HiERARCHlCAL smus AND ATTITUDES TOWARD INTERORGANIZATIDNAL COOPERATION mo coonnmmom A COMPARATIVE ANALYSIS Thesisfiorhthe Degree of M; A. MICHIGAN STATE UNIVERSITY WEI-YUAN WERNER ACHENG 1975 ' Mllfllflflzllfllfllivlflllfljflfll”MIME! _ 4' 1, amSmG BY 7 MIKE & SUNS’ . 300K BINDERY INC. 3 =9'2."£fi§'.§955§ ' 41 Att What COO! with blOC Atti admi: “Orkq Orgar Inizec' SiOna inDEr Major ABSTRACT HIERARCHICAL STATUS AND ATTITUDES TOWARD INTERORGANIZATIONAL COOPERATION AND COORDINATION: A COMPARATIVE ANALYSIS BY Wei-yuan Werner Cheng This thesis is part of a larger project studying 41 welfare agencies in a medium-sized Midwest city. Attitudes of 474 respondents were measured by two blocks of questions.. In one block, respondents were asked to what extent ten obstacles preventing cooperation and coordination among agencies. In the other, they were asked to what extent they would like to cooperate and coordinate with other agencies in eleven areas of service. Each block of questions was further grouped into three scales. Attitudes of personnel in different hierarchical statuses, administrator, professional, semiprofessional, and clerical worker, and in agencies in terms of seven dichotomized organizational properties were compared. Seven dichoto- mizedorganizational properties used were: size, profes- sionalization, technology, interaction (with other agencies), inner communication, formalization, and centralization. Major hypotheses tested were: professionals were more lj CC 50! mun era and fav: and dete diff Seve ShOW( cles and ‘ than Wei-yuan Werner Cheng likely to have favorable attitude toward interorganizational cooperation and coordination than clerical workers, admin- istrators and semiprofessionals were in between. Personnel in agencies of smaller size, higher professionalization, technology, inner communication, lower formalization and centralization, and higher interaction were more likely to support cooperation with other agencies than those in agencies of larger size, lower professionalization, technology, interaction, inner communication, and high formalization and centralization. It was assumed that per- sonnel in agencies of larger size, higher formalization and centralization, lower professionalization, inner com— munication, and technology would be under stronger bureau- cratic control which lead to departmentalism, timidity, and conservativeness which might impede the creation of favorable attitudes toward interorganizational c00peration and coordination. Two way analysis of variance was employed to determine differences between attitudes of personnel of different statuses on the one hand and of agencies of the seven dichotomized properties on the other hand. Results showed that professionals were less likely to find obsta- cles in interorganizational cooperation and coordination and were more willing to see cooperation to larger extent than administrators and semiprofessionals. Clerical Wei-yuan Werner Cheng workers were least to find obstacles, but they were also least willing to cooperate. In order to compare the differences between atti- tudes of personnel in various kinds of agencies, a total of 42 tests of two-way analysis of variance were employed. Nine showed that there were significant differences, eight supported the original hypotheses. Of the seven organi- zational properties, technology, formalization and central- ization were most influential. Although the influences of the other properties were not great, analysis showed that organizational properties have impact on the attitudes of personnel in welfare organizations. HIERARCHICAL STATUS AND ATTITUDES TOWARD INTERORGANIZATIONAL COOPERATION AND COORDINATION: A COMPARATIVE ANALYSIS BY Wei-yuan Werner Cheng A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Sociology 1975 ACKNOWLEDGMENTS This thesis could scarcely have been completed without the assistance of many others. I should first mention Dr. Philip M. Marcus, who first stimulated my interest in interorganizational relations. I would like to thank him for his allowing me to participate in his research project and let me analyze part of his data for this thesis. I thank him for his continuous encouragement and advice throughout my preparation and writing of the thesis. I would also like to thank Dr. William A. Faunce and Dr. Harry Perlstadt for their highly valuable criti- cism, guidance and advice, and their patience in reading my drafts. I appreciate the help from the research group: Margaret Adams, Judith Bates, Clarence Chien, Linda Dammers, Robert Hughes, Reed Kendall, Jacqualine Kron, Kerry McBride, Bernard Offerman, Marc Rideout, Ann Workman Sheldon, and Michael Tinnon, especially Ann Workman Sheldon, the project manager whose generous help is most grateful. I wish to express special appreciation to Tom Nicol who did most of the computer programmings for this thesis and gave suggestions for statistical analyses. ii I would like also to express my thanks to Ching-li Wang for his advice in statistical methods and his continuous concern of my work. To those who often showed their con— cern, like Manuel Lopez and Joe Yankech, I am also thankful. I wish to express my gratitude to the National Science Council, Executive Yuan, Republic of China to offer me a two-year fellowship for graduate study in the United States. I am indebted to Dr. Kwan—hai Lung, former Chair— man of the Department of Sociology, College of Law, National Taiwan University, to recommend me for the fellowship. I also appreciate Professor Chung—mo Han, Dean of the College of Law and Professor Tseng-huei Fann, Chairman of the Department of Sociology, National Taiwan University to arrange a living allowance for my family back in Taiwan during my stay in the United States. To Dr. William A. Faunce and Mr. Morris Wei-hsin Tien, I am grateful for their careful and patient correc- tion of my English in this thesis. Finally, I want to express my deepest appreciation of my mother Mrs. Pei- hsin Chen Cheng who often encouraged me to pursue my graduate study abroad, and my wife Ting-an whose love, understanding and spiritual support enable me to surmount all the impasses I met during the period of my writing this thesis. iii . ‘.J" L- 4|” .v ' .‘b I? TABLE LIST OF TABLES . . . . . . INTRODUCTION . . . . . . . THEORY AND LITERATURE. . . ASSUMPTIONS AND HYPOTHESES SETTING OF THE STUDY . . . The Sample . . . . . . Measurement. . . . . . Method of Analysis . . FINDINGS O O O O O O O O 0 Status and Attitude. . Size and Attitude. . . OF CONTENTS Professionalization and Attitude . Technology and Attitude. . . . . . Interaction and Attitude . . . . . Communication and Attitude . . . . Formalization and Attitude . . . . Centralization and Attitude. . . . DISCUSSION AND CONCLUSION. APPENDIX C O O O O O O O . BIBLIOGRAPHY . . . . . . . iv Page 22 26 26 29 39 44 44 52 55 59 62 67 70 74 80 90 92 Table 10. ll. LIST OF TABLES Table Page 1. Classification of Respondents by Sex and Status . . . . . . . . . . . . . . . . . . . . . 27 2. Classification of Respondents by Educa- tion and Status. . . . . . . . . . . . . . . . . 28 3. Pearson Correlation Coefficients Between Items in Obstacles Preventing COOperation and Coordination . . . . . . . . . . . . . . . . 32 4. Pearson Correlation Coefficients Between Items in the Extent of Cooperation and Coordination . . . . . . . . . . . . . . . . . . 33 5. Pearson Correlation Coefficients Between Items in Job Influences. . . . . . . . . . . . . 36 6. Two-Way Analysis of Variance for Attitudes Toward Interorganizational C00peration and Coordination (Hypothetical Data) . . . . . . . . 41 7. Average Scores and One Way Analysis of Variance on Influences on Job for Staffs Of Different Statuses. 0’ o o o o o o o o o o o o 49 8. Pearson Correlation Coefficients Between Job Influences and Attitudes Toward Inter- organizational Cooperation and Coordination. . . 50 9. Average Scores and Two Way Analysis of Variance on Attitudes of Staffs by Status and Size 0 O O O O C O O O O O O O O O O O O O O 53 10. T Tests on Attitudes Between Staffs of Same Status in Agencies of Large and Small Sizes. . . . . . . . . . . . . . . . . . . 54 11. Average Scores and Two Way Analysis of Variance on Attitudes of Staffs by Status and Agency Professionalization . . . . . . . . . 56 Table 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. Page T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Profes- sionalization. . . . . . . . . . . . . . . . . . 57 Average Scores and Two Way Analysis of Variance on Attitudes of Staffs by Status and Agency Technology. . . . . . . . . . . . . . 61 T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Tech— nologies . . . . . . . . . . . . . . . . . . . . 63 Average Scores and Two Way Analysis of Variance on Attitudes of Staffs by Status and Agency Interaction . . . . . . . . . . . . . 64 T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Inter- action . . . . . . . . . . . . . . . . . . . . . 66 Average Scores and Two Way Analysis of Variance on Attitudes of Staffs by Status and Agency Communication . . . . . . . . . . . . 68 T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Communi- cation . . . . . . . . . . . . . . . . . . . . . 69 Average Scores and Two Way Analysis of Variance on Attitudes of Staffs by Status and Agency Formalization . . . . . . . . . . . . 72 T Tests on Attitudes Between Staffs for Same Status in Agencies of High and Low Formali— zation . . . . . . . . . . . . . . . . . . . . . 73 Average Scores and Two Way Analysis of Variance on Attitudes of Staffs by Status and Agency Centralization. . . . . . . . . . . . 75 T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Central- ization. O I O O O O O O O O O O O O O O O O O 0 76 Pearson Correlation Coefficients for Organizational Properties. . . . . . . . . . . . 78 Chi Square for Organizational Properties . . . . 79 vi INTRODUCTION In recent years, sociologists have had a growing interest in the study of interorganizational relations. Organizations within a community are interacting by either cooperation or competition. Their relations can be pre- sented by an "organization set," which is composed of a network of organizations in interaction with the focal organization or a group of organizations (Even, 1966). In order to compete for limited resources, gain larger profit, provide better services, or use resources more properly, they may compete, cooperate or form coalitions. These phenomena have been described by many writers on inter- organizational relations (e.g., Levine and White, 1961; Litwak and Hylton, 1962; Warren, 1967; Marcus et al., 1974a). Interorganizational c00peration among welfare organizations is not new, but has become more common since the beginning of the war on poverty in the 19603, when, to solve complicated problems, joint efforts were needed, and were encouraged and pressured by governments of all levels and local communities. Other reasons for increasing cooperation and corrdination among welfare agencies are r1, t4 the seeking of additional effectiveness and efficient utilization of resources, increasing accountability, improving service to clients, avoiding duplication and fragmentation, etc. Yet, these efforts have not been very successful (Marcus et al., 1974b). Attitudes of wel— fare personnel toward these joint efforts could be criti- cal. A study on the attitude of welfare personnel toward interorganizational cooperation and coordination would benefit the clients, welfare workers, as well as welfare sponsors and taxpayers. It is believed that attitude could be influenced by personalities, social and cultural factors, and also by organizational structure. There are some criticisms that the structural approach in the study of organizations ignores individual factors (Argyris, 1972), studies syn— thesizing the social-psychological and structural approaches are needed. A few sociologists (e.g., Blau, 1960; Aiken and Hage, 1966; Blauner, 1964) deal with the two approaches simultaneously. This study dealt with attitudes of staffs and organizational structure in welfare agencies. Organi- zations of similar structure were treated as a whole, they were not dealt with individually. This thesis tried to find how organizational properties and personal orienta- tions influence the achieving of favorable attitudes toward interorganizational c00peration and coordination. THEORY AND LITERATURE W. I. Thomas (1966:258—9) said that attitude is "a process of individual consciousness which determines real or possible activity of the individual in the social world." He said that attitude of individual and prevailing attitude of social organization and culture (he called "value") are mutually influenced. An individual's attitude comes from three sources: wants, group affiliations, and personality (Krech et al., l962:Ch. 6). Attitude can be seen from at least two dimensions, its direction and its intensity (Riley et al., 1954; Scott, 1968). A favorable attitude toward interorganizational cooperation and coor— dination is firstly willingness to cooperate and coor- dinate with other organizations and secondly willingness to see cooperation and coordination to a larger extent. Interorganizational cooperation and coordination might be initiated from inside and outside organizations. Govern- ments and community leaders might pressure welfare agencies to cooperate and coordinate in order to function more efficiently and effectively. The same demands might be also initiated by agencies themselves. Litwak and Hylton (1962) point out that agencies will develop and continue coordination if they are partially interdependent (if they are totally interdependent, they would merge) and are aware of this interdependence. Resources of each agency are limited, exchange of resources between two agencies will benefit both to fulfill their goals. Levine and White (1961) say Organizational exchange is any voluntary activity between two organizations which has consequences, actual or anticipated, for the realization of their respective goals or objectives. Organizational exchange is based on organizational inter- dependence. But not all welfare agencies are aware of their interdependence. Still others are unwilling to cooperate and coordinate, because they have conflicting values or because the demands of efficiency require organi— zational specialization. For example, some national agencies, like The American Cancer Society, resist affili- ation in local community chest and Catholic welfare agen- cies exclude planned parenthood agencies (Litwak and Hylton, 1962). Besides, organizations want to own a domain and acquire power in interorganizational relations, they also attempt to buffer themselves from external pres- sures (Thompson, 1967:20). All these lead to "department— alism" of organizations. ”Departmentalism" in organiza— tions is a condition in which the personnel of organizations or subunits within organizations insist on their viewpoints which are based on the interest of their own units. Departmentalism becomes an obstacle to cooperation and coordination between different agencies. Each agency (or subunit in an agency) is competing for resources, commod- ities and reward. Schein describes the results of unit competition as an obstacle for group interaction, he notes, The fundamental problem of intergroup competition is the conflict of goals and the breakdown of interaction and communication between the groups; this breakdown in turn permits and stimulates perceptual distortion and mutual negative stereotyping (Schein, 1970:99). Although Schein mentions only intergroup relations, his argument could also refer to interorganizational relations. These conflicting values are derived from different orien- tations of personnel and the structure of organizations. Blau and Scott (1962:64) delineate the concepts of "professional orientation," and "bureaucratic orientation." Personnel of professional orientation are those who retain their identification with their professional group, are highly committed to their professional skills and look for social support from professional colleagues outside and inside the organization. Those of bureaucratic orientation come to identify with the very organization by which they have been employed and they are conforming to its programs and procedures. They are more concerned with gaining the approval of administrative superiors inside the organiza- tion than that of professional colleagues outside. The coordination between personnel of different orientation faces a fundamental gap. When the one of professional orientation asks for a shortcut to complete a task, the one rel eff ski tio: 1‘91; mp} non. org; mode comb Prof one of bureaucratic orientation insists that official pro— cedures should be followed. It is believed that personnel of professional orientation will have more favorable atti- tude toward interorganizational cooperation and coordina- tion, but personnel of bureaucratic orientation, even if they are not against interorganizational cooperation and coordination, will cooperate and coordinate in the less extent. However, it does not mean bureaucratic orienta- tion is absolutely bad to professional organizations. Udy (1959b) notes that organizations have rational and bureaucratic characteristics which are negatively associ- ated. Litwak (1961) compares two models of organizations, one is weber's legal-rational model, another is the human- relation model. He notes that Weber's model Would be more efficient for organizations which emphasize technical skills (he calls "traditional knowledge"), secondary rela- tions, rules and deal with uniform tasks; the human- relation model is more efficient for organizations which emphasize social skills, primary relations and deal with non-uniform tasks (e.g., research). But, many professional organizations, Litwak points out, should adopt a third model, he calls "professional bureaucracy," that is a combination of the above two models. He notes that many professional organizations should deal with both uniform and non-uniform tasks and they need both technical and social skills. It might also say the adaptation and goal gr no st ac mi. St. to for the merr if leg 1(er achievement functions of organization need to be relatively more rational and flexible, and its integration and pattern maintenance functions need to be relatively more bureau- cratic and stable. Professional orientation first was acquired from professional education and training. But, orientation might be changed when personnel hold different jobs and under verious organizational climates. Personnel of dif- ferent hierarchical statuses are usually holding different orientations. Homans (1961:Ch. 16) argues that personnel of different hierarchical statuses in organizations or groups have different patterns of conformity to group norms. _He classifies statuses into three levels: high status, middle status, and.the established low status. He actually mentions awfourth level, the non-established low status which has the same behavior pattern as those of the middle status. Homans states that personnel of the middle status (and the non-established low status) are more likely to conform to the norms of the group. He notes that mem- bers of established low status particularly do not con- form to group norms, because they get little reward from the group, they think they are not accepted by their fellow members and are more alienated. They would leave the group if they have any chance. Those in high status also show less conformity but for different reasons. In order to keep their high status, they should show their extraordinary abilities to provide valuable services to others. Other members in return would allow them some leeway in minor things. The members of middle status are those who are actually in the middle status or only believe themselves to be so, those in the non-established low status still see chances to be accepted by fellow members and to climb the hierarchical ladder, therefore, they are those who believe they are in the middle status. Whether they are actually in the middle or only believe that they are, they are more likely to conform to group norms. Homans (1961:352-3) states, the middle man can less easily stand the loss of his little all . . . . At any rate, if the man of middle status conforms to the group's judgment and the group turns out to be right, his position as an accepted member is confirmed, and it needs confirming more than does that of an upper-status member. If he conforms and the group is wrong, he does not lose anything; he has only been a boob with the rest, who are in no condition to turn on him. If, on the other hand, he refused to conform, and the group's judgment turns out to be right, he may really hurt himself in status; he is not so far from the bottom that a single misstep will not bring him appreciably closer to it. And if, finally, he refuses to conform, the group's judgment turns out to be wrong, and he, accordingly, is right, he will indeed gain status; but it will take more than one such achievement to get him to the top. Therefore, the middle man plays safe. On the other hand, the members at the t0p have little to gain in conformity, and the members in the bottom have little to lose in non- conformity, therefore, both of them show less conforming behaviors (Homans, 1961:353-4). Patterns of status and conformity would be dif- ferent in various organizations, especially those hiring professionals. The professionals in organizations seem to be a special interest group. Dalton's (1950) famous report of the conflict between staff and line illustrates that the professional, whose status is in the middle of organizations, has larger authority than do the middle line managers in some circumstances. The sources of the authority of the professionals, as mentioned by Greenwood (1957), are (l) a basis of systematic theory, (2) profes- sional authority recognized by the clientele, (3) community sanction of this professional authority (mandate), (4) an ethical code, and (5) a professional culture sustained by formal professional associations. While some established professionals enjoy their discretion, some do not. Anyway, professional authority and bureaucratic authority sometimes conflict. Scott (1966) points out four areas of role con— flict associated with the differences of professional con- trol versus bureaucratic control: (1) the professional's resistance to bureaucratic rules; (2) the professional's rejection of bureaucratic standards, (3) the professional's resistance to bureaucratic supervision; and (4) the pro- fessional's conditional loyalty to the bureaucracy. This is especially so in service organizations, where the ability of service organizations to neutralize the personal ideologies of their members is limited. Staffs of different 10 organizational locations, occupational and professional trainings, and social backgrounds are likely to develOp different ideologies toward their jobs (Hasenfeld and English, 1974:11). Consequently, in service organizations in general, and welfare organizations in particular, administrators, professionals, semiprofessionals, and clerical staffs would show various patterns of attitudes because of differences in their training, responsibilities and backgrounds. Welfare agencies rely heavily on semi— professionals. They are "characterized by either lack of a systematic theoretical knowledge base; lack of monopoly over their field of practice; or fragmented association" (Hasenfeld and English, 1974:20). The semiprofessionals are ”locked" into organizations, and their professional career patterns coincide with their organizational careers (Hasenfeld and English, 1974:413). As a result they are more likely to be under bureaucratic control and are more compliant to bureaucratic authorities and procedures, they are like the middle men described by Homans. On the other hand, the professionals, who have their own standards of conduct and whose self-esteem and achievement are not totally determined by the evaluation of their agencies, are less conforming to bureaucratic authorities. The clerical staffs are the men of lowest status in agencies, they are also less conforming to authorities. It is expected that they are less willing to cooperate and 11 coordinate but will use official regulations as excuses to avoid additional loads to their jobs. The administrators, like their semiprofessional colleagues, are marginal men if they are professional themselves. It is difficult for them to follow their professional ethics under bureaucratic demands from above. Hawkes (1961) notes that administrators in psychiatric hospitals have to deal with three kinds of relationships: they are firstly the relationships with political, legal and financial supports, secondly the relationships with technical and professional supports, and thirdly the rela- tionships with the problem population. Although Hawkes studies chief executives and not administrators in general, one would suppose that the administrators are multi- orientated. The administrators also are more committed to administrative jobs, focusing on the maintenance of internal cohesion.and consensus rather than on the pursuit of occupational specialities, they internalize more the bureaucratic regulations and procedures of the organization. They are more dedicated and loyal to the organization and are "the homeguard" of their organization (Gouldner, 1957: 444-9). In a study of public assistance workers and super- visors in 12 public assistance agencies and a group of professional social workers, Thomas (1959) discovered that, of seven items concerning professional ethics, the professional social workers are more committed to the 12 ethics of the profession, the public assistance workers are least committed to the ethics, and the supervisors are between the two. Thomas's findings support my previous contention that the administrators are the marginal men. However, orientation toward professional ethics might also be due to seniority in the profession in addition to status in the hierarchy. Blau (1961) reports that, for the social workers of less than 3-year experience, the more secure they are the more they are service-oriented. But this is slightly reversed for workers with more than three-year experience. Blau notes that they are "too secure had no incentive to do more than their official duties required" (Blau, 1961:363). Both W. I. Thomas (1966:258-9) and Blau (1960) state that the prevailing attitude of a group could affect the attitudes of its members. The more professionals have been hired in an agency (that is the higher the profes- sionalization), the more professionally orientated the whole personnel would be. However, professionals them- selves are not all professionally orientated. In some circumstances the professional lose their discretion and under strong bureaucratic control, especially when they are small in number and are in less influential positions in their agencies (Vinter, 1963). This is not unusual in the field of social welfare, since social work is not an established profession. Greenwood (1957) mentions that 25 at at 0t te. CU. 13 social workers are still struggling for recognition as professionals and not all social workers are enthusiastic about the professionalization of social work. High for- malization and centralization in organizations are two of such circumstances. Formalization, according to Aiken and Hage (1966) is the degree of work standardization and the amount of deviation that is allowed from standards. Pugh et a1. (1963) note that formalization includes state- ments and Operations of procedures, rules, roles which are related to decision seeking, conveying of decision and instruction, and conveying of information. The work of the established professional is judged and audited by pro- fessional standards, it is self-regulated and less formal- ized by the employing organization, but the work of not established professional is judged and audited by the employing organization (see Hall, l975:Ch. 4). When officials should follow regulations strictly, they adapt to these regulations by following established routines to check whether all regulations are followed (Merton, 1968: 255). Merton (1968:255) notes that bureaucratic officials are more likely to resist change in established routines, at least those changes which are felt to be imposed by others. Personnel who are unwilling to change and insis- tent on established routines (or red-tape) are more diffi- cult to c00perate and coordinate. Merton also points out that personnel who are over concerned with strict adherence 14 to regulation would induce timidity and conservatism. They are difficult to deal with interorganizational cooperation and coordination because instead of solving problems on the spot, they often submit their cases to superiors for approval which impede the effectiveness and efficiency of interorganizational c00peration. Centralization refers to how decision-making system is concentrated in t0p echelons. Hage and Aiken (1967a; Aiken and Hage, 1966, 1968) define centralization as the degree of participation in decision making in organizations. Personnel in agencies of higher centralization have less discretion. They are under higher bureaucratic control, as those in agencies of high formalization. Although Blau (1970) finds formalization and centralization are not necessarily related, and his findings are opposite to those of Aiken and Hage's, Blau's study was on government financial agencies which might have less interorganiza- tional relations (measured by number of joint programs) than those of welfare agencies studied by Hage and Aiken. The findings of Hage and Aiken are more relevant to this study. Whether they are related or not, formalization and centralization limit discretion of the professional. In some agencies, t0p administrators make most decisions and they do not want to decentralize their power, in that case, personnel become discouraged to do more than they are required; they lose their own initiative. Even when 15 they are ordered to cooperate with other agencies, without discretion and self-initiative, they would be less inter- ested in interorganizational c00peration and coordination. They might be originally in favor of interorganizational cooperation and coordination but later feel impotent and become apathetic. Communication could benefit the achieving of con- sensus within organizations. Etzioni (1961:138-41) points out in normative organizations (e.g., professional organi- zations),downward expressive communication is emphasized in order to establish expected attitudes among their per- sonnel. Communication would improve c00peration and coor- dination within an agency or with other agencies. For achieving better cooperation and coordination between any two units, mutual understanding between their personnel, mutual goals and mutually acceptable procedures must be set up. There are several types of coordination and cooperation mechanisms. Cooperation and coordination can be achieved through either formal or informal channels. Under some circumstances, officials of the units concerned hold coordinating meetings to solve common issues face to face instead of passing written communications. If the meetings become relatively regular with appointed staffs, it turns out to be a cross—functional group. A cross- functional group is a unit of an organization or a number of organizations that has been organized to fulfill special 16 goals. It is usually called a team (for special project), ad hoc committee, or task force. The cross—functional groups are composed of personnel of different units neces- sary to fulfill the task (or tasks), members of the cross- functional group are able to solve complex problems con- cerning different units within the cross-functional group rather than through all levels in the hierarchy of the units concerned. It is a device to meet the needs of modern organizations facing fast changing and complicated goals which could not be reached without combined efforts. When the establishment of a cross-functional group is not necessary, coordination can also be achieved through liaison offices or officials. Personnel in cross- functional groups would have a better understanding of interorganizational cooperation and coordination. Aiken and Hage (1968) find that internal communication (measured by number of committee and committee meetings) is posi- tively associated with interorganizational cooperation. Personnel in such cross-functional groups might have a favorable attitude toward interorganizational cooperation and coordination. Although this favorable attitude need not be a favorable attitude toward their counterparts in other organizations. Conflict and competition can also rise within the cross-functional groups (Marcus et al., 1974a). 17 Size is another important property of organization. Although studies on the impact of size of organizations are far from being conclusive and Hall et a1. (1967) argue that size is not the most influential factor, their study was based on different types of organizations, the influence of size might not be prominent when other factors (e.g., technology) were not controlled (Hall, 1972:119). Thomas (1959) discovered that in small welfare agencies, personnel are more committed to professional ethics, but he does not think that size, per se, is influential. Since large agencies are mostly located in large urban communities and small agencies are mostly located in small urban and rural communities, he states that community types rather than size is the major factor influencing the attitude of wel- fare personnel. However, Blau (1968) and Meyer (1968a) report size is positively correlated with hierarchical differentiation, which would reduce social interaction among subunits and distort the error-correcting function (Blau and Scott, 1962:167), and would be disadvantageous to c00peration and coordination. Besides, coordination between staffs of lower levels in two units is more eco- nomical with regard to time in agencies with fewer hier- archical levels than in agencies with more hierarchical levels. Hierarchical differentiation, therefore, would increase the difficulty in c00peration and coordination within and between organizations. Size is also positively 18 related to the degree of bureaucratization (Hall et al., 1966) and formalization (Meyer, 1968b). Both, according to our previous discussions, are not favorable climates for positive attitudes to interorganizational cooperation and coordination. From a study of 254 government financial departments, Blau (1968) discovered that the number of hierarchical levels and the proportion of managers is F? higher among professional staffs. Their explanation of this unexpected finding is that the higher proportion of managers among professional staffs would improve coor- fig dination and transfer information to the top. In pro- fessional organizations (e.g., research and development units), the professionals are usually rather loosely supervised (Hall, 1972:142) and the proportion should be low rather than high. Because Blau et al.'s findings were derived from public personnel agencies and public financial agencies, which are rather bureaucratic organi- zations, the generalization of their findings is limited. Their argument might not apply to other professional organizations. Their findings could also be explained as follows: the high proportion of managers among profes- sional staffs was set up for the sake of transferring demands from the top, and for closer supervision, and meeting problems caused by increasing division of labor and specialization. 19 Technology in welfare agencies defined in this study is the modes and number of services. "Technology" is related to "complexity," an important prOperty of organization. "Complexity" is measured by the number of specialities, professional training and professional activities of personnel in Hage and Aiken's (1967a) study. It is measured by the division of labor, differentiation, and dispersion of organization in Hall's (1967) study. The complexity in Hage and Aiken's measurement could be labeled as "professionalization." The complexity of Hall's measurement is the division of labor without concerning the technology of organization. Perrow (1967) and Thompson (1967) see technology as critical attribution to the structure of organization. Thompson (1967:54-55) states that technology also attributes to the types of coordination in organization. Therefore, the complexity of organization should include both technology and division of labor. Since the inconsistency of the use of the term "complexity," I would use technology instead. According to Hasenfeld and English (l974:5), there are two kinds of technology for welfare agencies-- the people processing and the people changing (treatment). In the people processing agencies, the core technology of the agencies is the classification and disposition of clients, which are based on agency or government regula- tions, discretion for professionals in these agencies is 20 limited (Hasenfeld, 1972), and these agencies are more bureaucratically orientated. The people changing or treatment agencies are responsible for changing people. These agencies are concerned with effecting new and dif— fuse models of behavior, new self-images and personalities, and consequently, these agencies rely very much on profes- sionals to achieve their goals and are rather profes- sionally orientated. Professionals have much more dis- cretion in these agencies (Vinter, 1963). Besides, in people changing agencies, the professional will interact with his colleagues in other organizations, thus yielding interorganizational linkages (Hall, 1972:318). Not all agencies of professional orientation support interorgani- zational cooperation and coordination. The number of services provided by agencies is also critical to inter- organizational relations. When welfare agencies provide multiple services, they depend more on environment. The more services they provide, the more resources they need. They are more inclined to interorganizational cooperation and coordination. But, when they provide a rather specialized service, they might have a "fixed market" for themselves and have no intention to work with other agencies, even though they are professionally orientated. Litwak and Hylton (1962) have taken the American Cancer Society as an example. From the preceding discussions, er 0t ti pr or. £0: be les and son ser to bur ate Pro‘ inte thej 21 both number of services and kinds of services determine whether an agency would work with others or not. In conclusion, attitudes toward interorganizational cooperation and coordination are affected by the orienta- tions of their personnel. Personnel of professional ori- entation are more likely to cooperate and coordinate with other agencies. And, personnel of bureaucratic orienta- tion are less likely to do so. Although the orientation of personnel could be determined by previous training and present job, organizational structures can also affect the orientation of personnel. Personnel in agencies of high formalization, high centralization, and larger size would be more bureaucratically orientated, and consequently, be less enthusiastic toward interorganizational c00peration and coordination. Other structural properties also have some influences. Personnel in agencies concentrating on services of people processing and having less opportunities to participate in cross-functional groups are also more bureaucratically orientated and are less likely to cooper- ate and coordinate with other agencies. Agencies that provide few services or specialized service have also less intention to work with others despite the orientation of their personnel and the agencies. ASSUMPTIONS AND HYPOTHESES This study examined the attitudes of personnel in welfare agencies toward interorganizational cooperation and coordination. Their hierarchical status and seven organizational properties were used as independent vari- ables. The seven organizational properties were size, professionalization, technology, communication, formali- zation, centralization and interaction (with other organi- zations). Interaction was added because it is interesting to find whether agencies which interact with others more often would create favorable attitudes toward interorgani- zational cooperation and coordination. Interaction was also treated as an independent variable although it might be argued that it might be an intervening variable. Two kinds of attitudes were treated as dependent variables-~willingness to cooperate and coordinate with others and the extent of cooperation and coordination respondents would like to see. The two kinds were expected to measure the "direction" of attitude and the ”intensity" of attitude respectively (Riley et al., 1954; Scott, 1968). It is assumed that hierarchical status and organizational structures which lead to bureaucratic 22 23 orientation would also yield negative attitudes toward interorganizational cooperation and coordination. Hier- archical status and organizational structures which lead to professional orientation would also create positive attitudes toward interorganizational cooperation and coor- dination. Organizational properties which facilitate cooperation and coordination would also be good climates for positive attitudes toward interorganizational coopera- tion and coordination. Nevertheless, attitude might also influence organizational properties. Lefton and Rosenger (1966) report that orientation toward clients would affect organizational structure. When an agency becomes more client-orientated, some arrangements (e.g., new subunits, new services, etc.) would be made in order to provide better services for clients. Although attitude and organi— zational properties might be mutually influenced, in this study only one-way of the influence was studied. The influence of attitudes on organizational properties was ignored. I would also concentrate my concern mainly on the attitudes prevailing in organizations and in a partic- ular status as a whole. Hypotheses that were tested were: 1. Professionals are more in favor of interorganiza- tional cooperation and coordination. Clerical staffs are least in favor of interorganizational 24 cooperation and coordination. The administrators and semiprofessionals hold attitudes in between. The personnel in agencies of smaller size are more in favor of interorganizational cooperation and coordination than those in agencies of larger size. The personnel in agencies of higher professional- ization (higher proportion of professionals in agencies) are more in favor of interorganizational cooperation and coordination than those in agencies of lower professionalization. Personnel in agencies of higher technology (pro- viding more services and/or emphasizing treatment) are more in favor of interorganizational coopera- tion and coordination than those in agencies of lower technology (providing less services and/or emphasizing people processing). Personnel in agencies of higher communication (higher ratio of personnel participating in cross- functional groups or units) are more in favor of interorganizational c00peration and coordination than those in agencies of lower communication. Personnel in agencies of higher formalization (organizational rules and procedures are more influential on jobs) are less in favor of inter- organizational cooperation and coordination than those in agencies of lower formalization. 25 Personnel in agencies of higher centralization (participating less in organizational decision- making) are less in favor of interorganizational cooperation and coordination. Personnel in agencies of higher interaction (with other organizations) are more in favor of inter- organizational cooperation and coordination than those in agencies of lower interaction. {:3 ‘— -.l_ - E3 ‘ ’1- SETTING OF THE STUDY The Sample This study was part of a larger study of 41 welfare agencies in a medium-sized city in the Midwest (see Marcus et al., 1974a, 1974b). Only a small portion of the ques- tions were examined. Personal factors, except the hier- archical status of respondents, were ignored. There were a total of 474 respondents. They were divided into four hierarchical statuses--administrator, professional, semi- professional, and clerical staff. Directors of agencies in the sample were not included in this study. Criteria for the classification were based upon the characteristics of the jobs and the education of the job occupants. Of the total 474 respondents, 103 were adminis- trators, 182 were professionals, 139 were semiprofessionals, and 46 were clerical workers. Four respondents did not give their job titles and descriptions and, hence, were discarded from further analysis. Since the respondents were not selected at random, the clerical staff seemed to be underrepresented. There were 153 male and 300 female respondents. Of the male, 50 were administrators, 63 were professionals, 38 were semiprofessionals, and 2 were 26 27 clerical. Of the female, 50 were administrators, 116 were' professionals, 94 were semiprofessionals, and 40 were clerical. Seventeen did not give their sex (see Table 1). Of the administrators, 29 percent hold advanced degrees, 50 percent have bachelor degrees; of the professionals, 38 percent have advanced degrees and 57 percent have bachelor degrees. Of the semiprofessionals, 18 percent have bachelor degrees, 45 percent have some college edu- cation. Of the clerical staff, 15 percent have bachelor degrees, 33 percent have some college education (Table 2). There were 80 percent whites and 8 percent blacks and 11 percent others. Most respondents were mature adults (only 18 percent were under 25), but most were low in seniority in the social services (47 percent less than two years, 39 percent 2-5 years). Table l.--Classification of ReSpondents by Sex and Status.* Adminis- Profes- Semipro- Clerical Total trator sional fessional Staff Male 50 63 38 2 153 Female 50 116 94 40 300 Total 453 *Seventeen respondents did not give their sex. 28 one lwooavme Awooaomma Awooavwma lmooavmoa Hence Asa Ce Ass is led Cm Awe is csoaxco lame cow imam ems Ame Co 1mm Cm mmmH can Hooeom swam lwmm Vma Awme CNS Ame is lwma ems mmmmflaoo meow AAmH is Amea vmm Amnm .qoa Amom me .m.m a .«.m Awe Co Asa Cm Awmm Cam ream Com mmmumme omoam>na mHOB mmmum HMGOHmmmm HMGOMm HOHMHU H HMUAHmHO loumwfiom Immmoum Imflcweoc .msumuw one coflumosom an mucmpcommmm mo aoflueoAMAmmeHoun.~ magma 29 Measurement The attitudes of the staffs were measured by two blocks of questions, one concerns the extent of each of the obstacles to c00peration and coordination of services and programs among agencies, while the other concerns the extent of cooperation and coordination with other agencies the respondents would like to see. The first block of questions included 15 items and the second block included 14 items. Respondents were asked to check one of the five fixed-alternatives of each item from "very great extent," "great extent," “some extent," "slight extent," to "no extent." Respondents were given scores from five points to one point for each item. The reason for taking the two blocks of questions into account is that: since respondents who scored high in the first block of questions saw more obstacles or saw obstacles to a larger extent to inter- organizational c00peration and coordination, they might be less likely to c00perate and coordinate with other agen- cies. They were either discouraged by the obstacles or might only use the obstacles as excuses to disguise their unwillingness. It was also believed that respondents who scored high in the second block of questions might be more likely to cooperate and coordinate to a larger extent. It was expected that the 15 items of the first block of questions would measure the "direction" of attitude and 30 the 14 items of the second block of questions would measure the "intensity" of attitude. It was found that the items in the same block of questions were not highly correlated. If one took the average scores as respondents' attitude scores, the results would be misleading, since scores on the items would be mutually eliminated. Scott (1968) notes that attitudes have multiple dimensions. It is difficult to find the attitude in a single scale. Hence, items in each of the blocks of questions which were relatively highly corre- lated and theoretically were mutually related were clustered together. Items of the questions in the first block, con- cerning the obstacles for c00peration and coordination, were grouped into three kinds of obstacles, they were "cost too high," "lack of control," and "don't need." Items of the second block of questions, concerning the extent of cooperation and coordination, were also clustered into three areas of cooperation and coordination, they were "case cooperation," "resource sharing," and "informa- tion exchange." Items which were not suitable to any cluster were discarded. Items of the six clusters of the two blocks of questions are presented below: 1. To what extent does each of the following prevent coordination of services and programs among agencies? 31 (1) Cost too high la too much time lb cost too high 1c fund not available (2) Lack of control 2a cannot get c00peration 2b no way to sponsor 2c difficult for staff working together 2d lack control of others' staff (3) Don't need 3a clients don't need 3b don't help major job 3c other agencies don't need 2. For each of the following to what extent would you like to see more cooperation and coordination with other agencies? (1) Case cooperation la share information about case lb share intake (2) Resource sharing 2a share facilities 2b joint-use of equipment 2c share staff 2d joint operation of program (3) Information exchange 3a joint public relations 3b share planning 3c share information about resources 3d more referrals from other agencies 3e joint staff training The correlations of items in each of the two blocks of questions are shown in Table 3 and Table 4. Respondents were also asked to check one of five fixed-alternatives in a block of questions including 21 32 oo.a poms u.coc moaocmmm Hmnuo 0m mmq oo.H no“ “ohms mam: u.coo mm «m4 mmq oo.H new: 6.:06 muamflao mm SH. ms. SH. oo.H «mmum mo Houueoo gums am am. am. ma. mm. oo.H umeummou egos 0» mmmum mom uHsOAMMAe om mm. mmq «H. mm. ms. oo.H newcomm on mm; on an ha. MMJ ea. mma .Mfld mm. oo.H coHumummooo uom uoccmo Mm No. ma. ma. mm. ma. mm. mm4 oo.H magmaflm>m you seem Hmaommm 0H ma. mm. SH. ma. km. SN. mMJ .qu oo.a as“: 00» umoo nH mo. om. mo. Hm. mm. mm. «N. mm. qu oo.H mafia some 00» ma Um am Mm 6N 0N am MN OH DH Md .cowumcwnuooo can cowumummooo mcwuco>oum moaomumno CH mEouH com3umm mucowofimmmou cowumamuuou comumomnl.m manna 33 oo.H wcflchuu wwmum ucfiofl om mv. oo.H moflocowm worse on mamuummou once cm 0v. mmd oo.H moUHDOmmH 950nm cowumsu0mcw oumnm om mm. om. mo. oo.H mswccmam mumcm nm S. .3. mwl. Mom... 84 96333 333 650..” mm mv. mm. ov. .mm4 .mm4 oo.H Emugoum mo coaumuwmo ucflon cm me. mm. mm. we. we. .mm4 oo.a wmmum wumnm om mm. mm. vm. av. 0m. «0. co. oo.a ucmsmflsvo «0 oms ucfiofl pm me. mm. mm. 3. me. Mm”. flol. .olnl. 84 33383 3.3.0. mm mm. hv. «m. 0%. mm. 5v. Hv. me. mv. oo.H mxmucfl mumnm QM mm. mv. mv. mv. mm. mm. mm. mm. av. .NMJ oo.H mmmo “scam cowuweuowcw wumnm NH mm pm on an Mm pm ON AN Mm ad ma .GOwuwcwcuoou paw sceumuomooo mo ucmuxm on» Ca mamUH cmm3uom mucowoflmmoou soflumHouuou comumwmll.v manna 34 items concerning to what extent the 21 items influencing the respondents on their jobs. As with the attitude ques- tions, the 21 items were grouped into three clusters and some items were discarded. The three clusters dealt with three kinds of influences: they were "the authorities within agency," "authorities outside agency," and the "influence of the community" to which services were offered. It was expected that professionals and the clerical staff would be less influenced by both authorities and the community. The professionals have more discretion and the clerical workers, low in the hierarchy, do not care about the authorities and the community. But, because the administrators and the semiprofessionals are more bureaucratically orientated, their jobs would be more affected by the three types of influence. Items of the three clusters for job influences are presented below: 1. Influences of authorities within agency la agency rules lb board/commission lc director 1d upper-level official 2. Influences of authorities outside agency 2a financial resources 2b funding organization 2c laws 2d requirements of government funding agencies 35 3. Influences of community 3a clients 3b community groups 3c general public The correlations of items above are shown in Table 5. The seven organizational properties were dichoto— mized for analyses. The 41 agencies were dichotomized into high or low groups according to their scores in the seven properties. They were measured as following: The gigg of an agency is the total number of full- time staff of the agency. Part-time staffs prevail in welfare agencies and they were counted as their full-time equivalent. For example, two half-time or one 1/4 time and one 3/4 time were counted as one full—time staff (Hall et al., 1967). The size of the sample agencies ranged from 5 to 218, agencies with a staff of 25 or more were rated as ”large." There were 19 large and 22 small agencies. Professionalization of an agency is the percentage of professionals among the total staff. The percentage of professionals in the sample agencies ranged from 0 percent to 76 percent. Agencies with 46 percent or more professionals were rated as "high" in professionalization. There were 18 agencies of high and 23 agencies of low pro- fessionalization. 36 oo.a Deanna Hmumcmm 0m de oo.H mmsoum muHcsEEoo Am mm. MMJ oo.H mucoHHo Sm NH. «a. ma. oo.H mmwocmmm mcwwc5m ucmficuwbom mo mummEmuflsvmu om mH.. So. NH. .wmq oo.H msmH SN SH. SH. So. mmd .qu oo.H mcmHoHuHHom S mHmHOHSSo useficum>om om mm4 mm. oH. .mmq So. on. oo.H coHumuHammuo oeHwasS am Hm. SH. mo. «ma mm. «m4 .mmd oo.H mucua0mmu HmHuameHS Sm mH. So. so. Sm. mm. mm. mm. Sm. oo.H HmHOHmHo prmHuummms SH SH. NH. OH. em. mH. HH. mm. mm4 .mm4 oo.H uouumuHS oH SH. SH. So. .mmq SN. mm. «N. am. NMJ. qu. oo.H conmHesoo\Sumon nH So. HH. So. AN. mmq. mH. So. SH. .mmq .mmq .qu oo.H mmHsu soewmm SH on an mm mm um um pm SH SH oH nH SH .moocosamcH non cw mfimuH coozuom mucowoamwmou coauuamuuoo :Omummmll.m wanna 37 Technology of an agency is its scores in the degree of technology which was computed by Sheldon (1974) from a formula of her design. degree of technology = number of service X mode of service mode of service mode of service agency 1 in people processing agency 3 in people changing (treatment) Agencies with a score of six or more were rated as "high" in technology (for people processing and people changing (treatment) agencies, see Hasenfeld and English, l974:5; Hasenfeld, 1972; Vinter, 1963). This is a combination of the division of labor (used by Hall (1967), and Pugh et a1. (1963)) and mode of work (or types of technology used by Vinter (1963), Hasenfeld (1972), Perrow (1967), and Hage and Aiken (1969)). It was assumed that the degree (or complexity) of technology of each service in peOple changing agencies, e.g., family and child consultation, rehabilitation, alcoholism and drug treatment, mental health, etc., is three-times higher than each service of people processing agencies, e.g., public assistance, recreation, public health, etc. Although it seemed to be arbitrary, yet, it was believed that this was a conserva- tive estimation. Compared with mode of work (high in people changing and low in people processing) and number of service (high in five or more services and low in one to four services), 38 technology was highly correlated with mode of work (Gamma = .80), but technology was only moderately corre- lated with number of service (Gamma = .41). Of 17 people changing agencies, only one was rated as low in technology. Of 19 agencies of high technology, 16 were people changing agencies. It would say that the measurement of technology in this study was rather the measurement of mode of work than number of service. The interaction of an agency with other agencies was computed from the number of times the agency was men- tioned by directors of other agencies. Directors were given a list of agency names in the city by the inter- viewers. Directors were asked to give the names of five agencies they interacted with most often in fifteen dif- ferent areas. Only 33 agencies were on the list, there- fore only 33 agencies had scores on interaction. Those which got 35 or more choices were rated as "high" in interaction. There were 17 high and 16 low agencies in interaction frequency. The classifications were a revision of Sheldon's (1974) work, but the major ideas were hers. Communication of an agency was computed from the average scores of its respondents in the total scores of the following two closed-ended questions from the self- administrated questionnaires. 1. How many committees or planning groups you serve within your agency? 39 2. How many committees or planning groups you serve with people from other agencies? if 1-3, 1-3 points 4-8, 4 points no answer, 0 points There were 17 "high" and 19 "low" agencies in communication. Formalization of an agency was computed from the average of its respondents in the scores in the following questions. To what extent are written regulations and procedures followed in your agency? If very great extent 5 points great extent 4 points some extent 3 points slight amount 2 points not at all 1 point no answer 3 points There were 16 "high" and 20 "low" agencies in formalization. Centralization of an agency was computed from the average scores of its respondents on the following three items. In your job how frequently do you do each of the following? 1. staff hiring, promotion. 2. allocation of resources. 3. changing programs. if rarely sometimes points quite often 3 points almost always 4 points no answer 0 point point NH There were 17 "high" and 19 "low" agencies in centrali- zation. Method of Analysis Since the size of the sample was not large, it was unsuitable to handle many independent variables 40 simultaneously. Each attitude toward interorganizational cooperation and coordination was compared in seven tests of two-way analysis of variance. That is, each attitude was compared by staff's status on the one hand and by one organizational property on the other hand in each of the seven tests of two-way analysis of variance. There were six attitudes to be created, hence, there were a total of 42 tests of two-way analysis of variance (see Table 6). The model for the two-way analysis of variance is shown below: Y = B0 + Ble + BZXZ + B3X3 + B4X4 + B5X1X4 + BGXZX4 + B7X3X4 + E Y = Attitude scores if X1 = 1 Professional X1 = 0 others X2 = l Semiprofessional X2 = 0 others X3 = 1 Clerical Staff x3 = 0 others X4 - 1 In agencies which were high in the organiza- tional prOperty treated in the analysis of variance. X4 0 In agencies low in the property. For example, an administrator was in an agency low in the organizational property included in the analysis of vari- ance. The formula for his expected attitude score in one of the six attitudes was: ucmofimwcmwm uoc u .m.z 41 mo. v m« m n .m.c cofluomumucw "Eouuon H n .m.© 3oulcom3umn ”ewe m n .m.p GEsHooncoozumn ”mouo A.ouo .wmoHocnomu .m.z S.~ o.m S.m o.m 30H .eoHumNHHmcon *~.mH ummmoum .muflmv .S.SH o.m o.S S.S 0.4 anm mmHuummoum chowumnwsmmuo Hmcoflmmmm Hmcoam Houmnu 0th meHO IOHQHEwm Immmoum Immcwamfi momsumum .Amumo Hmowumnuommm. coaumchHo U can coflumummoou HmcoHumNHcmmuoumch num3oa moosufluud How mosmwum> mo mammHmcd >m3n0391|.m magma 42 For a professional who was in an agency which was rated as high in the organizational property included in the analy- sis of variance, his expected attitude scores was: Y=BO+B1+B4+B5 It should be noted that the above model is assumed to be linear. Although a more complicated model could also be constructed (see Mendenhall, 1968:Ch. 8), the use of the above model was dictated by the available computer package. The differences of attitudes among staffs of different hierarchical statuses would not be significant if the null hypothesis below could not be rejected. Ho: 1 - The differences of attitudes between staffs in agencies which were different in the dichotomous organizational property dealt with in the test would not be significant if the null hypothesis below could not be rejected. One should be careful in interpreting the findings of this study. Several reasons make this consideration necessary. 1. If the differences are significant statistically but are very small, whether the differences would be important in the interorganizational cooperation and coordination in reality are doubtful. 43 If a difference is not a result of accidents of sampling, it might still be misleading if response bias exists (Blau and Duncan, 1967:16). It is assumed that the attitudes of staff in the same status would be similar. If there is no con- sensus within each status the study of the atti- tudes of different statuses would be meaningless. It should be mentioned that the sample used in this study was not a random sample. It was expected to be a survey of the total personnel in agencies. This would make the danger of response bias even higher. FINDINGS Status and Attitude Of the three scores measuring the "direction of attitude" toward interorganizational cooperation and coor- dination, "cost too high," "lack of control," and "don't need," all showed significant differences in statuses according to F tests. Findings showed administrators and semiprofessionals were more likely to find obstacles to cooperation and coordination while professionals and clerical staffs were less likely to find obstacles. This phenomenon could be explained as follows: semiprofessionals in welfare agencies are doing mostly routine jobs and that makes them more bureaucratically orientated. That is, they adhere to procedures and regulations in their work. Since cooperation needs flexible administrative procedure and semiprofessionals have less discretion, they might see more obstacles to interorganizational cooperation and coordination. Administrators have responsibility for the success and failure of any c00peration and corrdination with other agencies. If cooperation and coordination failed, they are the ones to be blamed and have most to lose. Therefore, they are more cautious and are also more 44 45 likely to find obstacles to interorganizational cooperation and coordination. The unexpected findings regarding the clerical staff's attitude are more difficult to explain. This finding could be rationalized as follows: clerical staffs were expected to be even more bureaucratically orientated because, doing routine jobs, they are less trained and have less discretion. Most of them, however, do not have adequate knowledge of the problems in coopera- tion and coordination and many of them were unable to answer the questions. In any item there were about one third to one fourth of the clerical staffs who left it unanswered. Those who did answer the items in the problem were less likely to find obstacles perhaps because they either did not understand the problems in coordination and cooperation among agencies and, hence, found fewer obsta- cles, or they might think cooperation and coordination were good but could not affect them and so gave more nega- tive answers (that is, saw fewer obstacles). The above findings give answers to half of the first part of the study--they confirm that there are differences of attitude among staffs of various hierarchical statuses. However, this does not mean that administrators and semiprofessionals are more likely to be against coop- eration and coordination among organizations, it is only a matter of degree. The average score for administrators and semiprofessionals in "cost too high" was 4.01 and for 46 professionals and clerical staffs were 3.73 and 3.49 respectively. In "lack of control," average scores for administrators, semiprofessionals, professionals, and clerical staffs were 3.56, 3.56, 3.27, and 3.19 respec- tively. In "don't need," average scores for the four levels were 3.40, 3.19, 3.06 and 3.13 respectively. Scores for measuring the "intensity of the atti- tude” mostly did not show any significant difference among staffs of different hierarchical levels. Both "case cooperation" and "resource sharing" failed to detect any significant difference of attitudes among statuses. Only the differences of attitudes of personnel of different statuses in "information exchange" were significant statistically. In the attitude toward exchange informa- tion, professionals were more likely to exchange informa- tion with other agencies (average score was 3.59), their scores were followed by administrators (average score was 3.39), semiprofessionals (average was 3.27) and the cleri- cal staffs (average was 2.93). It is not unexpected that professionals are more favorable to the exchange of infor- mation among agencies. Professionals are more interested in the professional circle and they have more contacts with outside professionals. The difference between semi- professionals and administrators was small, both were in the middle of the four statuses. It is interesting to note that clerical staffs who saw least obstacles in 47 interorganizational cooperation and coordination were least likely to have information exchange. This finding supports the original hypothesis, however. The clerical staffs, who saw less obstacles in interorganizational c00peration and coordination, were also more committed to their own agen- cies. They, therefore, were less likely to see real coop- eration and coordination. Coinciding with the original hypothesis, they were more bureaucratically orientated. This would suggest that both the established low status clerical staffs and the not-established low status clerical staffs (Homans, 1961:341-358) are against c00peration and coordination with other agencies, although their responses are due to different reasons. The established low status staffs who are neither committed to the agency nor to the community have less to gain in interorganizational coopera- tion and coordination. They have less expectancy and less self-investment within the agencies, according to the theories of Homans (l96l:Ch. 16), Campbell et al. (quoted by Hampton et al., 1973:45-50) and Faunce (1972). They would see information exchange as an extra burden. On the other hand, the not-established low status clerical staffs were overcommitted to their agencies, they did not want to divide their commitment to other agencies. Personnel of different statuses are under different pressures on their jobs in terms of the three influences--' authorities within their agencies, outside their agencies 48 and from the community to which their agencies provide services. It was found that both professionals and clerical staffs were less aware of the pressures of these three influences, and administrators and semiprofessionals were more aware of the pressures from the three influences (see Table 7). It seems that the professional feel more secure in their jobs, have more discretion, hence are less affected by the three influences. The clerical staffs, low in status, care less about authorities within and outside. But the administrators should deal with different people (Hawkes, 1961), and the semiprofessionals feel less secure and their status is in the middle. These lead them to compliance to authorities and other pressure (Homans, 1961:352-3, Hasenfeld and English, 1974:413). It was assumed that personnel who saw more obsta— cles in interorganizational cooperation and coordination would be less likely to cooperate and coordinate with other agencies to a larger extent. Yet, data show that it was not the case. Table 8 shows three clusters of "the direction of attitude” are associated, so are three clusters of "the intensity of attitude." Three clusters of job influences are also related. But, personnel who said "cost too high," "lack of control," and "don't need" more often as obstacles to interorganizational cooperation and coordination were not necessarily also less willing to have "case c00peration," "resource sharing," and 49 m u .m.U cEsHOOIcmosuonm . . m mooo .v . mN.m om.m m¢.m Nh.m huwflflfifiou Sum.“ Hoo. v m . . . . mmwocomm Smm.m SN m mp m on S SS m muHmuso mmwuwuosusé Sooo. v a SS.m Ho.v mm.m m . mmHoamm< cH mHo.0H m m mmfiuwuosusm m xumfiu HmcommmomoummEmm HucOHmmmmoum uoumuumwnweufl m momsumum mmocosamcH How non co .mwmsumum ucmummmwo «0 mmmmum moocmsHmcH co moccaum> mo mammamcd has one was mouoom mmwum>¢lu.h manna 50 oo.H mocmsoxm cOHumEHOMGH 0m we. oo.H vcHHmnm mousOmmu mm mm. Hm. oo.H GOHumuwmooo ammo Mm 5H. HH. HH. oo.H com: u.:on om mo. mo. HH. me. oo.H Houucoo «0 xomH hm mo. So. NH. om. we. oo.H ann 00» umoo MN HH. NH. SH. mm. SS. HS. oo.H suHcassoo oH . . . . . . . . mmHoammm eH mH SH om me He we oo H memuso moHuHuonesm QH . . . . . . . . . mmHocmmm Ho mo NH mm mm me om we so H cH mmHuHHonusm mH 0m am Mm om am cm 0H nH mH .coHumchHoou can :oHumummooo HMGOHumNHcmmuououcH uum3oa mopsuHuu¢ can moocosHucH non cmw3umm mucoHonmmoo :oHMMHmHHoo camumomun.m «Hana 51 "information exchange" with other agencies. These were unexpected (Table 8). But, respondents who reported that their jobs were under greater influences of the authorities inside and outside their agencies--that is, they were under greater bureaucratic control and were more bureaucratically orientated--did see "cost too high," "lack of control," and “don't need" as obstacles to larger extent. It would say that "the direction” of attitude and "the intensity" of attitude are two dimensions of attitude, they are not necessarily to be related. It might also say that the cluster of obstacles of interorganizational cooperation and coordination was a measurement of "the cognitive” and ”the affective" dimensions of the attitude, while the cluster of the extent of cooperation and coordination was a measurement for "the behavioral” dimension of the atti- tude, using the categories of Upshaw (1968). Respondents who were cognitively and affectively more favorable toward interorganizational cooperation and coordination did not necessarily act more favorably. This explanation is similar with Blau's (1961) findings. He reports that although social workers often complain that bureaucratic procedures restrain their efforts to serve clients, they were apathetic toward a summer camp program which was more service-orientated. Among the three kinds of cooperation, information exchange is the most routine and prevalent cooperation in social services, it was also most favorably [Em 52 accepted (average score was 3.3) by respondents of the study. The case cooperation is also not unusual, and it was moderately accepted (average score was 3.1). But the above two areas of cooperation are only superficial, the more intensive cooperation is resource sharing which was not quite welcomed (average score was 2.4). Yet, atti- tudes among personnel in agencies of different organiza- 1 tional properties show differently. These will be dis- cussed below. Size and Attitude i; Of the six kinds of attitudes, Table 9 shows no significant differences between the attitudes of personnel in large and small agencies. But when each hierarchical status was examined, some interesting patterns emerged. In the attitudes toward "case cooperation," "resource sharing," and "information exchange," the semiprofessional in small agencies were more willing to cooperate with other agencies (significant at .05, .01, and .05 levels respectively, see Table 10). The semiprofessional in small agencies are more professional or service orientated than their counterparts in larger agencies. They are mostly working on front desks, facing the burden of cases, and feeling impotent in giving enough help to clients. While the semiprofessional in large agencies are more bureau- cratically orientated. They are just doing their jobs, and do not have to worry about what is beyond their 53 Table 9.--Average Scores and Two Way Analysis of Variance on Attitudes of Staffs by Status and Size. a Statusesb c Attitudes Size Adm Prof Semi Clerk Average F La 4.06 3.76 4.05 3.48 3.90 10.53**** Cost Too N S “lgh Sm 3.93 3.66 3.88 3.50 3.75 N.S. **** Lack of La 3.63 3.25 3.56 3.16 3.43 7.86N s C°ntr°1 Sm 3.46 3.30 3.52 3.21 3.38 N.S. La 3.37 3.00 3.18 3.06 3.14 S.32**** Don't Need N.S. Sm 3.45 3.17 3.25 3.21 3.27 N.S. La 3.02 3.11 3.14 3.06 3.10 N.S. Case N S C°°Perat1°n Sm 3.08 2.88 3.68 2.53 3.03 N.S. La 2.42 2.53 2.33 2.42 2.44 N.S. Resource N S Sharlng Sm 2.46 2.64 3.02 1.91 2.56 N.S. . La 3.39 3.58 3.18 3.03 3.38 5.99**** Information N S Excnange Sm 3.38 3.61 3.67 2.83 3.46 N.S. a b . . c La = Large Adm = Administrator Top: between-column Sm = Small Prof = Professional d.f. = 3 Semi = Semiprofessional Mid: between-row Clerk = Clerical Staff d.f. = 1 Bottom: Interaction d.f. = 3 **** P < .001 N.S. Not Significant 54 Table lO.--T Tests on Attitudes Between Staffs of Same Status in Agencies of Large and Small Sizes. Statusesb Attitudes Adm Prof Semi Clerk Cost Too High N.S. N.S. N.S. N.S. i" Lack of 3 Control N.S. N S. N.S. N.S. Don't Need N.S. N.S. N.S. N.S. i Case t=2.00 ”j Cooperation N 5' N'S' d.f.=120 N's' * Resource t=2.55 Sharing N'S° N'S' d.f.=120 N.S. ** Information t=2.12 Exchange N'S' N.S. d.f.=120 N'S' * bAdm = Administrator * p < .05 Prof = Professional ** p < .01 Semi = Semiprofessional *** p < .005 Clerk = Clerical Staff **** p < .001 N.S. Not Significant 55 reSponsibilities. In small agencies, semiprofessionals might also do some professional work which leads them to be more professionally orientated. The above results show size is not very influential in attitude except in one status, the semiprofessional. The difficulty in finding the influence of size may be attributed to some unforeseen factors. In this study, agencies with over 25 staff members are classified as large. Blau (1966) defines agencies over 20 as large but a size of 25 may still be too small. It might be worth- while to test the hypothesis again, raising the number of staff in agencies defined as "large" to over 50. Professionalization and Attitude Among the six kinds of attitudes, Table 11 shows a significant difference between attitudes of personnel in agencies of high and low professionalization in the atti- tude toward "lack of control." The difference, however, was in the opposite of the predicted direction. Personnel in agencies of high professionalization were more likely to see "lack of control" as an obstacle to interorganiza- tional cooperation and coordination. Why personnel of agencies of high professionalization were more likely than those in agencies of low professionalization to see "lack of control" as an obstacle could be explained as follows: personnel (especially administrators, see Table 11 and Table 12) in agencies of high professionalization are more 56 Table ll.--Average Scores and Two Way Analysis of Variance on Atti- tudes of Staffs by Status and Agency Professionalization. Profession-a Statusesb c Attitudes alization Adm Prof Semi Clerk Average F Hi 3.99 3.74 3.87 3.63 3.80 10.53 **** Cost Too N S “lgh Lo 4.03 3.70 4.05 3.39 3.89 N.S. . **** Lack of Hi 3.68 3.33 3.73 3.25 3.46 3.9: ** C°ntr°1 Lo 3.47 3.17 3.51 3.14 3.37 N.S. Hi 3.32 3.06 3.34 3.17 3.17 5.32 *** Don't Need N.S. Lo 3.46 3.05 3.15 3.11 3.19 N.S Hi 2.95 3.10 3.38 2.92 3.10 N.S. Case N S C°°Peratl°n Lo 3.12 2.95 3.20 2.70 3.07 N.S. Hi 2.28 2.62 2.50 2.12 2.49 N.S. Resource N S Sharing Lo 2.56 2.49 2.44 2.19 2.46 N.S. . Hi 3.37 3.68 3.52 2.99 3.54 5.99 **** Information Exchan e N'S' 9 Lo 3.41 3.44 3.20 2.89 3.30 N.S. a . . c Hi = High pr: between-column d.f.=3 L0 = Low Mid: between-row d.f.=1 Bottom: interaction d.f.=3 bAdm = Administrator *p < .05 Prof = Professional **p < .01 Semi = Semiprofessional ***p < .005 Clerk = Clerical Staff ****p < .001 N.S. Not Significant 57 Table 12.--T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Profession- alization. Attitudes Statusesb Adm Prof Semi Clerk Cost Too High N.S. N.S. N.S. N.S. r5] Lack of t=l.7l ‘ 1“ Control d.f.=101 ”'3' “'3' ”'5' *a Don't Need N.S. N.S. N.S. N.S. case . N s N s N s N s Cooperation ' ‘ ° ' ° ' ' ' Resource sharing N.S. N.S. N.S. N.S. Information t=l.79 Exchange N'S° d.f.=l74 N‘S' N.S. bAdm = Administrator * p < .05 Prof = Professional ** p < .01 Semi = Semiprofessional *** p < .005 Clerk = Clerical Staff **** p < .001 N.S. Not Significant a Against original hypothesis 58 concerned about the quality of their service and, there- fore, they are more concerned about the quality of their staff. They seem not to be willing to lose control in interorganizational cooperation. Besides, the profession of social work is not an established profession. Profes- sionals in the field of social work are more aware of their professional status than are the well established professionals (Hall, 1968). When interorganizational cooperation and coordination among welfare agencies include other well-established professionals (e.g., medi- Ll‘w' "i “i“3 cal doctors, lawyers, etc.), the established professionals would have more power as a result of the cooperation and coordination (Hall, 1972:323). Consequently, the social workers may feel that they would lack control over their tasks. When one deals with individual hierarchical levels separately, as has been done in dealing with agency size, it was revealed that in neither case was there a signifi- cant difference of attitudes between clerical staffs in agencies of high and low professionalization (Table 12). As to "information exchange," professionals in agencies of high professionalization were more willing to have infor— mation exchange than those in agencies of low profession- alization. This finding supported the hypothesis. Exchange of information is a means to extend services to clients and the community. Professionals in agencies of 59 high professionalization are more likely to exchange infor- mation than are their colleagues in agencies of low profes— sionalization. The finding is similar to Vinter's (1963) work that shows when professionals are present in large numbers and assume critical positions they would be influential in their agencies, but when they are in small number and in low-echelon "line" positions, they are less influential. One would suppose that the latter become more and more bureaucratically orientated and adapted to bureaucratic procedures. The influence of professional- ization seems not to be very great. Hall (1968) also reports that professionals are affected by the organiza- tional structure. He notes that the structural and atti- tudinal aspects of professionalization do not necessarily vary together. Therefore, personnel would have different attitudes even in agencies of similar professionalization. In conclusion, professionalization per se is not a crucial factor in the attitude of the respondents. Its influence might become more prominent when it combines with other properties. Technology and Attitude Technology was computed from the combination of number of services offered and the kind of technology (people processing and people changing, see Hasenfeld, l974:5) applied in the agency. In three of the six kinds of attitudes, the data showed there were significant 60 differences between personnel in agencies of high tech- nology and low technology. The three kinds of attitudes were "cost too high," "resource sharing," and "information exchange." In another one, attitude of "don't need" showed significant differences in interaction between status and technology in a test of two-way analysis of variance. an All supported the original hypotheses (Table 13). Person- nel in agencies of high technology were less likely to see "cost too high" as an obstacle to cooperation and coordina- tion than those in agencies of low technology. In agencies LJ of high technology, tasks are more complex, cooperation with other agencies becomes a necessity. Welfare workers either feel the necessity for cooperation and coordination with other agencies or are more familiar with such matters, hence, they have more favorable attitudes toward inter- organizational cooperation and coordination. The same interpretation can also apply to "resource sharing" and "information exchange." Since cooperation and coordination are common in everyday routines, personnel in agencies of high technology are more willing to support sharing and exchange programs. Personnel in agencies of low technology might not feel the necessity so greatly. The functions of their agencies might be simple, hence, a single agency can handle their jobs quite competently and easily. Neverthe- less, one should not emphasize their difference too much. It is still only a matter of degree. If one considers 61 Table l3.--Average Scores and Two Way Analysis of Variance on Atti- tudes of Staffs by Status and Agency Technology. b . a Statuses c Attitudes Technology Adm Prof Semi Clerk Average F Hi 3.89 3.71 3.79 3.65 3.76 10.76 **** Cost Too 3 94 * High Lo 4.10 3.75 4.14 3.32 3.93 3.06 * ' **** Lack of H1 3.53 3.30 3.55 3.32 3.42 7.87 N s C°ntr°1 Lo 3.59 3.23 3.56 3.06 3.40 N.S. Hi 3.11 3.07 3.20 3.19 3.12 5.43 *** Don't Need N.S. Lo 3.61 3.04 3.19 3.07 3.23 3.78 * Hi 2.93 3.08 3.52 2.78 3.12 N.S. Case N S C°°Perat1°n Lo 3.11 2.99 3.10 2.79 3.04 N.S. Hi 2.53 2.74 2.79 2.38 2.68 N.S. Resource 11.46 **** Sharlng Lo 2.37 2.35 2.29 2.00 2.30 N.S. . Hi 3.53 3.76 3.66 3.18 3.65 6.24 **** Information 21 65 **** Excnange Lo 3.30 3.36 3.08 2.73 3.20 N.S. aHi = High cTop: between—column d.f. = 3 L0 = Low Mid: between-row d.f. = l b Bottom: interaction d.f. = 3 Adm = Administrator Prof = Professional * p < .05 Semi = Semiprofessional ** p < .01 Clerk = Clerical Staff *** p < .005 **** p < .001 N.S. Not significant 62 individual statuses separately, one will find in the atti- tude of ”don't need," that a significant difference did not exist except for administrators. In attitudes of other kinds, differences did not appear in all statuses (Table 14). For example, in no case was there any signifi- cant difference between the attitudes of clerical staffs in both kinds of agencies. Clerical staffs in agencies of high and low technologies are mainly doing similar kinds of work. Their attitude did not show any difference because they lack the comprehension of the importance of interorganizational cooperation and coordination, or they had no interest in this matter at all. It is also inter- esting to note that with regard to "cost too high,” "resource sharing," and "information exchange,” adminis- trators in both kinds of agencies also showed no signifi- cant difference. Administrators are more established in their agencies, they are more committed to their own agencies, hence, they are less willing to cooperate and coordinate with other agencies, except for the benefit of their own agencies. Interaction and Attitude Regardless of status, personnel in low interaction agencies said ”don't need" was an obstacle more often than personnel in high interaction agencies. The result sup- ported the hypothesis (see Table 15). Dealing with individual status separately, it was found that 63 Table l4.--T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Technologies. b - Statuses Attitudes Adm Prof Semi Clerk Cost Too t=2.7l High N.S. N.S. d f.=135 N.S. *** Lack of Control N.S. N.S. N.S. N.S. , t=3.85 Don t Need d.f.=100 N.S. N.S. N.S. **** Case t=l.9l Cooperation N.S. N'S‘ d.f.=120 N.S. * Resource N.S. t= 2.36 t=2.25 N.S. Sharing d.f.=174 d.f.=120 ** ** Information N S t=3.18 t=3.12 N S Exchange ' ’ d.f.=174 d.f.=120 ' ° *** *** bAdm = Administrator * p < .05 Prof = Professional ** p < .01 Semi = Semiprofessional *** p < .005 Clerk = Clerical Staff **** p < .001 N.S. Not Significant 64 Table 15.--Average Scores and Two Way Analysis of Variance on Atti- tudes of Staffs by Status and Agency Interaction. “—ww— b . . a Statuses c Attitudes Interactions Adm Prof Semi Clerk Average F Hi 3.92 3.71 4.20 3.33 3.86 13.49 **** Cost Too N S “lgh Lo 4.13 3.92 3.65 3.44 3.58 5.28 *** . **** Lack of Hi 3.58 3.23 3.52 3.12 3.36 8.95 N S contr°1 Lo 3.61 3.48 3.64 2.93 3.51 N.S. Hi 3.18 3.03 3.19 3.00 3.10 5.91 **** Don't Need 17.66 **** Lo 3.78 3.41 3.40 3.06 3.52 N.S. Hi 2.87 3.04 3.12 2.82 3.02 N.S. Case N S C°°perat1°n Lo 3.30 3.11 3.31 2.83 3.22 N.S. Hi 2.47 2.55 2.30 1.99 2.43 N.S. Resource N S Sharing Lo 2.35 2.68 2.46 2.04 2.42 N.S. . Hi 3.39 3.62 3.20 2.89 3.42 7.31 **** Information N S ExChange Lo 3.29 3.52 3.09 2.68 3.25 N.S. a . . C Hi = High pr: between-column d.f. = 3 Lo 8 Low Mid: between-row d.f. = 1 b Bottom: interaction d.f. = 3 Adm 8 Administrator Prof = Professional * p < .05 Semi = Semiprofessional ** p < .01 Clerk = Clerical Staff *** p < .005 **** p < .001 N.S. Not Significant 65 administrators and professionals in agencies of low inter- action regarded "don't need" as an obstacle to cooperation and coordination with other agencies more often than their counterparts in agencies of high interaction (Table 16). Since their agencies have less interaction with others, they also have less interaction with personnel in other agencies and are less often regarded as important counter- parts in interorganizational cooperation and coordination by other agencies (since agency interaction was measured by the number of choices concerning areas of interorgani- zational cooperation and coordination the focal agency was given by other agency heads). If this is true, it is not strange that personnel in agencies of low interaction see "don't need“ as an obstacle to interorganizational coopera- tion and coordination. Dealing with each status, it was also found that there was no significant difference between the attitudes of clerical staffs in agencies of different interaction. It was found that semiprofessionals in high interaction agencies were more likely to see "cost too high" as an obstacle for interorganizational coopera- tion than those in agencies of low interaction. The result is the opposite of the hypothesis. This finding could be explained by the fact that the semiprofessional in agencies of high interaction have seen more interorgani- zational c00peration and coordination and are more aware of the actual costs of these matters. Also, if the 66 Table 16.--T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Interaction. - b . Statuses AttItUdeS Adm Prof Semi Clerk Cost Too t=3.56 High “'5' ”'5' d.f.=103 ”'5' ***a Lack of Control N.S. N.S. N.S. N.S. . t=4.l7 t=l.99 Don t Need d.f.=87 d.f.=156 N.S. N.S. **** * case . N.S. N.S. N.S. N.S. Cooperation Res°9r°e N.S. N.S. N.S. N.S. Sharing Inf°rmatl°n N.S. N.S. N.S. N.S. Exchange bAdm = Administrator * p < .05 Prof = Professional ** p < .01 Semi = Semiprofessional *** p < .005 Clerk = Clerical Staff **** p < .001 N.S. Not Significant a Against original hypothesis [1"— 1rd 67 semiprofessionals have been given more work because of the cooperation and coordination, they may feel they have no time to do it. The semiprofessional in agencies of low interaction are just the opposite, they do not compre- hend the actual situation of interorganizational relations. For other kinds of attitudes except in "don't need," the differences between personnel of the high and low inter- action agencies were not significant. Communication and Attitude Of the six kinds of attitudes, only “cost too high” showed differences between the attitudes of personnel in agencies of high and low communication. This result supported the original hypothesis that communication would benefit interorganizational cooperation and coordination (see Table 17). However, in other attitudes, the differ- ences were not significant. But, when each hierarchical status was examined, it was found that professionals in agencies of different communication showed some dif- ferences in the attitude of "cost too high" (Table 18). There was also a significant difference in the attitude of "information exchange." That is, professionals in agencies of high communication are less likely to see "cost too high" as an obstacle to interorganizational cooperation and coordination. They also were more likely to support greater information exchange with other agencies. But for 68 Table l7.--Average Scores and Two Way Analysis of Variance on Atti- tudes of Staffs by Status and Agency Communication. Communi- Statusesb c Attitudes cation Adm Prof Semi Clerk Average F ' **** Cost Too Hi 3.96 3.62 4.04 3.24 3.73 lg.§; * High Lo 4.06 3.83 4.01 3.68 3.93 N.S. ' ' **** Lack of Hi 3.60 3.32 3.70 3.14 3.43 7.88 N S C°ntr°l Lo 3.52 3.22 3.52 3.23 3.40 N.S. Hi 3.49 3.07 3.36 3.00 3.22 5.34 *** Don't Need N.S. Lo 3.32 3.04 3.15 3.24 3.15 N.S. Hi 2.98 3.04 3.52 2.74 3.06 N.S. Case N S c°°perat1°n Lo 3.11 3.04 3.17 2.86 3.10 N.S. Hi 2.33 2.62 2.56 2.03 2.47 N.S. Resource N S Sharing Lo 2.55 2.51 2.42 2.34 2.47 N.S. . Hi 3.30 3.70 3.42 2.92 3.47 5.98 **** Information N S ExChange Lo 3.49 3.48 3.23 2.96 3.36 N.S. a . . c Hi = High Top: between-column d.f. = 3 L0 = Low Mid: between-row d.f. = 1 b Bottom: interaction d.f. = 3 Adm = Administrator Prof - Professional * p < .05 Semi = Semiprofessional ** p < .01 Clerk = Clerical Staff *** p < .005 ***t p < .001 N.S. Not Significant 69 Table 18.--T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Communication. -_—‘4—..——-. _.-—._—_-- >.———.- - '— -—. —. b . Statuses Attitudes Adm Prof Semifi Clerk Cost Too t=l.9l High N.S. d f =180 N.S. N.S. * "1 Lack of a Control N.S. N.S. N.S. N.S. Don't Need N.S. N.S. N.S. N.S. Lj' case . N s N s N s N.S Cooperation ‘ ' ° ' ' ' Resource Sharing N.S. N.S. N.S. N.S. Information t=1.69 Exchange ”'5' d.f.=174 ”'3‘ ”'5' * bAdm = Administrator * p < .05 Prof = Professional ** p < .01 Semi = Semiprofessional *** p < .005 Clerk = Clerical Staff **** p < .001 N.S. Not significant 70 more intensive cooperation and coordination, no significant difference was found. Communication was computed by number of committee respondents served. Hage, Aiken and Marrett (1971) note, committee meetings represent a greater emphasis on horizontal information flows . . . there is also an increased horizontal flow of unscheduled task com- munication. Horizontal exchange of information, therefore, would lead to not only more cooperation and coordination but also more favorable attitudes toward them. But this might only apply to professionals. The semiprofessional and the clerical staff have less opportunities to serve on com— mittees and whether there are more or less committees existing in an agency makes no difference to them. For administrators, because they are more bureaucratically orientated, they were also less likely to be influenced by the type of communication in agencies. Formalization and Attitude Formalization is the extent of agency regulation followed by agency staffs. Although Aiken and Hage (1968) report that there is no relationship between formalization and organizational interdependence, in this study, formal- ization turned out to be quite influential, especially for semiprofessionals. In the six kinds of attitudes, the data showed there were significant differences between personnel in agencies of high formalization and low 71 formalization in two kinds of attitudes--"cost too high" and "don't need.” All supported the hypotheses. In agencies of high formalization staffs were more likely to see "cost too high" and "don't need" as obstacles for COOperation and coordination (Table 19). As mentioned by Merton (1968:Ch. 8), officials who adhere to regulations !? would take the "means" (regulation) as the "end" (organi- I zational goals). Therefore, in agencies of high formal- ization, agency staffs would be more likely to see obsta- cles to interorganizational cooperation and coordination $2 even though these would benefit the achieving of their agency goal to serve the community. More information regarding the impact of formalization was found when deal- ing with each status separately (Table 20). It was found that the semiprofessional were more influenced by the impact of formalization. The differences of attitudes between semiprofessionals in agencies of high and low formalization were significant in ”cost too high," "information exchange,” "resource sharing," and "case cooperation." For professionals, there were only significant differences in one attitude--"cost too high." For administrators, there were significant differences only in "don't need." Clerical staffs again did not show any significant dif- ference although their attitudes were mostly in the expected directions. 72 Table l9.--Average Scores and Two Way Analysis of Variance on Atti- tudes of Staffs by Status and Agency Formalization. .mlfl‘ b . Formal- Statuses c Attitudes ization Adm Prof Semi Clerk Average F Hi 4.09 3.89 4.17 3.74 4.03 11.14 **** Cost :00 27.47 **** 319 L0 3.90 3.56 3.71 3.40 3.64 N.S. . **** Lack of Hi 3.62 3.22 3.60 3.27 3.45 7.88 N 3 C°ntr°l Lo 3.49 3.32 3.47 3.16 3.36 N.S. Hi 3.54 3.10 3.20 3.33 3.25 5.38 *** Don't Need 4.54 * Lo 3.19 3.01 3.17 3.06 3.09 N.S. Hi 3.11 3.07 3.12 2.88 3.09 N.S. Case N S C°°Perat1°n Lo 2.93 3.01 3.53 2.76 3.06 N.S. Hi 2.43 2.47 2.32 2.03 2.39 N.S. Resource N S Sharlng Lo 2.44 2.66 2.75 2.20 2.57 N.S. . Hi 3.41 3.55 3.14 2.87 3.35 6.00 **** Information N S Excnange Lo 3.36 3.62 3.57 2.95 3.47 N.S. a . . c Hi = High Top: between-column d.f. 3 Lo 8 Low Mid: between-row d.f. = 1 b Bottom: interaction d.f. = 3 Adm = Administrator Prof = Professional Semi = Semiprofessional Clerk = Clerical Staff * P < ** P < *** P < **** p < N.S. NOt .05 .01 .005 .001 Significant ‘: I! I I 73 Table 20.--T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Formalization. b . Statuses AttitUdeS Adm Prof Semi Clerk Cost Too t=3.97 t=3.71 High , N's' d.f.=180 d.f.=135 ”'3' **** **** Lack of Control N S. N.S. N.S. N S. t=2.43 Don't Need d.f.=100 N.S. N.S. N.S. ** Case t=l.86 Cooperation N.S. N°S° d.f.=120 N.S. * Resource t=l.87 Sharing N.S. N'S' d.f.=120 N’S° * Information t=l.87 Exchange N'S’ N'S' d.f.=120 N.S * bAdm = Administrator * p < .05 Prof = Professional ** p < .01 Semi = Semiprofessional *** p < .005 Clerk = Clerical Staff **** p < .001 N.S. Not Significant 74 Centralization and Attitude The frequency of participation in decision-making from which centralization was computed also influences attitudes of the staff of welfare organizations. Comparing the attitudes of personnel in agencies of high and low centralization regardless of status (see Table 21), in the six kinds of attitudes, in "lack of control" the dif- ferences of attitudes were significant. When examining each different status respectively, the above significance did not exist in the status of semiprofessional (see Table 22). The above findings supported the hypothesis that in agencies of high centralization, personnel were more likely to say "lack of control" was an obstacle to interorganizational cooperation and coordination. Examin- ing each status, there were some interesting findings regarding the attitudes of clerical staff. Of the seven organizational properties, only centralization contributed to the significant difference of attitudes of the clerical staff in agencies of high and low centralization. The clerical staff in agencies of high centralization were more likely to see "cost too high," "lack of control," and "don't need" as obstacles to interorganizational c00peration and coordination. But, unexpectedly, the semiprofessional in agencies of low centralization were more likely to see "cost too high" as an obstacle and less willing to see "resource 75 Table 21.--Average Scores and Two Way Analysis of Variance on Atti- tudes of Staffs by Status and Agency Centralization. b . Central- Statuses c Attitudes ization Adm Prof Semi Clerk Average Hi 4.08 3.68 3.82 3.87 3.87 10.78 **** Cost Too N S ”19h Lo 3.94 3.74 4.08 3.28 3.84 4.26 ** ' **** Lack of Hi 3.68 3.45 3.52 3.46 3.55 2.22 ** C°ntr°1 Lo 3.46 3.20 3.57 3.04 3.35 N.S. Hi 3.46 3.05 3.22 3.42 3.27 5.34 **** Don't Need N.S. Lo 3.35 3.06 3.19 2.97 3.14 N.S. Hi 2.91 3.01 3.39 2.65 3.03 N.S. Case N S C°°perat1°n Lo 3.16 3.05 3.20 2.85 3.10 N.S Hi 2.39 2.36 2.83 1.95 2.44 N.S. Resource N S Sharing Lo 2.47 2.64 2.34 2.25 2.49 N.S. . Hi 3.28 3.48 3.56 3.06 3.40 6.00 **** Information N S Excnange Lo 3.49 3.62 3.19 2.88 3.41 N.S. a . . c H1 = High pr: between-column d f. = 3 L0 = Low Mid: between-row d f. = l b Bottom: interaction d f. = 3 Adm 8 Administrator Prof = Professional * p < .05 Semi = Semiprofessional ** p < .01 Clerk = Clerical Staff *** p < .005 **** p < .001 N.S. Not Significant 76 Table 22.--T Tests on Attitudes Between Staffs of Same Status in Agencies of High and Low Centrali- zation. Statusesb Attitudes Adm Prof Semi Clerk Cost Too t=l.83 t=2.13 High ”'5' ”'5' d.f.=l35 d.f.=43 *a * Lack of t=l.77 t=2.08 N S t=l.69 Control d.f.=101 d.f.=l80 ° ° d.f.=43 * * * . t=l.89 Don t Need N.S. N.S. N.S. d.f.=43 * case. NS NS NS NS Cooperation ' ° ' ° ' ' ° Resource t=l.94 Sharing ”'5' ”'5 d.f.=120 N°S° *a Information t=l.72 Exchange N.S. N'S° d.f.=120 N.S. *a bAdm = Administrator * p < .05 Prof = Professional ** p < .01 Semi = Semiprofessional *** p < .005 Clerk = Clerical Staff **** p < .001 N.S. Not Significant a Against original hypothesis 77 sharing," and ”information exchange" to a larger extent than their colleagues in agencies of high centralization. These findings were consistent in three areas and do not appear to be accidental. In this study, it was found that centralization was related to communi- cation (see Table 23). This was also unexpected. This unexpected finding could have some unexpected impact on the influence of centralization on attitudes. But the relationship between communication and centralization did not exist when comparing agencies of high communication and high centralization, and also low communication and low centralization as indicated in Table 24. The unexpected findings of centralization in the attitudes of semipro- fessional, hence, need further study. 78 moo. V Q *** Ho. v m «a mo. v m * oo.H cowumNHHmnucmo mo. oo.H coHumNfiHmEuom 44464. 4am. oo.H . conumoncsssoo mm.| ma. mH.| oo.H cowuomumucH 0H. 00.: «a. ma. oo.a mooHoccomB . . . . . . conumusam ¢O I MO *0 mo {mm 00 H IGOHmmmWOHm mo.| ma. mo.u «h«hm. mo. mo. oo.H muflm ceaumNfi coflumNfl soHumo c0auom nowadaflam namuucmo :Hmsuom unasssoo numucH smeaocnome scoflmmmmoum munm .mmwunmmoum HmcoflumNflcmmno now mucwfioflmmmoo cowumamuuou comummmul.mm mange 79 H u .m.U mo. coauMNmeuucmu oo. nodumNfiHmEnom oa. ow. newumowcsfieoo mo. 00. mo.H cofluomnmucH «v. Hm.m oo. oo.m wooHocnoma . . . . . downwaflam OH 00 OH mw N {hm M ICOflmMQMOHm Hm.H mm. oo. mm. ma. Ho. gnaw cOADMNw coflumnw cowumo cofiuom coflumNflam namuuamo namsuom unasssoo aumucH amoHocnome aconmwmonm munm .mmauummoum HmcowumeammHo “Om mumswm wnUII.vN GHQMB DI SCUSSION AND CONCLUS ION In this study, a sample of 474 respondents from 41 welfare agencies in a medium—sized Midwest city was in- F_? quired. Major issues in this study were to examine how hierarchical status and organizational properties influence attitudes toward interorganizational cooperation and coor- :‘j dination. It was assumed that personnel of professional — orientation would be more likely to support interorgani- zational cooperation and coordination. It was also assumed that professionals were more professionally orientated, clerical workers were more bureaucratically orientated, and administrators and semiprofessionals were holding both Orientations. Therefore, it has been hypothesized that professionals would be more likely to support interorgani- zational cooperation and coordination than those of their other colleagues would. Furthermore, professional orien- tation could be acquired in professional training, it could also be acquired from professional jobs and favorable organizational atmosphere. Personnel who initially held professional orientation might lose this orientation in certain organizational atmosphere, and vice versa. It has been hypothesized that personnel in agencies of higher 80 81 formalization, higher centralization, and larger size would be more bureaucratically orientated, and they were less likely to be favorable toward interorganizational cooperation and coordination; personnel in agencies of higher professionalization, higher inner communication, higher technology would be more professionally orientated E” and consequently more likely to support interorganizational g cooperation and coordination. Attitudes of respondents were measured by two ‘m J .‘v blocks of questions. One dealt with ten obstacles pre- venting interorganizational cooperation and coordination. Respondents were asked to what extent these obstacles pre- venting cooperation. These ten obstacles were grouped into three groups-~"cost too high," ”lack of control," and "don't need." Another block of questions concerned eleven areas of cooperation prevailing in the field of social work. Respondents were asked to what extent they would like to see cooperation and coordination in these areas. Items in this block were also clustered into three groups--"case c00peration,” ”resource sharing," and "information exchange.” Reapondents were also asked about influences on their jobs, these influences were also grouped into three kinds-~"authorities in agencies," ”authorities outside agencies," and ”community." The first block was supposed to measure "the direction" of attitude, the second block was supposed to measure "the 82 intensity" of attitude. The one concerning job influences was employed to measure to what degree respondents were under bureaucratic control. Data showed that professionals and clerical workers were less aware of the influences of authorities inside and outside agencies than were administrators and semipro- fessionals. Professionals and clerical workers saw fewer obstacles in interorganizational cooperation than did administrators and semiprofessionals. Professionals were also more likely to have information exchange with other agencies than administrators, semiprofessionals, and clerical workers. However, in "resource sharing" and "case cooperation" which require more intensive cooperation, the differences of attitudes among personnel of different statuses were insignificant. The result that clerical workers saw fewer obstacles in cooperation was unexpected. It was rationalized that clerical workers had no adequate knowledge about the real situation in interorganizational cooperation. The findings that respondents who saw fewer obstacles did not necessarily want more cooperation were also unexpected. These could be explained as either "the direction" of attitude and "the intensity" of attitude are two dimensions which are not necessarily to be correlated; or the two blocks of questions did not measure "the direc— tion” of attitude and "the intensity" of attitude but "the cognitive" and "the affective" aspects of attitude and 83 "the behavioral" aspect of attitude, concepts derived from Upshaw (1968). Those who were cognitively and affectively more favorable toward interorganizational cooperation and coordination did not necessarily behave more favorably. The influencc of organizational properties on the attitudes of personnel in welfare agencies seem not very _ great, except those of technology, formalization, and Fu] centralization. Only a few tests showed that there were significant differences between attitudes of personnel of agencies of different kinds in terms of the dichotomous H": y organizational properties. Size seems least influential and technology seems most influential among the seven organizational properties. To measure the differences between the attitudes of personnel in different kinds of agencies by virtue of organizational properties, there were a total of 42 tests of two-way analysis of variance; six tests for each organizational property. Only nine showed that there were significant differences; eight supported the original hypotheses and one showed that the Opposite was the case. As mentioned before, size is not a sensitive property, and professionalization can only be influential when the professionals are present in large proportion and hold critical positions in organization (Vinter, 1963). Interaction seemed to be an important prOperty, since personnel who are more familiar with interorganizational cooperation and coordination would have more favorable 84 attitudes toward them. Yet, only in one of six tests was there significant differences. Of the seven properties, technology, formalization, and centralization were more influential. In 19 technologically high agencies, 16 were treatment (peOple changing) agencies (Gamma = .80). Personnel in treatment-orientated agencies become more service-orientated. The professionals could maintain their commitment to their professional ethics and the administrators are also able to maintain their professional ethics if they are professionaksthemselves, or adapt or acquire the ethics if they are not professionals. So do the other staffs, especially the semiprofessionals. But, again, technology alone, according to Thompson (1967:19), is not sufficient to be deterministic. In agencies of higher formalization and centralization, personnel were less favorable to interorganizational cooperation and coordination. The findings were more prominent when each status was examined. The findings concerning the atti- tudes of semiprofessionals in agencies of high and low centralization, however, were unexpected, and require further studies. It is very unlikely that all organizational prop- erties will have similar influences on attitudes of employees, since organizational properties themselves are not necessarily highly correlated. Some properties would have impact on some kinds of attitudes while other 85 properties affect attitudes of other kinds. Hall and his colleagues (1967) mention that size is weakly related to other organizational properties. From Table 23, we can see that size is related only to interaction, and profes- sionalization is related only to technology. Communication is related to formalization and centralization. It is not a surprise to find that the larger the size of an agency is, the more interactions it will have with other agencies, and the higher the professionalization an agency is, the higher its technology will be. Communication is positively correlated with formalization and centralization. Since Hage and his colleagues (1971) report that communication is negatively related to formalization and centralization, our results were unexpected. But they find that formali- zation and centralization are only weakly related. In this study, formalization and centralization are not related significantly. In many agencies, it becomes prevalent that top administrators often call committee meetings. They employ the technique of "participative management" as a means to manipulate consensus within their agencies. It might say, in many highly formalized and centralized agencies, such meeting becomes even more prevalent and turns to be a place to show the authority of the top administrator who is the most dominant figure of the meeting. That is why communication is associated with formalization and centralization. 86 Organizational properties have different influences on the variation of the attitude of staffs of different statuses. In this thesis, the semiprofessionals were mostly influenced by organizational properties. The semi- professionals have little discretion and are bureaucratic- ally orientated like the clerical workers, but if the semi- professionals are not in established low status, they are more committed to their jobs and their organizations. Since they are working on the front desks and routinely processing clients, taking orders from their supervisors, facing clients everyday, they are the easy targets of com- plaining clients who pay less respect to them than to the professionals. The semiprofessionals are, hence, the most to be influenced by the structure of their agencies and the outer environment. The administrators and the profes- sionals have much more discretion and they are less influ- enced by the structure of organizations. The clerical workers, low in hierarchical status, may care less about matters beyond their responsibilities, and they are the least to be influenced. The above arguments were supported by the analysis of the influences on jobs. It was found that the semiprofessionals and the administrators reported that they were more influenced by the authorities within or outside agencies and the community, but the profes- sionals and the clerical workers reported they were less affected by the three influences. The reason for this 87 finding could be that the professionals have more dis- cretion and are more secure in their jobs and the clerical workers were less concerned about their agencies, hence, both felt they were less influenced. In this study, it has been found that organizational properties, such as technology, formalization, and central- ization, do affect attitudes of personnel toward inter- _ ! organizational cooperation and coordination in welfare agencies. Although the influences of other prOperties seem little, they do have impact upon attitudes of welfare ,i ( workers, especially upon semiprofessionals. Yet, some underlying factors were not dealt with in this paper, such as how organizational structure affects professional orien- tation and bureaucratic orientation, how interorganiza- tional relation yields conflict among agencies (it has been examined in other reports of this project, see Marcus et al., 1974a, 1974b). Still another issue requiring further studies is to examine how professionals hinder interorgani- zational cooperation and coordination. Some authors, like Freidson (1970) and Daniels (1973), suggest that a pro- fession (e.g., medicine) would become self-interested, which might impede its cooperation with other professions. I suggest that concepts of professional orientation and bureaucratic orientation should also be redefined. As suggested by Litwak (1960), bureaucratic control and professional control are both needed for professional (I) 88 bureaucracies (e.g., welfare agencies). Bureaucratic orientation might not be the opposite of professional orientation. According to Freidson (1970), the dominant medical profession distributes resources unevenly for the benefit of the medical profession, Daniels (1973) argues that public regulations are needed to control the practice of professional work, he points out that the profession- alization of work does not necessarily lead to altruism of professionals. The concept of bureaucratic orientation should also be redefined. We should differentiate orien- tation of tOp administrators who have broader contacts with different professions (Hawkes, 1961) and might have broader viewpoints than professionals, and the orientation of lower bureaucrats who are locked in their low positions. Other implications could explain why some organiza- tional properties attribute little to the attitude toward interorganizational cooperation and coordination. Welfare agencies might rather be "pushed" to cooperate and coor- dinate with other agencies because they need more resources, than be "pulled" to do so because the professional ethics requires them to expand their services to the community. Technology was most influential in this study because in agencies of high technology, personnel were more profes- sionally orientated, and these agencies also need more resources. The combination of "pushed" and "pulled" factors made technology most influential. Yet, favorable 89 attitudes toward cooperation and coordination could not be established automatically. Strict bureaucratic control leads to departmentalism, timidity, and conservatism which would impede the effectiveness and efficiency of inter- organizational cooperation and coordination. Decreasing formalization and centralization in organization, and giving more autonomy to personnel would make interorgani- zational cooperation and coordination more successful and create personnel's favorable attitude toward them. APPENDIX LIST OF AGENCIES STUDIED IN THIS THESIS Private Agencies American Red Cross Big Brothers/Big Sisters Boy Scouts of America Boy's Club Catholic Social Services Community Nursery C Community Center Family and Child Services, Inc. Legal Aid Bureau Urban League, Inc. Girl Scout Council Rehabilitation Industries, Inc. Salvation Army Tri-County Council on Alcoholism and Addiction Visiting Nurse Association Volunteers of America Y.M.C.A. Y.W.C.A. Credit Counseling Centers, Inc. Public Agencies County Health Department City Housing Commission Community Mental Health Board (CMHB) Area Office of Economic Opportunity Model Cities (MC) Department of Social Services Department of Vocational Rehabilitation State Employment Security Commission Social Security Administration Cooperative Extension Services Department of Veterans Affairs 90 91 Component Agencies B Center (CMHB) B.I.L.D. Corporation (MC) Comprehensive Drug Treatment Program (CM, CMHB) Health Services (MC) Housing Assistance/Community Resources (MC) County Mental Health Center (CMHB) L Center (CMHB) New Way In (MC) Senior Citizens, Inc. (MC) S Community Mental Health Center (CMHB) Youth Development Corporation (CM) Agyris, 1972 BIBLIOGRAPHY Chris The Applicability of Organizational Sociology. London: Cambridge university Press. Aiken, Michael & Jerald Hage 1966 1968 Blalock, 1972 ”Organizational Alienation: A Comparative Analysis." American Sociological Review, 31 (Aug). ”Organizational Interdependence and Intra- organizational Structure." American Sociological Review, 33 (Dec), pp. 912-929. Hubert M. Jr. Social Statistics. 2nd ed. New York: McGraw-Hill. Blau, Peter M. 1955 1960 1961 1965 1968 1970a 1970b The Dynamics of Bureaucracy. Chicago: University of Chicago Press. "Structural Effects.” American Sociological Review, 25. “Orientation Toward Clients in a Public Welfare Agency." Administrative Science Quarterly, 5, pp. 341-361. "The Comparative Study of Organizations." Industrial & Labor Relations Review, 18 (Apr). e Hierarchy 6? Authority in Organizations." American Journal of Sociology, 73 (Jan), pp. 453- 67. 'A Formal Theory of Differentiation in Organi- zation.” American Sociological Review, 35 (Apr), pp. 201-218. "Decentralization in Bureaucracies." In Mayer Zald (ed.), Power in Organizations. Nashville, Tenn.: Vanderbilt University Press. Blau, Peter M. & Otis Dudley Duncan 1967 The American Occupational Structure. New York: Wiley. Blau, Peter M., Wolf V. Heydebrand, and Robert E. Stauffer 1966 ”The Structure of Small Bureaucracies." American Sociological Review, 31 (Apr), pp. 179-191. 92 93 Blau, Peter M. & Richard Schoenherr 1971 The Structure of Organization. New York: Basic Books. Blau, Peter M. & Richard Scott 1962 Formal Organizations. San Francisco: Chandler. Blaunner, R. 1964 Alienation and Freedom. Chicago: University of Chicago Press. Crozier, Michel 1964 The Bureaucratic Phenomenon. Chicago: University of Chicago Press. Dalton, Melville 1950 "Conflicts Between Staff and Line Managerial Offices." American Sociological Review, 15, pp. 342-351. Daniels, Arlene K. 1973 "How Free Should Professions Be?" In Eliot Freidson (ed.), The Professions and Their Pros- pects. Beverly Hills, Calif.: Sage. Durkheim, Emile 1930 The Rules of Sgciolggical Method. Chicago: University of Chicago Press. 1933 The Division of Labor in Society. New York: The Free Press. 1951 Suicide. New York: The Free Press. Etzioni, Amitai 1961 A ComparativeAnalygis of Complex Organizations. New York:The Free Press. Even, William M. 1966 ”The Organization-Set." In James D. Thompson (ed.), Approaches to Organizational Design. Pittsburgh: University of Pittsburgh Press. Faunce, William A. 1962 ”Size of Locals and Union Democracy." American gournal of Sociology, 68 (Nov). 1972 "Self Investment in Occupational Role." Paper presented at Meetings of the Southern Sociological Society, New Orleans, April 6, 1972. 94 Freidson, Eliot 1970 "Dominant Professions, Bureaucracy, and Client Services." In William Rosengren and Mark Lefton (eds.), Organizations and Clients: Essays in the Sociology of Service. Columbus, Ohio: Charles E. Merrill] pp. 71-92. Freidson, Eliot, ed. 1973 The Professions and Their Prospects. Beverly Hills, Caiifi: Sage. Gouldner, Alvin W. 1954 Patterns of Industrial Bureaucracy. Glencoe, Ill.: The Free Press. 1957 "Cosmopolitans and Locals." Administrative Science Quarterly, 2 (Dec), pp. 281-306, 444-480. Greenwood, Ernest 1957 "Attributes of a Profession." Social Work. 2 (Jul). pp. 44-55. Hage, Jerald 1965 "An Axiomatic Theory of Organizations." Administrative Science Quarterly, 10 (Dec), pp. 289-320. Hage, Jerald & Michael Aiken 1967a "Program Change and Organizational Properties." American Journal of Sociology, 72 (Mar), pp. 503- 509. 1967b "Relationship of Centralization to Other Structural Properties.” Administrative Science Quarterly, 12 (Jun), pp. 72-92. 1969 ”Routine Technology, Social Structure, and Organizational Goal.” Administrative Science Quarterly, 14, pp. 366-375. 1970 §0c1al Change in Complex Organizations. New York: Randdm House. Hage, Jerald, Michael Aiken and Cora Bagley Marrett 1971 "Organization Structure and Communications." American Sociological Review, 36 (Oct), pp. 860- 871. Hall, Richard H. ~ 1962 "Intraorganizational Structural Variation." Administrative Science Quarterly, 7 (Dec), pp. 295- 308: 1963 "The Concept of Bureaucracy." American Journal of Sociology, 69 (Jul), pp. 32-40. 1968 "Profeséionalization and Bureaucratization." American Sociological Review, 33 (Feb), pp. 92-104. Ha Ha 95 1972 Organizations. Englewood Cliffs, N.J.: Prentice- Hall. 1975 Occupations and the Socingtructure. 2nd ed. Englewood Cliffs, N.J.: PrentiCe-Hall. Hall, Richard H., J. Eugene Haas, and Norman J. Johnson 1967 "Organizational Size, Complexity, and Formalization." American Sociological Review, 32 (Dec), pp. 903-912. Hall, Richard H. & Charles R. Tittle 1966 "Bureaucracy and Its Correlates." American Journal of Sociology, 72 (Nov), pp. 267-272. Hampton, David R., Charles E. Summer, and Ross A. Webber 1973 Organizational Behavior and the Practice of Management. Rev. ed. Glenview, 111.: Scott, EEEesman. Hasenfeld, Yeheskel 1972 "People Processing Organizations: An Exchange Approach." American Sociological Review, 37 (Jun), pp. 256-63. Hasenfeld, Yeheskel & Richard A. English, eds. 1974 Human Service Organizations. Ann Arbor: University of Michigan Press. Hawkes, Robert W. 1961 "The Role of the Psychiatric Administrator." Administrative Science Quarterly, 6 (Jun), pp. 89- 106. Also in Hasenfeld & English (1974). Hickson, J. J., D. S. Pugh and D. C. Pheysey 1969 "Operations Technology and Organizational Struc- ture." Administrative Science Quarterly, 14 (Sept). pp. 378-397. Holdaway, Edward A. & Thomas A. Blowers 1971 "Administrative Ratio and Organizational Size: A Logitudial Examination." American Sociological Review, 36 (Apr). Homans, George C. 1961 Social Behavior: Its Elementary Forms. New York: :—__ Harcourt, Brace & World. Katz, Daniel & Robert Kahn 1966 The Social Psychology of nganizations. New York: Wiley. 96 Krech, David, Richard S. Crutchfield, and Egerton L. Ballachey 1962 Individual in Society. New York: McGraw—Hill. Lefton, Mark 8 William Rosengren 1966 "Organizations and Clients: Lateral and Longi- Levine, 1961 Lipset, 1956 Litwak, 1961 1962 Lorsch, 1965 Marcus, Adams 1974a 1974b Mendenhall, William 1968 Merton, 1968 tudinal Dimensions." American Sociological Review, 31 (Dec), pp. 802-10. Sol & Paul White "Exchange and Interorganizational Relationships." Administrative Science Quarterly, 5 (Mar), pp. 583- 601. Seymour, et a1. Union Democracy. Garden City, N.Y.: Doubleday. Eugene & Lydia F. Hylton "Models of Bureaucracy Which Permit Conflict." American Journal of Sociology, 67 (Sept), pp. 177- 84. ”Interorganizational Analysis." Administrative Science Quarterly, 6 (Mar), pp. 395-420. Jay W. & Paul R. Lawrence "Organizing for Product Innovation." Harvard Business Review (Jan-Feb). Philip M., Ann Workman Sheldon, and Margaret J. ”The Dynamics of Interorganizational Networks." Paper presented at the North Central Sociological Association Meetings, Windsor, Ontario, May, 1974. "Increasing the Supply of Social Services." Paper presented at the Study of Social Problems Meeting, Montreal, Quebec, August, 1974. Introduction to Linear Models and the Design and Analysis_pf Experiments. Belmont, Calif.: Duxbury Press. Robert K. Social Theor and Social Structure. Rev. ed. New York: The Free Press. Meyer, Marshall W. 1968a "Two Authority Structures of Bureaucratic Organi- zation." Administrative Science Quarterly, 13 (Sept). pp. 211-228. . 97 1968b "Automation and Bureaucratic Structure." American Journal of Sociology, 74 (Nov), pp. 256-264. 1971 “Some Constraints in Analyzing Data on Organiza- tional Structure: A Comment on Blau's Paper." American Sociological Review, 36 (Apr), pp. 294-297. Miller, George 1967 "Professionals in Bureaucracy: Alienation Among Industrial Scientists and Engineers." American Sociological Review, 32 (Oct), PP. 755-768. Parsons, Talcott 1949 The Structure of Social Action. New York: The Free Press. Pennings, Johannes ~ 1973 "Measures of Organizational Structure: A Methodo- logical Note." American Journal of Sociology, 79 (Nov). pp. 686-704. Perrow, Charles 1967 ”A Framework for the Comparative Analysis of Organizations." American Sociological Review, 32 (Apr). pp. 194-208. 1972 Complex Organizations. Glenview, 111.: Scott, Foresman. ' Pugh, D. S. et a1. 1963 "A Scheme for Organizational Analysis." Adminis- trative Science Quarterly, 8 (Dec), pp. 289-315. 1968 “Dimensions of Organizational Structure." Administrative Science Quarterly, 13 (Jun), pp. 65-i05. 1969 "The Context of Organizational Structure." Administrative Science Quarterly, 14 (Mar), pp. 91-114. Riley, Matilda White, John W. Riley, Jr., and Jackson Toby 1954 "Consensus." In M. Riley et al., Sociological Studies in Scale Analysis. New Brunswick, N.J.: Rutgers University Press. Ritzer, George 1973 "Professionalism and the Individual." In Eliot Freidson (ed.), 1973. Ruzek, Sheryl K. 1973 "Making Social Work Accountable." In Eliot Freidson (ed.), 1973. rn 98 Schein, Edgar 1970 Or anizational_Psychology. 2nd ed. Englewood CIiffs, N.J.: Prentice-Hall. Scott, William A. 1968 "Attitude Measurement." In Gardner Lindzey and Elliot Aronson (eds.), The Handbook of Social Psyohology, Vol. 2. 2nd ed. Reading, Mass.: Addison-WesIey, pp. 204-273. Scott, W. Richard 1966 "Professionals in Bureaucracies--Areas of Con- flict." In Vollmer and Mills (eds.), Profession- alization. Englewood Cliffs, N.J.: Prentice-Hall, pp. 265-75. Selznick, Philip 1949 TVA and the Grass Roots. Berkeley: University of California Press. Sheldon, Ann WOrkman 1974 "Organizational Interaction in the Social Services System: A Two City Comparison." Dissertation proposal, Michigan State University. Stinchcombe, Arthur L. 1959 "Bureaucratic and Craft Administration of Pro- duction.” Administrative Science Quarterly, 4 pp. 168-187. 1965 "Social Structure and Organizations." In James G. March (ed.), Handbook of Organizations. Chicago: Rand McNally. Tannenbaum, Arnold et a1. 1974 Hierarchy in Orgapizations: An International Comparison. San Francisco: Jossey-Bass. Thomas, Edwin J. 1959 "Role Conceptions and Organizational Size." American Sociological Review, 24 (Feb), pp. 30-37. Thompson, James D. 1967 Organizations in Action. New York: McGraw-Hill. Udy, Stanley H. Jr. 1959a "The Structure of Authority in Non-Industrial Production." American Journal of Sociology, 64 (May), pp. 582-584. 1959b "'Bureaucracy' and 'Rationality' in Weber's Organization Theory." Amerioan Sociological Review, 24 (Dec), pp. 791-795. 99 1961 "Technical and Institutional Factors in Production Organization: A Preliminary Model." American Journal of SOCiOlOQZI 67 (Nov), pp. 247-54. Upshaw, Harry S. 1968 "Attitude Measurement." In Hubert M. Blalock and Ann Blalock (eds.), Methodology in Social Research. New York: McGraw-Hill. Vinter, Robert D. 1963 "Analysis of Treatment Organizations." Social Work, 8 (Jul). Woodward, Joan 1973 "Management and Technology." In D. Hampton et a1. (comps.), Organizational Behavior and the Practice of Management. Rev. ed. Glenview, 111.: Scott, Foresman. Warren, Roland 1967 "The Interorganizational Field as a Focus for Investigation.” Administrative Science Quarterly, 12 (Dec). Pp. 369-419. "42071410“