for the fact that in indivi- dual with organisational experience may seek a Job that allows for inter- action. lor does the perspective allow for the fact that experience in a enmity organisation nay lead the worker to increased use of oppor- tunity for interaction on a Job he already holds. The cautions demand that am causal priority assigned by the theorists we are testing, as well as the causality we suggest in regard to Job experiences, be taken as tentative. Realising this, our primary focus in data analysis will be upon determining relationships. We will discuss the possibility of a carry-over fru Job to non-Job am! will present the data in a fashion to allow for such causal analysis but must do so with the tentativeness that surrounds most assigi-ent of causation. Yoluntag Organisations Both sections of this study, the first testing theories of blue- 15 collar m-berehip ad the second focusing on the relationship of primary and secondary interaction, utilise the amenity volunteer organisation as a maJor variable. It is important, then, that we specify the manner in which we define such organisations. In addition, for the first sect- ion of the study, we will discuss types of volunteer organisations and organisational conditions under whida we expect interaction experiences that are non-family, non-neighborhood , and non-organisational to cor- relate with membership. He ascertained organisational membership by asking: 'irs you a member of sue organisations? (sport, regional, religious, cultural, recreational, fraternal)" he respondent than named the organisations to which he belonged so that, presumably, those not applicable could be screened out by the interviewer. ‘lhe data, then, include the respond- ent's recall of memberships, but excludes union membership (although per- haps not organisations within the union) and g_h_uro_h attendance (although religious organisations are included). Such an exclusion of church and union meflaerehip from the capilation of organisations is optimal for our research. The inclusion of church attendance inflates membership rates ails measuring a type of involvement that differs from community organisations. (In the three samples where these data are available, a large maJority report church attendance but in no case does the maJority belong to an organisation.) It is beneficial, also, that union member- ship m g is excluded from this item ad is treated separately. Union structure and participation, as well as the degree of voluntarisn, varies widely between nations so that its inclusion in membership rates would have an inconsistent effect and, more iqortantly, the theories we are testing attapt to explain enmity organisational membership and do not l6 focus on Job related organisations. If they did, further causal var- iables would be implied. fliers have been a number of attempts to classify volunteer organi- sations according to their goals, the functions they serve for larger society, the functions they serve for members, etc. The classification lost appropriate for our research question is that which labels organisa- tions as instnseental or grassive in nature. This scheme has been most thoroughly defined and developed by Nicholas Bsbchuk, although it is basic to new other models as well. Instrumental organisations are those oriented toward maintaining or changing the status quo, that focus on goals and activities that are outside the organisation itself, and that are not desipiad to provide i-ediats gratification for the members - althouh, of course, they often do. hpressive associations, on the other hand, are those adaose primary function is provision of iaauediate gratifi- cation to its members. (Babchuk and Edwards, 1965; (Jordan and Bsbchuk, 1959). In general, it has baa found that the working class tends to belong to expressive organisations while the middle class has hidier rates of membership in instrmaental associations (Bauskneeht, 1962). We would expect respondents in our samples to indicate most memberships in expressive organisations. There are several limitations specific to our study which make the use of this classification, and that of other similar models, difficult to include in our data analysis, but we can discuss the applicability of our predictions to each type of organisation. First, it should be noted that the explanations of blue-collar membership which we are testing do not, thaselves, specify adiich organisations are included within their realm. ‘By default, we amist assume that the limited interaction of 17 workers is seen as a barrier to membership in both types of organisa- tions. We might, however, go on to suggest that if the explanations are ever appropriate, it would probably be in reference to expressive organisations. than workers do Join organisations, they tend to be those of an expressive nature 3 we would expect that increased interaction Mensa would be most likely to facilitate involvement in those or- ganisations where there is already some normative support for membership. If this assumption is a valid one and the theories are most likely to be upheld aen applied to membership in an expressive organisation, then it is a less serious limitation of this study that the original question mentions , and probably receives information on, only expressive organi- sations. ilthagh the specific organisation to which the individual belongs and an understanding of the nature of each organisation in the four nations may result in the identification of some instrumental asso- ciations, the suggested sport, recreational, fraternal organisations cited in the question all call to mind maberships which are primarily for personal gratification to the individual. we must same then, because of inadequate information to prove otherwise, that our data are primarily, if not comletely, on expressive organisational membership. Again, we have further assumed that if the theories of blue-collar member- ship ever apply it will be in regard to these organisations. merience in interaction beyond the confines of family and neigh- borhood is not, of course, the only variable affecting membership in voluntary organisations. Examination of other factors will help to put the findiaus of this study in proper perspective , as identifyim some of the contributing influences. Hails other suggested explanations of nabsrship will be cited, it must be realised that explanations of vary- ing rates of organisational pgrticipaticn are outside the scope of this 'I". 18 study. Rates of participation are dependent upon mam factors within the organisation itself, as well as subJect to external influences, and our data do not furnish the information necessary to explain the more caaplex set of behaviors. we will focus on what occurs within the orgaIdsation only to the extent that knowledge of such might lead to the initial decision to Join. In noting factors that lead to membership we will first look at suggested benefits of membership which may motivate the individual to belong. We will then focus on correlates of member- ship-somaelassrelatedandsuenot-ahichmaybecausesofmember- ship. For these variables, like those in our study, the suggestion of causation is, of course , tentative. Arnold nos. (1956) has suggested that individuals can fulfill the personal needs of self expression and security via organisational mem- bership. In addition, they can satisfy axes of their personal interests through the collective action of. such organisations, can work toward social change and can gain a measure of «unity power. Babchuk and ldwards (1965) have suggested that individuals may Join organisations to benefit from the available effectual support, to engage in expressive activity and to implement special interests. In the second section of this study, we examine primary interaction variables that may help to explain expressive organisational mabership. Our data analysis will help ascertain the validity of the claims that individuals Join organisations to fill in for the inadequacies of their primary interaction, and that thepocess of industrialisation is the reason for these inadequacies. worship in instrumental associations is also considered to be beneficial to onsfis career advanc-sent and to one's social status in the 19 commnity, so that one Joins for anticipated personal benefit vie achievement within the organisation and valued service to the community. (Hausknecht, 1962). This explanation is probably not very applicable to the particular segment of the population included in this study. As we have noted, membership in instrumental associations does not predom- inate among blue-collar workers and, in addition, it has been found that achievement within organisations tends to parallel the occupational struc- ture (Ross, 1951.) so that those who do belong would be less likely to ad- vance within the association. in awareness of the improbability of achievement within the organisation is likely to further deter membership of this type. This suggested benefit of organisational membership is, then, class related. Ham of the correlates of membership, where causation is only implied and tentative, are also class related. Host of the factors that are positively related to marchership are either characteristics of middle and upper classes or definitions of them. Membership, for example, is higher among those with hidler incomes, with prestigeful occupations, with high levels of education, with resi- dance in high status neighborhoods and with home ownership. (Hausknecht, 1962; Hagedorn and Labovits, 1968; Hyman and Wright, 1971; Bell and Force, 1956). These findings of membership correlates have been sup- ported in repeated research. They do not, however, focus on blue-collar workers and offer little to explain membership among such individuals. Other class related explanations do focus directly on this seglnwt of the population and are more central to this study. Gene (1966) and Hausknecht (1966) have suggested that those who participate are in some way marginal within their social contexts. They may be socially mobile, upward or downward, or they may be residaltially mobile. In either case, all: 5m lam ttri 1!: £101 20 it is suggested, the individual is not well integrated into fennel re- lationship networks of his area and therefore turns to organisations. A further suggestion, although it sometdlat begs the question, is to refer to class related norms which support or discourage organisational membership. In this case, it may be noted that norms of the working class do not praeote manbership and that the low rates may be explained by a self perpetuating cycle of no membership and therefore no norms (Hausknecht, 1962). In addition to these class related correlates of organisational sabership, there are others that are not specific to arw given class. In this realm research has found membership to be positively associated with middle age, urban residence, length of residence and with being male, finite and Jewish (or Protestant, secondly). (Bell and Force, 1956; Hyman and Wright, 1958; Bauslmecht, 1962). Organisational membership, then, can be partially uplained by a large umber of factors. Some of these are benefits that come to the individual as a result of one's melnbership. Others are social charac- teristics, class related or not, for which the assigment of causation is indirect and tentative. Noting the factors which explain membership will also help to put our findings in proper perspective. weak correla- tions must be seen, at least in part, as a result of the fact that our independent variables only partially explain the influences toward organisational mmnbership. Haw other factors also help account for the variance in these rates. Primag Interaction and Organisational Hmbership In the second part of the study, we focus on the workers' rate of interaction with family and neighbors and examine the relationship of such 21 interaction to organisational membership. Here we ask, not whether Job experiences explain membership, but to what extent more primary forms of interaction do. There are two competing theories as to the relation- ship between primary and secondary forms of social interaction. In the first, interaction is seen as substitutive or comensatory, in the second as cumlntive or generalising. The first theory of social interaction and its particular pattern suggests that interaction is of a W or substitutive nature. the assumption underlying such a construct is that individuals have a fairly constant level of interaction needs which met be filled. The individual interacts so as to attain and remain at that balanced level. than an individual is lacking interaction opportunity. or experience in one arena, he conensates by involvnent in others. If this theoretical construct were empirically verified we would find that individuals |'specialise" in one interaction arena to the neglect of others. Follow- ing the theory, we would expect the data to show that those who parti- cipate in voluntary organisations have little involvement in primary form of interaction. On the other hand, those for whom primary inter- action is strong will not, in addition, Join formal organisations. There are two lines of thinking that build on this general compen- satory theory. may vary according to their assumptions about preferred interaction arenas. loch assumption, that either the primary or the secondary is preferred, leads to different interpretations of the data. larly industrialisation and modernisation literature (Hirth, 1938; Rose, 1956; that ad revs, 1961;) offers the suggestion that modern society has Jeopardised and extinguished primary relationships. This has led, they suggest, to efforts to fulfill these needs through volun- tary organisations. The assumption here is that primary relationships 22 are the most desirable and that individuals will choose voluntary organisations only when interaction in personal arenas is not available. Heissner's study (1971) offers a second set of assumptions under- lying the compensatory theory of interaction. Heissner (and Hagedorn and Labovits as well) suggests that interaction in voluntary organisations requires skills that are more socially valued than the skills required for personal interaction. For this reason, individuals will participate in the more highly rewarded arena, voluntary organisations, when they can (i.e., when they possess the required interaction skills). than they cannot, they will cmpensate by high rates of interaction in primary arenas. Heisaner's assertions are tested by looking at the relationship of Job experiences (the arena for gaining the needed skills) to the primary-secondary interaction pattern. A second theory suggests that interaction is of a cumulative nature so that interaction in one arena is positively correlated with inter- action in other arenas. There is support for this theory in the fact that primary relationships have not, as modernisation literature pre- dicted, declined in industrialised society (Sussman and Burchinal, 1962 3 mama, 19593 Axelrod, 1956; Dotson, 1951). Research points out that whereas the rate of voluntary organisation membership may have in- creased, it has not replaced or had to compensate for primary ties. Instead, data show that the primary and secondary interaction arenas vary’ together. Unfortunately, the research of these authors does not focus specifically on the blue-collar worker so is supportive only of the general explanations of mmabership idlich we will test for our sample. Of the studies which offer support for the assertion that interaction is of a cumulative nature, Allardt, Axelrod and Teele's work serve as good illustrations. 13 f’; 23 Erik Allardt (1957), who has studied leisure activities and social participation of Finnish youth, found participation in leisure activities to be of a cumulative nature—those who were active in informal arenas were also active in formal organisations. His study included association mmaberahip, peer groups, visiting friends, sitting in. cafes, and reading. He found his thirteen activities to be positively correlated, with the exception of two, at a significance level of .05. Axelrod (1956) has also found that fennel group association and the number of informal contacts tend to vary together. His data illustrate these findings: Table l. Extent of Fennel Group Association by intent of Informal Contacts, In Percentages S of informal contacts in two-month period 0:1; 5-12 12-19 20+ Ion-Member 56 1.1 32 32 llmnber m. 59 68 68 Total W 16‘0” 106 I Cases 71 229 225 221; James E. Tools (1962) has investigated the correlation between interaction with friends, relatives, in church, clubs, and social hobbies. He discovered that even though the intercorrelations were quite low- only three were simificant at the .05 level-419 could tentatively con- struct e culnllative scale of social participation in which the ”compon- ent variables (friends, clubs, and hobbies) appear to be linked by the actor's freedom to decide whether and with whom he will become involved in socially interactive situations.” (p. 39). His final scale, after elimination of those items which duplicate each other, is as follows: ah I II III Sees no friends See friends Sees friends He social hobbies No hobbies Hobbies, voluntary orgs, No voluntary orgs. No voluntary orgs . or both The frequencies of these three categories for the males of his sample were as follows: Number of Cases Percent I hl 17.2Z II 99 141.15 III 99 Ill.“ 1 third possibility, besides these two theories, is that no rela- timship exists among these various types of interaction. The theoretic- al support for this pmdiction is based upon sane aspects of role theory and suggests that the various roles of an individual in modern society are separate and unrelated, so that experiences in one do not carry over to influence involvement in other arenas. In such a case, the data would show no relationship instead of the positive correlation enacted for the cuanllative theory or the negative correlation of the canpensatory theory. In our study, we examine four types of primary interaction, that with family and kin, neighbors, friends, and fellow workers off the job. is we analyse the relationship between each of these and organisational membership, we will be able to determine whether the interaction patterns of our ample of blue-collar workers best approximate the compensatory or the cumlative model. The Four Nations is noted in the introduction, the data for this study, and the larger 25 one from which it is drawn, were obtained by sanpling automobile workers from.four nations at‘varying levels of industrialisation.* These samples allow us to investigate whether the impact of the work situation on membership in volunteer organisations is the same in all industrial situ- ations, regardless of the culture. we can also ascertain whether primary and secondary interaction exhibit the same relationship to each other for each of the sampled groups. In effect, we can ”control for culture" and investigate whether our predicted relationships still exist. The major prediction of the cross-cultural comparison is based on Alex Inkeles' (1960) Industrial Hen Theory. This theory emphasises the uniform.and immediate impact of industrialisation on the lives of workers, an impact strong enough to override traditional social structures. Based on this approach, we will hypothesise that the carry-over from.Job to organisational membership will be found in each of the four nations. Inkeles' (1969) study of 6,000 men from six developing nations of- fers data which are supportive of his theory about the impact of indus- trialisation on the lives of individuals. no found that occupational experiences in complex organisations, especially in factory work, con» tribute to modern attitudes and actions on the part of the worker. Inkeles writes, "In reviewing the results of our research on.modernisa- tion, one must be struck by the exceptional stability with which.variebles such as education, factory experience, and urbanism maintain the absolute and relative strength of their impact on individual modernisation despite the great variation.in the culture of the men undergoing the experience and in the levels of development characterising the countries in which.they live.“ (p. 225) ‘Hhereas Inkelea predicts an immediate and uniform.impect of indus- trial expariencas on the attitudes and activities of the worker, Hilbert s The four nations are the united States (OLDS), Italy (FIAT), Argentina (III) and India (PAL). I. l flu in: can he‘i wen u; 02m £qu 26 E. liners (1965) suggests an evolutionary theory thich smswhet medi- fies Inkeles. In there's evolutionary theory, industrialisation has an impact thich parallels a model of concentric circles. That is, its impact is most observable and uniform at the center of 'organised pro- duction.‘ Industrialisation has less impact as we move further from this center, as there is more opportunity for unique cultural and structural factors to intervene and alter the impact of industrialisation. If we were following these suggestions fro Moore, we would expect the greatest similarity between nations to be in experiences at the work place. We would expect the least sinilarity between nations in regard to patterns of primary interaction, where the impact of industrialisation is com- bined with new dimensions of the unique cultures. Although we will look more closely at lbore's thinking in our data analysis, our hypotheses predict a similar impact of industrialisation in each nation, as pre- dicted by Inkeles. That is, in each of our four nations, regardless of its level of iuiuatrislisation, we expect that experiences gained on the Job will facilitate membership in voluntary organisations, off the Job. The Research Basie In order to investigate the relationship of work experiences and organisational membersMp, our first set of independent variables are five aspects of the Job. The first measures the ability of the worker to move around while on the Job; the second and third, the presence of our two types of interaction - required and informal. The fourth and fifth variables ascertain a qualitative evaluation of the workers rela- tionships on the Job. his five Job variables are as follows: 2? l. Basical mobility on the Job 2. Required interaction on the Job 3. Informal interaction on the Job h. label- of good friends on the Job 5. Ability to discuss personal problems with fellow workers In addition, because these five Job experiences and opportunities are cumulative in their effect on the worker, we have constructed a total interaction index based on these five variables and will use it in analysis of the carry-over from the Job to off-work interactions. 1. ical ilit on The Job The ability to move about exposes the worker to a broader range of potential relationships than that experienced by the worker who cannot move Idlile on the Job. This information, gained by asking I'liow far can you move freely thile on the Job?“ is an indirect measure of possibility for interaction as well as a probable indicator of the amber of others to whom the worker is exposed. heissner, the has used this variable in his study, has suggested its interpretation as an indicator of 'diecretion'. That is, the worker the can move freely has the opportunity to decide whether or not to interact with others, to exercise discretion. In heissner's reasoning, use of discretion is one of the skills which may be carried over and utilised in the decision to Join a voluntary organisation. 2. mg Interaction on The Job For blue-collar workers, required interaction offers experience in secondary relationships, a type of interaction thioh also charac- terisea eo-unity organisations but which differs from the worker's experimlcas in fully and neighborhood interaction. urge the j dent! the: otu 11111 the 0M the the he Flu at: “Or 28 The worker gains experience in relating to bureaucracies and large structures as he enters into interaction which is required by the Job. For inforamtion on this type of interaction, the respon- dentewsre asked thequestion, 'howmarwmuatyoutelktowhileon the Job” 3. Informal terection on fine Job In this itu, the workers were asked to indicate the number of others they 'can talk to while on the Job." We have used this infor- mation as an indicator of the quantity of informal interaction which the Job situation permits. his chance to interact with fellow workers on a personal level, we have reasoned, offers the worker experience in interaction which extends beyond the characteristically small circles of fully and neighborhood. Such an opportunity increases the individuel'e interaction skills with a range of individuals and thus helps to overcome the inability to cope well with the broader environment. This inability, as we have seen, has been cited as ex- planation for the workers low rate of membership in voluntary organi- ..time h. lumber of Good Friends at work, and 5. Ability to Discuss Personal Problems with Fellow Workers With these two variables, we are gaining information on an oval- uative dimmlsion of interaction at work. We are investigating primary relationships and .ount of personal involvement with fellow workers. By asking his workers to indicate how many good friends they have in thairworkgroupand tosmwerthequestion, 'Canyoutalktefellow workers about very important problemet', we are investigating the ”1 ion inc? M be: Pm rel 29 extemion of personal relationships beyond the "inside" world dis- cussed by Plus (1961:). he second set of independent variables measure social interact- ion in the primary arenas. There are four types of such interaction inclnded in the study. 1. all and [in The presence of interaction is measured by noting whether family or kin is mentioned when the respondent is asked With whom do you spend weekends and holidays?‘ We have assumed that virtually every- one spends sue time with f-ily m-bers. We are tapping here whether or not the family is chosen by the worker for interaction than he has leisure time and a choice of companions. 2. mane Here the respondent was asked 'Do you exchange visits with neighborhood friends?" In this question we are interested in said)- borhood interaction which extends beyond the casual chatting that proximity often promotes. 3. mi For information on interaction with friends, the respondent was asked, 'Apsrt frma relatives, did you spend vacation time with friends?“ For the PAL sample, where this question was not asked, the same item as for family interaction we used, coding whether or not friends were also mentioned. h. Fellow Workers , off the Job The respondent was asked I'Do you meet fellow workers that are good friends outside work‘:"' Here we are measuring the presence of relationships which extend from Job to off the Job, not casual and 30 unplanned neetings with acquaintances fron work. Our dflent variable, used as an indicator of cos-unity involve- nent or interaction is that of nsnbership in one or more voluntary organisations. In: asking this question, union membership was excluded with the reasoning that this work related, often required, membership was not sufficient indication of the annuity involvenent we are inter- n ested in. me worker was asked, 'Are you a nenber of sons organisa- * tions? (tort, regional, religious, cultural, recreational, fraternal)." His answer was coded as yes or no, not tabulating separate responses for those belonging to nore than one such organisation. LJ The data analysis also includes five control variables. for each of the associations we have Impothesised, further analysis is nade to ascertain whether the expected relationship holds regardless of the fol- lowing social characteristics of the worker. 1. Education 2. Age 3. Residential Background, urban or rural h. Skill level 5. Occupational Task (a variable which includes the cats- gories of assembly, nachine opera- tor, test-inspsction-repair, and skilled trades. Each of the M will be tested for each of the four nations. Following the theory of Inkelss, we will predict that the expected rela- tionships will be obtained for each of the four nations. he first naJor lupcthesia focuses on the effect of the 1_v_o_r§_ 31 situation on nembership in formal voluntag organisations. is we have sea, the low laborship rates in voluntary organisations by blue- collar workers are often explained by their lack of skills and experi- ences in interaction—skills and experiences which night be gained on the Job. It is then suggested that those workers who do gain such experiences on the Job will utilise the skills and Join voluntary organ- isations. To test these assertions, we predict that the greater the opportunity to acquire and exercise such skills on the Job, the higher the rate of nonbership in voluntary organisations, controlling separ- ately for education, age, occupation, skill level, and urban background. 1. boss workers with less spatial constraint on the Job will be more likely to belong to voluntary organisations. 2. Those workers with more required interaction on the Job will he wore likely to belong to voluntary organisations. 3. those workers with greater opportunity for informal inter- action on the Job will be nore likely to belong to voluntary organisa- tions. h. mm workers with the greater nunber of good friends at work will be nore likely to belong to voluntary organisations. 5. These workers who can discuss personal problens with fellow workers will be sore likely to belong to voluntary organisations. 6. Dose workers with the highest score on the index of combined Job experiences will be nore likely to belong to voluntary organisations. The secmd major lupothesis deals with the prinary interaction patterns of the blue-collar nrker. Here, we are concerned with the relatiflig between seconds}: and m interaction arenas. Con- trolling separately for education, age, occupation skill level and urban 32 background, we predict: l. A positive relationship between membership in voluntary organisations and interaction with family and kin. 2. A positive relationship between membership in voluntary organisations and interaction with neighbors. 3. A positive relationship between membership in voluntary organ- isations and interaction with friends. h. A positive relationship between membership in voluntary ”sensations and interaction with fellow workers, off the Job. To test the degree of association between the variables rule's Q and One will be used, depending on the specific variables involved in the analysis. Dy conbining these two measures, we can, according to Jones nous (1971), use variables that are either dichctomous or not and hsve cuparabls statistics. In addition, we will present contingency tables so the original date are available to the reader. 0mm M THE ORIGINAL STUD! his data used for testing the hypotheses of this study are fraa a larger study conducted by william H. Form. The broad study plan allowed for gathering information from workers in one technologically advanced iMustry--the automobile industry. Pour such plants were supled, in nations of varying levels of industrialisation. It is a strength of this research that whereas the level of economic devop- ment of the nations varies, the industry, occupations, and work organi- sation of each plant remains sufficiently constant to investigate the impact of industrialisation on social interaction. Of the four nations, the most advanced was the United States where the Oldusobile (OLDS) plant in Lansing, Michigan was sampled. Italy was rated the second most highly industrialised nation, characterised by a small and stable, but growing, industrial base. To represent Italy, the Fiat (FIAT) plant in Turin was sampled. Argentina, a nation with a relatively recent trend of industrialisation, ranked third among the four. In Argentina, Industrias Kaiser Argentina (m) in Cordoba was the plant from which data were gathered. The nation ranked fourth was India where the industrialisation is at the lowest level and is located in only a few cities. India 's Frasier Automobile Limited (PAL) in Bombay was the plant sampled in this least industrialised nation. Table 2 gives social and economic characteristics of the four cities at about the time the data were gathered. (Pom, unpublished manuscript). 33 3h Table 2. Social and Economic Characteristics of Four Cities SBA ’ Greater Characteristic Lansi_ng__ Turjiln Cordohgfi Bombay Siset Metropolitan area 298,959 1,015,511. 635,000 11,152,000 Per cent of labor force b in.msnufacturing 31 61 35d h3° Bean sise of mamfacturing f establishesnt 108 79° 11 20 Mean years of education, adult population 12.0 11.8 5.9 3.0 Per cent of labor force in services 67 38 614 118 a. 0.8. Bureau of the Owens, 1962: Table 121. b. 1311!, 1961.. c. Cittd di Torino, 19593137. Mean number of uployees for metal manufacturing was 173. These data are for the province of Turin. d. Direccion General de Estadistica, Censos e Investigaciones, 1967. The Argentine census of 1960 reported an average 7.1 workers per manufacturing plant in Cordoba and 9.2 for the nation as a whole. e. Covert-ent of India, 1962:32-35. f. In the state of Maharashtra half of the factories uployed 20 workers or less. Usually only factories hiring 10 or more ms- ployees reported to the census. In all of India 79 percent of 132?”... in 1962 hind fewer than 119 workers. See Labour Bureau, 1 . Each of the plants chosen had been in operation for at least ten years, had a native born management, and involved a comlste range of functions from desipi to production. As the samples were chosen, deparhaents with ainilar technologies, representing the core manufactur- ing processes and the range of skill levels, were utilised. Hhere skilled workers were under-represented (OLDS and In), additional cases were drawn in order to have enough for analysis. Table 3 describes the 35 four plants and their work forces (Form, 1967). Table 3. Plant and Labor Force Background Data Characteristics OIDS FIAT IKA PAL 1. Age of plant 1902 1911 1955 19116 (year established) 2. Mean age of worker 112 35 30 32 3. Tears of education 9 S 7 6 (median) 14. Per cent married 9h 76 71 81 5. number of children 2.80 1.07 1.55 2.21 (mean) 6. wife deployed 29 26 15 5 (per cent) 7. ‘30 It fir't full-time not (loan) 17 13 15 17 8. Number of years in the Labor force 25 22 15 15 9. Tears of service in present factory 13 9 5 8 10. lumber of positions held during mtire period of uploy- unt (Hm) 3.8 3e3 3e? 3e9 11. Occupational background 3 (per cent) Industrial-manufactur- 108 2 59 62 '17 Business and Services 3 12 32 28 Agriculture 6 19 2 1.11 lo previous «ploy- msnt 10 h 11 Subtotal 100 100 100 100 12. Father's occupation- faraer (per cent) '10 3h 21 3h 13. Rural or small town birthplace (per out) 68 66 36 he que: m. In t "I! 0923 m1 he me 36 T'bl. 3 (Mt’de) 1h. Upward worklife mobility (per cent) 51 1:8 112 116 The data for the larger study were collected, in 1963, by use of a questionnaire, administered by the researcher in the homes of the work- ers. (accept for PAL, where workers were interviewd in the factory.) In 01.08 and FIAT about seven percent of the sample were either not available or refused the questionnaire. In In and PAL the correspond- ing rate was about four percent. The questions included measurement of the workers adaptation to a wide range of social contexts, including work routine, social relations on the Job, union participation, and involvement in fuily, neighborhood, camunity, and nation. In addition, data were gathered to construct an index of anemic. Several of the itnes used in w particular study are recoded from what were originally open-ended itnls. As described earlier, a cumulative index of the Job variables, equally weighted, was constructed in order to investigate the correlation of organisational membership and the entire set of Job interaction variables. The level of industrialisation in the nation, as well as the char- acteristics of each of the four cities and plants, is important for this study in that the level of industrialisation and urbamsation shape the predominance of voluntary associations within the nation and the extent to which such organisational participation is an integral part of the social context within which an individual lives. In addition, the industrial level of the nation shapes the extent to which primary forms of interaction are likely to predominate in the lives of individuals. We would, for example, upset that the family and kin networks would be more predominant in the traditionally based nation, India, than in 37 the more highly industrialised nation, the United States. On the other hand, we would expect both opportunity for an actual membership rates in voluntary organisations to be the highest in the United States. The plat itself, and the manufacturing system within, are also important in helping to understand the findings of this study. To a large extent, the structure of the Job situation in each of the four is similar since all four are autasobile manufacturing plants. Tech- nological differences, however, exist between the plants. The particular technoloa leads to varying interaction experiences of the worker while on the Job. Such variance in interaction among the four plants, as well as that within; each of the plants, provide the indqiendsnt variables of this part of the analysis. hrlier analyses of the data conducted by Fern have explored quest- ions about the effect of these technological factors on interaction experiences on the Job. He has two general findings that provide the setting for this study. First, he found (1971) that neither quantity nor quality of interaction on the job varied significantly with a mother of factors where the opposite is commonly assumed to be the case. That is, occupation of the father, rural or urban background of the worker, ccuamity of socialisation or residence, past occupation exper- ience, age, or caste (in India) are of secondary importance when com- pared to the inset of the plant technolog. A second finding by Term (1972) is that technological factors do have an important shaping effect on social interaction at work. The two conditions of the technology which had the greatest impact on inter- action opportunity were the man-machine relationship and the population density within the plant. In analysing the effect of these dimensions of the work situation, Porn found that interaction rates were hidier when 38 the worker had control over the machine and when the density of workers was high. mess findings about interaction while on the Job provide helpful background to our study, where we ask what effect these experiences have on further interaction off the job. Most importantly, they sub- etantiate the claim that our study is in fact examining an “on the Job. phenomenon not interaction which is a reflection of sass other, personal trait. As with all cross-cultural studies, questions arise concerning the caparability of the data from each of the four nations. The difficulty is in the need to attain a balance between comarability via identical questiomaire itus and comparability via meaning or intuit of the questim. In this study, essentially the same instrument was used for each nation and we have seemed that the responses are indeed compar- able. Suoh an assumption is, of course, a questionable one and inter- pretation of the data met be undertaken with the recognition that the items may lack cmceptual equivalence. In a few instances where identi- cal questions have not been asked in each of the four nations, we hsve had to settle for comparable information from alternative questions. In an effort to avoid difficulties of interpretation, we have utilised variables that are behavioral rather than depending on labels and cats- gories applied in our own understanding of industrial positions. Because this study is based on an already existent set of data, gathered with a slightly different research question in mind, some in- fornation is lacking. There are, for example, poor data on the types of voluntary organisations to which the workers belong and what is avail- able does not lend itself well to canparability between the nations. 39 There are also a number of background characteristics of the workers that would have helped in our analysis, particularly in the area of type and extent of family interaction. The study would be stronger with better differentiation between opportunity for and realised inter- action on the Job. A further difficulty in interpreting the data has arisen from the inadequacies of generalising from one plant to an entire nation. To be most accurate, we must bear in mind that our sample is from auto- mobile plants, not nations, and that the plants thmasslves are not necessarily representative of other plants in the nstion. In addition, each plant is in a large urban area, a setting which increases the inset of industry in the modernisation process. He camot, with these data, isolate the impact of the industrialisation process on the in- dividual from that of the urbanisation process. (He can, and do, control for the worker's urban background but this, of course, is only a partial solution.) “horses we will caspare findings for each of the four plants in nations, and test the hypothesis that there will be no difference between them, we must do so with these cautions about the level of analysis and independence of variables in mind. In this light, it can be seen that the stronger focus of the study is that which ex- mines the relationships between Job experiences , primary interaction, and voluntary association membership within each of the nations. In the next chapter, we examine the distribution of these variables for each of the four samples. CHAPTER 'fiIREE DISTRIBUTION OF TEE VARIABLES In this chapter we will sxauine each variable separately to deter- mine its distribution within the sample. Looking first at membership in voluntary organisations, we see that membership among our respondents quite closely parallels the rates found in other United States studies. As noted in Chapter One, Hausknecht (1962) found his working class sample to have a low rate of membership in voluntary associations - his blue-collar occupations of skilled, semi-skilled , and non-fare labor show that only twenty-five percent of the sample belong to an association. Similarly, Hanan and wright (1971) found that forty-one percent of the labor class (including skilled, semi-sldlled, and unskil- led) who earn above 337000 belong to an organisation and twenty-eight percent the earn less than this figure belong. Bennett Berger (1960) found that thirty percent of his working class sample belong to an organisation. Almond and Verbs (1963) have investigated membership rates across several nations and found the following rates: United States 57% Great Britain 147% German Md It“! 29% Hence 25% (pass 2117) They have concluded from their data that voluntary organisation member- ho pose that added If U: mic “hip from Table 1 big. I!“ . 111 ship is more prevalent in highly industrialised nations than it is in the less industrialised. In a more detailed analysis of Almond and Vsrba's data, along with the inclusion of infomtion on Canada, James Curtis (1971) has found the same relationship between membership rates and industrial level of the nation. He has concluded that the United States and Canada, when compared to other nations, can most certainly be given the label "nation of joiners.” In these nations, membership (non-union) was about fifty percent. It should be noted, for the pur- pose of comparing these results with the membership rates of our sample, that neither the comparative study of Almond and Verbs, nor the data added by Curtis, report the data specifically for blue-collar workers. If the occupational differences were the same in less industrialised nations as they are in the United States, we would expect the member- ship rates of our sample to be lower than those reported above. The following table presents the membership rates for our sample from each of the four nations: Table 11. Organisational Hasbership Rates in Each of the Four Nations 093 mu m m. J t J J; x r 1 1 Bo organisation- ’ al worship 5b 165 67 205 6b 200 66 172 Belongs to one 146 1111 33 101 36 115 3h 90 or more organisations Total 100 306 100 306 100 315 100 262 Bra these figures, we can see that although the 01.03 sample shows 0 higher melabsrship rate than the previously cited studies, we have less than fifty percent membership in each instance. The cross-cultural fife: States distri the i less rates one, hove! alone the I ‘ V We 142 differences hold for our data only when we compare rates at the United States plant (01.08) to the other three nations. For each nation, the distribution is such that we can compare Job experiences for two sise- able groups, those that do and do not report organisational manbership. Distribution of Job Mdences The issue of Job experiences of the blue-collar worker has been less researched than the question of their organisational membership rates. Martin Heissner, in one of the few studies which parallels this one, uses Job experiences as a major set of variables. Bis variables, however, are not identical and his sample is not of blue-collar workers alone so direct oomarisons cannot be made. The following table shows the distribution in our sample of the five Job variables we have used, for each of the four nations: Table 5. Distribution of Job Variables for Each Nation OLDS FIAT m PAL _::_7 _- ._S r as f as r as 1 Physical Mobility 0-5 yds. 2h 73 311 105 M: 137 12 30 6 or more 6 22 66 l 6 1 8 88 2 160 % 156 fig 00 3 2 Required Interaction 0-1 people 60 182 S2 15 8 NOT 26 67 2 or more fig 12% 8 8 AVAILABLE l 100 00 2 lumber Can Talk To 0-2 couple 31 9h 33 100 22 66 h? 123 3 or more 212 6% 202 8 228 1 00 2 00 9 2 2 Table 5‘! mar3§ rela lent E201 PAL ass: fer of get int 13 ‘14 6" 9 9 81‘ h3 Table 5 (cont'd.) Discuss Personal Problems No 177 35 110 119 Yes 61 18% 112 12; 65 20515;); :62 As these data indicate, a majority of the workers in each nation can move six or more yards while on the job, although for the IKA sample relatively fewer can. For the PAL sample, almost all report such move- ment, indicating the requirements of the crude assembly line there. Required interaction with two or more people does not describe the work uporiulcss of the majority in either the OLDS or FIAT samples. Among PAL workers, however, where the technolog is less advanced and the assembly line less stringently regulated, seventy-four percent state that their job requires interaction with two or more others. he dif- ferences in distribution of these two variables correspond to the level of industrialisation of the nation and do not fit the expectations sug- gested either by Inkeles or Moore. Instead of finding the required interaction and movement to be regulated by the presence of industri- alisation, m _s_e_, we assume it reflects the level of the technolog within the plant. PAL is again the exception to the general trend when we look at the number of others the worker can talk to tile working; whereas in the three most industrialised settings almost seventy percent can talk to three or more others, only fifty-three percent of the PAL workers report such an interaction opportunity. but is, although their work situation has more required interaction than the other nations, it is interaction with a narrower range of people . M4 The table also indicates that PAL workers are more likely than the others to report at least two good friends at work. All four nations are about the same in the percentage of workers who can discuss personal problems with fellow workers. (In workers are somewhat less likely to have such discussions than those in the other plants.) 1h may, we can see that the three most highly industrialised nations are quite similar in tons of the Job experiences of the auto- mobile workers. me proportion able to move around and engage in required or optional interaction, as well as the percentage reporting good friends at work, is roughly the sane. We see, however, that the experiences of the PAL workers do not follow the cue patterno-they move around more, must talk to others more, but can talk to fewer others. ihey report nore good friends on the Job. Personal Interaction Variables Most studies of the social interaction of blue-collar workers have characterised the: as heavily involved in personal or primary relationships. run-q Rausmoht (1961s) has suggested that the central focus of the working class is the hone and the family, and that the strongest and the longest lasting relationships are with one's kin. Interaction with others tends to be modeled after the family—primary and personal, with little co-itment to secondary relationships. Axelrod's (1956) study in Detroit reflects the high rate of such interaction. He finds that relatives are the source of most frequent association; friends and neighbors follow, and co-workers have least interaction. Social status for Axelrod's analysis is composed of education, income, occupation, and ethnic background. hS Table 6. Interaction Rates in Axelrod's Detroit Area Study, in Percentages (Axelrod, 1956) Frequency of Go- How Wm Workers Frequencies 1 (low) 60% 3h 37 18 10b 2 614 So he 19 108 3 62 h? 37 20 97 h 7h 55 28 22 111 S 65 Sb 3? 23 87 6 (high) 58 62 hl 28 106 Dotson's (1951) study is also supportive of the importance of in- formal kin ties for the blue collar worker. He found that about two- fifths of all respondents named no intimate friends outside of kin groups-but noat had active social lives within this area. For thirty percent of the fifty families he studied, leisure activities were canpletely dauinated by kin mile for another fifty-six percent, kin represented the maJor, but not the only, source of social activity. Cohen and Hodges (1963) have found forty-one percent of their upper- lower class and fifty-nine percent of their lower-lower class group claim most frequent contacts to be with relatives. ‘lhese are substan- tially higher than either of upper classes studied. The data from Cohen and Hodges suggests that non-work friends are the second most frequent group cited as the source of infernal contact. Axelrod's data supports this for all but the lowest status. Herbert Gans, in the Urban Villagers (1962), discusses the importance of kin and friendship groups for his working-class sample. This “peer group society" was composed of same age and sea: kin and a few friends. Infoml interaction with this group was found to be frequent. Ben- 1:6 nett Berger (1960) has added that such interaction is usually spon- taneous, unplanned, and highly informal. whereas informal interaction with relatives is quite frequent, and with friends and neighbors moderate, we find that interaction with fel- low-workers is very low. Cohen and Hodges found that only twelve and two percent of their upper-lower and lower-lower classes, respectively, cite friends fron.work as individuals they have over to their house. Axelrod has found eighteen, nineteen and twenty percent for his three i- lowest statuses and Alan.Blun (l96h) cites data which claim_ that twenty- two percent identify friends from work. Table 7 shows the percentage distribution from our sample on these a variables for each nation: Table 7. Distribution of Primary Interaction Variables for Each Hation oms mu m m. J t 1 f j t S 1 Family and Kin :6 b5 :29 h3 :32 5g 16g 62 1g? 08 1:005; 30"'6 loo"2 "' 30:6 100 31! § ‘2'100' —"6L2 2 neighbors "° $3 2’; 2% 1272 ii :32 2121 1‘33 les 2 150 293 30 100 30 00 3 Friends No St 153 12 218 ho 125 7b 195 lbs 6 l 6 g 39 28 85 60 18% 2 61 3 303 00 3 00 2 2 Pellow'wbrkers No 26 79 148 1148 29 92 22 32 81: lbs :9 225 52 158 I; 3 68 1&8 100 30 100 30 100 31 100 2 This table shows that in.no nation is the highest rate of interact- ion reported to be with family and kin. Instead, for three of the four h? nations (.11 but In), the highest rate is with neighbors. Although this unexpected distribution does not interfere with our initial re- search question—the relationship between primary and secondary inter- action-it does contradict what other research has found. we explan- ation for the finding may be in the item used for gaining the data on family interaction. It will be recalled that the respondent was asked to indicate with whom he spent week-ends and holidays. It may be that the respondent answered the question from the perspective of his family unit rather than of himself as the unit, or he may have assumed family and responded in regard to tho else he spent such time with. In addition, there is a discrepancy in that interaction which previous research has found to be the least prevalent, that with fellow workers off the Job. In workers report higher interaction rates with fellow workers than with an other category. For both OLDS and PAL, interaction with fellow workers is second highest, after that with neighbors. Again, such findings do not interfere with our research question about patterns of interaction, but they do differ from the expected distribution on the variables. For each plant except IKA, interaction with friends was reported by the smallest proportion of workers. Distribution on The Control Variables: Age, Education, Still Level, Occupation and Urban Bum Table 8 presents the percentage distributions for the five control variables we will use. Three variables-«gs, education, and urban back- ground—provide general information about the automobile worker. 'hvo other variables, skill level and occupational task within the plant, are directly tied to the work place. We would expect, of course, that ha these two job classifications would explain a major preportion of the individual's experience while at work. 'we have included them in.an effort to give substance to these descriptive categories. That is, we will ascertain whether job experience has an effect on organisation- a1 membership above and beyond the influence of skill level or the specific occupational task. Table 8. ‘Distribution of Control Variables for Each Nation onus r112 1x1 PAL ii I ,x r 4% r .5 f Educatione Low 39 119 S9 180 ho 126 21 Sh Medium - 26 78 S 109 3? 98 Age 30 yrs or 18 5h 39 119 188 h5 119 less 0'0! 30 lg: 301 13% Egg. 100 1 g 100 2%2 Residential Background - 3 w 1————zss :2 Boa—hi 6 Ho Ho: 2° 60 2 31 151 53 10h 81 Rural, 68 202 68 20h h8 Small City 6 6 Metropolis 22 2 _22 49 52 91 22 100 29 100 300 100 3 100 197 Skill Level unskilled 2h 72 31 95 h2 13b 36 95 Semi- skilled L9 151 5% 156 35 111 39 122 Skilled 21 2 1 55 22 [g 25 5 100 30 100 30 99 31 100 2 2 Occupational Task Assembly 23 71 30 92 30 93 3h 88 ‘Hhchine Operator 2b 7b 31 9h 30 96 28 7h Test, Insp- action, Repair 21 65 19 59 21 6s 8 22 Skilled 51 26 20 61 12 61 22 £8 30 100 556 100 31 100 2 h9 Table 8 (cont'd.) *mucation, by years, is as follows for each of the four samples: OLDS low 0 - 9 years high 10 years and over FIAT low 0 - 5 years middle 6 - 8 years high 9 - 13 years & low 0 - 6 years middle 7 - 9 Years high 10 years and over LA; low 0 - 11 years middle 5 - 7 years high 8 years and over In addition to the earlier descriptions of the autonobile plants and the nations sampled, we can now describe the sample in terms of social characteristics of the worker and his experiences on and off the job. The workers of IRA and PAL have higher proportions of workers under thirty and tend to be smewhat better educated than those in the other two samples. (The exception here is the educational level of our: workers where, because of different categorisations, comparison is difficult.) The FIAT and OLDS plants have a somewhat greater propor- tion of workers who do not have urban backgrounds, whereas for IRA and PAL the proportion of workers with metropolitan backgrounds is nearly equal to those whose commnity of socialization was a small town or rural area. The sampling design for each plant was such that each skill level was adequately represented. For OLDS and In this required additional sampling from the skilled workers so this category is over-represented in preportion to the whole but now contains sufficient respondents to allow for analysis. This stratified sampling technique also results in 50 a larger proportion of those whose occupational task is a skilled or craft function than would be the case with a simple random sample. CHAPTER FOUR RELATIONSHIP BETWEEN ORGANIZATIONAL WERSHIP AND SOCIAL INTERACTION EXPERIENCE ON THE JOB Martin Meissner (1971) and others have raised questions about the relationship of job experiences and off-the-job social interaction. Heissner'a research, with a ample of 206 industrial workers from a wood-products manufacturing conpaw, has upheld the expectation that those whose job experiences include opportunity for social interaction and discretionary decision-making are more likely to belong to volun- tary organisations in the community. his finding thus partially sup- ports other claims of Allardt (1957), Tools (1962), and Axelrod (1956) in saying that social interaction is of a cumulative nature. Re have, like Heisaner, hypothesised a carry-over of job skills and are, in part, testing Heissner's claim for a particular segment of workerso-blue-collar workers in the autauobile industry. We have seen that this segment of the population has a low rate of membership in voluntary organisations and theorists have pointed to limited interaction experience as the explanatory factor. We will investigate whether such suggestions, when applied to the work situation, are sound and able to explain melabership rates in voluntary organisations. In detenining whether such a relationship exists, we will look first at each separate job variable as it relates to organisational mad- bsrship. An index .2 the combined effect 1.111 than be discussed. In each case, we will be asking whether the data show that, as predicted, 51 52 there is a positive association between job experience and organise- tional mubership so that we can conclude that, in fact, the experi- ences in the foraer lead to the latter. Tables 9 through 13 show the mulbership rates for each category of each job variable, for the four nations. They also show the game value for each relationship. 1. Mical mobility Responses to the question "How far can you move freely while on the job' proved to be the best single predictor of membership in vol- untary organisations, although in no case does the value of Q indicate a very high correlation. The percent of workers who belong to organi- sations according to spatial mobility and the resultant values of Q are shown in table 9: Table 9. Physical Mobility While on The Job by Percent “10 Belong to Voluntary Organisations ems mm m m. 1 I J 1 1 r j r O - 5 FIN! 33 73* 27 105 30 137 27 30 More than 5 51 227 36 199 112 178 35 230 Q II .35 .22 .25 .20 a For each of the following tables, the amber of respon- dents given indicates the row total. That is, for example, 73 OLDS workers could move 0-5 yards. 3331 of them belonged to an organisation. We have nude several suggestions as to wtv the ability to move free- 1y while on the job correlates with manbership. In addition to exposure to a comparatively wide range of individuals, it perhaps indicates the degree of decision-making or discretion required of the worker. The 53 individual who can move around while working is exposed to a broad range of potential relationships, has the Opportunity to interact with others, and must decide whether or not to take such an opportun- ity. He is, than, broadening his exposure and gaining experience in making decisions and in initiating interactions. This means that those workers who can move around have, at least in part, overcome a limitation often attributed to the blue-collar workers and often cited as the type of deficiency which limits membership in organisations. We can see from the data that these expectations of the link between ability to move around while on the job and membership in an organisation gain only weak support, although the relationship is in the expected direction. 2. Mud Interaction mile On The Job For the blue-collar workers, required interaction offers experi- ence in secondary relationships, a type of interaction which also characterises ccuminity organisations but which differs from the person- a1 experiences of family and neighborhood. The worker gains experience, we reasoned, in relating to bureaucracies and large structures as he enters into interaction which is required by the job. Several attempts to explain why the membership rate in voluntary organisations is not as high for blue as white collar workers have focused on the blue-collar workers. lack of these kinds of skills. It has been suggested that the blue-collar worker is relatively "person centered" so that interaction with others in bureaucratised or fennel settings is difficult. The worker is said to relate to personalities and individuals rather than to roles and organisational structures. (Miller and Riessman, 1961; Hausknecht, 1961;; Gordon and Anderson, 1961;). (O- P"! 5’: In analysing the actual findings in our data, however, we find that the predicted relationship between required interaction and organisation- al membership existed only for OLDS, where the association is a weak one. Table 10. Required Interaction mile on the Job by Percent Who Belong to Voluntary Organisations ‘ OLDS f x FIAT 1 J m f 1 PAL f 0-1 people 113 182 35 158 NOT 37 67 2 or more 50 12’; 31 1118 AVAILABLE 33 195 Q " .13 -.08 - -e09 We can suggest several explanations for these unexpected findings. First, we should bear in mind that the expected positive relationship between mmabership and required interaction we; found mong OLDS workers-oso that the explanation might be found by closer analysis of the cultural differences or levels of industrialisation. Further data, not available in this study, are necessary in order to test this explan- ation. Secondly, the expectation might have been upheld had we been “,1. to control for the authority relationship of those required to interact. Those in control of interaction are likely to have different experi- ences with secondary relationships than those who are, perhaps, passive- 1y receiving orders. The third explanation is found in the body of knowledge about the types of organisations most common to blue-collar workers. That is, their preference for expressive over instrumental organisations (Babchuk and Edwards, 1965) may make secondary relation- ships and required interaction a relatively unimportant factor in the decision to join an organisation. 55 3. lhiormal Interaction‘While on the Jab Informal interaction on the jdb also leads to development of skills in interacting with those outside the family and neighborhood. Some sociologists have speculated that the blue-collar worker is suspicious and uncomfortable in the “alien! world outside his neighborhood, that he is lacking the ability to cope well with the broader environment. (Cohen and Hodges, 1963; Hauskneoht, 1961;; Miller and Riessman, 1961). Alan Blum (l96h) points out that the social networks of blue-collar workers are typically small and homogeneous, leading to a fear and dis- trust of strangers and an avoidance of the heterogeneous social contacts typically found in voluntary associations. The chance to interact on a personal level with fellow workers provides Opportunity to gain contacts outside the small circle of family and neighborhood and to increase one's interaction skills with a range of individuals. Such a skill, again, is also important in voluntary organisations and can, once gained, be carried-over to that arena. The relationships were generally weak but in the predicted direct- ion for OLDS and FIAT. In IRA and PAL there was essentially no rela- tionship. The Opportunity for informal interaction on the jdb was a particularly strong correlate of organisational membership for the ODDS workers. 56 Table 11. Number Can Talk To While on the Job by Percent Who Belong to Voluntary Organisations onus FIAT 1x1 PAL 1L_ t 3% f _x I .36 1' O - 2 people 36 9h 29 110 36 66 33 123 3 or'more So 212 3b 202 37 228 35 139 Q - .28 .12 .01 .oh The number of others that the worker can talk to while on the job is a such better predictor of organisational mulbership than the rate of required interaction, indicating that informal interactions in these "outside worlds" is a more important job experience than the required interactions. The importance of the more personal job interactions is supported by the findings for the final two variables. h. Number of Good Friends, and 5. Ability to Discuss Personal Problems The question concerning good friends on the job adds the dimension of quality or depth to the number of relationships with other workers. The ability to discuss personal problems with other workers indicates the worker's degree of personal involvement in relationships with work mates. Alan Blue (1961;) has suggested that the working class appears to maintain old friendships and to be resistant to new ones. In addi- tion, he says, blue-collar workers are not likely to derive personal friendships fraa their work situations. The worker creates an ”inside“ world which is closed to and protected from the 'outside' world, which includes work. We are investigating, with these two variables, the expansion of 57 personal relations beyond this inside world. Not only do mmbers of contacts offer new experiences for the blue-collar worker, but pri- mary relations or quality of interaction affect his set of social skills as well. Once the individual has expanded his primary relations beyond his protected world, he has gained an interaction skill which is appli- cable both on and off the job. The relationships of each of these variables with organisational mabership were in the expected direction with the exception of IKA where the mmber of good friends showed a negligible positive relation- ship and the ability to discuss personal problems a slight negative relationship with organisational membership. In no case, however, did the Q values for the other three samples indicate a very strong rela- tionship. Table 12. Number of Good Friends on the Job by Percent Who Belong to Voluntary Organisations Jams! 1mm: x1vaf j“’11.! 0 - 1 1.3 176 29 11.0 35 17b 29 68 2 or more 50 130 36 166 38 11:1 36 191; 0 - .11. .15 .07 .15 Table 13. Ability to Discuss Personal Problems With Fellow Workers by Percent Who Belong to Voluntary Organisations 01.03 rm IKA m. X f x t 31 f g 1 Cannot Discuss Personal Problmas ho 119 27 177 38 110 1) 9!; Can Discuss Personal Problems 50 187 12 129 36 205 37 168 Q . s19 e33 -e06 e16 58 6. The Index An additive index, combining four measures of experience on the job, was constructed according to the suggestions of Johann Geltung (1969). Required interaction while on the job was omitted because of the lack of its consistency with other job variables, suggesting that it is another dimension of job experiences than that upon which we have focused. In this index each variable was given equal weight and the respondent was assigned a sero or a one for each negative or positive response to the question of his interaction. The following tables show the correlation matrix of the index items, the distribution of the index scores, and the percent of each index category that belong to an organisation. A high score on the 0-}: scale indicates more of the job experiences that we expect to lead to organisational melnbership. Table 1):. Correlation Matrix of The Job Interaction Index Itsns to Organisational Membership: The Values of Q 01138 1 2 2 y FIAT l 2 2 Q 1. Pkwsical Mobility .75 .22 .31: .82 .27 .18 2. Can Talk to .h3 .08 .26 .29 3. Good Friends .0? .63 1:. Personal Problems - - £10112}; Pllegy l. Hwysical Mobility .149 .01 .23 .11: .19 .01 2. Can Talk To .03 .01: -.0h .10 3. Good Friends .29 .38 h. Personal problems - - 59 Table 15. Distribution of Job Interaction Index for Each Nation 01.08 I gun I gm 1 gm f 0 (low) 6 17 11 3h 1: 13 1 2 1 11: 1:3 15 ’45 15 M 8 21 2 26 80 30 92 31; 108 29 7S 3 36 1.11 2h 73 31 99 37 96 1. (him) 18 55 20 62 15 98 26 68 100 306 100 306 99 315 101 262 Table 16. Job Interaction Index by Mambo rship in Voluntary Organisations for Each Ration IndexScore: O 1 2 3 b Total 01.03 115111515151 hamster-7112 70 30 5h ’43 5h 60 36 20 51: 165 P W’M’HHHfiHh-tt Gasma- FIAT Noahilmber- 79 27 67 30 72 66 67 1:9 53 33 67 205 P Hubership 21 E g; E 28 26 2; 2g 31 22 22 101 3 56 92100 3100 210030 Games- 31' IRA Nommaber-69 96113066716665522563200 Hubership 1331 13: 3(6) 1; 0033 [35% 1005).: 92% 10098 25 3% 13.15 Nomaberhm02661h7153666359h066172. (la-aa- :15 60 The correlations between the index of job experiences and organi- sational membership are positive for each of the four samples but in no case are they substantial associations. For each nation there are several separate job variables that have higher correlations with mem- bership than the index does. In each nation, physical mobility while on the job is a better predictor of organisational membership than is the index. Thus, although the relationships are in the expected direc- tion, they are not sufficiently strong to explain the membership rate of blue-collar workers. The lack of a stronger relationship between index score and membership is explained by the correlations between the index items which are , in several instances, quite low. The job variables we have identified then, are not reinforcing each other in offering the worker experience in interaction which is enabling in organisational membership. Conclusions and Semen of The Zero Order Relationshig Although experiences on the job are positively linked to organisa- tional membership, they are not a major factor in explaining membership of blue-collar automcbile workers. Even when the correlations have been in the expected direction, the associations have been weak ones. From this we see that theories which suggest that blue-collar workers fail to join voluntary organisations because of a lack of interaction experiences beyond the confines of family and neighborhood are not sup- ported when investigation is focused on the job as an arena for gaining such experiences. That is, workers who have such interaction experi- ences on the job show only a slightly higher rate of organisational membership than workers who lack such experiences. In conclusion, the 61 predictions stated in Hypothesis I are, at best, only weakly supported. Cmpared to the other nations, the poorest fit to our expecta- tions was in the IRA ssuple. Of the four measures available for m, three were negligible associations (Q - .01, .07, .06). Only one variable, physical mobility, showed a notable positive association (Q = .25). The best fit with our predictions was found in the OLDS sample. _ Here, all five correlations were positive ones. For the three quanti- n tative measures of interaction, the OLDS associations were considerably higher tran those found in the other nations. This finding, of the stronger support for our hypotheses in the United States' plant, sug- L j gests that the theories which this study is designed to test may be - culturally-bound so that they explain only the membership rates of blue-collar workers in this society. We conclude, then, that there are different and more important shapera of omnity interaction in other nations than the ones theorists have identified for American workers. We might also suggest that voluntary organisations have varying mean- ing and operate in varying social contexts in each of the four nations. In particular, here, we refer to the greater prevalence of voluntary organisations in highly industrialised nations than in those nations with less industrialisation (Gist and Fava, 19614). ‘lhis latter sug- gestion gains some support from the fact that the FIAT sample had the second best fit with the expectations. PAL,-then ranked third in terms of "best fit". Except for one reversal, of IRA and PAL, the "goodness of fit' corresponded to the level of industrialisation of the nation. The theorists explanations of determinants of worker membership are most amropriate in a highly industrialised nation but even there they are not adequate when applied to social interaction at work. 62 Oontro For Social Characteristics of The Worker In this section, we will investigate the effect of several social characteristics on the relationships we have hypothesised. Our analysis will be directed toward determining whether the relationship between Job experiences and organisational membership is stronger for some sub- groups of each nation's ample than it is for the sample as a whole. As we have seen, the theories which explain low membership rates of blue-collar workers by pointing to lack of interaction experience beyond the confines of the neighborhood have not been strongly supported by our data. Only in the 01138 sample is there any omsistent positive relationship between Job experiences and organisational membership, and here the associations are weak. We will analyse the effect of the con- trol variables, to determine whether the Job to membership relationship is strengthened by age, education, couunity of residence, skill level or occupation of the worker. An overview of the effects of the control variables on the relation- ship between interaction experience and mabership shows that the general picture does not change when controls are introduced. Although there are exceptions, which we will discuss, the control variables do not alter the maJor trend of only weak associations in the expected direct- ion. thm there are major alterations as a result of the controls, they do not follow an perceptible pattern. Tables 11 through 21 give the gal-a value of the Job to membership relationship for each control 's categories. Io facilitate comparison, the sero-order correlations are given again. 63 ensue H4. nod 4.3.3.: 6% Canaan-385. 386363... How. Meow Boo-$055. Sew. Hoe meow 2.3.2: are «specs on 0 08m :5. kin». 2H. gaasnoiwefiow we. amp-m 5. Eu BB 5. :2. PE. 5. a. E gape-H moon—“3Q .nos ... .5 .3 .2. ...8 .rm .3 33 LE .8. .8. 53.... $3 $5 58 when! Her? ‘ IO¢H°5 no e0“ . e F e0“ '0 W: 'e ”H 2°." "flHHOUHO e EH ..P -e NH. ALUV A303 .. T63 2583. can eat“ 30 eHN . eUO IeOU saw eeeum sou. sou eeeow our e0“ IeOfl A.~mv A.H»V A.owv A.orv 2553. soon 3.8- eor U e8 eUH .00“ so: eHH cocoa eUN 'emfl ex e8 ALE “Luv To: ALB geese. wee-cob. EUHQO eNU ..e eHN or“ eUO e0“ e8 e8 seeHo emu- -eHm eHU Ce 8: Toe 33 * ..on goes chum son—3e. 555 Hot once-goo. use: no a well 4330 on .3 canteen gene-H 32.53% on are ...oe 8.x. oeueflaeaposaw 33333? do». are 3.3 ease-«on. nee!- noose-e: ac .5.- a... are a «nude no». «we «3.9.62.3. soap-«Hos. cease-B god damned: eon ego-«wasp anodes-Ev. ease Hm. nod damp-owes on own-333.25» 33335.1 new use: the 9.96. no». 9-3. :3.on are 4-wee- ou o Ilse. ulna E. E 953. 013. does». 93. come». 93. 953. 043. E 8 IFIIFIMPILIE ECHOUH. 30.0“.qu cub. cg emu- .eCN e; 05 eg e8 53.. 38 ray is manage Haven-canon .mu .3 ....un .Hm zoe been-ewe 3.5 ...o» , 33 Toe .. “:03 255.50” a.” 8.; 80 em“ eNM at” seeHU eseHU eNN eNW -eHm A.~mv A.H~V “.0HV A.orv 2.533. coca @339 Lb LE .3 .m» ...om .3. .8. .3 Abs ALB Ab: “Luv Eocene 33.32:. 3639:. ...fi. .3 .5 .5 0.3. 1.5 Lb .5 53 33 T53 35 s go o :55 new are 33.633. 33.3.3905 seduces god 4.3.36 33 03-32328“? assessor»? 6S 8.3.. 5. gov 33.3... 3. 03.33385 3.6.331 no“. any 385.535 3.9.63.5... as. was 3.32: an. 45.... on o Oahu :5. E Mb“... .8223»... wan-pull. afield... as...“ in... 3?.me 9.3mm who... wrong 3.»... «3. spam 3.36 9.3L.“ 3km... 3.3305 303.53 .um .w» .3 .mn .3 L.» .0» .8 83.. $3 :5 33 3365.3 Hanogagob .8 ...or ...o.~ ....om 30». 3.5.3.. .33 .2. 3.5V 3:03 .. T63 25...... 95 5E. ea .8 .3 L» .5 .2. ...o» ...o» .8 33 AL»; 3.8V 3.05 95d... moon $4.25. .0m .3 .3. .Ho 0.0» .Hm Lu .3 :5 3.5 Ab: :3 3.98. wogogw gng. our ..8 .NW .5 -.HW 00¢ GNU .8 33 Ca Tog 33 t as. o 4.5. HS. 2.. 33.63... snaffle: Gong no: 4.3.3. .3 03.33303“. 385.35.? .33.. no. now 4.3.3... UH 03.3.3993: 3oavogv no". 333 9...»: Saw. no». a...» 3.3.2: as. «5.... on o 08m :5. Ha. «mum. can mean... an: «In... an: E... can ... ME. .33. B B I an ax aw an ax a. mu an an an 93:2. 38.5% .nm .5 .5 .8 ...S .5 .3 ....» ...C .3 .3 .fi 53.. 75 $3 3.8V .33.... F323: E. ...op ...8 ...5 ...om .5 :3 ...-ES. .3 a...“ ...mo ”g AOHUV A‘oomv - A-.°WV amcwa mo Aoo=¢.a.v zuauou on: euww e0 .Hm 2.5g“. moon Eng. .5 owmasun madman-H waovwmau .Hr .uo A.~mv .5 SS .om 2.3 -oow .8. Lb .mw .mm .mm 3.5 “LS 3.00 Twmv .8 $3 .5 ...5 lo» .um “.05 om: -.OH on 30HH Ab: oUO 30H? IoNO 0N“ filooav ...p» A65 .3 ALB .Hu A.Hov .0m .3. .om * era 0 «spam won five uoaouonaou oonHoH-auon coetaou new «nan-cw. can own-nununponnw anadaduruu agape my. new «nap-duo. dw.ounonuu-«poanw xwauonuwpv non mint oaaausawonuw anew macaw. ocean . 9.95. How Mao: z.¢»ou. arc 00 . z. 3 . .7.le m” bank 3. OD .7.le mfl .* 33:: xofitfl .on .5 .8 .5 53 .3 E. ...mm .mm mancunoa Hanna-own H05 '08 0W“ 08 .000 AQHUV.OS CNN .08 -OPU 2.5.9: 3.x ea .8 .8 .8 ...om A.~3 .mo ...5 ...5 .mm 2‘56 000“ ‘qHOHQ. GNP IoOfl QHH 0N0 ACHtv oNU 0&0 -QUU or“ you-g. -00“ are 0°“ '00” AOHWV ouw 0”” CW” 0”“ A.-v 7.03 Abs :3 A.wuv audvo NH Acona.a.v 03am». 9.05m. boo. maquponw :ddpwunw .ru mmpppuoa H5¢aunownoa 253. can eowx ac I.~o 2:3. 003 3433.. .Ho uwuo. mafia. _" . ...- ... :w E E plixbo. auwnw mm 1 tafl. 3% a)...» mm ‘ .8 0H? .8 A0va .oow ado H.000 IoHM AoNOV SH. bdbgg I. our low.” oHH 'OUU AIOOWV .om .om .rm A.ouv .~m -.Hu .mo .00 A.orv 0H.” -00? oHU AoONV 3000 0H” 0?“ OHW AOHWV glow 1.9— .Hm Albav .Nw .wo .HH Iowa ALbV wdouw¢3n .or 67 * are a <~H¢o Hon amauouuvuv «um nououonaou ooanmwunuou. do¢¢¢o= gov «nannvwa can oumnuuunaponnw 68 To determine the effect of the controls on the hypothesised rela- tionship, we have looked for instances were the controls have an ef- fect on several Job experiences, rather than on isolated interaction experiences. To attain this overall picture, we first investigated which of the controls led to a substantial increase in the value of Q for at least three of the five relationships between interaction experience and membership. We looked at increases in Q of .20 or nore, then at increases of .10 or more, over the original, sero-order correlation. In each instance, because our question is whether the twpothesised relationship is true for an sub-groups of workers, we looked only at igreases in the value of Q, an increase being an indi- cation that the predicted relationship is nore true for that sub-group than for the uncle. For both the 01.06 and the In samples none of the control categor- ies made an increase of .20 or of .10 for three of the five Job experi- ences. That is, although there were scattered and isolated instances where the controls had a strugthening effect, they indicated no sub- group of the smple for which the originally hypothesised relation- ships were consistently stronger. For these two nations, then, the theories explaining low organisational membership do not hold for the sanple as a whole nor for an sub-group of that sample. The control variables made the biggest difference in the FIAT sample. Here, there are five sub-groups of the sample for which the relationship between interaction experience and membership is stronger than it is for the whole ample. For four of these five, the increase in the value of Q is .20 or more, indicating that when there is an increase in the correlation for a particular sub-group, it is a substan- tial one. 69 The four sub-groups of FIAT workers for vdzich the originally hypothesised relationships are considerably stronger (.20 or more) are those respondents under thirty, unskilled or skilled (i.e. at either extras of the skill continuum,” those belonging to an occu- pation classified as skilled. The fifth, which made an increase of .10 to .20 for three of the five Job experiences, was low education. The controls also had sans effect on the PAL sample, although not H as nany categories of controls made a difference as in the FIAT sample. Here, low education and the occupational group, test-inspectien-repair, both showed inproveuents in the value of Q of .20 or acre for three of t. J the five Job experiences. The sub-category of workers with an urban background showed a Q value which exceeded the sero-order correlation by .10 or more. In ccqaring these findings for the FIAT and PAL samples, we see that the only point of overlap is with those sub-groups of workers who have a low level of education. In both these plants the predicted relatiouhipa of Hypothesis I cone closer to being true for this small group than they are for that sample taken as a whole. Still, it is only for PAL that the correlations are quite strong. (The correlation of nubership with required interaction is .hl; with informal inter- action, .314; with ability to discuss personal problems, .81.) With this one exception, then, the controls do not have a consistent impact across the four plants. where the predictions of Hypothesis I come closer to being true with the use of controls their effect is unique to the plants, FIAT and PAL. 70 Conclusions About ”thesis I In this chapter we examined the data which tests Hypothesis I, the prediction that there is a positive relationship between five measures of social interaction on the Job and membership in a con-unity volun- tary organisation. The hypothesis was developed in response to a timber of theories which attempt to explain low membership rates of ._ blue-collar workers. These theories point out that blue-collar workers n limit their social interaction to family and neighbors and do not relate well to formal structures or secondary relationships. Low rates of organisational nubership is said to result from this lack of appro- : J priate interaction experience. 1h this chapter we have tested the assertion that when such experience i_s gained by the worker, on the Job, it will provide him with these necessary, and usually absent, skills which he will then utilise further by Joining an organisation. w. have concluded, after analysis of data from each of four nations, that these relationships are generally not found. That is, the predictions were only weakly supported in amr instance and the overall pattern was present only for the 01.08 sample. An index which combined four indicators of interaction on the Job had no more sub- stantial correlation with organisational membership than did the vari- ables taken separately. Further, when a umber of social character- istics of the workers were introduced as controls, the relationships were, in general, unchanged. These findings raise questions as to the soundness of the efforts to explain membership rates among blue-collar workers. If, in fact, inexperience in interaction beyond the confines of the family prohibits membership, its presence at work is not sufficient to overcome the 71 barrier. Those with the experience, measured in either quantity or quality, do not have substantially higher membership rates than those without such enerienoe. It is evident, then, that theories about blue-collar workers and their interaction patterns when taken in isola- tion, are inadequate to explain the low membership rates. Further research is needed in order to provide more complete explanations and to ascertain what other variables may, along with interaction skill, help to explain membership. In the next chapter, we investigate whether organisation membership can be explained by the pattern of primary interaction. The theories which explain low membership rates by point- ing to the individual's lack of experience in interaction other than that of a primary nature have not been supported by the data. we now determine whether the nature of the primary interaction itself offers a better explanation. CHAPTERFIVE RELATIONS'IIIa RENEE)! ORGANIZATIONAL MEMBERSID’ AND PRIMARY FORE 0F INTERACTION The first part of this study was designed to test the theories which explain low organisational membership rates among blue-collar workers by pointing to their lack of interaction beyond their own families and neighborhoods. As we have seen, when interaction of the worker does extend to the Job, the membership rates in cannunity organisations are still low. Those who have interaction experience on the Job are only slightly more likely to Join organisations in the com- munity. m a second part of the study, we now focus on the worker's interaction with family and neighbors and examine the relationship of such interaction to organisational membership. Job experiences do not explain membership; we now see if more primary forms of interaction do. In Chapter One we reviewed several ometing theories about the rela- tionship between primary and secondary interaction. 0n the one hand interaction is seen as compensatog in nature so that when interaction is unsatisfactory in one arena, the individual compensates by involve- ment in another. There is disagreement here as to which arena, primary or secondary, is the preferred alternative and, as a result, it is difficult to empirically determine which is compensating for the other. A second perspective suggests that interaction is patterned so as to be curmlative in nature. According to these theories, those who are active in one arena are also active in the other; the higher the rate 72 73 of primary interaction, for instance, the higher the rate of second- ary. A third alternative, as pointed out in Chapter One, is that there is no relationship between these two maJor categories of social interaction. In this chapter, we analyse our data to determine which of these theories best describes the interaction of the blue-collar workers in each of the four nations. We have selected four types of primary ‘“ interaction -- family and kin, neighbors, friends, and fellow workers n off the Job. Tables 22 through 25 show the relationship of each of these to organisational membership. Table 28 presents the correlation , matrix of the four types of primary interaction. .J 1. Family Interaction First responses to the question 'With when do you spend weekends and holidays?" were coded according to whether the fmnily or kin were mentioned. In three of the. four nations, all except OLDS, the cor- relations between interaction with the family, as indicated by this item , and organisational membership are negative. That is, for FIAT, IKA, and PAL, the substitutive or compensatory model of social inter- action is supported - the higher the rate of family interaction, the lower the rate of organisational membership. As can be seen from the table, this pattern is stronger for FIAT and IKA workers than for those in PAL. For OLDS, there is essentially no relationship (Q .. .05). 7h Table 22. Family Interaction by Percent Who Belong to Voluntary Organisations onus FIAT IKA PAL J _ f x f 1 t i r not Family" us 139*, 39 132 uz 169 36 179 Fiuily h? 167 29 17h 30 1h6 31 83 Q - .05 -.22 -.25 -.10 i For each of the following tables, the number of respon- dents given indicates the row total. That is, for example, 139 01.08 workers indicate no family interaction, 15% of them belong to an organisation. 2 . Neigborhood Interaction For information on neighborhood interaction an item was used which asked “Do you exchange visits with neighborhood friends?" Table 23 shows a positive relationship for each nation except FIAT, where there is no relationship. The association between neighborhood interaction and organisational membership is particularly strong among workers in PAL (Q - .hS). For onus, 1x1, and PAL we find support for a cumulative pattern of interaction between neighborhood and organisational member- ship. That is, those who exchange visits with neighbors are more likely to belong to organisations. Table 23. Neighborhood Interaction by Percent Who Belong to Voluntary Organisations onus FIAT IKA PAL 5 t _T:;71: ___f of I 2L_ r ___ not neighbors 35 3b 3b 7h 3h 178 20 us neighbors ha 262 33 156 to 126 to 168 Q . e25 -e01 .111 .115 75 3. Interaction with Friends 'me relationship of organisational membership and interaction with friends is the least consistent across the four nations. Here the question was ”Apart from relatives, did you spend vacation time with friends'."' As seen in Table 2b, FIAT and PAL show moderately high positive association between interaction with friends and organ- isational mubership. In shows essentially no relationship 3 01.08 a low negative one. Thus the cumulative theory is supported for the FIAT and PAL samples. 01.08 and IKA show no strong relationship between interaction with friends and organisational membership. Table 2b. Interaction With Friends by Percent Who Belong to Voluntary Organisations onus FIAT IKA PAL 111 f ._z 1 of r .5 1 Hot Friends so 153 29 218 36 125 30 195 Friends 1.5 130 m. 85 37 189 ha 67 Q . ’.08 e 31 e02 e 37 h. Interaction with Fellow Workers, Off the Job For information on this interaction the respondent was asked the question .'Do you meet fellow workers that are good friends outside work?" Analysis of the data show that, for each of the four nations, there is a positive correlation between interaction with fellow workers off the Job and organisational authorship. Here, the Tule's Q is the weakest for IRA and sanewhat equal in strength for the other three nations. In each nation, the cumlative theory best represents the relationship between the variables. That is, those who interact T"! 76 with fellow workers off the Job are also likely to belong to volun- tary orgamsations. A relationship such as this is an extension of the findings of Iipaet from the context of union organisations to organisations in the co—nity. Lipsst, in Union Democracy (1956), found that those indivi- duals the interested informally with fellow workers away from the Job and union activities, were more likely to be involved in union. (hr data show that this linkage applies to other voluntary organisations as well. Just as Lipset has found evidence of the broad effects of interaction with fellow workers, our study found the link between infor- mal interaction with work mates and membership in formal organisations to be a substantial one. Our data show that the effects are as far reaching as non-Job related organisations . Table 25. Interaction With Fellow Workers Off the Job by Percent Who Belong to Voluntary Organisations OLDS FIAT IKA PAL 1% _f s I 1 t x f not Fellow Workers 3b 79 26 1118 32 92 25 8h Fellow Workers 50 227 39 158 39 223 39 118 Q - .32 .29 .15 .31 Lipset's data raise a question as to whether the findings of a positive association between interaction with fellow workers off the Job and m-bership in an organisation might be partially explained by the fact that the organisation is one place where the individual gets together with his work mates. Lipset found that union activities did, indeed, provide such opportunity for workers, so that those with whom the worker interacted informally were also in his union organisations. 77 In examining the respmses to a separate questionnaire item in our study which asked the respondent if his fellow workers were in his organisa- tion, we find that an explanation such as this is most appropriate for the samples fraa the two most highly industrialised nations. Of those to belong to organisations, the following percentages of the respond- ants report that there are fellow workers in their organisations: ems 7M . ... p In 1181 PAL 38x We do not find, however, that the rates of seeing fellow workers off ; J the Job, irrespective of membership, vary in a similar fashion; they are as high for the IKA and PAL couples. The rates of seeing fellow workers off the Job are as follows: 01.03 714% FIAT 52$ IKA 71$ PAL 681 By caparing these two sets of figures, it is seen that the workers from the less industrialised nation are as likely to see fellow workers off the Job but are not as likely to have such encounters in the con- text of voluntary organisations. As we have noted in other instances, the relationship holds most strongly for the 01.08 sample. For these data, a large proportion of the relationship between fellow worker interaction, off the Job, and membership may be explained by the op- portunity to see fellow workers in the organisation. Lipset's data on these relationships are most strongly supported, then, in the 01.08 suple. 78 There is a further relationship between interaction with fellow workers off the Job and membership in an organisation. Data analysis for Hypothesis I showed that interaction which occurs on the job has only weak linkages with organisational membership. This finding was not in accordance with the findings of Lipset, where union activity was related positively to experience on the job. Now we might suggest that the Job is not as unrelated to membership as it initially appeared. Rather, Job has an effect when it is the source of friendships which extend to off the Job interactions. That is, when the relationships with fellow workers on the Job and off the Job coincide, membership rates are high. The two interactions together lead to organisational membership. Thus interaction skills gained on the Job have an indir- ect linkage, shaping membership primarily when accompanied by inter- action with fellow workers off the Job. To test the strength of this relationship we have again examined the association between organisa- tional membership and interaction with fellow workers off the Job, this time with a control for good friends on the Job. In this manner we can see whether it is the dual setting of friendship-oon and off the Job-uthat provides the strongest link to membership. Table 26. Interaction With Fellow Workers off the Job by Organisational Malnbcrship, Controlling for Good Friends on the Job 3 The Values of Q QLDS _FIAT IKA PAL __ No Good Friends (hi the Job .18 .33 -.OO .10 Good Friends 01 The Job .52 .21 .36 .38 79 Table 26 (cont'd.) Zero-order correlation between interaction with fellow workers off the Job and organisational membership: .32 .29 .15 .31 Table 26'shows that for each plant except FIAT, organisational membership is more likely if the worker has friends on the Job as well as interaction with fellow workers off the Job. These two variables are reinforcing each other as they correlate with organisational meme bership. As usual, the relationship is particularly strong in the OLDS sample. 'We have found support in the data for the importance of interact- ion with fellow workers. First, the relationship between membership and interaction, off the Job, with fellow workers is positively correlated so that the cumulative theory of social interaction gains support. Secondly, we found that this relationship is partially supported by the presence of fellow workers in the organisations to which the individual belongs. Thirdly, we found that the relationship of interaction to membership is strongest where interaction with fellow workers occurs both on and off the Job. Each of these three findings is in accordance with the data of Lipset but focuses on community organisations instead of union related organisations. Each of the three is most strongly sup- portsd with the data from the OLDS plant, suggesting that the descript- ions of workers' experiences may be culturally bound and less relevant to other nations. 80 Patterns of 12.1291 Interaction: A Continuum of Variables When looking at the four types of primary interaction and their relationship to organisational membership we can see that there is a pattern which is similar in each of the four nations. That is, three of the four types of primary interaction have similar rankings based on their correlation with organisational membership. alch a ranking en- ables us to ascertain the applicability of the suspensstory and suin- 1ative theories of interaction. As previously stated, the least consistent variable between the four nations is interaction with friends. It does not fit well into the cross-cultural pattern. The remaining three - interaction with fasily, neighbors, and fellow workers - do have such cross-cultural similarities in tens of their relationships with organisational mem- bership. The main pattern is such that interaction ulth fellow workers off the Job has the strongest correlation with membership, canpared to f-ily and neighbors. Interaction with neighbors has the second strong- est relationship, and interaction with family the least positive asso- ciation with membership. (In three of the four nations , this relation- ship is a negative one, OLDS being the exception with a small positive association.) The only exception to this pattern of fellow workers - neighbors - faily is for PAL, where the position of fellow workers and neidlbors are reversed. Family still has the least positive relation- ship, like the other nations. When this ordering is interpreted in tense of the counting theories of social interaction patterns, we see that both theories are applicable, depending on the type of primary interaction. Neither the compensatory 81 nor the cumulative is adequate in all instances. In addition, it can be suggested that the compensatory theory - when one is there, the other is absent - best describes the relationship between organisa- tional membership and that fon of primary interaction which is least sililar to it and has the fewest ties to it - that of interaction with the family. lhose who report interaction with the family are less likely to belong to an organisation. is we move further fru this I'most primary type of interaction' , toward interactions that are more often a mixture of primary and secondary interactions, the cumlative theory of interaction because more appropriate. For the relationship between membership and interaction with fellow workers, the culmlative theory is most accurate. nose who interact with fellow workers are likely to also belmg to organisations. Table 21. flowery of Primary Interaction Variables by Organisational Membership: The Values of Q oms rm mm m. Family .05 -.22 -.25 -.1o leighbors .25 -.01 .11; .hS Primds -.08 .31 .02 .37 Fellow Workers .32 .29 .15 .31 The data, then, point to an interesting and important relation- ship between these two theories of interaction patterns. Interaction leads to further interaction of a similar m; the individual, on the other hand, tends to choose between dissimilar types of interaction and to become involved in either one or the other. A finding such as this raises questions about the dimensions that underlie social interaction, in particular questions about what it is that leads to the mutually 82 occlusive nature of some forms of interaction while others tend to ac- oanpany each other. The answers to questions mch as these will be found on a level other than the column time-budget analysis. That is, the patterns smeared in this study when no limitations were made as to time span. This indicates that some forms of interaction are not undertaken together by one individual even when there are no time con- straints in the question asked. Similarly, the information will not be gained by asking questions which ascertain formal goals of organisations an! simple rates of interaction with various categories of individuals. Instead we midzt explore these basic dimensions by ascertaining latent and manifest functions of interaction for the participant, by investi- gating similarities of interaction sldlls demanded (much as we have in the early sections of this study), or by exploring variables such as Heissner's 'diecretion". In addition, research is needed within each mltural context to ascertain the most socially valued arenas of inter- action, those that are therefore likely to be preferred by the individual. The findings of this study, in addition, raise questions as to the appropriateness of the discipline's traditional categorisation of inter- action typee, whether it be primary-secondary or informal-formal. When we are clearer as to the comon dimensions shared by some fonas of interaction and the distinctiveness between others, we can move toward a classification based on less superficial criteria. These suggestions, that we learn what determines the relationships between various forms of interaction and that we revise our classification, imply a reinterpre- tation of the issue involving the competing theories of interaction patterns. Instead of asking which of the claims is the appropriate one, we‘ should test the assumptions of these theories. is we have seen in Chapter (he , these theories make assertions about an individual's need 83 for security and his preference for family 2; for formal interaction. They make statements about the decline of the family as well as of supportive comnity institutions. Each of these presents an empir- icsl question which needs to be answered with data rather than taken as an assmption. 111 this way we can move closer to understanding the nature of social interaction and the forces itich determine the pattern. There is, in addition, a need to ascertain in which instances each of .3... the theories is explanatory and to search for a theoretical frmework that not only encompasses both but explains the nature of social inter- action patterns to a further degree. We are still left with the inconsistent findings of the variable, interaction with friends. The inconsistencies m be explained by the source of friendships in each of the nations. In the IRA and PAL eagles, there is a high correlation between interaction with friends and with neighbors (.57 and .61; respectively). This most certainly indicates that the friends of these workers are also neighbors, so that to report interaction with one is to include the other. This expecta- tion is mpported by the literature dealing with blue-collar workers, which suggests that their major source of relationships is from family and neighborhood. Such a relationship, however, is not found in the 01138 and FIAT plants, where the correlation between interaction with friends and with neighbors is -.01 and .08, respectively. Instead, for these two plants, friendships are more likely to be related to inter- action with fellow workers. The correlations for the two nations, re- spectively, are .21; and .31. This is supported by the findings of upset, where fellow workers are, indeed, a major source of friendships. 81: Again, we find the OLDS plant to more closely correspond to our expec- tations about the effect of the Job in an industrial setting than do the other plants. In each of the four nations, there is a notable correlation between interaction with fellow workers and with neighbors. This relationship is supported by our knowledge of class segregation into residential areas, so that neighbors are composed of those of the same occupational level and, quite likely, those from the same plant. That these cats-- gories of interaction coincide is further supported by the fact that, with the exception of FIAT, the correlations of these variables and membership are quite similar. Table 28. Correlation Matrix of the Primary Interaction Variables: The Values of Q 0 L D S F I A '1' Wm. 1".Hkr Family mime smu- m - 022 -035 .21 - 002 -ehg .. 21 'W" " -e01 e38 - e08 e50 M“. - sell - 031 Fellow workers - - I K A P A L m1 N M! F.Wkr 1‘ Ne ds LHkr rail, "' "'e09 -e26 -.16 " -.25 .lem e07 him}... - e57 e37 - eéh e62 M“. - e33 - s26 Fellow Workers - - Controllng For Social Characteristics of The Worker In this section we will further investigate the relationship be- tween organisational membership and the four types of primary interact- ion by introducing the eontrol variables age, education, residential background, and skill level. Occupational task has been omitted since primary interaction here is outside the realm of the Job. Skill level 85 is included as an indicator of stems. In each case we will determine . whether the associations with the controls are essentially the same as the original correlations. As in section I, we will look for changes in the relationships between variables which are sufficiently strong to change the value of Q by .20. When controls were introduced into the Job experience relationships, the analysis was directed toward identi- fying those instances where the control had an effect on several of the Job experiences, not on isolated interactions. In this section, however, we would expect that the important controls may be different for each of the types of primary interaction. In the first part of analysing the effect of controls here we will focus on each of the four types of interaction separately and discuss how they are affected by the social characteristics of the workers. Then we will examine the data to see if the continuum of variables based on relationship to organisational membership remains the same in spite of the control. The values of gala for each relationship are given in tables 29 through 32. To aid in analyses, we have repeated the sero-order correlations as well. Lynn The original relationships between family and organisational mem- bership were negative for three of the nations and slightly positive for the fourth (01.08). In general, the overall relationships do not change with the controls, although there are isolated exceptions. In none of the four nations does age have an impact so as to change the direction of the original. Still level does only for FIAT where unskilled workers have a negative correlation considerably stronger than for the sample as a whole (Q I -.hh). The correlation for OLDS' skilled 86 workers is negative instead of positive as was the original relation- ship, but does not differ from the sero order correlation by as reach as .20. The original relationship in the FIAT sample is altered in three other intances. Those with metropolitan backgrounds and those with middle level education both have a positive correlation, .18 and .21: respectively, between fmuily interaction and membership. Those with rural or mall town backgrounds, on the other hand, have strongly nega- tive correlations, -.1:3. The first two changes indicate an opposite correlation than found for the whole. The third is in the same direc- tion but is much more strongly negative. Two of these social characteristics also make important changes in the PAL sample -- middle education level and rural or small town back- ground. For the middle educated group the relationship is strongly positive, .314, whereas the overall relationship indicated a correlation of -.10 and that of the lowest educated is a much more negative -.53. For the rural - small background group, the correlation was more strongly negative, -.30. Thus, among both the FIAT and PAL samples we see that those of the medium educational level who interact with their families are also more likely to belong to voluntary organisations. Those with rural backgrounds, however, are even more likely to choose to interact in one arena or the other. There is one instance where educational level makes a difference among m respondents. Those who are highly educated indicate essential- ly no relationship between family interaction and organisational mubsr- ship. As the other types of primary interaction are discussed, it will be seen that the basic impact of the controls remains the same as that Just 87 described. Education has the greatest impact of the four social characteristics utilised here and the greatest number of changes occur in the FIAT and PAL samples. Compared to the other three types of in- teraction, that with fellow workers is most affected by the introduction of these controls. Table 29. Interaction With Family by Organisational Membership for Each Control Variable, for Each Ration: The Values of Q ems FIAT _I_15A_ y; Education Low -.05 -035 ‘030 -053 H.611]! - . 2h -. 39 .31; m ' .11 -.28 e0]. -e17 Age UN” 30 e01 e21 -e16 -.10 0"? ” e011 'elb -eho 'eOS Still Level “ham“ e10 -e -e37 "elz Susi-Med .01 - .06 -.16 -. Skillod -.10 -. 311 -. 2]: -.1h Residmtial Bkgd. MCI-'3‘. Tom .12 -ehB -018 -0 ” H.tmpont.n -.O7 e18 -a 33 e08 Zero-Order Comlltlon: e05 "e22 "e25 -.lO 2 e N01 01" There are no overall changes from the sero-order correlations, where for each nation except In there was a positive relationship be- tween interaction with neighbors and membership in an organisation. Again, age had no changing impact upon the original relationships. Where- as with fuaily interaction the relationships for OLDS did not change, there are three instances of changes for OLDS in the relationship to in- teraction with neighbors. The sero-order correlation of .25 disappears 88 for those of the lowest educational level and becomes 1.00 for the un- skilled. Those from rural or small town backgrounds in the OLDS sample show a positive relationship of .148 between the two variables. For the FIAT ample there are again changes as a result of educa- tional and skill level. whereas the original correlation was a weak negative association, we find that it is more strongly negative for the middle educational level and more positive for the highly educated, -.2? and .25 respectively. The relationship is also a moderately strong pos- itive one for the unskilled, .29. For both family and neighborhood interaction, the relationship to organisational membership has been dif- ferent from the whole sample for those workers who have attained a medium level education and those who are unskilled. Thus the FIAT workers of medium educational level are m likely to belong to an organisation if they also interact with family but l2; likely to belong if they in- teract with neighbors - the opposite of the overall findings for the four nations, including the findings for those FIAT workers with less or more education. more is no instance where the controls have an effect on the rela- tionships found for IKA, one instance where there is a change in PAL. In the PAL sample, those who are skilled show less of a positive rela- tionship between neighborhood interaction and membership than is true for that sample as a whole. Table 30. Interaction with Neighbors by Organisational Membership for Each Control Variable, for nch Nation: The Values of Q 2123.. y. .233. EL. Education LOW .00 e05 -006 o 33 ”061‘“ - -e27 e20 e29 High .ho .25 .28 .hz 89 Table 30 (cont'd.) Age Under 30 .h3 -.09 . 20 .58 Over 3) .21: .01: .05 .31 Sdll Level Unskillod 1.00 . 29 .01 .39 m-mad e20 "e13 e 22 e50 $.11“ .13 -. 18 s 22 e18 Residential Bkgd. Raul-msTm .158 eO’J -e03 e38 Hetmpolltln ell-l " e01 e 26 e51 Zero-Order Correlation: . 25 -.Ol .11: .L5 3. Friends In all but one instance the changes in the relationship between organisational membership and interaction with friends come as a result of education as the control variable. The relationship between the two variables varies widely for each educational level of FIAT. The original correlation for that sample was .313 it is .52, -.08, and .02 for the low, medium, and highly educated, respectively. The positive correla- tion found originally is explained mainly by the lower educated. For that group, those who interact with friends are likely to also belong to an organisation. Among PAL workers, both the lower and medium educational levels are cmeiderably less likely to accomparw interaction with friends with membership. For the medium educated of IKA, there is more likelihood that the two will coincide. There are only two instances where age has the effect of changing the correlations by at least .203 both changes are among 01.08 workers under thirty years of age. Here, that age group has a more negative correlation between friends and membership than is true for that sample as a whole. Table 31. Interaction With Friends by Organisational Membership for Each Control Variable, for Each Nation: The Values of Q OLDS FIAT ILA; £4.11 lducaticn I” e09 e52 'eos "e01 Medium - - .08 . 2h .12 High -. 2h .02 -.16 . 29 Age undfir 30 "em .28 “.02 e26 Over 30 -.O3 .26 .09 . 38 Still Level Unskillod e01! eh? e06 e149 wiflilled -011 e 31 007 e 32 Stilled -.13 .11: -.15 .3) Residential Bkgd . Rural-Sm.Town -.13 . 3h - .15 .140 Metropolitan .02 .17 . 18 . 23 Zero-Order Correlation: -.08 . 31 .02 . 37 1:. Fellow Workers, Off ‘me Job As would be expected, the control of occupational skill level has its greatest effect on the relationship between organisational member- ship and interaction with fellow workers off the Job. In spite of the seven changes of at least .20, however, none are changes inthe direct- ion of the original correlations; in each instance those who interact with fellow workers are likely to also belong to an organisation. In the 01.08 sample, these two variables are even more likely to vary with each other among the unskilled than in the sample as a whole. no tendency is less among the semi-skilled of this sample (and among those under thirty). In the FIAT sample, there is a stronger positive 91 relationship among those workers at the opposite end of the continuum, the skilled, instead of the unskilled as in 01.05. For both the IRA and PAL samples, the stronger positive correla- tion is among the groups of semi-skilled. In addition, the other two skill levels of PAL drOp in value of gamma at least .20. There are several other changes in the findings for IKA and PAL. For IKA, the originally positive relationship is strengthened for the highly educated up but is quite a bit more negative (-.22) for those of the middle educated 11 group. Among those workers, interaction with fellow workers off the Job and organisational membership are not likely to coincide. Another change in the sign of the correlation occurs among the few _ 3 least educated of the PAL sample where the correlation is -. 36. For that sample, residential background changes the findings so that the correlations are considerably less positive among those from rural or small town backgrounds and quite strongly positive (.67) among those with metropolitan backgrounds. For PAL, then, the coincidence of inter- action with fellow workers and membership is true primarily for the more educated from metropolitan areas. Table 32. Interaction With Fellow Workers Off the Job by Organisational Membership for Each Control Variable, for Each Nation: The Values of Q OLDS FIAT .21.}. LA; Education Law .37 . 28 .27 -. 36 Medium - . 28 -. 22 .35 High . 26 . 32 .M: .31 Age Under 30 . 11 .12 .08 . 38 Over 30 .35 .31: .25 .17 92 Table 32 (cont'd.) 3:111 Level unskilled . 56 . 21 . 01 .00 Sui-Skilled .12 .21 . 32 .59 fillled .36 .59 .15 .10 Residential Bkgd . Rural-Sutton .140 .22 .10 .Oh Metropolitan .28 .b0 .19 .67 Zero-Order F mmflon 3 e 32 e 29 e 15 e 31 In sumary we can see that the changes that result frost the intro- duction of controls do not follow aw consistent pattern either within or between nations. Most often the changes are such that the correla- ; tion is still in the sme direction as the original but is a stronger relationship for that particular sub-group. For less tum twentybfive percent of the changes that resulted from the controls the new correla- tion carried the opposite sign fruit that of the sero-order. Host of the effects of the controls were a result of introducing the worker's level of education. Of the thirty-five instances where the correlation changed by at least .20, sixteen of these were when education was controlled . Again, however, the particular educational level which deviated was not consistent. As was the case in analysing the data for Hypothesis I, the find- ings for FIAT and PAL were the nest affected by the controls. Euca- tional levels were most important in the FIAT sample, skill level made the biggest difference for PAL. The type of primary interaction most affected by the controls, in terns of its relationship to organisational membership, was that of inter- action with fellow workers off the Job. As noted previously, the control of workers skill level accounted for most of the changes here. Age 93 level had the least affect on any of these relationships between pri- nary interaction and membership in a community organisation. Having found no consistent or large effects as a result of control- ling for social characteristics of the worker, there is one question which remains to be answered. Does the relationship between cumulative and comatog types of interaction remain the same with the intro- duction of the controls? That is, is interaction with the family the least likely to accompany organisational membership 3 does interaction with fellow workers off the Job tend to coincide most closely with mem- bership? Is the position on the continuum of interaction with friends still inconsistent between the four nations? In analysing the ordering of the types of primary interaction ac. cording to their correlation with organisational memberships, looking at the relationships for each control category separately, the most clear- cut finding is that family quite consistently has the least correlation with membership. In spite of the workers education, age, skill level or residential background - or his nation - he will be likely to choose to interact in either one of these arenas or the other, seldan in both. (As with the sero-order correlations, there is a low positive correla- tion here for the OLDS sample.) As in the original correlations, interaction with both neighbors and fellow workers tends to be positively correlated with meanbership. The cumlstive theory of interaction best describes the pattern here; if an individual interacts with neighbors or fellow workers off the job he is likely to belong to an organisation in the community. his relative position of these two variables on the continuum changes with some categories of the controls but they both tend to min positively 9h correlated with membership. The position of interaction with friends on the continuum raaains inconsistent between the nations, and often within each nation depend- ing on the control category. In the 01138 couple, the correlation with mubership is quite consistently even more negative than that of family interaction. In general, interaction with friends is likely to be the most positively correlated variable in the FIAT sample and the PAL F?“ sample. In the IRA sample, the correlation is so as to be slightly more positively correlated with membership than is interaction with the faily. The most consistent position for interaction with friends, then, is near that of family interaction for OLDS and IRA and at the most pos- itive end of the continuum for FIAT and PAL. we can conclude then that the findings of the original relation- ships rasin basically the same. 1‘hat is, the compensatory theory is best in describing the relationship of funin interaction and organi- sational meabership. Individuals in each nation, regardless of their social characteristics tend to interact with family _o_r to belong to organisations. The cumulative theory, on the other hand, best fits for the relationship between membership and interaction with neighbors or with fellow workers. Here , if the individual is involved in one arena he tends to also be involved in the other - regardless of nation or social characteristics. CHAPTER SH MARY AND CONCLUSIONS OF THE STUDY The study was designed to test two maJor hypotheses about the relatimship of social interaction experience to membership in a com- munity voluntary organisation. In the first section of the study, we focused on the relationship between interaction experiences on the Job and membership. In the second, the aphasia was on the relation- ship between various fonns of primary interaction and belonging to an organisation. The sample which furnished the data was of autonobile workers in four nations of varying levels of industrialisation. In this chapter we will review these hypotheses and the conclusions that resulted from our data analysis. In addition, we will discuss the bearing these conclusions have upon the larger theoretical structures from which the hypotheses are derived. Mthesis One Theorists have often speculated about why blue-collar workers have such low rates of membership in commity volunteer organisations. A trend in these explanations is to suggest that the workers are “person centered" so that interaction of a secondary nature is diffi- cult for them. They lack the interaction skills to function well in bureaucratised or formal settings and therefore avoid these interact- ion 'arenas and rely primarily on flaily and neighbors. Such explanations 95 96 are sumonly put forth and accepted among those concerned with blue- collar workers but have not been put to empirical test. This study has, in part, filled this gap in knowledge. his clear way to test the assertion that blue-collar workers avoid voluntary organisations because they lack secondary interaction in an organisational setting is to see whether those who _d_g have such experi- mces than do belong to voluntary organisations. If the lack of inter- action experience is the barrier, than absence of the barrier should lead to organisational n-bership. In our study, we focused on the work place as the arena for the acquisition of the necessary interact- ion. We then tested the theory of blue-collar membership by ascertain- ing whether those who had secondary interaction and experience with formal structures had higher membership rates that those who did not. mess relationships between interaction on the Job and membership were mined in auto plants of four nations. They were, in level of indus- trialisation, OLDS (United States), FIAT (Italy), IKA (Argentina), and PAL (lhdia). The data, ilen used to test the first lupothesis, did not offer stng support for the suggested relationship between interaction on the Job and newership in a voluntary organisation. Although, in general, the correlations were in the expected direction, they were weak associ- ations and were not strong enough for us to consider the theories sup- ported. m. data frau the 01138 sample most closely showed the predicted relationships. Controlling for the social characteristics of the workers (age, education, occupational task, skill level, and residential background) did not substantially alter these findings. We concluded that whereas there may be sane link between the two sets of variables, 9? interaction on the Job could not explain organisational membership. Of the five Job variables used, that which measured the worker's ability to move around while on the Job was the best predictor of mem- bership for each nation except FIAT, where only the ability to discuss personal problus had a higher correlation. n1... relationships indicat- ed that those workers who could move freely were slightly more likely to be a muber of a co-unity volunteer organisation than those who did not have this Option. Such a relationship is supportive of the theories of organisational membership in that the spatial mobility offers the individual a wide range of potential relationships and presents him with the opportunity to interact with others outside the realm of family and neighbor. These are the skills that are said to facilitate organisa- tional mnbership. Building on this finding, we midrt suggest that an important underlying dimension here is that the worker have the elmaant of 933-- E ovm‘ the interaction in which he engages. Although this presents a new set of variables which need to be further researched, there is sue support for the suggestion in our data. mat control over inter- action is inportant is borne out by the fact that the mere presence of others with them to interact (our question of how new you “can talk to.) is positively related but in no case as strongly associated with membership as is physical mobility. The idea is further supported by the finding that interaction per as, as measured by those with whom the worker must interact - no choice or control involved - is the least effective indicator of organisational membership. Social interaction, it seas, is a skill which may be important to membership only when the worker has an element of control over its initiation. 98 The two measures of quality of interaction were also weakly but positively correlated with organisational mmubership. Our assumption in including these variables (number of good friends at work and ability to discuss personal problmus with fellow workers) was that the strength or "depth" of relationships at work is also an important dimension of Job experience. n1.» expectations were upheld but, as was the case with all the Job variables, the relationships were not strong enough to be considered a main explanatory factor of mmubership. no data did show, however, that good friends on the Job increased the predietive value of interaction with fellow workers off the Job. This finding will be reviewed as we discuss Hypothesis Two. In caparing the four nations, we see that Alex Inkeles' (1960) theory does hold in that the Job situation proved to have a similar relationship to organisational membership in each of the four nations. here was a difference, however, between the strength of the relation- ship of awe and the other samples and this leads us to more closely examine the evolutionary theory as discussed by Moore (1965). As pre- viously mentioned, he predicts a similar impact of industrialisation across nations but introduces a time dimension whereby the effects of industrialisation come gradually rather than imediately. In addition, he has suggested a model of concentric circles so that industrialisation has its strongest impact at the work place itself and a weakening impact as we move out from the center or work place. This theory, while in new ways more intuitively correct, is hard to verify sinee "out from center" is difficult to operationalise. In this limited case, however, we can explain the stronger relationship found among OLDS workers by referring to Moore's theory. According to his thinking the most indus- 99 trialised of our nations, the United States, would be more completely shaped by that industrialisation than would those nations at a lower level. We midrt infer that the secondary organisations said to ac- cmpany industrialisation would be a more readily available and viable armra for expansion of one's social interaction in the United States than, for exmupla, in India, and that skills gained on the Job are more readily carried over to that arena when the opportunity is a more integ- ral part of the nation's organisational structure. These predictions are quite well borne out by the ranking of nations according to their fit with our predictions. The data on OLDS workers most completely bears out our predicted relationships. FIAT, the next most industrial- ised nation, shows the second strongest link between Job and. non-Job interaction. IRA and PAL are reversed from what would be expected ac- cording to Moore's thinking-ea closer overall fit being found for work- ers from India than from Argentina. The most consistent finding is the relatively greater strength of the relationships in the OLDS sample. Each of these points, the varying importance of the five Job inter- action variables and the ordering of the four nations, help us move toward a clearer understanding of organisational membership among blue- collar mrkers and of the impact of industrialisation within a nation. It must be recalled, however, that all the associations between Job variables and mmnbership were weak ones. In general, it is concluded that theories which point to secondary interaction in formal organisa- tions in order to explain worker's low membership rates have not been supported in this study. I mmesis Two Four types of primary, informal interaction were the focus of the 100 second maJor motheeis of the study. In this section competing des- criptions of interaction patterns were reviewed and the data were exnimd to determine the applicability of them. One description asserted that interaction is of a comatog nature so that an indi- vidual engages in secondary interaction when his needs are not fulfilled with more primary interaction - or vice versa. ‘me second suggestion is that interaction is cumulative so that interaction in one leads to r?- interaction in the other, the primary and secondary interactions accom- paw each other. To investigate the claims of the two theories, we examined four types of interaction - with family and kin, neighbors, friends, and fellow workers off the Job. In each case, we examined the _ relationship between the primary interaction mrd membership in a volun- tary organisation. As in Hypothesis One, we examined data from the workers of four nations. me data showed that instead of one or the other of the two pat- terns describing the relationship between primary interaction and organ- isational membership, the particular association depends on the arena of primry interaction. Interaction with the family proves to be nag- atively correlated with mmubership; workers who are involved in one of these interactions tend to not be involved in the other. Here the compensatory theory best describes the interaction patterns. At the other end of the continuum that showed in the data is interaction with fellow workers off the Job. Here it was found, for workers from all. nations, that the cumulative pattern is most applicable. Those who re- port interaction with fellow workers off the job are also more likely to belong to a voluntary organisation. Between these two extremes is interaction with neighbors, a variable that correlated positively with 101 mmabership but not in as strong a relationship as fellow workers. Interaction with friends was not in a cormnon position on the scale across the four nations. In the m and PAL samples, it was similar to neighbors in its relationship to mmnbership. In OLDS and FIAT, it was similar to fellow workers. This finding, and those dealing with the interaction with fellow workers off the Job, show important similarities to the work of Seymour n Lipset (1956). As he discovered in his study of the International Typographical Union, the Job is a source of informal relationships away from the workplace. Also, those in our samples who do interact with fellow workers off the Job showed a higher rate of mmsbership in caa- id‘J munity voluntary organisations. This relationship was further strength- ened when the worker reported both the presence of good friends on the Job and interaction with the: off the Job. These workers, especially in the 0133 plant, were most likely to belong to an organisation. These relationships are all as would be expected from Lipset 's study. Our data have shown that the effect of interaction with fellow workers extends not Just to the union and its related organisations but to or- ganisations in the ccrmmurity as well. In cur-nary, with regard to the second maJor hypothesis we concluded that both patterns of interaction occur depending on the context of interaction. When interaction is with the family, a type of interaction quite dissimilar frau that in organisations, the compensatory theory holds 3 interaction in these two arenas tends to be mutually exclusive. As the interaction becomes less home-centered and more orimrted toward the cumunity, it tends to be accompanied by organisational membership and the cumulative pattern is most applicable. 'nris pattern appears in 102 all four samles, regardless of the level of industrialisation. The relationships of each of the four types of primary interaction to meshes-ship did not substantially change with the introduction of the social characteristics of the worker as control variables. The Four Nations Throuflaout the data analysis, it has been noted that the relation- ships found in the four nations are generally in the same direction. In no case did we case to a major conclusion that did not apply in all the nations. when the controls had varying effect across the four there was no consistent pattern either within or between nations and the differ- ences could not readily be attributed to the level of industrialisation of the nation. The case for 011): however, was consistently nore sup- portive of the hypotheses. This suggests that previous analyses of voluntary organisational membership of blue-collar workers are somewhat culturally bound. In mpotheeis I the overall conclusion was that there was no more than weak support for explanations which suggest that the lack of exper- ience in secondary interaction in formal organisations accounts for the blue-collar worker's low membership rates in voluntary organisations. Again, there was somewhat stmnger support in the more industrialised nations, particularly 01138. This suggests an evolutionary effect sin- ilar to that of Wilbert Hoore, but the conclusion of only weak confirm- ation of Hypothesis I was true for each of the four nations. Similarly, the conclusions about Hypothesis II are the same for each nation. In each case, that prinary interaction nost similar to organisational membership follows the cunnlative pattern whereas that 103 nest unlike voluntary organisations tends to be of a compensatory or substitutive nature. lications Of The Stu The findings of this study must be fit into the broader context of voluntary organisational membership of blue-collar workers in order to fully understand their importance. In this section, we will discuss __, some of the implications of the relationships which our data have shown. q The major focus of this study, to test explanations of blue-collar membership rates, provides information for the larger issue of how we explain any membership in secondary organisations. These larger issues can be examined in the light of the findings of this study. We have, in this study, focused on the interaction experience and skill which mables the individual to interact in formal organisations and in secondary relationships. We can now evaluate the nerit of using such a variable. We will discuss interaction in the light of other factors thoudIt to enable maubership and in reference to the broader question of wlv individuals Join voluntary organisations. An assumption of this study, as we noted in Chapter One, has been that there i_g sane individual skill in interaction finch can be acquired in one context and applied in another, i.e. , the voluntary organisation. In our data analysis we found only weak support for the expectation that experiences on the Job do explain membership rates. We can now suggest a broader context in which to exmine these data. By examining other factors that influence membership, we can see that interaction experi- ence has value as an explanatory variable when it is viewed in the con- text of other variables working in the same direction. These other 10h variables which affect membership along with interaction skills may be other individual traits or activities or may be characteristics of the organisation itself. lb regard to other individual traits, there is support in our data for the claim that cone types of informal interaction lead to organisa- timal membership. We have found, specifically, that interaction with fellow workers off the Job and with neighbors is positively related to m-bership. This suggests that interaction which incorporates both prinary and secondary eleaents may lead to umbership. ‘mese variables facilitate, or accaspaxv, nubership as does interaction experience gained on the Job. Lipset has also found that such informal relation- shipe facilitate membership. We have previoust noted other individual traits which affect mulbership, such as individual needs for affection and security, collective action, and social status. Interaction skill, that, is best seen as one of a mmber of variables which explain organ- isational meabership. Some of these other variables are characteristics of the individual. There are also traits of the organisation which cowine with char- acteristics of the individual to explain the broader issue of member- ship. Ross (19510 has found that the hierarchy within the organization provides opportunity for the advancement of some individuals. This, then, is an organisational trait vmich may lead some to Join. In addi- tion, the functions the organisations serve for society may also explain mubership. his belief that voluntary organisations may work for the good of dnocraoy as described by de Toqueville and Lipset, along with the pruotion of a non-political channel of col-suni cation may lead some individuals to Join organisations. In addition, some organisations offer direct opportunity for political involvement and advancement and 105 mm offer opportunity for the individual to gain political skills. The broader question of who Joins organizations may be partially ex- plained by characteristics of the organisation as well as by character- istics of the individual. Interaction skill, the}, is one of several variables which help to explain membership. It is accompanied by traits of the individual and of the organisation. is still, however, might raise several cautions _ about the use of such a variable in future research. Lipset (1956)., for q example, has pointed out that organisations can function for society at large by preserving duocracy and yet not offer the Opportunity for the .. _ individual's use of skill within the organisation. is have assumed i J that such individual drills are in fact relevant within the organisation but the findings of Lipset correctly point out that this may not neces- sarily be the case. In addition, it met be noted that the use of organisational mem- bership as an indicator of interaction in organisations is a weak one. mereas we have defended, in Chapter One, its use within this study, further research should attupt to smasure the actual participation and rate of interaction within the organisation. Again, Lipset makes the point that the organisation may not in fact draw on such individual charactristics for its functioning and other organisational literature points out discrepancies between rates of membership and participation. It has become increasingly evident as the data are analysed that the explanations of membership cone-only put forth are inadequate to explain worship rates except for the middle and upper classes. In addition, we have found from our data analysis that the barriers seen as preventing the working class from following this pattern do not, in 106 fact, explain their failure to Join voluntary organisations. These findings lead us to further investigate the assumptions of such theories of nubership and to att-pt to find in than an explanation for their narrow range of applicability. As we noted in the previous section there are a number of specific explanations for n-bership in cemunity voluntary organisations. The litereture suggests that organisations provide the individual with an opportunity for self-expression and security. This explanation appears F to be inappropriate to the working class in face of our knowledge of the strength of fanilial relationships anong such individuals. The applica- bility of such an explanation to that category of individuals to are E, already secure in a context which fills such affective needs is quest- ionable. his is not to suggest that voluntary organisations do not, in fact, provide opportunity for self expression and emotional security but, rather, to question the appropriateness of the explanation for the blue-collar senent of the population. Other explanations of laborship suggest that the individual Joins organisations to gain a neasure of power and support for interests he shares with others. The literature relies on such findings as explana- tion for one' a motivation to Join. Whereas , again, organisations may, indeed, fulfill this function, there are several bases on which it can be questioned . First , we question whether opportunities for power or acquisition of political skills serve as an explanation for the indi- vidual's decision to Join the organisation. Personal power appears more likely to be a consequence of membership than a motivator for it, especially in those organisations not manifestly political. axch explan- ations are questionable on another basis, that of their class-limited nature. Ross (1951:), for exanple, has found that personal advancement 107 within the organisation is closely tied to occupational position. If political power did serve as inducement to membership, it is likely to explain only membership of those in prestigeful occupations, spin ex- cluding that segment of the population upon which we have focused. Similarly, opportunity for career advancaaent .and social status in the broader ccsnunity are likely to be relevant only to the occupations of the upper classes and to the instrunental organisations to which they . belong. r1“, This review of the discipline 's attempt to explain membership points to the fact that Just as the rate of membership is found to be . class related so, too, are the genetic” for such nenbership. That - J is, the upper classes tend to belong and the literature provides “upper class explanaticns'for such membership. Further, the theories about blue-collar workers which this study examines have also been developed frcla the sane, upper class, perspective. his focus is on barriers to nenbership of the working class instead of the motivations for nenber- ship, as with the upper classes. The explanations assmue, then, that the blue-collar workers would (and should) belong 11 only they were able. They never raise the question, as for the upper classes, of Why should they belong”. The literature fails to cite positive motivating factors for this sealant of the population, only barriers to the worker's abil- ity to be like the middle and upper classes. It is essential, then, that we nove our analysis of organisational masher-ship into the broader scope of asking what motivations exist for the membership of blue-collar individuals. It is expected that, as we move into this line of question- ing, we will discover few reasons for high rates of nubership. The benefits of belonging to new organisations do indeed seen to be class 108 related and this factor in itself is more likely to explain low member- ship rates than is personal barriers to fomal interaction. There is a further set of biases within the voluntary organisation literature, derived fru the value of such organisations to society as a hole as well as to the individual member. This has been a cannon theme since the writings of de Toqueville. Building on his thinking, sociologists have described the benefits of such organisations to the u, process of democracy. Such thinking suggests that organisations serve n as a check on the govemaent, as an impartial source of information and ideas. Similarly, Durkheim has pointed to the ability of the voluntary . organisation to act as a buffer between the individual and larger [ j society. Lipset has described the prevalence of voluntary organisations - as a major factor in explaining the democracy of the International Typo- graphical Union. The connections cited between such organisations and the presence of democracy are made with the explicit or implicit positive evaluation of their existence. Implied, but rarely examined, is the worth of the voluntary organisation in achieving democracy. We have found, in examination of these elements of the voluntary organisation literature, that this value Judgement spills over into thinking about ”bership as well. The connection is such that voluntary organisations should exist for the good of democracy and individuals should belong for the same reason. The literature seems to imply, then, that not only .9311 the blue-collar individual belong if he were able but, in addition, be M belong. The literature has suggested individual rationale for Joining and now implies a moral obligation as well. The theories, we suggest, have not only been developed from a social class-related per- spective but have also emerged from an unquestioned value assumption 109 that individuals “ought to“ Join for the good of democracy. The ex- planations of blue-collar membership which we have examined have assumed that such individuals should and would belong and have then sought barriers to explain win, in fact, they do not. Our data have led us to the realisation that such an approach offers little in explanation of win blue-collar workers Join voluntary organisations. is noted in Chapter One, theories of social interactim suggest that the pattern is cumlative or compensatory, that frequency of inter- action in different settings can be described by either one theory or the other. The literature presents these theories as mutually exclusive perspectives and there is disagreement over which of the two is the correct description of the relationship between interaction settings . We will argue that there mat be changes in the nature of the research questions which deal with this issue and that there must be refinements in our conceptualisation of the variables upon which such research focuses. Our data have pointed out the fact that the cumulative and compen- satory perspectives need not be regarded as mutually exclusive but, rather, that there are instances in which each is appropriate. We must now move to a point of asking under which conditions and for what specific types of interaction each of these holds. To answer questions such as these, it is essential that we have a clearer understanding of the factors which account for differential interaction rates among seg- ments of the population. The need for understanding of that underlies organisational membership has been documented in the previous section. Research into the bases of interaction with family, neighbors, and friends needs to be similarly developed. We might ask here what charac- llO teristics of the individual and of the interaction opportunity itself explain rate of involvement. In addition to more research into what factors explain differen- tial interaction rates, we mat also develop, on a conceptual level, a better understanding of types of interaction. is noted earlier, there are difficulties with the ideal types of primary-secondary and informal- formal. Neither of these adequately describes the wide range of types of social interaction, nor can they deal with the fact that most inter- action is likely to be some mixture of elmsents. Research such as that described above, which investigates the functions of such interaction and the factors that lead to interaction, may help us to refine these categories so that they more closely describe the experiences of indi- viduals and the combinations of interaction types which they exhibit. We have seen, as we look at the study in its broader context, that the findings have important implications for the study of voluntary organisations in general. We have noted that interaction skill as a factor explaining organisational membership is most useful when viewed in conjunction with other factors that also lead to membership. We have then, on s more abstract level, investigated the broader perspec- tive of how organisational membership might be explained. we have argued that such explanations are both social class bound, in that they view membership from its functions for the upper class, and value bound in that they imply that one ought to belong. We have suggested that these two limitations in the conceptualisstion of why individuals Join organisations make it difficult to speak to the question of blue-collar membership. We have then discussed the issue, arising out of the find- ings of Hypothesis II, of differential rates of social interaction and 111 the apparent need for further research and conceptual refinement. There is one further point that must be made in reord to the findings of the study as a whole. As we have noted throughout the data analysis, the hypotheses are more strongly supported by the data from the OLDS sample than for the other three plants. This suggests to us that the sociological perspective on voluntary organisations, on Job experiences, lid on primary interaction may be limited in applicability beyond the culture of the United States. This suggests the need for further research and conceptualisstion of the relationships as they exist in other nations. ammonium Allardt, ma, Pmtti Jartti, Paina Jyrkila, and YrJo Littunen 1957 "On the Cumulative Nature of Leisure Activities" Acts Sociologica 3: 165-172. H Almond, Gabriel and Sidney Verbs 1963 The Civic Culture. Princeton, New Jersey: Princeton University Press. Axelrod, Morris ‘ 1956 'Urban Structure and Social Participation" American Sociological Review 21 (February) : 13-18. ,I Bsbchuk, Nicholas and John Edwards 1965 "Voluntary Associations and the Integration Hypothesis" Sociological Inquiry 35 (Spring): 1119-162. Bell, Wendell and Maryanne Force 1956 "Urban Neiyaborhood Types and Participation in Formal Associations“ American Sociological Review 21 (February): 25-31:. Berger, Bemett H. 1960 Working Class Suburb. Berkeley: University of California Press. Blauner, Robert 1961: Alienation and Freedom: The Pactory Worker and His Industry. Chicago: University of Chicago Press. Blum, Alan L. 19614 "Social Structure, Social Class, and Participation in Primary Relationships" in Arthur B. Shostak and William Oomberg (eds.). Blue-Collar World. Englewood Cliffs, New Jersey: Prmtice-Hall. Cohen, Albert and Harold Hodges 1963 I'Characteri.stics of the Lower Blue Collar Class" Social Problems 10: 303-311. Curtis, James 1971 "Voluntary Associations Joining: A Cross - National Comparative Note” American Sociological Review 36 (October): 872-880. Davis, James A. 1971 Elementary airvey Statistics. Buglewood Cliffs, New Jersey: Prentice-Hall. 112 113 Dean Lois 1551: "Social Integration, Attitudes, and Union Activity“ Industrial and Labor Relations Review 8 (October): 138-58. Dotson, Floyd 1951 "Patterns of Voluntary Association Among Urban Working-Class Families" American Sociological Review (October): 687-693. Pom, William H. and Kirk Dansereau 1957 "Union Masher Orientations and Patterns of Social Integration" Industrial and Labor Relations Review 11 (October). Pom, William H. 1969 "Occupational and Social Integration of Automobile Workers in Four Countries: A Cooperative Study" International Journal of Comparative Sociology 10 (March and June): 95-116. Pom, William H. 1971 "The Accosnodation of Rural and Urban Workers to Industrial Discipline and Urban Living: A Four Nation Study" Rural Sociology 36 (December): baa-508. Pom, William H. 1972 "Technology and Social Behavior of Workers in Four Countries: A Sociotechnical Perspective" American Sociological Review 37 (December): 727-737. ram, "1111“ He "Automobile Workers in Four Countries: The Relevance of system Participation for Working Class Hovements" (an unpublished manuscript). Priedlander, Frank 1966 "Importance of Work Versus Nonwork Among Socially and Occupation- ally Stratified Groups" Journal of Applied Psychology 50 (Decmaber): h37-hbl. Oeltung, Johan 1967 Theory and Methods of Social Research. New York: Columbia University Press. Cans, Herbert J. 1966 The Urban Villagers. New York: The Free Press. Gist, Noel and Sylvia Fave 1961: Urban Society. New York: Thomas Y. Crowell. Gordon, C. Wayne and Nicholas Babchuk 1959 "A Typology of Voluntary Associations" American Sociological Review 2h (February): 22-29. Gordon Milton and Charles Anderson 19 "The Blue Collar Worker at Leisure" in Arthur Shostsk and William Gomberg (eds.) Blue-Collar World. 33ng Cliffs, New Jersey: Prentice-Hall. 11h Ragedorn, Robert and Sanford Labovits 1963 "Participation in Community Associations by Occupation: A Test of Three Theories" American Sociological Review 33: 272-282. Hausknecht, Murray 1962 The Joiners. New Jersey: The Bedminster Press. Hausknecht, Murray 1961: "The Blue-Collar Joiner" in Arthur mostak and William Comberg (eds.) Blue-Collar World. Inglewood Cliffs, New Jersey: Prentice-Hall. Hyman, Herbert and Charles Wright 1971 "Trends in Voluntary Association Membership of American Adults: Replication Based on Secondary Analysis of National Sample Surveys" American Sociological Review 36 (April): 191-206. Inkeles, Alex 1960 "Industrial Man: The Relation of Status to kperience, Perception and Value" American Journal of Sociology 66 (July): 1-31. Lipset, Seymour Martin, Martin A. Trow, and James 3. Coleman 1956 Union Democracy. Olencoe, Ill.: The Free Press. Meissner, Martin 1971 "The ‘Long Arm of the Job: A Stu of Work and Leisure" Industrial Relations 10 (October : 239-260. Miller, 8.1!. and Frank Rieesman 1961 "The Working Class aibculture: A New View" Social Problems 9 (Sinner): 86-97. more, Wilbert E. 1965 The Impact of Industry. mglewood Cliffs, New Jersey: Prentice-Hall. Rose, Arnold 1956 Sociology: The Study of Human Relations. New York: Alfred A. Mp1. Ross, Aileen 19511 "Philanthropic Activity and the Business Career". Social Forces 33 (March). Seaman, Melvin 1967 "On the Personal Consequences of Alienation in Work" American Sociological Review 32 (April): 273-385. aissman, Marvin B. 1959 "The Isolated Nuclear Family: Fact or Fiction" Social Problems 115 Sussmsn, Marvin and Lee Burchinal 1962 "Kin Family Network: Unheralded Structure in Current Conceptual- isations of Family Functioning" Marriage and Family Living 2h (August): 231-2140. Teele, James 1962 "Measures of Social Participation" Social Problems 10 (Sumner): 31-39e Walker, Charles R. and Robert H. Guest 1952 The Man on the Assembly Line. Cambridge, Massachusetts: Harvard University Press. Wilanslq , Harold E I 1961 "Orderly Careers and Social Participation: The Impact of Work -- History mSocisl Integration in the Middle Masses" American 'r' Sociological Review 26 (higust): 521-539. Wirth, Louis 1938 '(‘Urbanism as a Way of Life" American Journal of Sociology 1:1: 5 j Jilly e g 5.:- HICHIGQN STQTE UNIV. L IBRQRIES II \ ll mill M !II\ "I “WI W II” N (II “N! ||>|1 m W m 1 ll 31293105232429