Cc.) Copyright by GAIL LUND BARWIS 1977 PUBLISHER AND JOURNALIST ATTITUDES TOWARD THE CONCEPTS AND COLLECTIVE BARGAINING PROVISIONS OF NEWSROOM DEMOCRACY By Gail Lund Barwis A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Communication Arts and Sciences Mass Media Program 1977 ABSTRACT PUBLISHER AND JOURNALIST ATTITUDES TOWARD THE CONCEPTS AND COLLECTIVE BARGAINING PROVISIONS OF NEWSROOM DEMOCRACY By Gail Lund Barwis With the l9705 has come a movement among reporters to secure greater rights in the production of news and protection of their work product. One avenue for effecting these rights--or newsroom democracy as it has come to be known--is the collective bargaining agreement between the publisher and The Newspaper Guild (TNG), the union which represents approximately half the reporters in this country. The bargaining proposals formulated by TNG to secure newsroom democracy for editorial employes were the focus of the study. Since the adoption of these rights issues as part of its bargaining program in l970, the Guild has achieved its demands to varying degrees but has made no assessment of its successes and failures. The study was designed to measure publisher and editorial employe attitudes toward the general concepts embodied in newsroom democracy and toward contractual agreements on newsroom democracy. Basically, three major issues are involved: (1) Voice, or greater participation in policy making; (2) Integrity, or protection of the work product and the employe's professional integrity; and (3) Privilege, or protection against forced disclosure of news sources. G. L. Barwis Newsroom democracy challenges the traditional publisher- reporter relationship. It suggests that there should be a contractual means by which the employe can protect and improve the nature of his individual product and thereby affect the nature of the employer's overall product. But in allowing the employe greater control over his work, newsroom democracy limits the rights of management. The study attempted to answer the following questions: (1) To what degree do publishers and editorial employes find each of the newsroom democracy issues acceptable as general principles and as contract provisions? (2) How do the attitudes of publishers and editorial employes differ on contractual newsroom democracy? (3) Do certain characteristics exist among publishers and editorial employes which would suggest more favorable attitudes toward contractual newsroom democracy? The data were collected by means of a survey questionnaire mailed to all publishers currently holding contracts with The Newspaper Guild and to a random sample of Guild members. The hypotheses were tested by one-way analysis of variance, difference of means T-tests and Pearson correlations. The data support the following conclusions: (l) Editorial employes are more favorable toward the general principles of newsroom democracy than are publishers. (2) Both editorial employes and publishers agree with the general principle of not requiring journalists to distort facts or knowingly create false impressions. G. L. Barwis (3) Editorial employes are more favorable toward incorporating newsroom democracy provisions in contracts than are publishers. (4) Editorial employes view corresponding newsroom democracy principles and contract provisions synonymously while publishers do not. (5) Reporters' attitudes toward The Newspaper Guild strongly influence their attitudes toward contractual newsroom democracy: those who are strongly supportive of the Guild are highly in favor of news- room democracy. For Bill and Our Lord ii ACKNOWLEDGEMENTS I wish to offer my thanks and appreciation to those who made the writing of this dissertation possible: to all the publishers and editorial employes who participated in the survey; to the members of my Guidance Committee, George A. Hough III and Maurice R. Cullen Jr. of the School of Journalism, Fred S. Siebert, professor emeritus, Benjamin Wolkinson of the School of Labor and Industrial Relations, and John D. Abel of the Department of Telecommunication, director of the disserta- tion, all of whom offered advice and encouragement throughout the study; to David J. Eisen and Charles A. Perlik Jr. of The Newspaper Guild who provided invaluable assistance and information; to Robert Bruner who first interested me in the work of the Guild; to my loving parents, Louis S. and Virginia K. Lund; and most importantly, to my husband, Bill, to whom this work is dedicated, and to our child-to-be, who have made the task of completing this study a joy and a privilege. "Except the Lord build the house, they labor in vain that build it." Psalm l27:l TABLE OF CONTENTS Chapter 1. STATEMENT OF THE PROBLEM ................. l Changes in Newsroom Control .............. l Purpose of the Study .................. 5 Questions to Be Answered ................ 7 Definitions Used in the Study ............. 8 Parameters of the Study ................ 9 II. LITERATURE REVIEW .................... 10 Worker Participation .................. 10 Management Rights and Collective Bargaining ...... 14 The Newspaper Experience ................ 21 U.S. Experience ................... 21 International Experience ........... j. . . . 26 Current Status of Newsroom Democracy .......... 30 Voice ........................ 31 Integrity ...................... 38 Privilege ...................... 43 Newsroom Democracy Publication Characteristics . . . . 49 First Contract Date .............. A. . 49 Circulation .................... 49 Group Affiliation ................. 50 Research Hypotheses .................. 50 III. DESIGN OF THE STUDY ................... 56 The Sample ....................... 56 The Survey Instrument ................. 56 The Pretest ..................... 59 Covering Letters and Mailing Procedures ....... 61 Survey Returns .................... 61 Editorial Employes ................. 61 Publishers ..................... 62 Testable Hypotheses .................. 63 Hypotheses 1A - 1R .................. 63 Hypotheses 2A - 2R .................. 70 Hypotheses 3A - 3R .................. 75 Hypotheses 4A - 4C .................. 81 Hypotheses 5A - 5C .................. 84 Hypotheses 6A - 6C .................. 86 iv Hypotheses 7A - 7C .................. 87 Hypotheses 8A - 8C .................. 89 Hypotheses 9A - 9C .................. 90 Hypotheses 10A - 10C ................. 92 Data Analysis ..................... 93 IV. RESULTS AND DISCUSSION .................. ' 95 Descriptive Statistics ................. 96 Validity and Reliability ................ 102 Validity ....................... 102 Reliability ..................... 105 Results of Hypotheses Tests .............. 108 Test of Hypotheses 1A - 1R .............. 108 Test of Hypotheses 2A - 2R .............. 114 Test of Hypotheses 3A - 3R .............. 119 Test of Hypotheses 4A - 4C .............. 123 Test of Hypotheses 5A - 5C .............. 128 Test of Hypotheses 6A - 6C .............. 132 Test of Hypotheses 7A - 7C .............. 137 Test of Hypotheses 8A - 8C .............. 141 Test of Hypotheses 9A - 9C .............. 144 Test of Hypotheses 10A - 10C ............. 147 Summary ........................ 149 V. SUMMARY AND CONCLUSIONS ................. 151 Hypotheses 1 - 3 .................... 151 Hypotheses 4 - 10 ................... 157 Limitations of the Study ................ 159 Size ......................... 159 Scope ........................ 159 Definition ...................... 160 Measuring Instrument ................. 160 Generalizability ................... 161 Conclusions ...................... 162 Suggestions for Future Research ............ 164 Applications of the Study ............... 169 APPENDICES APPENDIX A. Model Contract Language ................. 171 B. Supplementary Tables ................... 173 C. Procedures for Drawing Editorial Employe and Publisher Samples .................... 187 0. Publisher and Editorial Employe Questionnaires ...... 189 E. Questionnaire Covering Letters .............. 199 F. Publisher Comments ..................... 205 G. Editorial Employe Comments ................ 218 H. Key to Item Statements and Corresponding Model Contract Language .................... 220 V LIST OF REFERENCES GENERAL REFERENCES ........................ 230 vi 10. 11. 12. 13. LIST OF TABLES Editorial employe job classifications .......... Working relationship between editorial employe and publisher ....................... Relationship between editorial employe and The Newspaper Guild ......................... Publisher title classifications ............. Daily newspaper circulation figures among publisher respondents ...................... Working relationship between publisher and The Newspaper Guild ......................... Difference of means T-test for publishers and editorial employes on general principles of newsroom democracy Difference of means T-test for publishers and editorial employes on contractual provisions of newsroom democracy ....................... Difference of means T-test between correlation for publishers and editorial employes on general principles and contractual provisions of newsroom democracy ....................... One-way analysis of variance F-test comparing publishers' mean scores on Voice-A with newspaper circulation ...................... One-way analysis of variance F-test comparing publishers' mean scores on Voice-B with newspaper circulation ...................... One-way analysis of variance F-test comparing publishers‘ mean scores on Integrity-A with newspaper circulation .............. One-way analysis of variance F—test comparing publishers' mean scores on Integrity-B with newspaper circulation .............. vii 99 99 100 100 101 102 111 116 121 125 125 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. One-way analysis of variance publishers' mean scores on with newspaper circulation One-way analysis of variance publishers' mean scores on with newspaper circulation One-way analysis of variance publishers' mean scores on number of years a contract One-way analysis of variance publishers' mean scores on number of years a contract One-way analysis of variance publishers' mean scores on number of years a contract One-way analysis of variance publishers' mean scores on number of years a contract One-way analysis of variance publishers' mean scores on number of years a contract One-way analysis of variance publishers' mean scores on number of years a contract One-way analysis of variance publishers' mean scores on relationship with The Newspaper Guild One-way analysis of variance publishers' mean scores on relationship with The Newspaper Guild One-way analysis of variance publishers' mean scores on their relationship with The Newspaper Guild One-way analysis of variance publishers' mean scores on their relationship with The Newspaper Guild One-way analysis of variance publishers' mean scores on their relationship with The Newspaper Guild viii F-test comparing Privilege-A F-test comparing Privilege-B F-test comparing Voice-A with the has been in effect ..... F-test comparing Voice-B with the has been in effect ..... F-test comparing Integrity-A with the has been in effect ..... F-test comparing Integrity-B with the has been in effect ..... F-test comparing Privilege-A with the has been in effect ..... F-test comparing Privilege-B with the has been in effect ..... F-test comparing Voice-A with their F-test comparing Voice-B with their F-test comparing Integrity-A with F-test comparing Integrity-B with F-test comparing Privilege-A with 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. One-way analysis of variance F-test comparing publishers' mean scores on Privilege‘B with their relationship with The Newspaper Guild ...... One-way analysis of variance F-test comparing publishers' mean scores on Voice-A with their independent or affiliated status ............ One-way analysis of variance F-test comparing publishers' mean scores on Voice-B with their independent or affiliated status ............ One-way analysis of variance F—test comparing publishers' mean scores on Integrity-A with their independent or affiliated status ......... One-way analysis of variance F-test comparing publishers' mean scores on Integrity-B with their independent or affiliated status ......... One-way analysis of variance F-test comparing publishers' mean scores on Privilege-A with their independent or affiliated status ......... One-way analysis of variance F-test comparing publishers' mean scores on Privilege-B with their independent or affiliated status ......... One-way analysis of variance F-test comparing editorial employes' mean scores on Voice with their relationship with their employer ......... One-way analysis of variance F-test comparing editorial employes' mean scores on Integrity with their relationship with their employer ...... One-way analysis of variance F-test comparing editorial employes' mean scores on Privilege with their relationship with their employer ...... One-way analysis of variance F-test comparing editorial employes' mean scores on Voice with their relationship with The Newspaper Guild ...... One-way analysis of variance F-test comparing editorial employes' mean scores on Integrity with their relationship with The Newspaper Guild . . . . One-way analysis of variance F-test comparing editorial employes' mean scores on Privilege with their relationship with The Newspaper Guild . . . . ix 138 138 139 139 140 140 142 145 146 40. 41. 42. B-1. 8-2. B-3. 8-4. B-5. B-6. B-7. B-9. B-10. B-11. One-way analysis of variance F-test comparing editorial employes' mean scores on Voice with the number of years employed as a journalist ...... One-way analysis of variance F-test comparing editorial employes' mean scores on Integrity with the number of years employed as a journalist ...... One-way analysis of variance F-test comparing editorial employes' mean scores on Privilege with the number of years employed as a journalist ...... Percentage of contracts containing newsroom democracy provisions ....................... Percentage of contracts containing Voice, Integrity and/or Privilege by date of first contract ..... .- . Percentage of contracts containing Voice, Integrity and/or Privilege by circulation ............ Percentage of contracts containing Voice, Integrity and/or Privilege by affiliation ............ Frequency data for publishers and editorial employes on general principle Voice items used in the questionnaire ..................... Frequency data for publishers and editorial employes on general principle Integrity items used in the questionnaire ..................... Frequency data for publishers and editorial employes on general principle Privilege items used in the questionnaire ..................... Frequency data for publishers and editorial employes on contractual Voice items used in the questionnaire . . Frequency data for publishers and editorial employes on contractual Integrity items used in the questionnaire ..................... Frequency data for publishers and editorial employes on contractual Privilege items used in the questionnaire ..................... Inter-item correlations for publishers and editorial employes on general principle and contractual Voice items and indices ................ 148 148 149 173 174 174 B-12. Inter-item correlations for publishers and editorial employes on general principle Integrity items and indices ...................... 183 B-13. Inter-item correlations for publishers and editorial employes (Ni contractual Integrity items and indices ........................ 184 B—14. Inter-item correlations for publishers and editorial employes on general principle Privilege items and indices ...................... 185 B-15. Inter-item correlations for publishers and editorial employes on contractual Privilege items and indices ........................ 186 xi CHAPTER I STATEMENT OF THE PROBLEM Changes in Newsroom Control Freedom of the press traditionally has been equated with the publisher's right to determine the content of the news product. Yet in this decade a sharing of that right has become evident in many newsrooms as reporters are being given a consultative role in the generation, selection and presentation of news. The changing pattern of control in the newsroom is the subject of this study. The movement toward a greater say by'editorial employes in the news process had its origin in the wave of media criticism that followed the coverage of the 1968 Democratic National Convention in Chicago. The treatment received by the press during that time from government and police officials and from media managers prompted a group of Chicago reporters to form the Association of Working Press, which began publication of the Chicago Journalism Review in October 1968 as an outlet for criticism of media management shortcomings. The militant, protective mood of Chicago reporters soon spread across the country and began to penetrate the ranks of the largest collective association of reporters in the country, The Newspaper Guild.* The "reporter power" advocates within the Guild *The Newspaper Guild represents approximately 38 per cent of the reporters in this country. It has a total membership of about 30,000, of which half are editorial employes. membership succeeded in 1970 in having the union adopt a new bargain- ing posture which encompassed the non-economic professional issues of voice in the product, protection of the employe's integrity and work product, and protection against forced disclosure of news sources-- areas that had previously been considered the exclusive domain of management. Commenting on the new mood among reporters, former Newsweek Editor Edwin Diamond noted that "Ten or fifteen years ago unions battled to win wage increases and protest mergers, but the way a publication was run...remained the prerogative of the owner. The journalist's attitude was, typically, acquiescent; after all, wasn't it management's bat and ball--and ball park?" (Diamond, 1970). The tenor of the times had changed. The young newspaper activists of the late 19605 and early 19705 came out of the college protest era, Diamond noted. Those who marched for civil rights or sat in at the dean's office (or covered those events for their college newspapers) were now turning reformist on their own pro- fession. At its 1970 annual convention, The Newspaper Guild Collective Bargaining Committee reported: The committee believes that gearing the Guild to the seventies calls for expanding ANG's mandatory bargaining program to focus locals' attention newly or in larger measure on a more direct voice in the product their members turn out, on greater protection of newspaper employes in the exercise of their First Amendment rights, on greater protection for the employe's right to protect his integrity...[and] on greater protection of members in all their activities as citizens. (ANG Proceedings, 1970, p. 115) The Guild's International Executive Board (IEB) then proposed new Voice, Integrity and Privilege Against Disclosure provisions to be incorporated in the Guild's collective bargaining strategy. The main areas of concern were (1) employe voice or input in the selection of supervisory personnel, representation on editorial boards and com- mittees dealing with newsroom practices and policy; (2) protection of employe integrity through the use of bylines, assignment of work, presentation of material and defense of the work product; and (3) employe privilege against forced disclosure of news sources and indem- nification of employes refusing to disclose sources. These rights issues were an abrupt departure from The News- paper Guild's bread-and-butter trade unionist policy that had been in effect since the early days of the Guild. Yet it was also a return to a philosophical approach which had lain dormant since 1937. In December 1933 The Newspaper Guild* was formed as a direct consequence of efforts by reporters to have some say in drafting the Daily Newspaper Code, which was written earlier that year in compli- ance with President Roosevelt's National Recovery Administration (Leab, 1970). The Code was to govern the entire range of industrial and trade activities in the industry as set forth in provisions of Section 3(a) of the National Industrial Relations Act (NIRA), passed by Congress in June 1933. Initially, publishers opposed the formation of a code, but in July of that year the American Newspaper Publishers Association (ANPA) decided it would be in its best interest to draw up its own code for the industry rather than have one written by others. *The name of the organization was changed from the American Newspaper Guild (ANG) to The Newspaper Guild (TNG) by action of the 1971 convention. (TNG Proceedings, 1971) Although public hearings were held on the proposed code as drafted by ANPA, reporters' groups which had organized to present counter pro- posals to the ANPA code found their arguments had little effect on the final version. In the weeks that followed the code hearings, momentum gained among reporters for the formation of an organization which would represent their interests--something the Daily Newspaper Code had failed to do. On Dec. 15, 1933, the American Newspaper Guild was form- ed in Washington, D.C. (Leab, 1970). Publishers had fought against including the newspaper industry under the provisions of the NIRA. When that Act was declared uncon- stitutional by the U.S. Supreme Court and the National Labor Relations Act (NLRA) of 1935 took its place, publishers fought against inclu- sion of the industry under this Act as well, contending a violation of their First Amendment free press rights. Their protestations were unsuccessful, however. In a landmark dismissal case, The Associated Press v. National Labor Relations Board, 301 U.S. 103 (1937), the U.S. Supreme Court made the final determination that publishers' freedom of the press was not violated by employes who chose to join the Guild, nor were publishers subject to any special immunity from the application of general 1aws--in this case, the NLRA. Although the Guild had succeeded in organizing editorial employes, it did so equivocating between trade unionist and pro- fessional philosophies. In 1935 R.S. Gilfillan, a Guild regional vice president, wrote in Journalism Quarterly: There has been much petty surmising over whether the Guild is a labor union or a professional group. I don't think there are many members of the ANG who believe it is either of these.... The Guild is a guild and if its negotiations with publishers take on the appearance of trade unionism, it is still a guild. If it seeks to set up and maintain a professional standard, it is yet a guild and that alone. The philosophical coexistence of bread-and-butter and pro- fessional issues, however, was abandoned by the Guild in 1937 in favor of a trade unionist approach, as evidenced by its comment as amicus curiae in the Transcript of Record of the AP v. NLRB case: Petitioner [AP]...remains the complete master of its business. Its prejudices are still supreme....If an employe exhibits a bias in reporting news events which is distasteful to petitioner, petitioner may reject his work, may alter it or may discharge the offending employe. The privilege which petitioner demands “to print the news it wants as it wants to" remains wholly unimpaired. Not until the institution of the newsroom democracy clauses in 1970 did the Guild again actively support professional, ethical concerns . Purpose of the Study The bargaining proposals formulated by The Newspaper Guild to secure newsroom democracy rights for reporters are the focus of this study. The term newsroom democracy has been used herein to refer to the Guild-proposed contract provisions covering Voice, Integrity and Privilege. Since the adoption of these issues as part of its bargaining program, the Guild has achieved its demands to varying degrees yet has made no assessment of its successes and failures. The provisions were unanimously adopted at the 1970 convention by vote of the dele- gates, but neither Guild members nor publishers with whom the Guild bargains were questioned on their attitudes toward the general con- cept of providing reporters with such rights, nor on the desirability of securing such rights through contractual agreements. This study is designed to measure publisher and editorial employe attitudes toward a general concept of newsroom democracy and toward a contractual agreement on newsroom democracy. It attempts to identify those areas where agreement exists among publishers, among employes, and between both on specific newsroom democracy issues. Those issues on which the attitude assessments are based are taken from the contract language proposed by the Guild (see Appendix A). The purpose of the study is to determine whether agreement exists among publishers and editorial employes on specific newsroom democracy issues and whether those areas of agreement--in the opinion of publishers and employes--ought to be formalized as contract pro- visions. In other words, are publishers willing to incorporate news- room democracy in collective bargaining agreements? Are editorial employes in favor of such action? The issues raised by the attempt to include newsroom demo- cracy in editorial contracts apply both to reporter-publisher rela- tions in particular and to labor-management relations in general. Voice, Integrity and Privilege are all integral parts of reporter professionalism and are tied to the concepts of worker participation and management rights. The Guild‘s proposals would in effect ensure reporters the right to self-assessment of the quality of the repor- torial product. The concept of newsroom democracy challenges the traditional publisher-employe relationship. It suggests that there should be a contractual means by which the employe can protect and improve the nature of his individual product and thereby affect the nature of the employer's overall product. But in allowing the employe greater control over his work, newsroom democracy by its very nature limits the rights of management. This study will help to assess the attitude among editorial employes and publishers as to the feasibility of legal, or contractual, newsroom democracy. It will also offer insight into the future course of publisher-reporter relations in the newspaper industry. And in a broader context it will establish new definitional parameters for worker participation and management rights. Questions to be Answered This study attempts to answer the following: 1. To what degree do publishers find each of the newsroom democracy issues acceptable in principle? 2. To what degree do publishers find each of the newsroom democracy issues acceptable as a contract provision? 3. To what degree do editorial employes find each of the newsroom democracy issues acceptable in principle? 4. To what degree do editorial employes find each of the newsroom democracy issues acceptable as a contract provision? 5. How do the attitudes of publishers and editorial employes differ on contractual newsroom democracy? 6. 00 certain characteristics exist among publishers and editorial employes which would predispose them to more favorable attitudes toward contractual newsroom democracy? Definitions Used in the Study l. Newsroom democracy refers to Guild-proposed provisions covering Voice, Integrity and Privilege. These three areas have been subdivided into eighteen issues for the purpose of analysis (three dealing with Voice, seven with Integrity and eight with Privilege). They are listed in Appendix A. 2. Editorial employe refers to any person working in the editorial department covered by the Guild contract.* This study con- cerns itself with responses from those employes working only in the editorial department. Throughout the study the terms reporter, news- writer, newsperson, journalist and editorial employe are used inter- changeably. It should be noted, however, that editorial employe is actually a broader term than reporter or newswriter. It may encom- pass such job descriptions as copy editor, photographer, desk man, city editor, clerk, librarian, etc., provided the job description falls within the purview of the editorial department and Guild juris- diction. 3. Publisher is used generically to refer to the principal employer. The publisher may also be known as the general manager or business manager at some publications. Only those publishers who hold contracts with The Newspaper Guild are included in this definition. 4. Publications with which the Guild holds contracts include the following: *The Newspaper Guild has contracts covering employes who work in more than 66 different divisions or departments in various publi- cations. Editorial is one such department. a. Daily and Sunday newspapers of general circulation b. Wire and news services c. Foreign language newspapers d. Weekly, semi-weekly and tri-weekly newspapers e. Magazines f. Radio and television stations 9. Labor papers and services h. Religious papers and services i. Miscellaneous publications and services. Parameters of the Study The Newspaper Guild bargaining proposals on Voice, Integrity and Privilege are used as the definition of newsroom democracy in this study in order to empirically assess the changing labor-management picture in the newsroom. The findings and conclusions are based solely on this language and on the working relationship between edi- torial employes and publishers who are covered by Guild contracts. No attempt has been made to extrapolate these results to non-Guild publish- ers or to other definitions of newsroom democracy. The study assesses the attitudes of publishers and reporters involved with the Guild. It does not attempt to assess the theo- retical legitimacy of the Guild's philosophy in seeking such rights. CHAPTER II LITERATURE REVIEW Newsroom democracy, or participation by editorial employes in newsroom decision and policy making, can be viewed as one form of industrial democracy, a term broadly used to describe increased parti- cipation by employes in decisions which affect their work. Industrial democracy, therefore, is intimately linked with the issue of manage- ment rights; and when workers are represented collectively by a union, the issue is further linked with the collective bargaining process. This chapter first reviews the literature on the theory of worker participation, then on the relationships among participation, management rights and collective bargaining and finally on how those relationships have been dealt with in the newspaper industry both here and abroad. Worker Participation The general theory underlying industrial democracy is that by providing for employe input, the industrial system is improved. In- cluded under "improved system" are such factors as (1) increased job satisfaction among employes, (2) better employer-employe relations, (3) higher quality and production levels and (4) greater interest among workers in being associated with the product. According to Daniel and McIntosh (1972), the pressure upon management to give workers a greater say in decision making has come 10 11 from three main sources: (1) the increasing acceptance by management itself of the proposition that workers are more committed to aims, objectives or goals that they themselves have played a part in setting than they are to those that are imposed from above; (2) the demand of those in subordinate positions for greater influence in shaping their lives (in other words, the belief that lack of involvement is a denial of the worker's democratic freedom); and (3) inevitable loss Of' industrial efficiency and control by failure to make policies and practices consistent with social values outside the industry. The classic General Electric Hawthorne plant study in worker participation in decision making (Roethlisberger and Dixon, 1964) and many subsequent studies support the first proposition. According to Blumberg (1968), "There is hardly a study in the entire literature which fails to demonstrate that satisfaction in work is enhanced or that other generally acknowledged beneficial consequences accrue from a genuine increase in workers' decision-making power." Among the studies cited by Blumberg is one by French (1960) of production workers in a Norwegian shoe factory. Experimental groups were given the power of participation in certain areas of man- agerial decision making with respect to the production process, and their work and attitudes were compared with those in control groups. French found that although the experimental groups were not more productive than the control groups, satisfaction with the company and attitudes toward labor relations tended to be higher among the exper- imental groups. Within the experimental groups, those whose atti- tudes had changed most tended also to believe that participation was a legitimate worker's right. 12 In an experimental study by Morse and Reimer (1956) office workers in two sections of a large insurance company were given sub— stantially increased powers of decision making while those in two other sections were given substantially less power. Their results showed that workers in the autonomous sections experienced signifi- cant changes from the pre-experimental situation, including greater feelings of self actualization on the job, greater satisfaction with supervisors, greater satisfaction with the company, and greater in- trinsic work satisfaction. Workers in the hierarchically-controlled sections experienced significant decreases in each of these areas. Bavelas (1942) compared the efficiency of sewing machine operations in a garment factory where some workers were allowed to discuss and decide upon production and others were permitted only to discuss production. He found that the discussion and decision group was clearly the more efficient. Bavelas' study was replicated by Lawrence and Smith (1955) among groups of office and manual workers in a garment factory. They found that the discussion and decision group increased production significantly over its pre-experimental level and over that of the discussion-only group. Fleishman and Harris (1962) studied the grievance and turn- over rate among production workers in a U.S. truck manufacturing plant. They found that workers under foremen whose supervision included participation by workers had far fewer grievances and far less turnover than workers under foremen whose mode of supervision was characterized by authoritarianism. Viteles (1953) summarized the conclusions reported by various researchers on employe participation by stating that employe 13 participation in decision making in a democratic atmosphere increases motivation and raises the levels of employe production and morale. Studies by Herzberg (1959), Argyris (1964, 1974) and Likert (1961) have focused on the importance of job content, interest and relations with superiors and peers as determinants of involvement, satisfaction and attitudes. They suggest that job enrichment, parti- cipative styles of management and increased employer-employe consul- tation improve worker effectiveness. The demand for greater voice by the worker in the kinds of work he will perform has been growing both in this country and abroad. Carmichael (1974) pointed out that the traditional labor-management relationship is receiving closer scrutiny in this era of accelerating social change. With the emergence of a more affluent, better educated and more critical worker, Carmichael contended business leaders throughout the world are experiencing increased pressure for parti- cipation in areas once reserved for management alone. One of the most successful examples of worker participation in America is the so-called Scanlon plan, first introduced by Joseph N. Scanlon at the Empire Steel Company in 1937 (Lesieur, 1958). The plan involves both a plantwide incentive system which allows employes to realize increased earnings from increased effort and labor-saving ideas, and the establishment of joint employe-management committees to discuss ways to increase production efficiency. Under the plan, information sharing takes place on a broad scale and flows in both directions. The plan is now being used successfully by more than fifty companies in the United States. American involvement in industrial democracy is relatively 14 new compared to that of European nations. In Sweden forms of indus- trial democracy began to emerge in the 19205. In West Germany the concept of co-determination was introduced after World War II. In Yugoslavia a national system provides that all employes of an enter- prise have ultimate authority with regard to basic policy, personnel, and technical issues of the firm. And in Great Britain industrial democracy has existed in a number of industries for many years. Management Rights and Collective Bargaining Worker participation challenges the appropriateness of the traditional patterns of management, a challenge which was first created by the growth of trade unionism. In Great Britain, the United States and the Scandinavian countries, unions have developed an extensive bargaining role with the employer. Although the formal limits of negotiation are the sub- jects of continuing conflict, there is initially no limit to the range of managerial decisions that might become the subject of union bargaining pressure. But on the other hand, traditional procedures, and certain court and administrative decisions (such as the National Labor Relations Board's rulings on what are mandatory subjects of bargain- ing) have placed restrictions on the union's bargaining range. In Western European countries such as Austria, WestGermany, France, Italy and Holland, the unions have concentrated on negotiating industrywide agreements and on bringing pressure to bear on the government to regulate industrial practices by law. In these coun- tries the State has compelled employers to establish Work Councils through which the interests of employes can be protected. Such 15 councils, including those representing journalists' societies, have become increasingly powerful. According to Davey (1972) and Basagni and Sauzey (1975) there is a powerful current now flowing in the direction of increased participation in the management of enterprises worldwide. Yet in the American tradition union representation in official managerial bodies is rejected (Carmichael, 1974). Daniel (1972) suggested that part of this rejection may lie in the fact that the union is not viewed by employes as a channel of involvement. Trade unions, he contended, traditionally have limited themselves to narrow economic goals. But with changing societal needs, the pattern must now change. In studies conducted in Great Britain, Goldthorpe (1968) reported that in four out of five occupational groups, employes felt that unions should limit themselves to their specifically economic functions. Only among the craftsmen was the idea of greater worker control largely upheld as a union objective. These same workers also suggested that a larger share in the running of the firm not only was the employe's right but also would make for improvements in operating efficiency. Chamberlain (1967) conducted an extensive study of union pene- tration into management areas in six industries: auto, steel, rubber, electrical equipment, meat packing and utilities. He found that while the unions have been interested occasionally in such areas of management as line products and research and development, there have been no plant programs to share authority in these areas. He reported that unions have made proposals in these areas but have not backed them with bargaining strength. The proposals were withdrawn 16 when rebuffed by management, Chamberlain reported, yet there was no indication that such reception resulted in their losing interest in these fields or in renouncing future action. Chamberlain found that among the six major industries studied, the primary areas of management penetration have been personnel practices and production. He suggested that the union's interest in management matters is a quest for security, not power. He quoted one union leader as saying: Any [union] leader who speaks of democracy in industry as a drive is indulging in intellectualization which the worker thinks little of in such terms, though he may think of it in terms of his having greater protection, greater voice in the decisions that affect him--not for the sake of participating but for what participation can do for him. Another union official suggested that Workers want something more than the assurance of a full belly. Even if security is assured, they don't want to be pushed around. They want participation as well. If democracy means anything, it must come in industry to give any meaning to democracy in the political sphere. The Steelworkers Organizing Committee, noted Chamberlain, addressed the question of needs in terms of equality between manage- ment and worker: Underlying the aspirations for full participation...is labor's basic position of equality. This is more than equality at the bargaining table. It is an intellectual equality....The idea that workers are intellectually inferior to management is the source of many of our difficulties....The intellectual equality of management and labor is a prerequisite to the development of industrial relations along constructive lines. A study by Miles (1965) of a large number of managers revealed that most managers approved of soliciting advice from workers, but at the same time they felt their employes were unimagin- ative and unlikely to come up with worthwhile ideas. These managers 17 looked upon participation as a means of raising morale and of making employes more docile and content. Chamberlain reported several reasons why management is reluctant to share its authority: (1) it destroys unified final authority, (2) management's legal responsibility to its stockholders precludes it from granting concessions that dilute its responsibility, (3) it endangers efficiency, (4) unions seek greater authority without being willing to assume concomitant responsibilities, (5) management is suspect of union's motives and (6) management fears that small concessions will lead to greater ones and will threaten the business- man's concept of private property. In 1945 President Truman convened a National Labor-Management Conference to define the areas and issues between the two groups. Yet no agreement could be reached. Each side presented its own case. Although labor conceded that "at the present time there are operations which can best be performed by management alone," it added the caveat that "these areas and operations are not fixed. Some future occasion may demonstrate the desirability of shifting them to the category of decisions in which labor plays an advisory role or where labor cooperates on a full partnership basis." Management, however, set down what it considered to be its impenetrable rights. Among the areas listed in 1945 (and still adhered to by most managers today) which are being challenged by newsroom democracy provisions are (1) the determination of products to be manufactured or services to be rendered to customers by the enterprise, (2) selection of employes for promotion to supervisory and other managerial positions, and (3) the determination of job 18 content, the allocation and assignment of work to workers, and estab- lishment of quality standards for judgment of workmanship required. Chamberlain contended that the move to industrial democracy is inevitably evolutionary: Management who seek to confine unions to traditional areas of collective bargaining are fighting not only unions as economic organizations but also the historical continuity of a still developing philosophy--a philosophy which they themselves have accepted as desirable in the political sphere. Union participa- tion must be viewed not simply as a device for satisfying the goals of employes, unrelated to other social phenomena; it is encouraged by a prevailing social philosophy which has deep historical roots and which is still growing. The direction which unions have taken has been guided not by self-choice but by the democratic philosophy of the society of their origin. Derber, Chalmers and Stagner (1958) surveyed fifty-one Illinois firms on the extent to which collective bargaining had penetrated the managerial function. They found that in only a few functions, mainly concerning job security and work effort, had unions generally succeeded in obtaining a voice. They hypothesized that (l) the degree of union penetration into management rights would be related to differences in the quality of the union-management rela- tionship, (2) where there was high union influence there would be high penetration, (3) similar organizations would have similar patterns of agreement, and (4) there would be more penetration in older contracts. These hypotheses were confirmed (except (3) which was not tested). Their study showed that management rights had not been deeply penetrated as the 1945 conference managers had feared. The employe's allegiance to the goals of his union and/or his employer is not always clear-cut. Gruen (1954) found that dual allegiance is often a complicating factor in an employe's attitude toward his union or company policy. He suggested that an employe's 19 allegiance in a given situation depends on how the worker perceives a discrepancy or a convergence between his own needs and the perceived goals of either organization. Kerr (1950) found a significant correlation between attitude toward company and attitude toward the union in a plant where management had wholeheartedly accepted the union. Kerr suggested that the correlation would be negative in a plant characterized by warfare between the two groups, since the conflict would tend to force workers to choose allegiance to one at the expense of the other. Sayles and Strauss (1953) found that anxiety arises among workers when they feel they are attacking management unfairly. They reported that the average worker may resent being dominated by his employer, but that he is grateful for his job. Sayles and Strauss also studied the attitudes of rank and file union members toward their union. They found that the average member is convinced he needs his union for protection against arbi- trary management action and for obtaining economic security, but that few workers have a feeling of identification with their union. Atti- tudes toward union goals in one large industrial plant showed 68 per cent of the membership were neutral or had no opinions concerning the union's goals. In a critique of participative management studies, Strauss (1966) pointed out that the effectiveness of worker participation varies with the nature of the work, the locus of information, the need for quick decision making, culture and personality. He suggested that if the employer has most of the needed information and quick decisions must be made, then he is the one to make them. But 20 if information is spread throughout the organization, coordination is highly desirable. And if time pressures are not great, the answer may be some sort of committee participation. Strauss noted that workers from different backgrounds have different needs and that college graduates expect to be consulted. With 'UHB growth of professional unions, as among nurses and teachers, areas once considered management prerogatives are now being opened for collective bargaining. Such issues as the nature and content of curriculum, selection of textbooks and other teaching materials, and assignment of teachers to non-teaching functions are regarded both as vital professional concerns and as "working conditions" (Davey, 1972). Findings from 1a study by Alluto and Belasco (1974) of attitudes among teachers and nurses toward unions suggest that issues such as job autonomy, participation in decision making, longevity of career, and other conditions of professional practice are likely to have a major influence on the level of attitudinal militancy with respect to the enterprise. Chandler (1964) conducted an exhaustive study of union response to the management rights issue of subcontracting. She found that management in smaller plants seemed less concerned over the matter of sharing control with the union and felt less certain that the union had curtailed management's effectiveness. She noted that "On the whole, it would be fair to character- ize management rights as an area marked by very little systematically gathered empirical data." Her study of numerous firms engaged in subcontracting revealed that management position on the rights 21 question was a function of technological and structural factors as well as of organization position with regard to potential change in the area. Far from being an automatic response, managerial adherence to the legalistic model had organizational roots. The Newspaper Industry U.S. Experience Traditionally publishers have held that editorial decisions are strictly management prerogatives. Several studies in the industry have been conducted to determine where editorial decision-making power lies and have shown that it remains in the hands of editors and publishers. There has been little policy making going on at the reporter level. Breed (1955) in his classic study of social control in the newsroom stated that ideally there would be no problem of either control or policy on the newspaper in a full democracy. The only controls would be the nature of the event and the reporter's effective ability to describe it. In practice, he reported, we find the publisher does set news policy and this policy is usually follow- ed by members of his staff. He concluded that structural changes in the newsroom must begin with the publisher and that one of the leverages to be used in making this come about was the union. Argyris (1974) conducted a three-year study of the human dynamics of an unnamed large metropolitan daily newspaper (generally believed to be the New York Times). He suggested among other factors that younger reporters tended to be more aggressive and would ask for significant changes in managerial activities. Argyris reported 22 that tensions in the newspaper "living system" may be distorting its news and slanting its editorials, and he argued for a more open, par- ticipative management based on mutual trust. He suggested that the problems of internal conflict leading to problems of external credibil- ity may be endemic in the profession. Flegel and Chaffee (1971) investigated the influences on reporters' work by editors' and readers' opinions. They found that the views of editors and readers, at least where they differed from the reporter's opinions, were much less influential than the reporter's own beliefs. Bowers (1967) studied the degree to which publishers involve themselves in decisions affecting the amount and nature of the news. He found that (l) the closer the geographical proximity of the subject matter, the more active the publisher is in news direction; (2) the larger the circulation of the newspaper, the less active the publisher is in news direction; (3) publisher activity is higher in areas which conceivably might affect the revenue of the newspaper directly or indirectly; (4) publisher activity is higher in areas which might affect the revenues of the newspaper on those papers under 15,000 circulation, followed in order by newspapers 15,000- 74,999 in circulation and newspapers 75,000 and above; (5) publisher activity has no particular variation from one geographical region to the next; (6) slightly less than one quarter of the publishers were judged by their managing editors as so inactive in the newsroom as to be listed as never or seldom active; and (7) more publishers' direc- tions are expressed than implied when they concern specific news stories. In another classic study Rosten (1937) found that Washington 23 correspondents "sense" what is expected of them without specific orders and govern themselves accordingly as to the selection and treat- ment of the news. Sixty per cent of the 127 correspondents he studied agreed that "It is almost impossible to be objective. You read your paper, notice its editorials, get praised for some stories and criti- cized for others. You sense policy and are psychologically driven to slant your story accordingly." Sigelman (1973) in his study of two southeastern daily news- papers examined how certain attitudes among employes were inculcated through selective recruitment and socialization. He found that over time reporters begin to sense company news policy and to adhere to it. He also found that veteran reporters often have attitudes that are highly supportive of the newspaper, and that it is the experienced reporter whose opinions are recognized in editorial conferences. The question of greater participation in decision making in the newspaper industry is tied to the question of professionalism-- the role in which the newswriter views himself. It is perhaps this self-image more than any other factor which relates to the employe's desire for more of a say in the performance of his duties. Johnstone, Slawski and Borman (1972) conducted a survey of 1,313 journalists on their self~concepts of professionalism and found five types of influencing factors: (1) amount and type of education, (2) age and longevity of service, (3) the community and organizational setting, (4) current job responsibilities, and (5) patterns of pro- fessional and social integration. They compared these concepts with neutral and participative orientation toward journalism and found that participant concepts of the press were associated with higher levels 24 of formal education, those working in large cities, those who tended to belong to professional organizations and associations, and those who were newer in the field. Neutral orientations toward journalism responsibilities were held by older, more established journalists, by those working on smaller newspapers, and those with less formal educa- tion. In a later study (Johnstone, Slawski and Borman, 1976) they pointed out that "As a result of special training, the professional comes to feel that his special abilities entitle him to carry out his work free from interference or pressure from persons either outside or inside the profession. Professional autonomy and freedom from supervision are considered to be of greater importance to the pro- fessional." They noted that some journalists are able to achieve a high degree of autonomy and self-regulation of their work, but many, particularly those in large organizations, are subject to a range of controls. To the extent that journalists consider themselves pro- fessionals, they argued, they regard their freedom to perform to their own standards as especially important. Unlike medical and academic institutions, the term profession- al has a double meaning in the newspaper industry. One is legalistic or statutory and the other is sociological or psychological in nature. In the legalistic sense the National Labor Relations Board has ruled that reporters are not professionals (The San Antonio Express-News Corp. v. San Antonio Typographical Union #127, 223 NLRB 97 (1976)). The Board, after reviewing the decision of its regional director, rejected the publisher's contention that certain categories of employes (reporters, staff writers, columnists, copy editors, 25 editors, editorial writers and cartoonists, referred to collectively as journalists) are professional employes. The Board's ruling was based on the facts that (l) the employer had no policy of hiring and employing only journalists who have a degree evidencing advanced edi- torial training in journalism or communications and (2) the employer did not maintain any formal training apprenticeship program which its employes had to complete prior to being journalists. The employer's position was that the ruling in Jersey Publishing Co., 76 NLRB 467 (1948), should be reversed and journal- ists accorded professional status in view of the dynamic changes in the industry in recent years. Such a ruling would have separated them from the rest of the bargaining unit with the International Typo- graphical Union (ITU), TNG or any other union, and hence would have emasculated their bargaining strength. The ITU and the Guild as amicus curiae argued that journal- ists are not professionals within the statutory definition of the term. They argued that the field of journalism is one of generalists. Section II(12) of the Taft-Hartley Act states: The term "professional employe" means (a) any employe engaged in work (i) predominantly intel- lectual and varied in character as opposed to routine mental, manual, mechanical, or physical work, (ii) involving the con- sistent exercise of discretion and judgment in its performance, (iii) of such a character that the output produced or the result accomplished cannot be standardized in relation to a given period of time, (iv) requiring knowledge of an advanced type in a field of science or learning customarily acquired by a prolonged course of specialized intellectual instruction and study in an institu- tion of higher learning or a hospital, as distinguished from a general academic education or from an apprenticeship or from training in the performance of routine mental, manual, or physical processes, or (b) any employe, who (i) has completed the courses of specialized intellectual instruction and study described in 26 clause (iv) of paragraph (a), and (ii) is performing related work under the supervision of a professional person to qualify himself to become a professional employe as defined in paragraph (a). The Board concluded that "the employer's journalists are not professional employes within the meaning of Section II(12)." It went on to add: We do not question the intellectual demand of the modern journalist's business nor the special responsibilities inherent in his position. We recognize that in the performance of their duties these employes must of necessity exercise judgment in deciding how to approach a given assignment and that the story which ultimately appears in a newspaper is often the result of vigorous intellectual work.... In reaching our conclusion, we do not denigrate journalism as a calling or question in any way the quality of the work product of journalists as a class. The majority contended that a distinction must be made between pro- fessionals within the meaning of the Act and the term professional used to denote a person doing admirable or challenging work. In other countries, however, reporter professionalism does not have two meanings. Journalists for the most part belong to pro- fessional societies similar to unions and in many cases those socie- ties have gained a voice in decision making. International Experience Connor (1974), who studied United States and foreign newsroom democracy, suggested that there are few American newspapers that permit anything approaching the liberty enjoyed by journalists in Western Europe. Yet many of the rights that have been won by European journalists are similar in nature to those proposed in The Newspaper Guild's Model Contract. The origin of reporters' freedom outside this country began in France after World War II when news organs in Paris that had 27 cooperated with the German occupation forces were turned over to journalists and businessmen for reorganization to ensure freedom of the press from government and financial control. Le Monde became the model from which spread the plan for working journalists to partici- pate in policy decisions and management of newspapers (Connor, 1974). Currently Le Monde is controlled by its writers organized under the Corporation of Journalists of Le Monde (CJM). According to a 1968 agreement, the CJM holds 40 per cent of the ownership of the paper, which is run on a non-profit basis for the dissemination of information. Jean Schwoebel has been the head of CJM since its inception in 1952. He said it is difficult to implement the concessions journalists have won in theory (Gruson, 1974). If management wants to exclude the supervisory financial council it simply does not consult with it and later explains the decision was routine. And the editor- ial committee is often faced with decisions already made. The employe's power functions only as a break on management. At Le Figaro (also in Paris) the editorial staff formed an editorial association in July 1965 and went on strike in 1968 for greater power in the operation of the newspaper. Today the corpora- tion is run by a twelve-member supervisory council--seven of whom represent noneditorial workers. Since all major decisions must have the backing of at least nine council members, the editorial staff has effective veto power. The council also chooses a five-member direct- orate, two of whom must be journalists. The directorate is charged with the more immediate problems of the paper's functions. Gruson (1974) suggested that change does not come rapidly. 28 The present structures at Le Monde and Le Figaro evolved over a twenty-year period. An arrangement similar to that at Le Figaro was established at Paris-Normandie in 1972 (Connor, 1974). In Italy the editorial union of journalists (National Press Federation) is making some inroads in its attempt to gain control of the editorial side of newspapers, according to Connor. Recently redefined bylaws at the West German magazine Der Stern state that none of the editors or workers can be forced to do, to write or to defend anything which is against what they believe. Their refusal to do so cannot be held against them. A seven-member elected staff committee represents the interests of the editors and safeguards the journalistic independence of the magazine. Any change in ownership has to be brought to the attention of the editorial committee before it is effective. The hiring or dismissal of the executive editor must also be brought to the committee's attention before the contract is signed or terminated. Stories can be short- ened only if financial reasons make it absolutely necessary. And in such cases the editorial committee must be informed beforehand. At Die Zeit, an independent German weekly newspaper, the by- laws state that the publisher cannot act against the wishes of the majority of editors who have held their position for more than two years. Major changes in the structure of the paper or editorial or- ganization must be discussed with the editors. In the Netherlands after twelve years of negotiations the Nederlandse Dagblad Pers, the organization of newspaper enterprises, and the Nederlandse Veriging van Journalisten, the journalists' 29 union, accepted the final version of the Model Charter for Editors- in-Chief and Editorial Staffs in May 1976 (Snijders, 1976). As of Jan. 1, 1977, every newspaper was required to introduce a version of this Model Charter after consultation with the three parties to the agreement: editorial staff, the editor and the management. After that an editorial council must be consulted on important decisions of a newspaper's policy. Management may not take action without prior advice from the editorial council on the appoint- ment or dismissal of the editor, changes in editorial principles and aims of the paper, or changes in the editorial staff. The charter clearly delimits the functions and responsibil- ities of both the editor and the manager. The editorial council is the organ for editorial participation in all matters pertaining to editorial work. It consists of the editor and at least five members elected by the editorial assembly. The right of the editor to change articles or to give orders regarding editorial work is not denied. But when a member of the staff objects to an order and does not accept the decision taken by the editor, he can appeal to the editorial council. The same applies where an editor has ordered changes in articles. If the article is bylined, the writer may request his name be removed. The International Federation of Journalists (IFJ) in 1972 passed a five-point resolution calling for (1) regular consultation with employes on editorial operations (The publisher's right to set policy is recognized but such policy should be written into con- tracts.), (2) staff veto power over appointments to "leading posi- tions" such as editor, (3) veto power over changes in ownership, 3O (4) a journalist's protection against being required to express an opinion contrary to his convictions, and (5) consultation with employes on all decisions affecting their work (Guild Reporter, Oct. 13, 1972). These resolutions cover many of the same areas found in the Guild's Voice, Integrity and Privilege clauses. At the 1973 meeting of the International Federation of News- paper Editors, concern was raised over the limiting of publishers' prerogatives while at the same time assuming they will take on all the risks (Connor, 1974). One French representative stated that "Since the publisher assumes all liabilities, he must also be able to exercise the right to make decisions concerning the main questions involving the editorial framework. Democracy must not be confused with license." Current Status of Newsroom Democracy In 1976 an analysis was made by the author of all Guild con- tracts which contained some language on Voice, Integrity and/or Privilege. The data were collected from 189 Guild contracts on file at the international headquarters in Washington, D.C. The Voice, Integrity and Privilege proposals were divided into eighteen issues, and the contract language for each issue was coded in four cate- gories: (1) no coverage, (2) coverage but with qualifications, omissions and/or restrictions on the Model Contract language, (3) Model Contract language or its equivalent, (4) Model Contract language or its equivalent but with additional language beneficial to Guild members. 31 Table B-1 (Appendix 8) lists the percentages of contracts containing newsroom democracy clauses. Voice That portion of newsroom democracy which involves reporter input in matters related to the quality of the product and the manner of presentation is labeled Voice. It is embodied in three areas of the Guild's Model Contract: Grievance, Coverage, and General Pro- visions: V-l The Guild shall designate a committee of its own choosing to take up with the Employer or his authorized agent any matter arising from the application of this contract or affecting the relations of an employe and the Employer. V-lA The Guild shall designate a committee of its own choosing to meet with a committee of the Employer's choosing to discuss and to make recommendations affecting all aspects of the operations of all Guild-represented departments, including news coverage. The Guild committee shall meet with the Employer committee once a month. (V-l and V-lA are contained in ARTICLE VI: GRIEVANCE.) V-2 The Guild shall be notified in writing of the name and qual- ifications of any person proposed for appointment to an ex- cluded position. Such appointment shall not be effective until discussed with the Guild. (V-2 is contained in ARTICLE I: COVERAGE, Section 2(b).) V-3 The Guild shall be represented in any editorial or news conferences or boards by representatives of its own choosing. (V-3 is contained in ARTICLE XXIV: GENERAL PROVISIONS, Section 8.) By incorporating matters "affecting the relations of an employe and the Employer" (V-l) under Paragraph 1 of the Grievance Procedure, the Guild seeks to open formal channels with the publisher on issues not specifically related to the application of the contract. Issues concerning reporters' rights, news gathering procedures, and professional ethics--which may not fall under specific provisions of 32 the agreement-~are Open for discussion with the publisher under this provision. The phrase "or affecting the relations of an employe and the Employer" was included in the Guild's first Model Contract (1940), but in 1970 when the Guild began to concern itself with Voice in the product, the phrase was incorporated in the official Collective Bargaining Program as a channel for discussing with the publisher matters of journalistic ethics and news policy. Provisions V-l and V-lA are contained in the first paragraph of the Grievance Procedure. The third paragraph of that Article in the U.S. Model Contract begins: Any matter involving the interpretation, application, admin- istration or alleged violation of this contractTéxcept renewal of this contract), including any question of whether a matter is arbitrable, not satisfactorily settled within 30 days of its con- sideration may be submitted to final and binding arbitration by either party. The Canadian Model Contract, however, begins Paragraph 3 of the Grievance Procedure as follows: Any such matter (except renewal of this contract), including any question as to whether a matter is arbitrable, which is not satisfactorily settled within 30 days of its first consideration may be submitted to final and binding arbitration by either party. A question exists as to whether or not Paragraph 3 of the U.S. Model Contract excludes issues "affecting the relations of the employe and the Employer" from arbitration. The wording of the Canadian contract clearly does not. ("Any such matter" would include both "any matter arising from the application of this contract" and "affecting the relations of an employe and the Employer.") In coding this first issue on the no-coverage to model-plus 33 coverage scale, a value of model language was assigned to contracts containing an avenue for discussion of matters not strictly related to the application of the contract, and a value of model-plus language for any contracts that had model language or a variation thereof for Paragraph 1 in the Grievance Procedure and had a Paragraph 3 which would permit a matter "affecting the relations of the employe and the Employer" to follow the grievance route through to arbitration (in other words, contracts that contained the phrase "any such matter" or its equivalent). To date no grievance has been filed concerning a V-1 matter. The questions of both the arbitrability of a matter "affecting the relations of the employe and the Employer" and the interpretation of Paragraph 1 in the light of Paragraph 3 remain open. In contracts where the phrase "affecting the relations" did [Tot exist, or where the Grievance Procedure was not available, the (Build suggested including a separate provision for a standing com- nnittee that would be able to raise the same kinds of discussions with the publisher (V-lA). However, fewer than 5 per cent of the contracts with Voice contain this alternative to the V-l language, and in most cases this standing committee is included in addition to the V-l language. Of the three Voice in the product statements sought by the Guild, the only one which has had a favorable reception from publish- eV‘s is a joint comnittee to discuss editorial matters or the inclu- Sl<>n of the phrase "affecting the relations of an employe and the E"lployer" under the Grievance Procedure. Only three contracts have Provision for discussing with the Guild appointment of supervisory 34 personnel, and all restrict the model language in some fashion. The third Voice item, that of representation on editorial boards and in news conferences, has not been accepted by any publisher, not even in modified form. Conceivably, questions regarding the appointment of supervisory personnel in the newsroom and matters concerning editorial policy could be raised under the first Voice provision. But publishers have demonstrated their reluctance to include separate provisions for editorial board representation and consultation on supervisory appoint- ments. The Guild's most active spokesman on Voice is John Carmichael, executive secretary of the Twin Cities Newspaper Guild. Carmichael in '1972 first secured a joint Guild-management committee at the Minnea- {aolis Star and Tribune to meet monthly to discuss matters "related to employment" but not in violation of the contract (Guild Reporter, Sept. 24, 1976, page 2). Carmichael believes the conmittee has been (effective and considers its greatest achievement to be an improvement iii communication: "The necessary ingredients are there. The top newsroom executives are present, and there are no limits to the questions that can be raised (Guild Reporter, Oct. 25, 1974, page 5)." After the Minneapolis Star and Tribune committee had been in effect for about four years, the publisher of the Twin Cities sister Papers, the St. Paul Dispatch and Pioneer Press, signed a contract in St-Eptember 1976 which provided for a similar joint management-Guild "unarker participation“ committee (Guild Reporter, Sept. 24, 1976, page 2). The Twin Cities are not the only places where Voice has been 35 an active concern. A new contract at the Saskatoon (Sask.) Star- Phoenix provides for periodic meetings between Guild members and management to discuss the paper's editorial content (Guild Reporter, Nov. 12, 1976, page 3). The Woodbridge (N.J.) News Tribune instituted a Voice in the product committee in October 1974. Among the issues raised at the initial meetings were beat coverage, changes in work assignments, a revitalized photo policy, changes in editorial policy, more communi- cation between management and reporters, and reinstitution of investi- gative reporting. Two of the primary movers of Voice in the product were the Guild units at the Denver Post and the Chicago Sun Times and Daily lVews. At the Post, an ethics and human rights committee was estab- lished to meet with management on hiring practices and editorial con- ‘tent. Three Post editorial workers meet monthly with three members of rnanagement (Connor, 1974). At the Chicago Sun Times and Daily News, a ten-month attempt ‘to secure Voice was finally quashed by management. In late 1971 the Sun Times Guildsmen put together a proposal that included an employes' veto of the appointment of new department heads, a say in the direc- trion of news coverage by the papers as determined at regular staff "Heatings, and equal space on the editorial page to air the staff's <fl£5nocratically determined disagreement with management opinions. The Guild's proposals were viewed negatively by management's Cfrief bargainer Robert McAllister, vice president of Field news- Papers, who declared: "We are not engaged in a college debate on social justice. at“ pr in Cu 36 We're engaged in negotiating a union contract. It's dollars and cents (Finley, 1973)." The company refused to move on anything relating to reporter power, and after ten months the Guild's Voice proposals had been re- duced to notifying a reporter of substantive changes in his copy. Larry Finley, a reporter for the Chicago Daily News and a member of the negotiating team, contended that part of the reason the Guild lost on the Voice question in Chicago was the wide publicity the proposals received. Another problem was the tenuous support which the proposals got. "When the push came, priorities were with the bread-and-butter, trade unionist issues, rather than with an amorphous attempt to solve a myriad of problems," Finley said. A third factor working against Voice in the product was the 'traditional Guild method of negotiating. According to Finley, "It seemed as though pinning the tag 'voice in the product' on a lot of 'ideas had diminished the ideas and had given some sort of undeserved reality and threatening importance to it." None of the provisions first sought in 1971-72 were won in the subsequent 1974-76 contract with Field Enterprises. A different version of Voice was imposed by management at the New York Times but with negative results. Early in 1976 the 'ritnes instituted a check sheet for reporters--a type of report card-- on which supervisors were to coment on reporters' work, and reporters Wt-Z'V‘e urged to discuss the items with their editors. For the most Part the report card was ill received by staff members (Editor and Publisher, Aug. 7, 1976; Aug. 14, 1976). They termed it "a joke," 37 "ludicrous," and "offensive.” Peter Millones, assistant managing editor who had responsibility for editorial personnel, explained the purpose of the check sheet: "The idea is to encourage discussion between reporter and editor on possible improvement in performance, where appropriate. The reporter can add any comments he or she wishes. This procedure en- courages the reporter and editor to discuss the future-~aspirations and goals that the management of the paper should be aware of." The report card method, however, did not seem to be the type of participation reporters in: the Times had in mind. David Bird, shop steward for the Guild, posted a notice shortly after the forms were issued, which stated: "Management has told the Guild that you do not have to sign [the form that evaluates your performance] and the Guild recomends that you do not." The report card's existence was short-lived. On Dec. 3, 1976, management reported it would cease using the existing form. The potential for Voice beyond a management-imposed check sheet exists in 60 per cent of Guild newspaper contracts, either through standing comittees or through the Grievance Procedure. The Guild IEB determined that the best way to secure Voice would be through the Grievance Procedure, rather than through a separate functioning Gui ld-management comittee. Therefore, those newspapers lacking a Voice conmittee still have the means for securing Voice. That means exists at papers like the New York Times, the Washington P081: and the Chicago Sun Times. There are within the Guild those who consider Voice a 38 significant issue and those for whom it is insignificant. Although there has been no active opposition to Voice as a bargaining subject, there has been widespread indifference to it. Integrity Under the seven provisions of the Integrity clause, the Guild seeks to protect the professionalism of the employe and the employe's product: I-1 I-2 I-4 I-5 I-6 I-7 An employe shall not be required to perform, over his pro- test, any practice which in his judgment compromises his integrity. An employe shall not be required to use his position as an employe for any purpose other than performing the duties of his position. An employe's byline or credit line shall not be used over his protest. Substantive changes in material shall be brought to the employe's attention before publication. An employe shall not be required to write, process or prepare anything for publication in such a way as to distort any facts or to create an impression which the employe knows to be false. If a question arises as to the accuracy of printed material, no correction or retraction of that material shall be printed without prior consultation with the employe concerned. An employe whose work or person is mentioned in a letter to the editor shall be informed of such letter immediately and shall have the right to respond to such letter simultaneously on the page on which it is published. If the byline statement (I-3) is deleted from the Integrity Clause, the number of contracts with Integrity is small--smaller than those with Privilege and much smaller than those with Voice clauses (see Table B-l). The byline statement may be treated as a special case. It 39 had been a part of the convention Recommendations of the Guild since 1938. In terms of numbers of years that it has been sought by the Guild, it is understandable that 129 of 189 contracts (68.2 per cent) have byline coverage, and in most cases (55.0 per cent) there are no restrictions on the model language. In most contracts where it exists, the byline clause has remained a separate "Miscellaneous" item rather than having been incorporated into a new Integrity clause. As with Voice, none of the Integrity provisions (with the exception of the byline statement) has gone to arbitration. The first statement of the Integrity clause is a broad one, covering any area that might compromise an employe's integrity. In a sense it may be viewed as the general case of which the remaining six Integrity provisions are specific examples. Table B-1 shows that nine <:ontracts (4.7 per cent) contain this provision, four of which have icestrictive language; four, essentially model language; and one, nnodel-plus language. The model-plus language omits the qualifier "over liis protest." Modifications include such wording as "The right of an canploye to express to his department manager or personnel manager concern over matters which he may feel to be a compromise of an acceptable or ethical practice is hereby confirmed." Under such a Statement the employe may express concern but still may be required to perform such work. Although there has been no arbitration case on this provision 01’ the Integrity clause, a case did arise in 1974 in which the failure of the Guild to secure I-l coverage in a contract was a factor in an arbitrator's decision to uphold management's disciplin- 3'1)! action against a reporter who refused to write a story on a 40 news release, contending that to write the story would have compro- mised his integrity (Rockford Newspapers, Inc., 63 LA 25 (1974)). The arbitrator ruled: This Arbitrator does not have before him an abstract moral and ethical question. His authority extends only to resolving the issue in terms of the Collective Bargaining Agreement between the parties. The evidence does show that despite a 70-day strike and 41 collective bargaining meetings, the Publisher would not accede to the Guild's contract preposal reading as follows: "ARTICLE II--Professional standards. Section 1(a) An employe shall not be required to perform over his protest any practice which compromises his integrity." The evidence is that in the 4lst negotiation meeting held on January 15, 1971, the Guild agreed "no professional standard language in the contract." It is fundamental that no party can obtain through arbitration what it was unable to gain through collective bargaining. This Arbitrator simply lacks authority to add any provision to this contract. He cannot now under the guise of interpretation or based upon any personal philosophy legislate a professional standard provision. The second Integrity provision, that of being required to per- ‘form other duties, may be interpreted from reporters being required 'to act as photographers and vice versa, to reporters being required 'to write publicity copy or perform non-editorial functions. In many contracts job assignment provisions are spelled out and the right of the employer to assign work is sometimes written ilito the management rights clause. The majority of contracts with restrictive language on I-2 make statements about reporters acting as photographers and vice versa. The editing of news copy is a necessary part of the newsroom Process. Integrity provision 4, however, attempts to notify the emPloye of major changes in his copy prior to his seeing the story appear in print. This provision calls into question the nature of the nt-Z‘ws product. It does not give the reporter veto power over editorial 41 changes. It merely informs him of those changes prior to publication. The byline is the reporter's signature on his work. If sub- stantial changes are made in that product before it is printed, the reporter may seek to disassociate himself from the altered product. If a contract contains both Integrity items 3 and 4, the reporter has the right to request removal of his byline. The same, of course, applies to a photographer's work. Much of the publisher protest against inclusion of I-4 is not that the publisher wishes to conceal editing from the reporter prior to publication, but that time constraints are often such that it is difficult to inform the reporter of changes in material. 0f the 38 contracts (20.1 per cent) that contain this provision, 34 have re- stricted the language by saying that reporters will be notified of substantive changes "in the event a byline is to be used" or that "reasonable" attempts will be made to inform reporters prior to publi- cation. In only 4 cases (2.1 per cent) have publishers permitted language which would require notifying reporters in every instance. The unqualified nature of the Guild's language on I-4 may be la linfiting factor in its acceptance. The qualifier, "in the event a tzyfiine is to be used," was stricken from this provision in 1972 yet remains in many contracts today. The I-4 statement has not been the snzbject of arbitration, and there is no precedent for what constitutes a "substantive change." The fifth Integrity statement is a protection against the l"¢-”—‘porting of false or misleading information. Most publishers sub- scribe to a written or unwritten ethical code which precludes the publication of any false or distorted material. Several publishers 42 have contended that writing into the contract a statement regarding such an ethical standard is unnecessary and even an affront to the publisher-reporter relationship. Of the 19 contracts (10.0 per cent) that do contain such a provision, however, only 2 have language that is restrictive of the model language, and 5 go beyond model language to state that the employer or anyone outside of Guild jurisdiction will not prepare or process any information that is false or misleading. The last two provisions in the Integrity clause deal with printing corrections, retractions and letters to the editor. 0f the contracts that provide for consultation with the employe before a correction or retraction is printed, more than half have restrictive language. As with the "substantive changes" statement, the restric- tion is generally one of reasonableness in attempting to notify the employe concerned. Only 8 contracts (4.2 per cent) contain provision of the right of reply by the employe, and of those 7 are restrictive-- requiring that the material must have been bylined, exempting cases where libel is alleged, or dr0pping the requirement for simultaneous reply. This final Integrity provision was added to the Model Contract after the 1972 convention. In terms of new Integrity issues (those in addition to byline protection), the Guild has made the greatest gains in securing noti- fication of the employe of changes in copy or consultation before a retraction or correction is printed. The three Integrity provisions with the highest frequencies (I-3, I-4 and 1-6) all concern the actual product. The most difficult issues to obtain coverage on have been protection against compromising one's integrity and writing false or misleading information. 43 Privilege The Privilege portion of newsroom democracy has been an active concern of the Guild for more than a decade. It is a right which both employes and employers have been seeking through the courts and through state statutes. In fact, the absolute shield law which the American Newspaper Publishers Association proposed before Congress after the U.S. Supreme Court decision in Branzburg v. Hayes, 408 U.S. 665 (1972), No person shall be required to disclose in any federal or state proceeding either (1) the source of any published or unpublished information obtained in the gathering, receiving or processing of infor- mation for any medium of communication to the public, or (2) any unpublished information obtained or prepared in gathering, receiving or processing of information for any medium of communication to the public. bears strong resemblance to the first portion of the Guild's clause: P-l An employe may refuse, without penalty or prejudice, to give up custody of or disclose any knowledge, information, notes, record, documents, films, photographs, or tapes or the source thereof, which relates to news, commentary, advertising or the establishment and maintenance of his sources, in con- nection with his employment. P-2 An employe may also refuse without penalty or prejudice to authenticate any material. Currently 51 of the 189 Guild contracts used in the study (26.9 per cent) have some protection against forced disclosure of news sources (see Table B-l). The Privilege clause was divided into eight items, although there are essentially four areas of concern: (a) the right of the employe to refuse to give up custody of or disclose materials or to authenticate any material, (b) the promise by the employer that he will not relinquish such material without the consent of the employe, 44 (c) notification of employe or employer and the Guild of a demand for surrender of material, and (d) indemnification of the employe if he is proceeded against under law for failure to disclose. The first of these four areas includes P-1 and P—2. The second area corresponds to P-3 and the third area to P-4: P-3 The Employer shall not give up custody of or disclose any of the above [P-1 and P-2] without consent of the employe. P-4 The Employer shall notify the employe concerned, and the Guild, of any demand on the Employer for such surrender or disclosure or authentication. The fourth area concerning indemnification includes P-5 through P—8: P-5 If the employe is proceeded against under law on account of his refusal to surrender or disclose or authenticate, the Employer shall move to join as a party to such proceedings; P-6 [The Employer] shall meet all expenses incurred by the employe, including fees and expenses of legal counsel retained by the employe; P-7 And [the Employer] shall indemnify such employe against any monetary loss including but not limited to fines, damages or loss of pay. P-8 In no case shall an employe suffer loss of wages, employe status or benefits under this contract as a result of his refusal to surrender or disclose or authenticate. It may appear odd that the number of contracts containing statements regarding indemnification is higher than the number con- taining statements on nondisclosure. This may be due to two factors: (1) the contract may be worded in such a fashion that the right not to disclose is assumed and the language merely spells out the pro- tection to be afforded the employe in the event a demand is made for his sources; (2) the publisher may have wished to avoid automatic liability in the event an employe refused to disclose and a legal suit followed. The publisher may have agreed to include in the contract 45 specific conditions under which he would assist an employe who refused to disclose his sources. One Privilege statement that has received little acceptance from publishers is the employer's joining as a party in a suit against disclosure. Only sixteen contracts (8.4 per cent) contain such a statement. The Guild Model Contract language assumes that the employer has complete faith in the judgment of his employe. The language makes no demands on the employe to reveal information to the employer, nor to follow the employer's legal counsel advice, yet it seeks full finan- cial and legal support from the employer. It gives the employe full control over the disclosure issue. Most publishers are unwilling to go this far, however. Re- strictions on the Guild's Privilege language include disclosing all information to the publisher and following the advice of publisher's legal counsel. As written in most contracts, the Privilege clause provides protection only for the employe who accepts the publisher's judgment in a disclosure case. The final Privilege statement, P-8, protects the employe against any losses (monetary, job status or benefits) regardless of the action he takes. This provision has been interpreted in varying ways by publishers. As written, the phrase "In no case..." seems to qualify all the preceding Privilege statements. That is, regardless of the employe's course of action subsequent to a refusal to disclose, he will suffer no losses. Yet in some contracts where the publisher has placed qualifiers on indemnification or has required the employe to disclose to the employer, P-8 is still included. It is not clear 46 in such instances which provision--the qualified or the all-encompass- ing one--is controlling. The growing number of contempt of court cases and the 1976 contempt of Congress case regarding protection of sources make Privilege an important newsroom democracy consideration. The New York Times, one of the publications involved in the landmark U.S. Supreme Court decision in 1972 on privilege against disclosure of sources to grand juries, is a Guild newspaper. The Times has since that ruling incorporated Privilege into its editorial agreement as a letter of memorandum. At the Fresno (Calif.) Bee, four employes were sentenced to jail in September» 1976 fbr refusing to disclose their news sources. The editorial employes of that afternoon daily, however, were not protected by a Privilege clause. The publisher had signed his first contract with the Guild in 1974 without such a provision. Currently twenty-four states have shield or privilege laws for reporters. 0f the fifty-one Guild contracts containing Privilege, thirteen are in Canada, the District of Columbia or states that do not have shield laws (Delaware, Colorado, Wisconsin, Massachusetts and West Virginia). Incorporating a Privilege clause in the contract raises the question of its usefulness: how often are demands made on reporters for information sources and what types of subsequent action are carried out? Vince Blasi, professor of law at the University of Michigan, conducted an extensive survey in 1971 of reporters' attitudes toward Privilege., From a nonrandom mail survey of 1,470 reporters across 47 the country, he received 975 replies (66.3 per cent). Most respond- ents reported that they did not know how they would react to a hypo- thetical grand jury subpoena. Some 180 (15.5 per cent) of the respondents said they had been subpoened in conjunction with their reporting. Of those, 72 (46.5 per cent) said they did not challenge the subpoena. Blasi attempted to find out whether a Privilege statute would make a difference in the quality and amount of news coverage. He discovered that only 35 per cent of his survey respondents knew if the state in which they wrote most of their material had a shield law. If shield laws can be written into state constitutions by statute, can they be written into labor contracts as well? Because fifty—one contracts currently exist with Privilege clauses, the answer may appear self-evident. Yet the question of the mandatory nature of bargaining on Privilege has not been determined by the NLRB. Does the clause violate basic management rights? Is Privilege a mandatory subject on which the publisher is required by law to bargain? These questions first surfaced in 1972 in negotiations between the Portland (Maine) Newspaper Guild and the Guy Gannett Publishing Co., publisher of the Portland Press Herald, Express and Telegram. The previous contract with Gannett expired in June 1972, and the Guild, in its initial proposal, submitted Model Contract language on Integrity and Privilege. After several negotiations the company declared that it was not interested in incorporating either Integrity or Privilege language in its contract. In a memo to its employes in February 1973 the company stated: 48 We do not feel this subject is properly a part of our collec- tive bargaining agreement. As a matter of fact the legal depart- ment of ANPA and our own counsel are agreed that these subjects do not fall under the National Labor Relations Board rules for compulsory items for bargaining but are rather only permissive items for bargaining upon agreement of the parties. In our opinion these proposals constitute attempts to have a hand, indeed considerable control, in determining how and under what circumstances our product will be produced and what its nature shall be. If you will think about this carefully, I can't believe that fair-minded men would expect us to permit this. The company shortly thereafter ceased to be willing to discuss the issue, and the Portland Newspaper Guild filed an unfair labor practice charge against the publisher in March 1973. One week later the publisher filed a counter charge against the Guild, claiming an unfair labor practice on the part of the union by attempting to bargain to impasse on permissive subjects. National Labor Relations Regional Director Robert S. Fuchs (Region 1, Boston) deferred the matter to the General Counsel of the NLRB, Peter G. Nash, as there was no precedent for determining the mandatory bargaining nature of such a clause. The issue remained in the hands of General Counsel Nash without a decision being rendered until July 1975 (three years after the Portland contract had run out) when the Guild withdrew its unfair labor practice charge. Thus, the issue over the mandatory nature of the subject matter has not been clarified. Also remaining unanswered since the Guild's withdrawal of its unfair labor practice charge is a determination of the legal pro- priety of indemnification of employes. A question remains as to whether a contract with an agreement not to disclose would be contrary to public policy in a state where a shield law did not exist, and whether it would be proper for a publisher to be required to pay legal fees for someone who refused a court order to disclose. 49 Newsroom Democracy Publication Characteristics In addition to examining frequencies of contracts containing each of the eighteen newsroom democracy issues, comparisons were made between publications with similar characteristics and the degree to which their contracts contained newsroom democracy. Among the charac- teristics examined were (1) date of the signing of the first editorial contract, (2) circulation figures, and (3) group affiliation. The contracts were divided into those that contained Voice, Integrity and/or Privilege and those that did not. The results are presented in Tables B-2, B-3 and B-4 (Appendix B). First Contract Date: Derber's study (1958) suggested that contracts of long standing tend to have more provisions for worker involvement than do more recently negotiated contracts. This hypothesis is borne out by the comparison of contracts containing Voice, Integrity and Privilege and the length of time those contracts have been in existence (see Table B-2). Across all three newsroom democracy issues, older con- tracts, particularly those signed prior to September 1945, contain the highest percentages of newsroom democracy. However, the differ- ences are not significant if the chi-square test is applied (Downie and Heath, 1965). Circulation: In a comparison of newsroom democracy coverage and newspaper circulation figures, the data suggest that small and moderate size dailies have the greatest proportion of Voice, Integrity and Privilege (see Table B-3). These empirical data also lend support to 50 the findings of Derber (1958) and Chandler (1964), but again the chi-square probabilities are not significant. Group Affiliation: Table B-4 shows that group publications tend to have less Voice but slightly more Integrity and Privilege than do independent newspapers. In this crosstabulation, the chi-square probability for Voice is .00, indicating that Voice clauses exist to a greater degree in independent newspapers. The chi-square probabilities for Integrity and Privilege are not significant. Researchyflypotheses The research hypotheses for this study are derived from industrial relations research on worker participation, industrial democracy, management rights and collective bargaining and also from the empirical data on current contract clauses summarized in the preceding section. The hypotheses and a brief summary of the rationale for their proposal follow: Hypothesis 1 Editorial employes will have more favorable attitudes toward each of the eighteen newsroom democracy issues as a general concept than will publishers. Hypothesis 2 Editorial employes will have more favorable attitudes toward each of the eighteen newsroom democracy issues as a contract provi- sion than will publishers. 51 Rationale for Hypotheses l and 2: Studies by Blumberg (1968), Daniel (1972), Roethlisberger (1964), Morse (1956), Fleishman (1962), Herzberg (1959), Argyris (1964, 1974), Likert (1961) and Sigelman (1973) form the basis for the first and second hypotheses. They suggest that employes are more satisfied with their jobs and are more productive when given greater participation in decision making. They also suggest that employes are receptive to and anxious for greater participation. The other side of the first two hypotheses, that publishers will be less inclined to favor newsroom democracy issues, has its rationale basis in the work by Daniel (1972), Chamberlain (1967), Derber (1958), Breed (1955), Argyris (1974) and Chandler (1964). Their studies suggest that employers are not anxious to relinquish to joint decision making areas traditionally held to be management's rights. Hypothesis 3 The correlation between attitudes on a general newsroom democracy concept and a contractual one will be higher across each of the eighteen issues for editorial employes than for publishers. Rationale for Hypothesis 3: Hypothesis 3 suggests that the correlation for editorial employes between their general and contractual scores on newsroom democracy will be significantly greater than the corresponding pub- lishers' correlations. The rationale for this hypothesis proceeds from the basic concept of management rights versus worker participa- tion as discussed in the work by Blumberg (1968), Daniel (1972), Argyris (1964, 1974), Likert (1961), Chamberlain (1967), Chandler 52 (1964), Derber (1958) and Breed (1955). Publishers may agree in prin- ciple to newsroom democracy but are reluctant to incorporate it as a contract provision. Management traditionally has been reluctant to define the limits of its rights in contracts (Chandler, 1964; Chamberlain, 1967). Editorial employes, on the other hand, are more likely to view general principles and contract rights as synonymous since they do not face the same legal constraints their employers would if the principles were made part of the contract language. For employes it is a rights guarantee; for the employer it may be a rights liability. Hypothesis 4 Publishers of smaller newspapers (less than 50,000 circula- tion) will have more favorable attitudes toward contractual newsroom democracy than will publishers of larger newspapers. Rationale for Hypothesis 4: The rationale for Hypothesis 4 is based on findings by Bowers (1967) that publishers of smaller newspapers tend to partici- pate more in newsroom decisions and especially when those decisions may affect the financial aspects of the newspaper. Derber (1958) and Chandler (1964) suggest that smaller firms are more likely to be favorable to greater worker participation. Johnstone, Slawski and Borman (1976) found that employes in smaller newspapers are often given more autonomy by management than those in larger organizations. Additional support comes from the data in Table B-3 showing the percentage of Voice, Integrity and Privilege currently in TNG con- tracts based on circulation. 53 Hypothesis 5 Publishers with long-standing contracts covering editorial employes will have more favorable attitudes toward contractual news- room democracy than will publishers with more recent contracts. Rationale for Hypothesis 5: Derber's study (1958) suggested that employers who have long- standing contracts are more likely to grant or to have granted greater worker participation than are employers with shorter-standing con- tracts. The empirical data from the analysis of current contracts also lend support to this hypothesis (see Table B-2). Hypothesis 6 Publishers who consider their relationship with The Newspaper Guild to be excellent or good will have more favorable attitudes toward contractual newsroom democracy than will publishers who consider their relationship to be fair or poor. Rationale for Hypothesis 6: Hypothesis 6 is based on the theory suggested by Chamberlain (1967) that employers who do not fear the union's motives are more likely to consider extending worker participation rights than are those employers who are suspect of the union's motives. Derber's work (1958) suggests the same hypothesis formulation. Hypothesis 7 Publishers who are members of a group will have less favor- able attitudes toward contractual newsroom democracy than will independent publishers. 54 Rationale for Hypothesis 7: In a study by Wackman (1976) members of groups were found to have similar behaviors with respect to campaign endorsements. Also, data from existing contracts show that agreements with members of the same groups tend to have the same or similar Voice, Integrity and/or Privilege language. The data in Table B-4 show that group publica- tions have far less Voice than independent newspapers but approxi- mately the same amount of Integrity and Privilege. Hypothesis 8 Editorial employes who consider their relationship with the employer to be excellent or good will have less favorable attitudes toward contractual newsroom democracy than will editorial employes with a fair or poor relationship. Rationale for Hypothesis 8: Hypothesis 8 suggests that workers who are dissatisfied with their working relationship seek greater change and participation than those who feel they have a good working relationship with their employer. Studies supportive of this hypothesis include those by Alutto (1973), Thompson (1973), Cullen (1965) and Gruen (1954). Hypothesis 9 Editorial employes who are active and supportive of their union will have more favorable attitudes toward contractual newsroom democracy than will editorial employes who are indifferent or negative toward their union. 55 Rationale for Hypothesis 9: Studies by Sayles and Strauss (1953) and by Brumback (1972) suggest that union members who are active and supportive of their union tend to favor the union's policies. Hypothesis 10 Editorial employes with shorter professional service time will have more favorable attitudes toward contractual newsroom democracy than will veteran editorial employes. Rationale for Hypothesis lO: Hypothesis 10 is based on the rationale that less experienced newspersons are more apt to seek greater participation, more rights and greater change within the organization. This suggestion is supported by the work of Argyris (1974) and Johnstone, Slawski and Borman (1972, 1976). Also, it should be noted that the newsroom democracy movement of the late 19605 and early 19705 was sparked in large measure by groups of fledging militant reporters. CHAPTER III DESIGN OF THE STUDY The Sample A questionnaire distributed by mail was used to gather data for this study. The population consisted of all editorial employes covered by The Newspaper Guild contracts and all publishers who currently hold Guild contracts covering editorial employes. A stratified sample of 916 Newspaper Guild members was drawn from the union's Guild Reporter (its byweekly newspaper) mailing list and from membership lists. For the sample of publishers, 190 names were drawn. The procedures used in drawing the samples are given in Appendix C. The Survey Instrument The publisher and editorial employe questionnaires are re- printed in Appendix D. The format for the two questionnaires was identical. Different colors (white for editorial employes, blue for publishers) were used to distinguish the two. A series of thirty-six attitude statements (two statements per index for an attitude measure on each of the eighteen newsroom democracy provisions) was printed in the center of the questionnaire, with a five-point Likert scale in the right and left-hand margins. The statements were written as measures of attitudes toward each of the eighteen provisions for newsroom democracy as defined by the 56 57 Model Contract language. All but three of the index pairs had one statement in which the strongly agree or strongly favor category indi- cated a favorable attitude and the second in which the strongly agree or strongly favor category indicated an unfavorable attitude toward newsroom democracy. The scales were adjusted for direction uniformity prior to coding, with a value of "5“ indicating a strongly favorable attitude toward newsroom democracy and a value of "1" indicating a strongly opposed attitude toward newsroom democracy. The wording of the statements followed as far as possible the language of the Model Contract. Words were changed or phrases shortened only to decrease obvious repetition or eliminate legalistic terminology. Many of the phrases used in the eighteen Guild provi- sions are ambiguous but they were left that way in the wording of the questionnaire statements to force respondents into an agree-disagree or favor-oppose scale. (Several respondents added qualifiers to the statements as they filled out their questionnaires. These are men- tioned in the Results and Discussion chapter.) The purpose of the study was to assess attitudes as they relate to the specific wording proposed by The Newspaper Guild. To qualify or clarify meanings suggested by the provisions would have distorted the analysis. The first set of fourteen statements dealt with the seven Integrity provisions of the Integrity clause. The statements were ordered in such a way that the first was intended to be universally simple to respond to: "All news copy should be presented without distortion of any facts and without an attempt to create false im- pressions." The remaining statements were arranged to minimize the chance of respondents recognizing that there were pairs of statements 58 measuring the same attitudes. The second set of six questions con- cerned the three Voice provisions, and the third set of sixteen questions concerned the eight Privilege provisions. The format for listing the statements in the Voice and Privilege sections was the same as for the Integrity statements. Respondents were asked to read the statement and then to circle on the left—hand scale the degree to which they agreed or disagreed in general with the statement. The key for the scale was as follows: SA = strongly agree A = agree N = no opinion 0 = disagree SD = strongly disagree. Next they were asked to consider the same statement as a contract pro- vision (changing the verb form "should" to "shall") and to circle on the right-hand scale the degree to which they favored or opposed including such a statement in a contract. The key for the right-hand scale was as follows: SF = strongly favor F = favor N = no opinion 0 = oppose $0 = strongly oppose. At the end of the questionnaire, respondents were asked to answer general categorical questions. For editorial employes, the questions concerned (1) job description, (2) whether they were covered by a Guild editorial department, (3) the number of years with their current employer, (4) the number of years as a newsperson, (5) the relationship with the publisher (excellent, good, fair, poor), and (6) the relationship with The Newspaper Guild (officer, active and 59 supportive, supportive, indifferent, would prefer not to be a member). For publishers the questions included (1) job title, (2) pub- lication's combined circulation, (3) type of publication, (4) whether the publication is a member of a group, (5) number of years publica- tion has had a contract covering editorial employes, and (6) the publisher's relationship with The Newspaper Guild (excellent, good, fair, poor). The Pretest The pretest consisted of ninety-eight attitude statements (eighteen on Voice, thirty on Integrity and fifty on Privilege). All of the general principle statements were grouped together and were followed by a similar series of contract provision statements. Re- spondents were asked to indicate their agreement or disagreement with each statement on a scale from O to 100. As in the final question- naire, the statements were organized according to the topics Integrity, Voice and Privilege. The same questionnaire (without the final categorical ques- tions) was distributed to eight publishers/managing editors. Four were returned. Twenty were distributed to reporters, and eleven were returned. The reporters' scores on the items for each attitude index were summed and each individual item was correlated with the other items and with the total. The analysis suggested that qualified statements be eliminated and that most of the positive-negative statements were measuring the same attitude (that is, where one item of a positive-negative pair was rated 0, the other was rated 100). 60 Some items were rewritten where reporters and/or publishers found difficulty with the language. For example, the phrase "making an employe whole" was changed to "assuming all responsibility for legal costs and financial losses." One of the major complaints received in the pretest was the excessive length of the questionnaire. Such complaints, as well as the correlation test results, suggested shortening the questionnaire to two items per index and printing the statement once with a differ- ent scale on either side of the statement. The correlation between scores on the general statement and the contractual statement were high for editorial employes. This suggested that any potential value in separating the evaluation of a statement as a contract provision and as a general principle could not justify doubling the size of the questionnaire. An insufficient number of publishers completed both parts of the questionnaire to draw comparisons from their pretests. Another major modification of the questionnaire was the change from a 0 to lOO-point interval scale to a 5-point Likert scale. 0n the pretests responses were 0-25-50-75-100 with rare exceptions. Also, some respondents questioned their ability to make loo-point distinctions in their agreement or disagreement with a statement. The major criticisms voiced by those taking the pretest led to the final design of the study which listed the "general principle" Likert scale to the left of the statement and the "contract provision" Likert scale to the right of the statement. None of the thirty-six items were consistently avoided by respondents. However, in the pretest, because of its length, many did 61 not complete the second half which concerned contract provisions. Covering Letters and Mailing Procedures The questionnaires were printed lO-point Univers medium on both sides of 82-by-l4 inch 20-pound bond. This was done to conserve weight in mailing. The questionnaires were folded twice, with the title of the study facing up, and were inserted in the center of the covering letter. The covering letters (reproduced in Appendix E) were of a similar format for publishers and editorial employes. Pub- lishers' letters were printed offset on 24-pound MSU School of Journalism stationery and were individually addressed. Editorial employes' letters were mimeographed on 20-pound bond and were addressed "Dear Newsperson." All editorial employes' covering letters were accompanied by a second letter written by David J. Eisen, director of research and information at The Newspaper Guild, explaining how the names were selected and urging support of the research study. All covering letters were individually signed by the research director. The letters, questionnaires and a stamped self-addressed return envelope were folded into MSU School of Journalism bond envelopes and were posted with commemorative stamps. Survey Returns Editorial Employes: The target sample size for securing a 60 per cent usable return was 592. A total of 56 questionnaires were returned undeliv- erable. If it is assumed that half of these were potential editorial employes from the target sample of 592, the effective target sample was reduced to 564. Thus, a 60 per cent return would yield a sample 62 size of 338. Mailings were made on the following dates: May 9 and 10 (N=9l6); June 10 and 11 (N=420); and July 14-17 (N=296). By Aug. 3, 1977, the number of usable questionnaires received was 352, or 62.4 per cent. The percentage of usable returns from each mailing (based on the total number of questionnaires sent) was approximately 22.4 per cent on the first mailing, 26.2 per cent on the second mailing and 14.9 per cent on the third mailing. Publishers: The original sample was 190, of which seven were returned undeliverable and seven were returned not applicable, as there was no contract in existence with the Guild or the nature of the publica- tion was such that the questions were inappropriate, reducing the effective sample to 176. Three mailings were made, the first two addressed personally to publishers, general managers, presidents, or editors-in-chief, on the following dates: May 11-13 (N=l90); June 1 and 2 (N=ll9). The third mailing was sent personally to executive or managing editors or labor relations vice presidents at those publications where pub- lishers had not responded. Eighty-six questionnaires were sent in the third mailing on July 1. By Aug. 3, 1977, 90 usable questionnaires were received (51.1 per cent). An additional 24 responses in the form of letters or comments were received from publishers reluctant to categorize their attitudes on a five-point scale. These letters and comments are included in Appendix F. (Editorial employe comments are printed 63 in Appendix G.) The percentage of usable returns from the three mailings was approximately 21.1 per cent from the first mailing, 21.0 per cent from the second, and 27.9 per cent from the third. Testable Hypotheses Hypotheses 1A - 1R The first group of hypotheses (1A through 1R) employs a difference of means T-test to compare publisher and reporter responses to the eighteen general newsroom democracy concepts. The first of these hypotheses is written out in detail below. The remaining seventeen, all of which follow the same pattern, are written in abbreviated form. Null Hypothesis 1A: There is no difference between the mean scores of publishers and editorial employes on VG-l (the desirability of a joint committee to discuss newsroom policy and practice). H : VG-l = VG-l p e o where VG-l = mean score* for publishers on the general con- p cept of a joint committee VG-l = mean score for editorial employes on the gen- e eral concept of a joint committee *The mean score for publishers is computed by summing the scores (1-5) for each publisher on the two index items measuring VG-l. These indi- vidual index totals are then summed across all publishers and that sum is divided by the number of publishers answering both index statements. Where only one index item of a pair is answered, the same value is assigned to the second and the scores summed accordingly. The same pro- cedure is used to calculate the mean score for editorial employes. 64 Alternative Hypothesis 1A: The mean score for editorial employes on VG-l (joint committee) will be higher than the mean score for publishers. H VG-1e > VG-1p 1A: where the legend is the same as in the null hypothesis. Null Hypothesis lB: There is no difference between the mean scores of publishers and editorial employes on VG-2, the right to consultation on the appointment of supervisory personnel in the newsroom. Alternative Hypothesis lB: The mean score for editorial employes on VG-2 will be higher than the mean score for publishers. Null Hypothesis 1C: There is no difference between the mean scores of publishers and editorial employes on VG-3, representation by editorial employes on editorial boards. Alternative Hypothesis 1C: The mean score for editorial employes on VG-3 will be higher than the mean score for publishers. Null Hypothesis lD: There is no difference between the mean scores of publishers and editorial employes on IG-l, not requiring an employe to ' perform a duty which he feels compromises his integrity. Alternative Hypothesis lD: The mean score for editorial employes on IG-l will be higher than the mean score for publishers. Null Hypothesis 1E: There is no difference between the mean scores of publishers and editorial employes on IG-2, not requiring an employe to perform duties other than those of his job description. 65 Alternative Hypothesis 1E: The mean score for editorial employes on IG-2 will be higher than the mean score for publishers. Null Hypothesis 1F: There is no difference between the mean scores of publishers and editorial employes on IG-3, not using the byline or credit line over the employe's protest. Alternative Hypothesis 1F: The mean score for editorial employes on IG-3 will be higher than the mean score for publishers. Null Hypothesis 1G: There is no difference between the mean scores of publishers and editorial employes on IG-4, notifying an employe of substantive changes in copy prior to publication. Alternative Hypothesis lG: The mean score for editorial employes on IG-4 will be higher than the mean score for publishers. Null Hypothesis 1H: There is no difference between the mean scores of publishers and editorial employes on IG-5, not requiring an employe to write false or misleading copy. Alternative Hypothesis 1H: The mean score for editorial employes on IG-5 will be higher than the mean score for publishers. Null Hypothesis 11: There is no difference between the mean scores of publishers and editorial employes on IG-6, printing corrections or retrac- tions only after consulting with the employe concerned. flternative Hypothesis 1 I : The mean score for editorial employes on IG-6 will be higher than the mean score for publishers. 66 Null Hypothesis 1J: There is no difference between the mean scores of publishers and editorial employes on IG-7, permitting employes the right of simultaneous reply to letters concerning their work. Alternative Hypothesis 1J: The mean score for editorial employes on IG-7 will be higher than the mean score for publishers. Null Hypothesis 1K: There is no difference between the mean scores of publishers and editorial employes on PG-l, employe privilege against disclosure of sources. ‘ Alternative Hypothesis 1K: The mean score for editorial employes on PG-l will be higher than the mean score for publishers. Null Hypothesis 1L: There is no difference between the mean scores of publishers and editorial employes on P612, employe privilege against authentication of sources. Alternative Hypothesis 1L: The mean score for editorial employes on PG-2 will be higher than the mean score for publishers. Null Hypothesis 1M: There is no difference between the mean scores of publishers and editorial employes on PG-3, agreement that the employer will maintain the confidentiality of the employe's sources. Alternative Hypothesis 1M: The mean score for editorial employes on PG-3 will be higher than the mean score for publishers. Null Hypothesis 1N: There is no difference between the mean scores of publishers and editorial employes on PG-4, agreement that the employer will notify the employe and the union of a demand for disclosure of sources. 67 Alternative Hypothesis 1N: The mean score for editorial employes on PG-4 will be higher than the mean score for publishers. Null Hypothesis 1-0: There is no difference between the mean scores of publishers and editorial employes on PG-5, agreement that the employer will join as a party in a disclosure case. Alternative Hypothesis l—O: The mean score for editorial employes on PG-S will be higher than the mean score for publishers. Null Hypothesis 1P: There is no difference between the mean scores of publishers and editorial employes on PG-6, agreement that the employer will pay all legal fees in a disclosure case. Alternative Hypothesis 1P: The mean score for editorial employes on PG-6 will be higher than the mean score for publishers. Null Hypothesis lQ: There is no difference between the mean scores of publishers and editorial employes on PG-7, agreement that the employer will pay all losses suffered by an employe in a disclosure case. Alternative Hypothesis lQ: The mean score for editorial employes on PG-7 will be higher than the mean score for publishers. Null Hypothesis 1R: There is no difference between the mean scores of publishers and editorial employes on PG-B, agreement that the employer will pay all employe losses in a disclosure case regardless of the employe's course of action. Alternative Hypothesis 1R: The mean score for editorial employes on PG-B will be higher than the mean score for publishers. 68 Testing Hypotheses 1A - 1R: Method of Analysis: Difference of Means T-Test The first hypothesis is concerned with differences in atti- tudes between the two sample groups (publishers and editorial employes) on eighteen general newsroom democracy concepts. The means of each group are used as the basis of comparison. The sample mean in each case (publishers and editorial employes) is an estimate of the population mean (all publishers and editorial employes covered by Guild contracts). Since it is possible that two samples from the same population would be different due to the natural variability of the population, a difference in sample means does not necessarily imply that the populations from which they were drawn actually differ on the characteristic being studied. A T-test was computed to establish whether or not a difference in scores between the two samples is significant; that is, whether the difference is indicative of a true difference between the groups. Since samples are being compared, a decision to accept or reject the null hypothesis (that there is no difference between the groups) cannot be made with absolute certainty. The decision must be based on probabilities. A significance level of .05 was chosen as the smallest probability that will be accepted as reasonable. This means that the probability of the null hypothesis being rejected erroneously would be five times in 100. The difference of means T-test is generally applicable to a normally distributed random variable where the mean is known and the population variance is estimated from the sample. A two-tailed test 69 was used (Nie et al., 1975). Test of Significance: T-Test Significance Level: .05 Assumptions of the Difference of Means Test: 1. The subjects are randomly drawn from a normally distributed population. 2. The sample mean is a normally distributed variable. 3. The difference of two normally distributed variables is a normal random variable. 4. The sample means are based on interval level data. The five-point Likert scales were treated as interval-level data. Some social scientists argue against treating Likert or other summated rating scales as interval-level measures because the distance between the values cannot be fixed. Coombs (1953) argued that there is an ordered metric level that falls between the ordinal and interval levels of measurement and consists of ordered categories of known but unmeasurable and not necessarily equal distances. Abelson and Tukey (1970) argued that often more is known than just rank order and that if only non-parametric techniques of analysis are applied to those rankings, the flexibility of the analysis is severely and unnecessar- ily restricted. Labovitz (1970) offered a convincing argument for assigning intervals to any ordinal data. He showed that scoring systems randomly generated by a computer resulted in negligible error when comparing any assigned scoring system with any “true" scoring system. Errors were determined by comparing Pearson r's. He 70 suggested the following benefits can be derived from treating ordinal variables as though they were interval: (1) the use of more powerful, sensitive, interpretable statistics; (2) retention of more knowledge about the characteristics of the data; (3) greater versatility in the statistical manipulation. Hypotheses 2A - 2R The second group of hypotheses (2A through 2R) also uses a difference of means T-test to compare publisher and reporter atti- tudes on incorporating the eighteen newsroom democracy provisions in contracts. As with the first set of questions, Hypotheses 2A through 2R are of the same form. The first one is written out in detail and the remaining seventeen are presented in abbreviated form. Nu11 Hypothesis 2A: There is no difference between the mean scores of publishers and editorial employes on VC-l (the contractual provision for a joint committee to discuss newsroom policy and practice). Ho: VC-lp = VC-le where VC-lp mean score* for publishers on the contractual provision for a joint committee VC-le mean score for editorial employes on the contractual provision for a joint committee *The mean score for publishers is computed by summing the scores (1-5) for each publisher on the two index items measuring VC-l. These indi- vidual index totals are then summed across all publishers and that sum is divided by the number of publishers answering both index statements. Where only one index item of a pair is answered, the same value is assigned to the second and the scores summed accordingly. The same pro- cedure is used to calculate the mean score for . editorial employes. 71 Alternative Hypothesis 2A: The mean score for editorial em loyes on VC-l (contractual provision for a joint committee will be higher than the mean score for publishers. HZA: VC-le ) VC-lp where the legend is the same as in the null hypothesis. Null Hypothesis 28: There is no difference between the mean scores of publishers and editorial employes on VC-2, the right to consultation on the appointment of supervisory personnel in the newsroom. Alternative Hypothesis 28: The mean score for editorial employes on VC-2 will be higher than the mean score for publishers. Null Hypothesis 2C: There is no difference between the mean scores of publishers and editorial employes on VC-3, representation by editorial employes on editorial boards. Alternative Hypothesis 2C: The mean score for editorial employes on VC-3 will be higher than the mean score for publishers. Null Hypothesis 20: There is no difference between the mean scores of publishers and editorial employes on IC-l, not requiring an employe to perform a duty which he feels compromises his integrity. Alternative Hypothesis 20: The mean score for editorial employes on IC-l will be higher than the mean score for publishers. Null Hypothesis 2E: There is no difference between the mean scores of publishers and editorial employes on IC-2, not requiring an employe to perform duties other than those of his job description. 72 Alternative Hypothesis 2E: The mean score for editorial employes on IC-2 will be higher than the mean score for publishers. Null Hypothesis 2F: There is no difference between the mean scores of publishers and editorial employes on IC-3, not using the byline or credit line over the employe's protest. Alternative Hypothesis 2F: The mean score for editorial employes on IC-3 will be higher than the mean score for publishers. Null Hypothesis 2G: There is no difference between the mean scores of publishers and editorial employes on IC-4, notifying an employe of substantive changes in copy prior to publication. Alternative Hypothesis ZG: The mean score for editorial employes on IC-4 will be higher than the mean score for publishers. Null Hypothesis 2H: There is no difference between the mean scores of publishers and editorial employes on IC-5, not requiring an employe to write false or misleading copy. Alternative Hypothesis 2H: The mean score for editorial employes on IC-5 will be higher than the mean score for publishers. Null Hypothesis 21: There is no difference between the mean scores of publishers and editorial employes on IC-6, printing corrections or re- tractions only after consulting with the employe concerned. Alternative Hypothesis 21: The mean score for editorial employes on IC-6 will be higher than the mean score for publishers. 73 Null Hypothesis 2J: There is no difference between the mean scores of publishers and editorial employes on IC-7, permitting employes the right of simultaneous reply to letters concerning their work. Alternative Hypothesis 2J: The mean score for editorial employes on IC-7 will be higher than the mean score for publishers. Null Hypothesis 2K: There is no difference between the mean scores of publishers and editorial employes on PC-l, employe privilege against disclosure of sources. Alternative Hypothesis 2K: The mean score for editorial employes on PC-l will be higher than the mean score for publishers. Null Hypothesis 2L: There is no difference between the mean scores of publishers and editorial employes on PC-2, employe privilege against authentication of sources. Alternative Hypothesis 2L: The mean score for editorial employes on PC-2 will be higher than the mean score for publishers. Null Hypothesis 2M: There is no difference between the mean scores of publishers and editorial employes on PC-3, agreement that the employer will maintain the confidentiality of the employe's sources. Alternative Hypothesis 2M: The mean score for editorial employes on PC-3 will be higher than the mean score for publishers. 74 Null Hypothesis 2N: There is no difference between the mean scores of publishers and editorial employes on PC-4, agreement that the employer will notify the employe and the union of a demand for disclosure of sources. Alternative Hypothesis 2N: The mean score for editorial employes on PC-4 will be higher than the mean score for publishers. Null Hypothesis 2-0: There is no difference between the mean scores of publishers and editOrial employes on PC-S, agreement that the employer will join as a party in a disclosure case. Alternative Hypothesis 2-0: The mean score for editorial employes on PC-5 will be higher than the mean score for publishers. Null Hypothesis 2P: There is no difference between the mean scores of publishers and editorial employes on PC-6, agreement that the employer will pay all legal fees in a disclosure case. Alternative Hypothesis 2P: The mean score for editorial employes on PC-6 will be higher than the mean score for publishers. Null Hypothesis 20: There is no difference between the mean scores of publishers and editorial employes on PC-7, agreement that the employer will pay all losses suffered by an employe in a disclosure case. Alternative Hypothesis 20: The mean score for editorial employes on PC-7 will be higher than the mean score for publishers. 75 Null Hypothesis 2R: There is no difference between the mean scores of publishers and editorial employes on PC-8, agreement that the employer will pay all employe losses in a disclosure case regardless of the employe's course of action. Alternative Hypothesis 2R: The mean score for editorial employes on PC-8 will be higher than the mean score for publishers. Testigg Hypotheses 2A - 2R: Method of Analysis: Difference of Means T-Test Test of Significance: T-Test Significance Level: .05 Assumptions of the Difference of Means Test: Same as Hypothesis 1 Hypotheses 3A - 3R Hypotheses 3A through 3R involve a comparison of the bivar- iate correlation of publishers' general and contractual scores with editorial employes' general and contractual scores. The first of these eighteen hypotheses is written out in detail and the remaining seventeen are given in abbreviated form. Null Hypothesis 3A: There is no difference between the correlation for publishers on general and contractual V-l (a joint committee to discuss news- room policy and practice) and the correlation for editorial employes on general and contractual V-l. H : V-l = V-l O rp re where 76 V-lr correlation (r) for publishers on contractual p and general V-l V-l correlation (r) for editorial employes on e contractual and general V-l Alternative Hypothesis 3A: The correlation on general and contractual V-l will be higher for editorial employes than for publishers. H : V-1 ) V-1 3A re Fp where the legend is the same as for the null hypothesis. Null Hypothesis 38: There is no difference between the correlation for publishers and editorial employes on general and contractual V-2, the right to consultation on the appointment of supervisory personnel in the newsroom. Alternative Hypothesis 38: The correlation on general and contractual V-2 will be higher for editorial employes than for publishers. Null Hypothesis 3C: There is no difference between the correlation for publishers and editorial employes on general and contractual V-3, represen- tation by editorial employes on editorial boards. Alternative Hypothesis 3C: The correlation on general and contractual V-3 will be higher for editorial employes than for publishers. Null Hypothesis 3D: There is no difference between the correlation for publishers and editorial employes on general and contractual I-l, not requiring an employe to perform a duty which he feels compro- mises his integrity. Alternative Hypothesis 3D: The correlation on general and contractual I-l will be higher for editorial employes than for publishers. 77 Null Hypothesis 3E: There is no difference between the correlation for publishers and editorial employes on general and contractual I-2, not requiring an employe to perform duties other than those of his job description. Alternative Hypothesis 3E: The correlation on general and contractual I-2 will be higher for editorial employes than for publishers. Null Hypothesis 3F: There is no difference between the correlation for publishers and editorial employes on general and contractual I-3, not using the byline or credit line over the employe's protest. Alternative Hypothesis 3F: The correlation on general and contractual I-3 will be higher for editorial employes than for publishers. Null Hypothesis 3G: There is no difference between the correlation for publishers and editorial employes on general and contractual 1-4, notifying an employe of substantive changes in copy prior to publication. Alternative Hypothesis 36: The correlation on general and contractual I-4 will be higher for editorial employes than for publishers. Null Hypothesis 3H: There is no difference between the correlation for publishers and editorial employes on general and contractual 1-5, not requiring an employe to write false or misleading copy. Alternative Hypothesis 3H: The correlation on general and contractual I-5 will be higher for editorial employes than for publishers. 78 Null Hypothesis 31: There is no difference between the correlation for publishers and editorial employes on general and contractual I-6, printing corrections or retractions only after consulting with the employe concerned. Alternative Hypothesis 31: The correlation on general and contractual 1-6 will be higher for editorial employes than for publishers. Null Hypothesis 3J: There is no difference between the correlation for publishers and editorial employes on general and contractual 1-7, per- mitting employes the right of simultaneous reply to letters concerning their work. Alternative Hypothesis 3J: The correlation on general and contractual I-7 will be higher for editorial employes than for publishers. Null Hypothesis 3K: There is no difference between the correlation for publishers and editorial employes on general and contractual P-l, employe privilege against disclosure of sources. Alternative Hypothesis 3K: The correlation on general and contractual P-l will be higher for editorial employes than for publishers. Null Hypothesis 3L: There is no difference between the correlation for publishers and editorial employes on general and contractual P-2, employe privilege against authentication of sources. Alternative Hypothesis 3L: The correlation on general and contractual P-2 will be higher for editorial employes than for publishers. 79 Null Hypothesis 3M: There is no difference between the correlation for publishers and editorial employes on general and contractual P-3, agree- ment that the employer will maintain the confidentiality of the employe's sources. Alternative Hypothesis 3M: The correlation on general and contractual P-3 will be higher for editorial employes than for publishers. Null Hypothesis 3N: There is no difference between the correlation for publishers and editorial employes on general and contractual P-4, agree- ment that the employer will notify the employe and the union of a demand for disclosure of sources. Alternative Hypothesis 3N: The correlation on general and contractual P-4 will be higher for editorial employes than for publishers. Null Hypothesis 3-0: There is no difference between the correlation for publishers and editorial employes on general and contractual P-5, agree- ment that the employer will join as a party in a disclosure case. Alternative Hypothesis 3-0: The correlation on general and contractual P-S will be higher for editorial employes than for publishers. Null Hypothesis 3P: There is no difference between the correlation for publishers and editorial employes on general and contractual P-6, agree- ment that the employer will pay all legal fees in a disclosure case. Alternative Hypothesis 3P: The correlation on general and contractual P—6 will be higher for editorial employes than for publishers. 80 Null Hypothesis 3Q: There is no difference between the correlation for publishers and editorial employes on general and contractual P-7, agree- ment that the employer will pay all losses suffered by an employe in a disclosure case. Alternative Hypothesis 3Q: The correlation on general and contractual P-7 will be higher for editorial employes than for publishers. Null Hypothesis 3R: There is no difference between the correlation for publishers and editorial employes on general and contractual P-8, agree- ment that the employer will pay all employe losses in a dis- closure case regardless of the employe's course of action. Alternative Hypothesis 3R: The correlation on general and contractual P-8 will be higher for editorial employes than for publishers. Testing Hypotheses 3A - 3R: Method of Analysis: Pearson Correlation and Difference of Means T-Test In this analysis the correlation for each pair of variables, such as VG-l and VC-l (general and contractual Voice 1) was calculated for publishers and a.separate correlation was calculated for editorial employes. The two scores were then transformed to z scores (standard scores) and a difference of means T-test applied (Snedecor and Cochran, 1967; Lordahl, 1967). Test of Significance: T-Test Significance Level: .05 Assumptions of the Difference of Means Test: Same as Hypothesis 1 81 Hypotheses 4A - 4C The fourth series of hypotheses (Hypotheses 4 through 10) test the effects of one independent categorical variable on the composite variables of contractual Voice (VC-l + VC-2 + VC-3), Integrity (IC-l + IC-2 + ... IC-7) and Privilege (PC-l + PC-2 + ... PC-8). To form the composite dependent variables, each of the items in that category was summed. A modification of this procedure was used in analyzing the results of those hypotheses dealing with publisher scores and is discussed in detail in Chapter IV. Null Hypothesis 4A: Publishers of newspapers with less than 50,000 circulation will have the same mean scores for contractual Voice as publishers of newspapers with circulations of 50,000 or more. H : V = V = V 0 pa pb DC where Vp = mean score on contractual Voice for publishers a of newspapers with less than 50,000 circulation Vp = mean score on contractual Voice for publishers b of newspapers with circulations between 50,000 and 200,000 Vp = mean score on contractual Voice for publishers c of newspapers 200,000 Alternative Hypothesis 4A: with circulations greater than Publishers of newspapers with less than 50,000 circulation will have higher mean scores for contractual Voice than publishers of newspapers with circulations of 50,000-200,000 and publishers with circulations greater than 200,000. V ;> V ) V pa pb pc H4A: where the legend is the same as in the null hypothesis. 82 Null Hypothesis 4B: Publishers of newspapers with less than 50,000 circulation will have the same mean scores for contractual Integrity as publishers of newspapers with circulations of 50,000 or more. I = mean score on contractual Integrity for publishers a of newspapers with less than 50,000 circulation I = mean score on contractual Integrity for publishers of newspapers with circulations between 50,000 and 200,000 I = mean score on contractual Integrity for publishers of newspapers with circulations greater than 200,000 Alternative Hypothesis 4B: Publishers of newspapers with less than 50,000 circulation will have higher mean scores for contractual Integrity than publishers of newspapers with circulations of 50,000-200,000 and publishers with circulations greater than 200,000. H : I ) I > I 48 pa pb DC where the legend is the same as in the null hypothesis. Null Hypothesis 4C: Publishers of newspapers with less than 50,000 circulation will have the same mean scores for contractual Privilege as publishers of newspapers with circulations of 50,000 or more. H : P = P = P P o a pb pc where P = mean score on contractual Privilege for publish- pa ers of newspapers with less than 50,000 circulation P = mean score on contractual Privilege for publish- ers of newspapers with circulations between 50,000 and 200,000 83 Pp = mean score on contractual Privilege for publish- c ers of newspapers with circulations greater than 200,000 Alternative Hypothesis 4C: Publishers of newspapers with less than 50,000 circulation will have higher mean scores for contractual Privilege than publishers of newspapers with circulations of 50,000-200,000 and publishers with circulations greater than 200,000. : P H4C Pp ) Db) Pp a C where the legend is the same as in the null hypothesis. Testing Hypotheses 4A - 4C: Method of Analysis: One-Way Analysis of Variance One-way analysis of variance is used to test the possible effects of one independent variable on a dependent variable. In Hypotheses 4A through 4C the dependent variables are Voice, Integrity and Privilege respectively and the independent variable is circula- tion size. Test of Significance: F-Test The F ratio is the ratio of the mean square between groups divided by the mean square within groups (i.e., the error mean square). When there are only two groups being compared, the F-test is equivalent to the T-test. Significance Level: .05 Assumptions of One-Way Analysis of Variance: 1. Subjects are randomly selected from the population. 2. Population means are randomly distributed. 3. Data are interval-level measures. 84 4. Relationships among the variables are linear. 5. Variances are equal among treatments of the independent variable. The assumption of homogeneity of variance is tested by Bartlett's test, which yields a chi-square. If the chi-square is significant for the number of groups minus one degree of freedom, there is reason to believe that the assumption of homogeneity of variance has not been met (Snedecor and Cochran, 1967). A discussion of the effects of violating assumptions of this test is included in the Descriptive Statistics section of Chapter IV. Hypotheses 5A - 5C Null Hypotheses 5A, 58, 5C: There is no difference in mean scores on contractual (A) Voice (8) Integrity (C) Privilege among publishers who hold contracts dating from December 1933 through August 1945 (32-44 years); from September 1945 through August 1965 (12-31 years); and from September 1965 through December 1976 (less than 12 years). H : 4A: v = v = v ° p33-45 p45-65 p65-76 48: I = I = I p33.45 p45-65 p65-76 4c: Pp = P = P 33-45 P45-65 Pas-76 where V = mean score on contractual Voice for pub- p33.45 lishers whose first Guild contract is dated between December 1933 and August 1945 85 V = mean score on contractual Voice for p45-65 publishers whose first Guild contract is dated between September 1945 and August 1965 Vp = mean score on contractual Voice for 65-76 publishers whose first Guild contract is dated between September 1965 and December 1976 Ip = same as Voice legend above, substituting Integrity for Voice P = same as Voice legend above, substituting p Privilege for Voice Alternative Hypotheses 5A, 58, 5C: The mean scores of publishers who hold contracts dating from December 1933 through August 1945 will be higher on contractual (4A) Voice (4B) Integrity (4C) Privilege than those of publishers with contracts dating from September 1945 through August 1965 and from September 1965 through December 1976. H : V ) V V 5A p33-45 p45-65 > p65-76 H : I I I 53 p33-45 > p45-65 > Pas-76 H : P P P 5C p33-45 ) p45-65 > p65-76 where the legend is the same as in the null hypotheses. Testing Hypotheses 5A - 5C: Method of Analysis: One-Way Analysis of Variance Test of Significance: F-Test Significance Level: .05 Assumptions of One-Way Analysis of Variance: Same as Hypothesis 4 86 Hypotheses 6A - 6C Null Hypotheses 6A, 6B, 6C: There is no difference in mean scores on contractual (6A) Voice (6B) Integrity (6C) Privilege among publishers who consider their relationship with The Newspaper Guild to be excellent or good and those who consider it to be fair or poor. H : 6A: V = V 0 p9 pp 68: I =1 pg pp 6C P = P pg pp where Vp = mean score on contractual Voice for g publishers who rate their relationship with The Newspaper Guild as excellent or good V = mean score on contractual Voice for p publishers who rate their relationship with The Newspaper Guild as fair or poor Ip = same as Voice legend above, substituting Integrity for Voice P = same as Voice legend above, substituting p Privilege for Voice Alternative Hypotheses 6A, 6B, 6C: The mean scores of publishers who rate their relationship with The Newspaper Guild as excellent or good will be higher on contractual (6A) Voice (6B) Integrity (6C) Privilege than those of publishers who rate their relationship as fair or poor. 87 H v > v 6A p9 pp H I > I 68 p9 “P H P > P 6C p9 pp where the legend is the same as in the null hypotheses. Testing Hypotheses 6A - 6C: Method of Analysis: One-Way Analysis of Variance Test of Significance: F-Test Significance Level: .05 Assumptions of One-Way Analysis of Variance: Same as Hypothesis 4 Hypotheses 7A - 7C Null Hypotheses 7A, 7B, 7C: There is no difference in mean scores on contractual (7A) Voice (7B) Integrity (7C) Privilege between publishers who are affiliated with a group and those who are not. H : 7A: V = V 78: I = I Pg pi 7C P = P Pg p,i where Vp = mean score on contractual Voice of publishers 9 who are members of a group 88 Vp = mean score on contractual Voice of independent i publishers I = same as Voice legend above, substituting p Integrity for Voice Pp = same as Voice legend above, substituting Privilege for Voice Alternative Hypothesis 7A: The mean scores of publishers of independent newspapers will be higher on contractual Voice than those of publishers who are members of a group. where the legend is the same as in Null Hypothesis 7A. Alternative Hypothesis 7B: The mean scores of publishers who are members of a group will be higher on contractual Integrity than those of independent publishers. H : I I 78 pg 7 p.i where the legend is the same as in Null Hypothesis 78. Alternative Hypothesis 7C: The mean scores of publishers who are members of a group will be higher on contractual Privilege than those of independent publishers. H:P)P 7C Pg P]. where the legend is the same as in Null Hypothesis 7C. Testi ngypotheses 7A - 7C: Method of Analysis: One-Way Analysis of Variance Test of Significance: F-Test Significance Level: .05 89 Assumptions of One-Way Analysis of Variance: Same as Hypothesis 4 Hypotheses 8A - 8C Null Hypotheses 8A, 88, 8C: There is no difference in mean scores on contractual (8A) Voice (88) Integrity (8C) Privilege among editorial employes who rate their working relationship with their employer as excellent, good, fair or poor. HO: 8A: ve = v = v = v e g V = mean score on contractual Voice of employes e who rate their relationship with their employer as excellent Ve = mean score on contractual Voice of employes 9 who rate their relationship with their employer as good Ve = mean score on contractual Voice of employes f who rate their relationship with their employer as fair Ve = mean score on contractual Voice of employes p who rate their relationship with their employer as poor I = same as Voice legend above, substituting Integrity for Voice Pe = same as Voice legend above, substituting Privilege for Voice 90 Alternative Hypotheses 8A, 88, 8C: Editorial employes who rate their relationship with their employer as excellent will have lower mean scores on contractual (8A) Voice (88) Integrity (8C) Privilege than will employes who rate their relationship as good, fair or poor. 8C: Pe ‘( Pe < Pef ( Pe e 9 P where the legend is the same as in the null hypotheses. Testing Hypotheses 8A - 8C: Method of Analysis: One-Way Analysis of Variance Test of Significance: F-Test Significance Level: .05 Assumptions of One-Way Analysis of Variance: Same as Hypothesis 4 Hypotheses 9A - 9C Null Hypotheses 9A, 9B, 9C: There is no difference in mean scores on contractual (9A) Voice (9B) Integrity (9C) Privilege among editorial employes who differ in their attitudes toward and support of The Newspaper Guild. 98: I = I6 96: P =Pe 91 = Ve = Ve. = Ve S 1 11 = I = I = I e s ei en : P = P = P es ei en mean score on contractual Voice of editorial employes who are officers of the union mean score on contractual Voice of editorial employes who are active and supportive of the union mean score on contractual Voice of editorial employes who are supportive of the union mean score on contractual Voice of editorial employes who are indifferent toward the union mean score on contractual Voice of editorial employes who would prefer not to be members of the union same as Voice legend above, substituting Integrity for Voice same as Voice legend above, substituting Privilege for Voice Alternative Hypotheses 9A, 9B, 9C: The mean scores of editorial employes who are officers of The Newspaper Guild or who are active and supportive of the union will be higher on contractual (9A) Voice (9B) Integrity (9C) Privilege than those of employes who are indifferent toward the union or who would prefer not H9A‘ veo > Ve a )ve S to be members. ) V ) V s ei en where the legend is the same as in the null hypotheses. Testing Hypotheses 9A - 9C: Method of Analysis: One-Way Analysis of Variance Test of Significance: F-Test Significance Level: .05 Assumptions of One-Way Analysis of Variance: Same as Hypothesis 4 Hypotheses 10A - 10C Null Hypotheses 10A, 108, lOC: There is no difference in mean scores on contractual (10A) Voice (108) Integrity (10C) Privilege between editorial employes who have less than 10 years' experience as journalists and those who have more than 10 years' experience. H: 10A: V =V 0 en ev 108: I = I en ev 10C Pe = P n 6n where Ve = mean score on contractual Voice of n editorial employes with less than 10 years' experience (n = novice) V = mean score on contractual Voice of v editorial employes with more than 10 years' experience (v = veteran) 93 H II same as Voice legend above, substituting Integrity for Voice .0 (D 11 same as Voice legend above, substituting Privilege for Voice Alternative Hypotheses 10A, 108, 10C: The mean scores of editorial employes with less than 10 years' experience as journalists will be higher on contractual (10A) Voice (108) Integrity (10C) Privilege than those of editorial employes with more than 10 years' experience as journalists. “iOA‘ Ve > ve n V H : I ) I 108 en 9v HIOC: PE > Pe n V where the legend is the same as in the null hypotheses. Testing Hypotheses 10A - 10C: Method of Analysis: One-Way Analysis of Variance Test of Significance: F—Test Significance Level: .05 Assumptions of One-Way Analysis of Variance: Same as Hypothesis 4 Data Analysis The data from the survey were analyzed on the Control Data Corp. 6500 computer at Michigan State University. The system of computer programs as printed in the Statistical Package for the Social Sciences (Nie et al., 1975) was used to input the data. 94 The following programs were used: 1. FREQUENCIES for descriptive statistics and as a check on the assumptions used in testing Hypotheses 1 through 10; 2. T-TEST to test Hypotheses l, 2 and 3; 3. PEARSON CORR to test Hypothesis 3 and as a reliability check on index item pairs; 4. ONEWAY (analysis of variance) to test Hypotheses 4 through 10. CHAPTER IV RESULTS AND DISCUSSION The survey was conducted between May and August 1977, with three mailings each to publishers and Guild members at four to five- week intervals in May, June and July. A total of 90 usable question- naires were obtained from publishers (51.1 per cent return) and 352 usable questionnaires from editorial employes (62.4 per cent return). The study was designed to compare the attitudes of publishers and editorial employes on (1) the general concepts proposed by the Guild in the area of newsroom democracy and (2) incorporating those concepts in contracts. It attempted to identify areas of agreement between the two groups and to assess which of the newsroom democracy concepts publishers favored and which concepts employes favored. The main questions to be answered by the study were: (1) Are publishers willing to formalize newsroom democracy through contractual agreements? (2) Are editorial employes in favor of such action? The Guild's proposals in newsroom democracy are examples of attempts toward greater worker participation and involve the redefi- nition of management rights. Newsroom democracy challenges the traditional employer-employe roles in the newsroom by giving the employe greater control over his work product, and hence, the employer's product. 95 96 In addition to examining which newsroom democracy issues were favored by publishers and employes, certain publisher, employe and publication characteristics were examined to determine their possible effect on attitudes toward newsroom democracy. Specifically, the characteristics considered were: (1) circulation size of the newspaper, (2) number of years contract has been in effect, (3) working relationship between the publisher and The Newspaper Guild, (4) publisher's affiliation with a group, (5) working relationship between employe and employer, (6) working relationship between employe and The Newspaper Guild, (7) number of years employe has worked as a journalist. Descriptive Statistics The following descriptive data were obtained for each of the thirty-six items on the questionnaire: mean, mode, median, kurtosis, skewness, standard error, standard deviation and variance. The number of cases, mean, skewness, standard deviation and variance for each of the thirty-six statements in the questionnaire on both the general principle scores and the contract scores are presented in Tables B-5 through B-lO (see Appendix B). The skewness of the scores on each item help to determine the degree to which a distribution of cases approximates a normal curve. The skewness takes on a value of 0 when the curve is completely symmetrical. A positive value indicates that the cases are clustered 97 more to the left of the mean with most of the extreme values to the right. A negative value indicates clustering to the right of the mean and the extreme values to the left. Tables B-5 through B-lO show that none of the thirty-six items has a completely symmetrical distribution. The direction of the skew- ness varies for publishers on general principle statements and contract provisions in both size and direction without any apparent pattern. For editorial employes, however, almost all measures in both the general principle and contract categories have a negative skew, indi- cating that the bulk of responses are to the right of the mean (that is, higher than the mean), and the extreme cases fall below the mean. The nature of the skewness should be kept in mind in interpreting the results of the data, since one of the assumptions of the T-test and the F-test is a normal distribution of the variable. The standard deviation is a measure of the dispersion about the mean and is the square root of the variance. Standard deviations range from a low of .256 for publishers on IG-5Al to a high of 1.617 for publishers on IC-5Al. (This example indicates that the dispersion of scores about the mean is lowest for publishers when considering the general principle of not printing false information and is highest when publishers consider putting that statement in a contract.) The fluctuation in standard deviations for publishers and editorial employes appears to be about the same. The standard deviation and variance data also should be taken into account in interpreting the results of the study, because homogeneity of variance between groups is another assumption of the T-test and the F-test. An F-probability test (where F = variance of editorial 98 employes 4 variance of publishers and 0< = .001) was carried out on each of the thirty-six general principle and thirty-six contract pro- vision statements to determine the homogeneity or heterogeneity of variance. The following twenty-seven of the seventy-two items were found to violate the assumption of equal variances: VG-lA, VG-lB, VG-BB IG-lA, IG-ZB, IG-4A, IG-SAl, IG-6A, IG-68 PG-lA, PG-3A, PG-BB, PG-4A2, PG-6A, PG-7A, PG-BB VC-lB IC-lB, IC-3A, IC-5A2 PC-lA, PC-2A, PC-3A, PC-4Al, PC-4A2, PC-6A, PC-7A. These results, combined with the skewness of each variable, indicate that a double violation of the T and F-test assumptions exists in more than a third of the statement variables. Yet due to the nature of the two tests, the violations do not cause serious problems in interpret- ing the data. Boneau (1971) has shown that the T and F-tests are remarkably robust in that they are only inconsequentially affected by a violation of these underlying assumptions. Furthermore, Boneau suggested that as the sample size exceeds 30, the effects of viola- tions of assumptions become negligible. Increasing the sample size has the effect of removing the skew. In addition to the descriptive statistics given in Tables B-5 through B-lO, a number of frequencies on the publisher and editorial employe samples were recorded. Table 1 lists the job classifications into which the editorial employe respondents fall. 99 TABLE l.--Editorial employe job classifications Category label Reporter Editor Copyeditor Photographer Other Total Absolute frequency Adjusted frequency (%)* 53.6 16.7 6.6 7.2 15.9 100.0 *Non-responses are not included in the percentage calculations. The average (mean) number of years an employe had worked for his current employer was 12.6; the median, 8.8. The average (mean) number of years an employe had worked as a journalist was 16.0; the median, 12.7. Employes were asked to categorize their working relationship with their publishers as excellent, good, fair or poor. The results are presented in Table 2. TABLE 2.--Working relationship between editorial employe and publisher Relationship Excellent Good Fair Poor Total Absolute frequency 87 177 60 23 347 Adjusted frequency (%)* 25.1 51.0 17.3 6.6 100.0 *Non-responses are not included in the percentage calculations. 100 Editorial employes were also asked to assess their relation- ship with The Newspaper Guild. These results are presented in Table 3. TABLE 3.--Relationship between editorial employe and The Newspaper Guild Relationship Absolute Adjusted frequency frequency (%)* Officer 41 11.8 Active/supportive 64 18.4 Supportive 172 49.4 Indifferent 57 16.4 Would prefer not to be a member 14 4.0 Total 348 100.0 *Non-responses are not included in the percentage calculations. Among publishers, respondents listed their titles according to the classifications found in Table 4. TABLE 4.--Publisher title classifications Job title Absolute Adjusted frequency frequency (%)* Publisher, president, editor-in-chief, general manager, business manager 36 41.2 Executive editor, managing editor 46 52.9 Other 8 5.9 Total 90 100.0 *Non-responses are not included in the percentage calculations. 101 The circulation breakdown of publisher respondents' newspapers is presented in Table 5. TABLE 5.--Daily newspaper circulation figures among publisher respondents Circulation Absolute Adjusted frequency frequency (%)* Less than 25,000 3 3.8 25,000-49,999 12 15.0 50,000-99,999 18 22.5 100,000-199,999 14 17.5 200,000-499,999 22 27.5 500,000+ 11 13.7 Total 80 100.0 *Non-responses are not included in the percentage calculations. 0f the 90 publisher respondents, 72 were publishers of daily newspapers; 4 of weekly newspapers; 3 of magazines; 3, managers of radio/television stations; and 8, "miscellaneous" or no response. Forty-one (47.7 per cent) of the publishers were affiliated with a group and 45 (52.3 per cent) were independent. The average (mean) number of years a publisher had held a contract with the Guild was 28.0 years; the median, 30.0 years. The assessment among publishers of their relationship with The Newspaper Guild is presented in Table 6. 102 TABLE 6.--Working relationship between publisher and The Newspaper Guild Relationship Absolute Adjusted frequency frequency (%)* Excellent 4 5.5 Good 37 50.7 Fair 26 35.6 Poor 6 8.2 Total 73 100.0 *Non-responses are not included in the percentage calculations. Validity and Reliability Validity Unless the measurements used in the study have validity, the results are suspect. The three major types of validity are content, construct and criterion (both predictive and concurrent): 1. Content validity: Does the measurement cover the content under analysis? 15 the range of material being investigated explored in its entirety? 2. Construct validity: Do the questions measure what they are supposed to be measuring? 00 the items on the questionnaire measure publisher and reporter attitudes toward the newsroom democracy concepts of Voice, Integrity and Privilege? 3. Criterion validity: (a) Predictive: Can the measures be validated against some outside criterion? Can publisher or reporter scores be predicted by some independent measure? (b) Concurrent: Can the predictive criterion be applied at the same time the measure— ment is made? The measuring instrument for this study is difficult to 103 assess in terms of content, construct and criterion validity. The main problem lies in the "given" nature of the definition of newsroom democracy. In relation to content validity, assessing attitudes on contract language necessarily involves the use of that contract language. If one were to seek answers on a scale that looked at the general area of employe rights, the content validity would be lost. The content is the contract language. It is the parameter within which the study must work. This limitation poses problems in achieving construct validity. A collective bargaining agreement is a legal contract. Words have particular meaning in the context of the contract. What may appear to be minor adjustments, omissions or additions to the language could greatly alter the obligations placed upon the parties to the contract. In constructing the questionnaire, an attempt was made to follow the model language proposed by The Newspaper Guild as closely as possible. Most of the positively worded "A" statements are either Guild model language or language approximating it as closely as possible. The reverse "8" statements were written in such a way as to attempt a pro-publisher mirror image of the Guild's model language. That is, the "8" items were worded to give publishers the rights that the Guild was seeking for its nmmbers. For example, an "A" item on the questionnaire would give the reporter the right to determine whether he would disclose his source of information. The correspond- ing "8" item would give the publisher the right to determine whether a reporter could refuse to disclose his source of information. Trying to create a balance between pro-Guild and pro-Publisher 104 statements posed problems in that some statements appeared implausible when reversed. For example, the reverse of "An employe shall not be required to prepare anything for publication in such a way as to dis- tort any facts or to create an impression which the employe knows to be false" would be "The employer may require an employe to prepare for publication material which may distort facts or create a false impression." Obviously no publisher would favor such a clause. Yet to rework the statement and delete the reference to distorting infor- mation would invalidate the measure both in content and construction. In three instances (V-2, I-5 and P-4) the item pairs do not consist of "A" and "B" statements, but rather consist of two similar "A" statements. A second problem in the construction of the "A" and "8” items is insuring that the "8" items do in fact measure the same attitude as the corresponding "A" items. This problem is discussed in detail in the Reliability section of this Chapter. Because this study is basically an exploratory one in an area that has not been investigated previously, the opportunity to deter- mine predictive validity of the questionnaire is limited. The only independent criterion on which publishers' answers might be compared is the actual language of the contracts they hold. Some of the hypo- theses were based on the data collected from contracts containing Voice, Integrity and Privilege. But the publishers who responded to the survey were not necessarily publishers whose contracts contained Voice, Integrity and/or Privilege clauses. It would be faulty to assume that because a publisher's contract had no Voice, Integrity or Privilege written into it that he was strongly opposed to it. 105 Other factors may have influenced the contract provisions: date of negotiations, other concessions by the publisher, the tenor of the Guild bargaining committee, the primary issues under consideration by both parties. Closely tied to predictive validity is concurrent validity. A comparison of contract language with publishers' attitudes may not offer a true test of validity because of the time lapse between the signing of the last contract and the time the survey was conducted. All of the situations presented above pose problems in evalu- ating the validity of the measuring instrument. The problems are inherent in the nature of the study (the definition of newsroom demo- cracy). Attempts were made to deal with these problems through the evaluation of the pretest and through reliability checks on the items. Reliability If the measuring instrument lacks reliability, it is necessar- ily invalid. To check on the reliability of the item pairs comprising the eighteen newsroom democracy indices, an inter-item correlation matrix was computed, the results of which are presented in Tables B-ll through B-15 (refer to Appendix B). Each item was correlated with its pair and with the sum of the two items. On all item pairs for editorial employes, the correlation (Pearson r) values were signi- ficant both in correlations between paired items and in correlations between each item and the index sum. For publishers, all r values were significant in correlations between each item and the index sum. However, there was no significant positive correlation between publisher scores on the general principle paired items for VG-l, 106 IG-l, IG-3, IG-6, PG-l and PG-8. Such findings suggested that for publishers these newsroom democracy general principles and their negative pairs were not measuring the same attitudes. Therefore, if the means of the "A" and "8" items comprising these six indices were added to form a composite index--as was the original intent of the study--the results would not be reliable, and hence, not valid. The probable explanation as to why the "A" and "B" statements were not drawing the same response from publishers is that the two statements are not viewed as opposites. For example, on VG-lA the publishers' mean was much higher than that of VG-lB. Publishers did not equate allowing reporters to discuss news policy matters with their right to make decisions regarding news policy. The reliability problem occurs only with publishers. For editorial employes, the "A" and "B" statements appear to be measuring the same attitude. All the items in the study were pretested on both editorial employes and publishers. But the number of publisher returns (four) was so small that an inter-item correlation matrix was not computed on them. On the contract provision paired items for publishers, all but three (VC-2, IC-5 and PC-4*) were either not significantly corre- lated or were negatively correlated. The negative correlations sug- gested that not only were the "A" and "8'' items not measuring the same attitude, but they were measuring opposite attitudes. The question to consider was why this should be so when editorial employes consider the "A" and "B" statements to be equivalent measures. *Note that the three pairs that were correlated are those with two positive "A" items rather than one positive and one negative. 107 The apparent explanation is that publishers want neither newsroom democracy language favorable to them or to their employes incorporated in a contract. The assumption made in reversing the scale on the "8" items was that publishers would favor inclusion of statements that guaranteed them control over Voice, Integrity and Privilege. But publishers, it appears from the data, would rather leave the contract open; that is, free of any binding statements. (Publishers' written comments concerning the study (see Appendix F) lend support to this possible explanation.) In so doing they retain the right to control Voice, Integrity and Privilege through the unwritten management rights dictum that what is not given away to labor in a contract is reserved for management. By scoring their questionnaires "strongly oppose" to items favoring publishers' rights, the "8" scores for publishers were higher than would have been pre- dicted from the corresponding "A“ scores. In order to take into account the invalidity of pairing publishers' item responses on the non-significant or negatively correlated indices, the thirty-six items were divided into "A" and "8" categories and were examined separately rather than as paired indices. The hypotheses in which these non-correlated items appear were tested separately for "A" and "8" scores. In order for a hypothesis to be confirmed, both the "A" and "8" results had to be significant at .05. In analyses dealing only with editorial employes, the composite indices were retained as the appropriate measures. 108 Results of Hypotheses Tests Test of Hypotheses 1A - 1R Null Hypotheses: There is no difference between the mean scores of publishers and editorial employes on (1A) VG-1* (lG) IG-4 (1M) PG-3 (lB) VG-2 (1H) IG-S (1N) PG-4 (1C) VG-3 (11) IG-6 (1-0) PG-5 (10) IG-l (10) IG-7 (1P) PG-6 (1E) IG-2 (1K) PG-l (10) PG-7 (1F) IG-3 (1L) PG-2 (1R) PG-8 Alternative Hypotheses: The mean score for editorial employes will be higher than the mean score for publishers on (1A) VG-l (lG) IG-4 (1M) PG-3 E18; VG-Z (1H; IG-5 (1N) PG-4 VG-3 lI IG-6 (l-O) PG-5 (lD) IG-l (lJ) IG-7 (1P) PG-6 (1E) IG-2 (1K PG-l (10) PG-7 (1F) IG-3 (1L PG-2 (1R) PG-8. Hypotheses 1A through 1R examine the difference between publishers' mean scores on a two-item index of each of the eighteen newsroom democracy proposals and the mean scores of editorial employes on the same indices. The alternative hypotheses postulate that a significantly higher positive response toward all eighteen proposals as general concepts will be obtained from editorial employes than from publishers. The wording of the items followed as much as possible the Model Contract language proposed by the Guild. All but three of the indices contained one statement in which the "strongly agree" category indicated a favorable response toward the model language, *For the statements corresponding to the symbols, refer to Appendix H. 109 and the second in which the "strongly agree" category indicated an unfavorable response. In computing the mean scores, all the unfavor- able response item scales were reversed, so that all statements indi- cating a strongly favorable attitude toward newsroom democracy were assigned a value of 5. For example, the positive-negative item pair for index 1-6 was as follows: (A) An employe should be informed of major changes in his copy before publication. (8) The employer need not consult with an employe regarding major changes made in his material prior to publication. In this example, a "strongly agree” response would be scored 5 for statement "A" but a "strongly agree" would be scored 1 for statement "8." The mean scores were obtained by summing the scores on the two index items for each of the eighteen proposals, then summing all publisher (or editorial employe) scores and dividing by the total number of publishers (or editorial employes). In those instances where only one response was recorded on an index, the score on that response was doubled and was included in the summation. Only respond- ents who failed to answer both items on an index were included in the missing data category. For those indices with non-significant corre- lations for the two items, separate analyses were computed on the "A“ and "8" items rather than on the indices. A two-tailed difference of means T-test was used to evaluate the significance of the relationship. To take account of possible differences in variance between the mean scores of publishers and editorial employes, a pooled variance estimate was used if the two- tailed probability of the F value was greater than .001 (confirming 110 the null hypothesis of equal variance for publisher and editorial employe scores). The separate variance estimate was used if the two- tailed probability of the F value was less than or equal to .001 (indicating unequal variances) (Nie et al., 1975). The results are presented in Table 7. Null Hypotheses 1A through 1R were rejected in all but one case: Hypothesis 1H concerning IG-5 statements which are: (A1) All news copy should be presented without distortion of any facts and without an attempt to create false impressions. (A2) No editorial employe should be required to prepare false or misleading material. For this index pair there was no significant difference between publisher and editorial employe scores. Of a possible maximum mean score of 10, the mean scores for publishers and editorial employes were 9.765 and 9.784 respectively. The fact that Hypothesis 1H cannot be rejected suggests that factual, undistorted reporting of news is of equal importance to publishers and editorial employes. 0n no other newsroom democracy statement do publishers and editorial employes concur so strongly. All the remaining newsroom democracy indices and items show a significant difference (p < .001) between publishers and editorial employes. In all cases, editorial employes scored more favorably toward the newsroom democracy principles than did pub- lishers. These data, which support rejection of Null Hypotheses 1A through 1R (with the exception of 1H) suggest that editorial employes have more favorable attitudes than publishers toward general concepts concerning input in newsroom practices, protection of the work pro- duct, and protection against forced disclosure of news sources. 111 oem.o mam mm ooo. No.o - Nmo. oe.~ moo.~- mop.“ am one eioH e~m.m mom mm ooo. mo.o . ooo. No.P oo¢.~: moo.~ mo goo omiom ome.e mom mm ooo. om.m - Fpo. Fm.F mun.o- moo.m om one mmo.m Fem mm ooo. um.oF- moo. on._ moo.mi ee~.m om no; Nio> Foo.m com um ooo. mm.a_- ooo. me.m nom.P- eoe.~ mo ago o~1o> Neo.m Rem mm ooo. eo.o . ooo. mo._ Nee.P- oom.m om one mocmsmmmwo mpaeucwsa suepwnmnoca mopm>ie sowpeamnosa mafim>iu cow: com: 2 «goose Posmcmo somguoeoo soosmzmc to mmFQPucwso Fmsmcmo co mmaopaem _ewsopwom use msmgmwFaoa soe umwuue memos mo mucwsmmmeo--.m mooup sowrenenosa mopm>iu com: cum: 2 «oooso pesocmo umocmucoouu.n m4muh sHVFeoeaosa mopm>im com: cum: 2 «oooso Pmsmcwo smocpucoouu.m m4m noe.m oem mm ooo. m~.N~- mop. Fm._ Fm~.~- opn._ an no; oon.m emm mm ooo. Ne.o . “No. em._ “mm.~- mm~.m on no; Nio> opm.m omm mm opo. Fe.m - ooo. oo.~ ooo. . moo.~ on one opio> No~.m Fem um ooo. mo.mp- ooo. mm.~ ~m~._- oeo.~ mu one mucmsmeewo cowme>osq xuw_wnenosa mope>ih sHPFenanosa moFe>im com: cum: 2 «oooso “unsucoo mongooewo soosmzm: mo mcoeme>oso ”mouoosucou co mesa—gem megouwom ecu msmsmppooa so» “maple momma eo mucosmmmwolu.o mooih mop. ooo. mnm. ooo. mN—. New. ooo. Nmo. ooo. mmm. mpo. sueeeeeeosa o_._ mo.m om.~ o~.P oo.~ oe.P on.m eo._ oe.F mopm>im mmm. a vom._u Nme. u wpm.Pu mmF. u omw.pn 5N0. u oum.~u oom.~: moo. . ewN.Pu mucmsmwwwo cam: mo¢.m omm.m mmm.m pow.~ onm.m nmm.m mme.e omm.m Noe.m Nem.m oom.m omN.N oom.m www.~ umm.m Fwe.m mmo.w opw.m mmo.¢ mmo.m oup.e mom.~ :mmz emm mmm Nu mmm omm mu «mm «mm es mmm mmm on Pmm on mmm com mm mm one omioa mo ago <~ioa mo ago opioa mm one <_-oo mm one ouioo mm one oca «nooso pumspcoo easeebeou--.m Mona» 118 .mmsopoem —mwsoawom u mm “msmcmmpnoa u noes NoN.e oem mu ooo. mo.¢ - mmo. me.p moo. . omo.m PR can oo-oa mmn.m Nmm mo ooo. o_.P—- moo. mo._ Npo._- mmp.m mm goo osa sue~_amnosa mope>ip sum—rampage mopm>im com: com: 2 roaoso pumeucoo voocwucounu.m mom200,000 26 Total 64 Source of D.F. variance Between groups 2 Within groups 61 Total 63 Std. dev. 2.503 2.726 2.248 Sum of squares 27.70 378.73 406.43 Std. error Bartlett's prob. .669 .647 .556 .441 Mean F-ratio F-probability squares 13.85 2.23 .116 6.21 Table 11.--One—way analysis of variance F-test comparing publishers mean scores on Voice-B with newspaper circulation Circulation N Mean (50,000 14 7.929 50,000- 200,000 24 8.000 >200,000 24 8.417 Total 62 8.145 Source of D.F. variance Between groups 2 Within groups 59 Total 61 Std. dev. 2.336 2.167 2.636 Sum of squares 2.93 338.76 341.69 Bartlett's prob. .645 Std. error .624 .442 .538 Mean F-ratio F-probability squares 1.47 5.74 .25 .776 126 TABLE 12.--One-way analysis of variance F-test comparing publishers' mean scores on Integrity-A with newspaper circulation L..— Circulation (50,000 50,000- 200,000 )200,000 Total Source of variance Between groups Within groups Total N 14 24 25 63 D.F. 2 60 62 Mean 16.714 18.333 17.200 6.899 6.819 5.598 17.524 Sum of squares 27.52 2,440.19 2,467.71 Std. dev. Std. error Bartlett's prob. 1.844 .575 1.392 1.120 Mean F-ratio F-prob. squares 13.76 .34 .714 40.67 TABLE l3.--0ne-way analysis of variance F-test comparing publishers' mean scores on Integrity-B with newspaper circulation k Circulation (50,000 50.000- 200,000 7200.000 Total Source of variance Between groups Within groups Total N 14 23 25 62 D.F. 59 61 Mean 21.643 6.160 20.522 21.920 21.339 5.401 5.204 Sum of squares 25.09 1,784.79 1,809.88 Std. dev. Std. error Bartlett's prob. 1.646 .777 1.126 1.041 Mean F-ratio F-prob. squares 12.55 .42 .662 30.25 127 TABLE 14.--One-way analysis of variance F-test comparing publishers' mean scores on Privilege-A with newspaper circulation Circulation N Mean Std. dev. Std. error Bartlett's prob. (50,000 13 20.000 7.326 2.032 .780 50,000- 200,000 22 18.909 8.530 1.818 >200,000 23 19.783 8.780 1.831 Total 58 19.500 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 2 12.77 6.38 .09 .913 Within groups 55 3,867.73 70.32 Total 57 3,880.50 TABLE 15.--One-way analysis of variance F-test comparing publishers' mean scores on Privilege-B with newspaper circulation Circulation N Mean Std. dev. Std. error Bartlett's prob. (50,000 13 27.692 6.356 1.763 .269 50,000- 200,000 22 28.318 4.325 .922 >200,000 23 28.783 5.760 1.201 Total 58 28.362 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 2 9.94 4.97 .17 .844 Within groups 55 1,607.46 29.23 Total 57 1,617.40 128 Test of Hypotheses 5A - 5C Null Hypotheses: There is no difference in mean scores on contractual (5A) Voice (58) Integrity (5C) Privilege among publishers who hold contracts dating from December 1933 through August 1945 (32-44 years); from September 1945 through August 1965 (12-31 years); and from September 1965 through December 1976 (less than 12 years). Alternative Hypotheses: The mean scores of publishers who hold contracts dating from December 1933 through August 1945 will be higher on contractual (5A) Voice (SB) Integrity (5C) Privilege than those of publishers with contracts dating from September 1945 through August 1965 and from September 1965 through December 1976. Hypotheses 5A, 58 and 5C examine whether the number of years a publisher has negotiated with the Guild influences his attitude toward incorporating newsroom democracy in contracts. Publishers were divided into three categories: those whose first contract dates from 1933 to 1945 (32 years or more), those with first contract dates between 1945 and 1965 (12 through 31 years), and those with first contract dates from 1965 through 1976 (less than 12 years). One-way analysis of variance was used to test the relation- ship. The results are presented in Tables 16 through 21. 129 TABLE l6.--One-way analysis of variance F-test comparing publishers' mean scores on Voice-A with the number of years a contract has been in effect First N Mean Std. dev. Std. error Bartlett's prob. contract (12 years 5 8.200 2.168 .970 .951 12-31 years 33 5.515 2.438 .424 332 years 36 4.833 2.455 .409 Total 74 5.365 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 2 51.11 25.55 4.32 .017 Within groups 71 420.04 5.92 Total 73 471.15 TABLE l7.--One-way analysis of variance F-test comparing publishers' mean scores on Voice-B with the number of years a contract has been in effect First N Mean Std. dev. Std. error Bartlett's prob. contract <12 years 5 8.200 1.643 .735 .530 12-31 years 33 8.030 2.270 .395 332 years 34 8.471 2.549 .437 Total 72 8.250 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 2 3.26 1.63 .29 .751 Within groups 69 390.24 5.66 Total 71 393.50 130 TABLE l8.--One-way analysis of variance F-test comparing publishers' mean scores on Integrity-A with the number of years a contract has been in effect First contract (12 years 12-31 years 232 years Total Source of variance Between groups Within groups Total N 32 35 73 Mean 20.500 17.719 16.829 17.520 D.F. 2 7O 2, 72 2, Std. dev. Std. error Bartlett's prob. 6.285 2.566 .997 6.227 1.101 6.152 1.040 Sum of Mean F-ratio F-prob. squares squares 71.28 35.64 .928 .400 686.94 38.38 758.22 TABLE 19.--One-way analysis of variance F-test comparing publishers' mean scores on Integrity-B with the number of years a contract has been in effect First contract (12 years 12-31 years 332 years Total Source of variance Between groups Within groups Total N 33 71 Mean 20.800 20.636 22.364 21.451 D.F. 2 68 1, 70 1, Std. dev. Std. error Bartlett's prob. 3.633 1.625 .441 5.776 1.006 4.930 .858 Sum of Mean F-ratio F-prob. squares squares 51.50 25.75 .923 .402 898.07 27.91 949.57 131 TABLE 20.--One-way analysis of variance F-test comparing publishers' mean scores on Privilege-A with the number of years a contract has been in effect First contract (12 years 12-31 years 232 years Total Source of variance Between groups Within groups Total 29 33 67 Mean 23.600 19.069 18.788 5.504 7.833 8.813 19.269 D.F. Sum of squares 2 102.59 64 4,324.58 66 4,427.17 Std. dev. Std. error Bartlett's prob. 2.462 .501 1.454 1.534 Mean F-ratio F-prob. squares 51.29 .759 .472 67.57 TABLE 21.--0ne-way analysis of variance F-test comparing publishers' mean scores on Privilege-B with the number of years a contract has been in effect First contract (12 years 12-31 years 332 years Total Source of variance Between groups Within groups Total Mean 27.000 28.621 28.750 3.317 5.301 5.553 28.561 D.F. Sum of squares 2 13.43 63 1,786.83 65 1,800.26 Std. dev. Std. error Bartlett's prob. 1.483 .501 .984 .982 Mean F-ratio F-prob. squares 6.72 .237 .790 28.36 132 The data do not support rejection of Null Hypotheses 5A - 5C. Although the F-probability for Voice-A is significant, the relation- ship is the reverse of the predicted direction. Publishers with the longest-standing contracts were the most negative toward incorporating Voice provisions in contracts, and those with the newest contracts were the most favorable toward incorporating Voice. When the Voice-B scores were compared, however, publishers exhibited no difference in attitudes toward writing into contracts protections against en- croachment of their decision-making functions. For Integrity and Privilege clauses there also appears to be no relationship between the length of time a contract with the Guild has been in effect and a publisher's willingness to include these newsroom democracy provisions. Hypotheses 5A through 5C were based in part on data from existing contracts (see Table B-2, Appendix B) which show that the oldest contracts do contain the greatest percentage of Voice, Integrity and Privilege clauses, although the relationships are not statistically significant. Test of Hypotheses 6A - 6C Null Hypotheses: There is no difference in mean scores on contractual (6A) Voice (68) Integrity (60) Privilege among publishers who consider their relationship with The News- paper Guild to be excellent or good and those who consider it to be fair or poor. 133 Alternative Hypotheses: The mean scores of publishers who rate their relationship with The Newspaper Guild as excellent or good will be higher on contractual (A; Voice 8 Integrity (C) Privilege than those of publishers who rate their relationship as fair or poor. These hypotheses were developed from the assumption that publishers who had a more favorable relationship with the Guild would be more accepting of the Guild's proposals in the non-economic, management-rights areas of newsroom democracy. As with Hypotheses 4 and 5, one-way analysis of variance was used to test the relationship. The results are presented in Tables 22 through 27. None of the F-probabilities are significant, indicating that Null Hypotheses 6A, 6B and 6C cannot be rejected. There appears to be no association between the publisher's relationship with the Guild and his attitude toward Guild newsroom democracy bargaining provisions. 134 TABLE 22.—-0ne-way analysis of variance F-test comparing publishers' mean scores on Voice-A with their relationship with The Newspaper Guild Relationship Excellent/ good Fair/poor Total Source of variance Between groups Within groups Total Mean 5.556 5.409 5.500 D.F. 56 57 Std. dev. 2.720 2.404 Sum of squares .30 380.20 380.50 Std. error Bartlett's prob. .453 .535 .512 Mean F-ratio F-prob. squares .30 .043 .836 6.80 TABLE 23.--0ne-way analysis of variance F-test comparing publishers' mean scores on Voice-B with their relationship with The Newspaper Guild Relationship Excellent/ good Fair/poor Total Source of variance Between groups Within groups Total Mean 8.228 8.450 8.339 D.F. 54 55 Std. dev. 2.491 2.038 Sum of squares .38 296.17 296.55 Std. error Bartlett's prob. .415 .336 .456 Mean F-ratio F-prob. squares .38 .070 .793 5.48 135 TABLE 24.--One-way analysis of variance F-test comparing publishers' mean scores on Integrity-A with their relationship with The Newspaper Guild Relationship N Mean Std. dev. Std. error Bartlett's prob. Excellent/ good 34 17.500 5.775 .990 .280 Fair/poor 23 16.956 7.125 1.486 Total 57 17.281 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 1 4.05 4.05 .101 .752 Within groups 55 2,217.46 40.32 Total 56 2,221.51 TABLE 25.--One-way analysis of variance F-test comparing publishers' mean scores on Integrity-B with their relationship with The Newspaper Guild Relationship N Mean Std. dev. Std. error Bartlett's prob. Excellent/ good 33 21.394 5.500 .957 .614 Fair/poor 22 20.773 4.966 1.059 Total 55 21.146 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 1 5.09 5.09 .182 .672 Within groups 53 1,485.74 28.03 Total 54 1,490.83 136 TABLE 26.--0ne-way analysis of variance F-test comparing publishers' mean scores on Privilege-A with their relationship with The Newspaper Guild .____ Relationship N Mean Std. dev. Std. error Bartlett's prob. Excellent/ good 31 19.194 7.989 1.435 .801 Fair/poor 21 19.714 8.415 1.836 Total 52 19.404 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 1 3.39 3.39 .051 .822 Within groups 50 3,331.12 66.62 Total 51 3,334.51 TABLE 27.--One-way analysis of variance F-test comparing publishers' mean scores on Privilege-B with their relationship with The Newspaper Guild Relationship N Excellent/ good 29 Fair/poor 22 Total 51 Source of variance Between groups Within groups Total D.F. 1 49 50 Mean 28.517 27.318 28.000 Sum of squares 17.99 1,354.01 1,372.00 Std. dev. 5.096 5.463 Std. error Bartlett's prob. .946 .735 1.165 Mean F-ratio F-prob. squares 17.99 .651 .424 27.63 137 Null Hypotheses: There is no difference in mean scores on contractual (A) Voice (8) Integrity (C) Privilege between publishers who are affiliated with a group and those who are not. Alternative Hypothesis 7A: The mean score of publishers of independent newspapers will be higher on contractual Voice than that of publishers who are members of groups. Alternative Hypothesis 7B: The mean score of publishers who are members of a group will be higher on contractual Integrity than that of independent pub- lishers. Alternative Hypothesis 7C: The mean score of publishers who are members of a group will be higher on contractual Privilege than that of independent pub- lishers. Hypotheses 7A, 7B and 7C were based on the assumption that publishers of independent newspapers would have more freedom to negotiate on management rights areas, specifically Voice, than would those who were members of groups. The direction of the hypotheses was also based on data from existing contracts (see Table B-4, Appendix B). One-way analysis of variance was used to test these hypotheses. The results are given in Tables 28 through 33. 138 TABLE 28.-~0ne-way analysis of variance F-test comparing publishers' mean scores on Voice-A with their independent or affiliated status Status N 33 37 Group member Independent Total 70 Source of variance Between groups Within groups Total Mean 5.061 5.649 5.371 D.F. 68 69 Std. dev. 2.645 2.541 Sum of squares 6.03 456.31 462.34 Std. error Bartlett's prob. .460 .817 .418 Mean F-ratio F-prob. squares 6.03 .899 .346 6.71 TABLE 29.--0ne-way analysis of variance F-test comparing publishers' mean scores on Voice-A with their independent or affiliated status Status N 32 36 Group member Independent Total 68 Source of variance Between groups Within groups Total Mean 7.688 8.694 8.221 D.F. 66 67 Std. dev. 2.416 2.227 Sum of squares 17.18 354.51 371.69 Std. error Bartlett's prob. .427 .644 .371 Mean F-ratio F-prob. squares 17.18 3.20 .078 5.37 139 TABLE 30.--One-way analysis of variance F-test comparing publishers' mean scores on Integrity-A with their independent or affiliated status Status N Mean Std. dev. Std. error Bartlett's prob. Group member 33 16.848 6.568 1.143 .537 Independent 36 18.306 5.898 .983 Total 69 17.609 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 1 36.55 36.55 .943 .335 Within groups 67 2,597.88 38.77 Total 68 2,634.43 TABLE 31.-~0ne-way analysis of variance F-test comparing publishers' mean scores on Integrity-B with their independent or affiliated status Status N Mean Std. dev. Std. error Bartlett's prob. Group member 34 21.324 5.492 .942 .794 Independent 33 21.576 5.244 .913 Total 67 21.448 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 1 1.06 1.06 .037 .848 Within groups 65 1,875.50 28.85 Total 66 1,876.56 140 TABLE 32.--0ne-way analysis of variance F-test comparing publishers' mean scores on Privilege-A with their independent or affiliated status Status N Mean Group member 32 18.781 Independent 32 19.625 Total 64 19.203 Source of D.F. Sum of variance squares Between groups 1 11.39 Within groups 62 4,166.97 Total 63 4,178.36 Std. dev. 7.478 8.860 Std. error Bartlett's prob. 1.322 .350 1.566 Mean F-ratio F-prob. squares 11.39 .169 .682 67.21 TABLE 33.--One-way analysis of variance F-test comparing publishers' mean scores on Privilege-B with their independent or affiliated status Status N Group member 31 Independent 32 Total 63 Source of variance Between groups Within groups Total D.F. 62 Mean 28.936 27.844 28.381 Sum of squares 18.77 1,716.09 1,734.86 Std. dev. 4.760 5.782 Std. error Bartlett's prob. .855 .289 1.022 Mean F-ratio F-prob. squares 18.77 .667 .417 28.13 141 The data show no significant differences between independent and group-affiliated publishers on any of the contractual categories. Hence, Null Hypotheses 7A, 7B and 7C cannot be rejected. In the Voice-B category the F-probability is .078, suggesting a marginal relationship between affiliation and attitude toward incorporating management-protective Voice clauses in contracts. Group-affiliated publishers were more inclined to favor the reverse of the Guild Voice provisions than were independent publishers, but the data when viewed with the Voice-A results offer no statistical rationale for accepting Alternative Hypothesis 7A. The data from existing contracts (see Table 8-4, Appendix 8) reveal that independent newspapers were far more likely to have Voice clauses in contracts than were group publications. The asso- ciations between affiliation and the inclusion of Integrity and Privilege clauses, however, were not significant. Test of Hypotheses 8A - 8C Null Hypotheses: There is no difference in mean scores on contractual (8A) Voice (88) Integrity (8C) Privilege among editorial employes who rate their working relationship with their employer as excellent, good, fair or poor. Alternative Hypotheses: Editorial employes who rate their relationship with their employer as excellent will have lower mean scores on contractual (8A Voice (8B Integrity (8C) Privilege 142 than editorial employes who rate their relationship with their employer as good, fair or poor. Hypotheses 8A, 88 and 8C were postulated in the light of studies which showed that employes who are satisfied with their work- ing conditions and their relationship with their employer are less likely to seek worker participation rights than those who are dis- content. One-way analysis of variance was used to test these hypotheses. The results are presented in Tables 34, 35 and 36. For Hypotheses 8, 9 and 10 the items comprising the Voice, Integrity and Privilege scores were not separated into “A" and "8" categories, since all the inter-item correlations for these pairs were significant for editorial employes (see Tables B-ll through B-15, Appendix 8). TABLE 34.--One-way analysis of variance F-test comparing editorial employes' mean scores on Voice with their relationship with their employer Relationship N Mean Std. dev. Std. error Bartlett's prob. Excellent 82 19.842 4.256 .470 .645 Good 164 19.976 4.476 .349 Fair' 56 20.804 4.845 .647 Poor 22 22.318 5.037 1.074 Total 324 20.244 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 3 137.28 45.76 2.233 .084 Within groups 320 6,556.45 20.49 Total 323 6,693.73 143 TABLE 35.--One-way analysis of variance F-test comparing editorial employes' mean scores on Integrity with their relationship with their employer Relationship N Mean Std. dev. Std. error Bartlett's prob. Excellent 71 53.183 7.451 .884 .937 Good 154 53.701 7.105 .573 Fair 52 56.077 7.436 1.031 Poor 22 55.864 7.668 1.635 Total 299 54.150 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 3 355.06 118.35 2.229 .085 Within groups 295 15,665.16 53.10 Total 298 16,020.22 TABLE 36.--One-way analysis of variance F-test comparing editorial employes' mean scores on Privilege with their relationship with their employer g Relationship N Mean Std. dev. Std. error Bartlett's prob. Excellent 67 65.269 8.812 1.077 .645 Good 148 65.182 8.318 .684 Fair 56 66.696 9.309 1.244 Poor 22 66.727 7.629 1.627 Total 293 65.608 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 3 128.42 42.81 .581 .628 Within groups 289 21,283.44 73.65 Total 292 21,411.86 144 The results show F-probabilities of .084, .085 and .628 for Voice, Integrity and Privilege respectively. This means that Null Hypotheses 8A, 88 and 8C cannot be rejected. It is likely, however, that some association along the lines hypothesized does exist for Voice and Integrity. Those employes with an excellent or good relationship with the employer tend to favor contractual Voice and Integrity less than those employes with a fair or poor relationship. Test of Hypotheses 9A - 9C Null Hypotheses: There is no difference in mean scores on contractual (9A) Voice (9B) Integrity (9C) Privilege among editorial employes who differ in their attitudes toward and support of The Newspaper Guild. Alternative Hypotheses: The mean scores of editorial employes who are officers of The Newspaper Guild or who are active and supportive of the union will be higher on contractual (9A) Voice (9B) Integrity (9C) Privilege than those of employes who are indifferent toward the union or who would prefer not to be members. Hypotheses 9A, 9B and 9C suggest that those who are supportive of their union will also be supportive of the provisions the union is attempting to secure. One-way analysis of variance was used to test these hypotheses. The results are presented in Tables 37, 38 and 39. 145 TABLE 37.--One-way analysis of variance F-test comparing editorial employes' mean scores on Voice with their relationship with The Newspaper Guild Relationship N Mean Std. dev. Std. error Bartlett's prob. Officer 37 22.297 5.021 .825 .123 Active and supportive 62 22.048 4.452 .565 Supportive 157 20.026 4.211 .336 Indifferent 55 17.836 3.479 .469 Prefer not to be member 13 18.923 5.171 1.434 Total 324 20.256 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 4 706.80 176.70 9.624 .000 Within groups 319 5,856.93 18.36 Total 323 6,563.74 TABLE 38.—-One-way analysis of variance F-test comparing editorial employes' mean scores on Integrity with their relationship with The Newspaper Guild Relationship N Mean Std. dev. Std. error Bartlett's prob. Officer 34 56.471 8.073 1.384 .390 Active and supportive 57 56.333 7.251 .960 Supportive 148 53.554 6.875 .565 Indifferent 48 51.500 6.703 .968 Prefer not to be member 12 54.333 9.576 2.764 Total 299 54.117 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 4 844.53 211.13 4.093 .003 Within groups 294 15,164.37 51.58 Total 298 16,008.90 146 TABLE 39.--One-way analysis of variance F-test comparing editorial employes' mean scores on Privilege with their relationship with The Newspaper Guild Relationship N Mean Std. dev. Std. error Bartlett's prob. Officer 34 68.147 9.848 1.689 .353 Active and Supportive 58 67.000 8.958 1.176 Supportive 142 65.866 7.756 .651 Indifferent 49 62.082 8.236 1.176 Prefer not to be member 12 62.000 9.545 2.752 Total 295 65.566 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 4 1,106.07 276.52 3.909 .004 Within groups 290 20,512.40 70.73 Total 294 21,618.47 The differences among groups are significant on the issues of Voice, Integrity and Privilege, and therefore, Null Hypotheses 9A, 9B and 90 can be rejected. A clear association does exist between an employe's relationship with the Gui1d and his attitudes toward Gui1d bargaining in the areas of Voice, Integrity and Privilege. Two anomalies appear in the mean scores, however: (1) those who ranked themselves as "indifferent" toward the Guild had the lowest mean scores on Voice and Integrity and (2) those in the "would prefer not to be a member" category scored higher on Inte- grity than either those who ranked themselves as "indifferent" or "supportive." One possible explanation for these deviations may be that emp1oyes who are indifferent toward the Guild are more indifferent 147 toward those protections offered by the Guild bargaining program than employes who would prefer not to belong to the union. There is no reason that employes who do not wish to be members of the Guild (or perhaps members of any union) would necessarily be opposed to the rights that the union is trying to secure for its members. Test of Hypotheses 10A - 10C Null Hypotheses: There is no difference in mean scores on contractual (10A) Voice (108) Integrity (10C) Privilege between editorial employes who have less than ten years' experience as journalists and those who have more than ten years' experience. Alternative Hypotheses: The mean scores of editorial employes with less than ten years' experience as journalists will be higher on contractual (10A) Voice (108) Integrity (lOC) Privilege than those of editorial employes with more than ten years' experience as journalists. Hypotheses 10A, 108 and 10C suggest that less experienced employes are more eager for change within the newsroom than their veteran counterparts. One-way analysis of variance was used to test these relation- ships. The results are presented in Tables 40, 41 and 42. 148 TABLE 40.--One-way analysis of variance F-test comparing editorial employes' mean scores on Voice with the journalist number of years employed as a Journalist years <10 310 Total Source of variance Between groups Within groups Total Mean Std. dev. 115 203 20.835 19.970 4.460 4.581 318 20.283 D.F. Sum of squares 1 316 54.84 6,507.68 317 6,562.52 Std. error Bartlett's prob. .416 .747 .322 Mean F-ratio F-prob. squares 54.84 2.663 .104 20.59 TABLE 41.--One-way analysis of variance F-test comparing editorial employes' mean scores on Integrity with the number of years employed as a journalist Journalist years <10 110 Total Source of variance Between groups Within groups Total N Mean Std. dev. 109 182 54.018 54.418 6.997 7.529 291 54.268 D.F. Sum of squares 1 289 10.86 15,548.23 290 15,559.09 Std. error Bartlett's prob. .670 .398 .558 Mean F-ratio F-prob. squares 10.86 .202 .653 53.80 149 TABLE 42.--One-way analysis of variance F-test comparing editorial employes' mean scores on Privilege with the number of years employed as a journalist Journalist N Mean Std. dev. Std. error Bartlett's prob. years <10 108 65.870 8.386 .807 .665 310 178 65.303 8.708 .653 Total 286 65.518 Source of D.F. Sum of Mean F-ratio F-prob. variance squares squares Between groups 1 21.61 21.61 .293 .589 Within groups 284 20,945.80 73.75 Total 285 20,967.41 As the tables show, the differences between the two groups are not significant, and hence, Null Hypotheses 10A, 108 and 100 cannot be rejected. However, it appears that some association along the lines hypothesized may exist. Less experienced editorial employes are more eager to incorporate Voice rights in contracts than are their veteran co-workers (F-probability for Voice = .104). The mean scores for novice and veteran editorial employes on Integrity and Privilege issues, however, appear to be approximately the same. Summary Of the stated hypotheses, the data support acceptance of the following alternative hypotheses: 1A through 16, 11 through 1R; 2A, 28, 20, 2E, 2F, 26, 2H, 21, 2K, 2M, 2N, 2-0, 20 and 2R; 3A, 38, 30, 30, 3F, 3G, 31, 3J, 3K, 3L, 3M, 3M, 3-0, 3P and 30; and 9A, 9B and 90. 150 Hypotheses l, 2 and 3 deal with the relationship between publishers and editorial employes and their attitudes toward newsroom democracy in general terms and as contract provisions. The results suggest that publishers are less accepting than their employes of increased employe voice in newsroom policy and decision making. They are willing to grant newsroom democracy rights to their employes-- whether stipulated in contracts or not--but not to the degree sought by The Newspaper Guild in its Model Contract bargaining language. Editorial employes, although more favorable than publishers in their acceptance of newsroom democracy, are not totally in favor of the Guild's proposals. The differences in attitudes between the two groups are significant in all but one general area: printing or writing false or misleading information. This exception suggests that perhaps a common ground exists for agreement between publishers and employes: the ethics of the profession. The remaining hypotheses (with the exception of Hypothesis 9) were not confirmed by the data. These hypotheses deal with character- istics of the publication (circulation, length of time a contract has been in effect, group affiliation), the publishers (attitudes toward the Guild), and the employes (attitudes toward the employer and length of time employed as a journalist). None of these character- istics appear to affect significantly either group's attitudes toward the Guild's contract provisions on newsroom democracy. The only editorial employe characteristic tested which does appear to affect reporters' attitudes toward newsroom democracy is the employe's relationship with The Newspaper Guild. CHAPTER V SUMMARY AND CONCLUSIONS At the beginning of Chapter IV the main questions to be answered by the study were stated: (1) Are publishers willing to formalize newsroom democracy through contractual agreements? (2) Are editorial employes in favor of such action? This study examined a new area of collective bargaining, one that defines the rights of employer and employe in the newsroom. It is one that determines who will make policy and under what circumstances. The significance of bargaining on the issues of Voice, Integrity and Privilege cannot be understated. Incorporation of these issues in contracts redefines, and in most instances restricts, management rights' clauses. They are issues dealing with a new and growing concern among white collar workers for greater participation in policy making. The answers to the main questions posed in Chapter IV and restated above have meaning only if attitudes toward newsroom democracy outside the context of contractual agreements are examined first. Hypotheses 1 - 3 Hypotheses 1A through 1R looked at the question of attitudes toward the general concepts of newsroom democracy. The results, 151 152 presented in Table 7, showed that editorial employes had significantly more favorable attitudes toward newsroom democracy than did publishers on all but IG-5, publishing false or distorted information. In eight of the eighteen newsroom democracy statements favorable to the employe, publishers had low attitude scores (less than 3), while in all but one case (VG-2A1) editorial employes had high attitude scores (greater than 3, 3 being the neutral point on the scale). The eight low scores for publishers (VG-1A, VG-2Al, VG-3A, IG-lA, IG-2A, IG-7A, PG-2A and PG-8A) show some common threads: First, all of the Voice statements--input in newsroom opera- tions policy, consultation on appointment of supervisory personnel and representation on editorial boards--were rejected by publishers as philosophically undesirable. These three issues are the crux of employe involvement in the operation of the newspaper. It appears that irrespective of a contract, publishers are opposed to a graying of the line between management and employe input into policy making. Second, the three Integrity issues which brought negative responses from publishers all deal with the relinquishing of manage- ment's right to determine employe conduct. The issues would place the responsibility for determining what is professionally ethical (IG-lA, IG-2A) and what is acceptable employe production (IG-7A) in the hands of the employes themselves; i.e., employe self-evaluation. Such issues, the results show, apparently are philosophically unpalatable for publishers. Third, the two unacceptable Privilege issues (PG-2A and PG-8A) are similar to the negated Integrity issues in that they provide for employe self-evaluation. But they have an additional 153 aspect which may contribute further to publisher distaste: they are sweeping in their coverage and offer the publisher no protection against poor employe judgment. Such a situation was suggested by one publisher respondent: There is no guarantee that every manager will be Solomonic-— or even immune to mental illness--but there is a virtual, statis- tical guarantee that in any large group of employes over a period of time there will arise certain other-than-normal aberrations including isolated cases of mental breakdown or erratic behavior. Now, there is no question about the ultimate responsibility for misconduct by an agent, as an agent, either in libel or other court action. Experience makes it clear that the standard one must deal with is that both the law and the damage-seeker go after the "company" where the money is. In sum, the list of propositions offered would make newsmen all free agents--and then extend the responsibility for them as NOT A FREE AGENT but as an agent of the newspaper management to absolute and arbitrary guarantees regarding their conduct.... One of a couple hundred employes whose wife has left him and who has been under explosive mental pressure decides to leave town--but first, Gotterdammerung! He can write any story he dreams of? Refuse to identify the "confidential sources" (which may not even exist)? An oblige to pay his bills, if any, as well as the damage liability?* How do employe attitudes differ from those of publishers? Responses from editorial employes were high on every item except VG-2Al: the employer should consult with the union's representatives prior to selection of management personnel in the newsroom. And they were only slightly more in favor of management consulting with employes prior to the selection of supervisory personnel. This same item drew the strongest negative response from publishers as well. It appears that both editorial employes and publishers feel that *Refer to Appendix F for additional publisher comments on the nature and acceptability of newsroom democracy. 154 decisions regarding selection of managerial positions in the newsroom ought to remain a management right. Hypotheses 2A through 2R examine the difference between publisher and editorial employe attitudes on incorporating newsroom democracy in contracts. All but four of the eighteen null hypotheses were rejected, indicating a significantly more favorable attitude among editorial employes than publishers toward incorporating newsroom democracy in contracts. Those items which showed no difference between publisher and employe attitudes were VC-3B, IC-7B, PC-28 and PC-BB (Hypotheses 2C, 2J, 2L and 2P respectively). The lack of significant difference exists in the comparisons of the “B" statements. As was suggested previously, part of the reason for the higher-than-hypothesized scores for publishers on "8" items may be due to publisher preference for omitting from the contract any such newsroom democracy language, whether it protects the employer or the employe. Hypotheses 3A through 3R examined the difference between publishers and editorial employes in the correlation of their general attitudes toward newsroom democracy and their attitudes toward contract provisions on newsroom democracy. It was hypothesized that employes' scores on contractual and general principles would be more highly correlated; that is, the corresponding scores would more closely agree than would publishers' contractual and general princi- ple scores. Fifteen of the eighteen alternative hypotheses were accepted. Null Hypotheses 3E, 3H and 3R could not be rejected, although the differences between editorial employes and publishers approached 155 a significance level of .05 (N1 3E and 3H (corresponding to items IG-ZB/IC-ZB and IG-SAl/IC-SAl respectively). For PG-8B/PC-88 the correlation on the general and contract statements was higher for publishers than for editorial employes. The results (If the test of Hypothesis 3 suggest that in general publishers are more reluctant than employes to make absolute and contractually binding issues with which they personally or pro- fessionally concur. This suggestion is borne out by several comments received from publisher respondents: ...My technical answers to the questions would uniformly be to preserve the rights of management in all areas in which you have made inquiry and that I would be opposed to putting in a contract with the union any provisions relating to the subjects outlined in your questionnaire. Should management negotiate all assignments, instructions, and situations with the staff in each instance? That question or all of its endless variations has been put to the management of ... many times down through the years....We have never been tempted, in any way, to turn our newsroom into an arena for 1awyers--real or inmginary. I like to deal in standard operating procedures that make sense among people who have mutual respect, but from which management and reporters alike can deviate when unusual or emergency conditions are apparent to all without falling under a legalized absolute.... Regimentation--even to seemingly sound general principles-- ends up in a jungle of refinements and regulations that stifle us all, management and union members alike. Not every good idea should be ensconced in absolutist and authoritarian form by either law or contract. Flexibility and case-by-case judgment are keys to quality. Iron rules and helter-skelter independence enforced by some will produce a weird end product...designed for self-gratifica- tion instead of for public service. 156 The Guild seeks contract language to cover all possible cases, but to reduce any policy to contract language introduces diffi— culties of definition because we are the party primarily respon- sible for the integrity of the newspaper and also we are the payer of the bills and we must look at all the obligations within the four walls of the contract. We are required to look at all of the newsmen as one anonymous body and not as individ- uals. We must consider that contract language would cover employes to be hired in the future whom we have not yet met. It would be best if as little as possible is included in a contract, with a strong management rights clause giving management the right to manage as they judge best. The key issues in the reluctance of publishers to incorporate newsroom democracy in contracts appear to be (1) restrictions on management's right to manage, (2) liability for employe actions, (3) absolute statements that do not allow for individual differences or unusual circumstances, and (4) vague language that may lead to arbi- tration or litigation. Until publishers are convinced that these factors can be satisfactorily resolved within the confines of the collective bargaining agreement, newsroom democracy will remain narrowly defined in the contract or will be practiced exclusive of the contract at the discretion of management. Editorial employes, who do not face the same contract re- strictions as publishers, are free to view newsroom democracy as both a basic right and as a contractual right. For an employe to overlook or disregard a contract provision giving him the right to determine the ethical nature of an assignment or the right to refuse to disclose his source would cause no legal clamor. Yet if an employer disre- garded one of these written provisions, the possibility of a grievance or arbitration would loom on the horizon. In many areas of a contract the responsibilities assumed by labor and management are equal. In 157 newsroom democracy the burden falls on management. And unless manage- ment can see some positive value in accepting this new burden, it is likely to continue to reject proposals in this area. Such a situation may lead to one of two changes: (1) the employe likewise accepts a burden in terms of his newsroom democracy freedoms, or (2) the burden imposed on management is lightened. Either or both are possible negotiation strategies. Hypotheses 4 - 10 Attempts to define the publisher or publication more likely to incorporate newsroom democracy in a contract proved unsuccessful. Hypotheses 4 through 7 examined the possibility of a relationship between the degree to which publishers agreed with contractual news- room democracy and (l) the size of their publications, (2) the number of years the publication's employes had been represented by the Guild, (3) the publisher's relationship with the Guild, and (4) the affiliated or independent nature of the publication. None of these characteristics proved to have a significant bearing on publishers' concurrence with or opposition to contractual newsroom democracy. In Hypotheses 8 through 10 three employe characteristics were examined for possible association with their views on newsroom democracy: (1) the editorial employe's relationship with his employer, (2) the employe's relationship with the Guild, and (3) the number of years the employe had worked as a journalist. It was shown (see Tables 37, 38 and 39) that the employe's relationship with the Guild is closely tied with his attitudes toward 158 contractual newsroom democracy, confirming Alternative Hypothesis 9. The more active and supportive of the union an employe is, the more he agrees with the Guild's newsroom democracy proposals. According to the survey findings, neither the employe's relationship with his employer (Hypothesis 8) nor the length of time he has worked as a journalist (Hypothesis 10) affects his attitudes toward contractual newsroom democracy. For both publishers and editorial employes the question of incorporating newsroom democracy in a contract appears to transcend the characteristics examined (with the exception of the employe-Guild relationship). Among publishers one possible reason for the lack of association is that contract language has the potential to affect equally the small or large publication. There are no special dispen- sations from grievance arbitration for the smaller publisher nor for the one affiliated with a group. Even though the Guild has had more opportunities to make inroads in securing acceptance of its pr0posals with publishers who hold long-standing contracts, most newsroom democracy issues arose after 1970. They, therefore, have been placed on the table of long- standing Guild-represented publications no more frequently than they have with newly Gui1d-represented publications. The issue of whether a publisher has a good or poor relation- ship with the Guild does not seem to influence his attitude toward signing a contract containing newsroom democracy clauses. A good relationship, it appears, does not necessarily imply a concurring one. Nor do good employer-employe relationships necessarily imply a lessening of employe interest in securing newsroom democracy in a 159 contract. The data do not support the notion that it is the militant, dissatisfied faction of editorial employes who are desirous of newsroom democracy. Similarly, it is not a less experienced faction that is seeking newsroom democracy. Publishers and Guild members alike may find it unusual that newsroom democracy has transcended all the variables tested in this study with the exception of the employe's relationship with the union. The latter association is understandable in that the more involved in union activities the employe is, the more he has been exposed to Guild policies and bargaining proposals. And with greater exposure may come greater acceptance as one's own the bargaining stand of the union. Limitations of the Study, Size Although the return of usable questionnaires was greater than 50 per cent for both publishers and editorial employes, the number of returns was small: 90 publishers and 352 editorial employes. This was due to the nature of the universe from which the samples were drawn. These numbers should be kept in mind when gener- alizing from the study. Scape The questionnaires were sent only to publishers who hold contracts with the Guild and to Guild-represented editorial employes. The study does not assess the attitudes of publishers and reporters who are not represented by the union. Non-union publishers and 160 reporters may have very different attitudes toward the same issues. Definition The main limitation is the definition of newsroom democracy. It is clearly defined in Chapter I as The Newspaper Guild's Model Contract language on Voice, Integrity and Privilege. Measuring Instrument 11w: pretest questionnaires were refined to include only two measures on each of the eighteen newsroom democracy issues. A major flaw was introduced in the measuring instrument when an attempt was made to use positive-negative item pairs for the eighteen indices. For editorial employes both the general and the contract item pairs correlated significantly. Yet for publishers fifteen of the eighteen contract item pairs either did not correlate significantly or corre- lated negatively. The decision to attempt positive-negative pairs was based on a faulty assumption that publishers would be as favorable toward contract items that guaranteed their rights as they would be opposed toward contract items that guaranteed editorial employe rights. If the assumption had been valid, reversing the negative "8" scores would have resulted in significantly correlated scores on the item pairs. A second rationale for preparing the item pairs as opposites was to avoid the appearance of a pro—editorial employe survey by having all positively constructed statements. The lack of correlation among publishers necessitated the evaluation of part of the data in separate "A" (positive) and "B" 161 (negative) categories. This resulted in the loss of a reliability- validity check on publisher contract scores, since comparative measures were eliminated. The publishers' composite "A" and "B" indices for all but three of the eighteen contract provisions had to be eliminated. A larger publisher' pretest sample might have caught this design error. Only four usable publisher pretests were returned, a number too small from which to draw meaningful interpretations. The use of only two items per index was in itself a weakness of the study. The decision to use two-item indices was prompted by pretest responses to the excessive length of the questionnaire when three-item indices were used. Increased reliability was sacrified to gain a higher response rate. An alternative would have been to limit the study to one or two of the three newsroom democracy areas of Voice, Integrity and Privilege, a move which would have necessi- tated the redefinition of newsroom democracy. Generalizability The results of this study should not be generalized to a different definition of newsroom democracy or to publishers or edi- torial employes who are not associated with the Guild. This does not mean, however, that the results have no applicability in other settings. The findings in this study may suggest other areas of employer-employe attitude assessment both within and outside the newspaper industry. 162 Conclusions A brief summary of the findings is presented below: (1) Editorial employes are more favorable toward newsroom democracy concepts than are publishers (Hypotheses lA-lG, lI-lR). (2) Both editorial employes and publishers agree with the general principle of not requiring journalists to distort facts or knowingly present misleading information (Hypothesis 1H). (3) Editorial employes are more favorable toward incorpor- ating newsroom democracy provisions in contracts than are publishers (Hypotheses 2A, 28, 20, 2E, 2F, 2G, 2H, 21, 2K, 2M, 2N, 2-0, 20, 2R). (4) There is no difference in attitudes between editorial employes and publishers on the inclusion in contracts of issues dealing with (a) representation on editorial boards (Hypothesis 2C), (b) permitting employes to respond to letters to the editor (Hypothe- sis 2J), (c) permitting employes to determine their right not to authenticate material (Hypothesis 2L), and (d) paying legal fees of employes who refuse to reveal their sources (Hypothesis 2P). (5) Editorial employes view corresponding newsroom democracy general principles and contract provisions synonymously while publishers do not (Hypotheses 3A-3D, 3F, 36, 3I-3Q). (6) There is no difference between editorial employe and publisher attitudes toward the general principle and contract provi- sion of (a) not requiring an employe to perform duties other than those of his job description (Hypothesis 3E), (b) not requiring an employe to distort facts or present false information (Hypothesis 3H), and (c) indemnifying an employe who chooses to refuse to 163 disclose his sources irrespective of the action the employe takes (Hypothesis 3R). (7) There is no difference in attitudes toward contractual Voice, Integrity and Privilege clauses among publishers of newspapers with circulations less than 50,000, between 50,000 and 200,000, and over 200,000 (Hypotheses 4A-4C). (8) There is no difference in attitudes toward contractual Voice, Integrity and Privilege clauses among publishers who have held contracts with the Guild since 1933, 1945, or 1965 (Hypotheses SA-SC). (9) There is no difference in attitudes toward contractual Voice, Integrity and Privilege clauses between publishers who consider their relationship with The Newspaper Guild to be excellent or good and those who consider it to be fair or poor (Hypotheses 6A-6C). (10) There is no difference in attitudes toward contractual Voice, Integrity and Privilege clauses between publishers who are affiliated with a group and those who are not (Hypotheses 7A-7C). (11) There is no difference in attitudes toward contractual Voice, Integrity and Privilege clauses among employes who consider their relationship with their employer to be excellent, good, fair or poor (Hypotheses 8A-8C). (12) Editorial employes who are officers of The Newspaper Guild or who are active and supportive of the union have more favor- able attitudes toward contractual Voice, Integrity and Privilege clauses than editorial employes who are indifferent toward the union or who would prefer not to be members (Hypotheses 9A-9C). 164 (13) There is no difference in attitudes toward contractual Voice, Integrity and Privilege clauses between editorial employes who have worked as journalists for less than ten years and those who have been in the profession for ten or more years (Hypotheses lOA-lOC). Suggestions for Future Research This study was exploratory in nature. It is the first attempt to evaluate publisher and employe attitudes toward the non- economic newsroom democracy issues of Voice, Integrity and Privilege as proposed by The Newspaper Guild in its collective bargaining program. Problems encountered in the execution and analysis of the study suggest modifications in design of any similar study to be undertaken in the future. The scope of this study also suggests areas for expansion and the testing of new hypotheses: (l) The question of what rights publishers are willing to include in contracts needs further examination. Any future study, however, should avoid the assumption that publishers will view contract statements guaranteeing their rights as favorably as they will oppose contract statements guaranteeing editorial employe rights. More reliable indices of attitudes toward newsroom democracy are needed. This might be accomplished by increasing the number of items per index and decreasing the number of indices to be examined. For instance, the researcher could conduct three separate surveys, one dealing with Voice, a second with Integrity, and a third with Privilege. The researcher should be cautious of attempting surveys among small populations of publishers and editorial employes. Both 165 groups tend to be reluctant participants in research studies, especially those with a survey format. A sample return of 50 per cent is unusually high. The researcher should be mindful of the low poten- tial return when designing the study. (2) Other measuring instruments need to be developed to assess publisher attitudes. A future study could use personal inter- views with publishers and compare data derived from those interviews with empirical data on collective bargaining sessions and contract settlements. The use of case studies would be another alternative to the mail survey. By this method, variables which could not be predicted or assessed in this investigation could be studied in detail. For example, the researcher could assess the tenor of the management- employe relationship in the newsroom by observing daily operations over a period of time and note how it affects the collective bargain- ing stand taken by management and by employes. He could also compare attitudes toward newsroom democracy among management and employes within the same organization and explore a possible relationship between their attitudes and the socialization process existing within the organization. Third, the researcher could study the daily functioning of newsroom democracy--or the lack of it--within particular organizations and compare actual operations with the contract language. (3) A survey of publishers and editorial employes who are not associated with The Newspaper Guild would provide interesting comparative data. The number of daily newspaper publishers who are not associated with the Guild far exceeds the number who are (by 166 approximately 11 to 1). This would offer the researcher a much larger population from which to draw a sample. Such a study could examine whether organizations which already have contracts and presumably have already explored the issues of newsroom democracy in contract nego- tiations are more or less militant toward those issues than are organizations which have never confronted the issues in any formal sense. (4) The definition of newsroom democracy might be broadened to include such areas as outside activities, codes of ethics and affirmative action programs. If the definition is expanded, the researcher would need to explore new means of assessing attitudes in order to keep the study manageable. (5) This study has shown that significant differences exist between reporter and publisher attitudes toward the concepts of newsroom democracy. Future studies could explore the reasons for this disparity. Possible variables to test include (a) internal conflict in the organization, (b) the employe's sense of professional- ism, and (c) militant, pro-union feelings among employes versus mili- tant, anti-union feelings among management. Areas where reporters and publishers agree also could be studied. Why do both groups agree so strongly on the necessity for accurate, unbiased reporting? Why are both seemingly opposed to employe representation on editorial boards? Why do both feel there must be some restrictions on employes' rights regarding privilege cases? (6) The ultimate goal of newsroom democracy is two-fold: a more satisfied employe and a better quality product. Future studies 167 could examine whether these goals are concomitant or at cross purposes. A cause-effect relationship between the two goals could also be examined. (7) This study showed no relationship between editorial employe attitudes toward newsroom democracy and their length of service time or their relationship with the employer. It must be stressed that these analyses were weak tests of the possible rela- tionships in that they were based on only responses to the questions "How long have you worked as a newsperson?" and "How would you assess your relationship with the publisher (excellent, good, fair, poor)?" Relationships between these variables and newsroom democracy attitudes may exist but need to be tested through more rigorous analyses. The possibility of intervening variables could be investi- gated. For example, is an employe's attitude affected both by his length of service time and by his relationship with the publisher? A study by Sigelman (1973) suggests this may be so. How does the socialization process affect the employe's attitudes? Are veteran employes more or less favorable toward newsroom democracy because of the length of time they have been influenced by the paper's policies? Do veteran reporters with relative autonomy have different attitudes toward newsroom democracy than novice reporters who work under constant supervision? (8) Other variables which might affect employe attitudes need to be examined. For example, how does the employe's sense of professionalism affect his desire for newsroom democracy contract protections? Does an employe's respect for the publisher (or lack thereof) influence his attitudes toward newsroom democracy? 168 (9) No relationship was found to exist between publisher _ attitudes toward newsroom democracy and the circulation size of the newspaper, the length of time a contract had been in effect, the publisher's relationship with the Guild, or the newspaper's indepen- dence or group affiliation. Each of these possible variables needs to be examined further. There may be confounding variables at work or other untested intervening variables. For example, the length of time the publisher has held a contract with the Guild and his relation- ship with the Guild may be confounding variables. The publisher's relationship with his employes, a liberal or conservative predispo— sition, his attitude toward unionism in general, the length of time he has been publisher and his previous experience in the newspaper industry may be variables worth exploring. (10) This study focused on the publisher and editorial employe. Future studies might examine the relationship between the publisher, middle management and editorial employes. Managing, associate and executive editors in the newsroom are likely to be irI closer' contact with the actual production of the paper and therefore may exert a greater influence than the publisher on the company's collective bargaining stand. Middle management may also function as an intervening variable in employe attitudes toward newsroom democracy. For example, does the authoritarian managing editor evoke more favorable atti- tudes toward incorporation of newsroom democracy in contracts among editorial employes? 169 Applications of the Study The results of this study suggest that there are common areas of agreement between publishers and editorial employes as to which newsroom democracy concepts are acceptable and which are not. Edi- torial employes are not wholly in favor of all Guild proposals, nor are publishers wholly opposed to them. The results offer both management and union a more informed base with which to come to the bargaining table on these issues. Yet more research is needed. The study showed that the door to negotiating in this area is not closed. But neither is it wide open. Publishers are generally opposed to placing parameters around their right to manage. They are not likely to agree to provisions which turn their decision-making rights over to their employes without some checks and balances on those rights. Nor are they likely to agree to restrictive provisions for which they see no positive return. A followup study comparing the quality of editorial output of newspapers with newsroom democracy and without it would provide empirical rationale for accepting or rejecting newsroom democracy proposals. Yet providing a bargaining strategy for one side or the other was not the purpose of this study. That is merely a supple- mentary use. The purpose was to explore the question of whether concepts of worker participation and self-evaluation (as defined in this situation by newsroom democracy provisions) can reasonably be expected to be included in collective bargaining agreements. This poses the two-fold question of (1) whether the employes want such 170 rights and (2) whether management is willing to grant such rights. It is clear from the results of this study that the inclusion of such rights is possible, even though they may curtail management's rights. But it is also clear that in most instances the provisions must be tempered with some protection for the employer against employe abuses. Chamberlain's assertion that the American labor movement is moving in the direction of increased industrial democracy, greater worker participation and a broadening of the issues that are bargain- able is supported by the results of this study. Incorporation of newsroom democracy in contracts, even in a restrictive form, is a prime example of the shift toward greater industrial democracy. APPENDICES APPENDIX A MODEL CONTRACT LANGUAGE V-1 V-lA V-2 V-3 I-1 I-2 I-3 1-4 APPENDIX A MODEL CONTRACT LANGUAGE Voice The Guild shall designate a committee of its own choosing to take up with the Employer or his authorized agent any matter arising from the application of this contract or affecting the relations of an employe and the Employer. The Guild shall designate a committee of its own choosing to meet with a committee of the Employer's choosing to discuss and to make recommendations affecting all aspects of the operations of all Guild-represented departments, including news coverage. The Guild committee shall meet with the Employer committee once a month. (V-l and V-lA contained in ARTICLE VI: GRIEVANCE) The Guild shall be notified in writing of the name and qualifi- cations of any person proposed for appointment to an excluded position. Such appointment shall not be effective until dis- cussed with the Guild. (ARTICLE I: COVERAGE, Section 2(b)) The Guild shall be represented in any editorial or news con- ferences or boards by representatives of its own choosing. (ARTICLE XXIV: GENERAL PROVISIONS, Section 8) Employe Integrity An employe shall not be required to perform, over his protest, any practice which in his judgment compromises his integrity. An employe shall not be required to use his position as an employe for any purpose other than performing the duties of his position. An employe's byline or credit line shall not be used over his protest. Substantive changes in material shall be brought to the employe's attention before publication. 171 I-5 I-7 P-1 P-2 P-3 P-5 P-6 P-7 P-8 172 An employe shall not be required to write, process or prepare anything for publication in such a way as to distort any facts or to create an impression which the employe knows to be false. If a question arises as to the accuracy of printed material, no correction or retraction of that material shall be printed with- out prior consultation with the employe concerned. An employe whose work or person is mentioned in a letter to the editor shall be informed of such letter immediately and shall have the right to respond to such letter simultaneously on the page on which it is published. Privilege Against Disclosure and Authentication An employe may refuse, without penalty or prejudice, to give up custody of or disclose any knowledge, information, notes, record, documents, films, photographs, or tapes or the source thereof, which relates to news, commentary, advertising or the establishment and maintenance of his sources in connection with his employment. An employe may also refuse, without penalty or prejudice, to authenticate any material. The Employer shall not give up custody of or disclose any of the above without consent of the employe. The Employer shall notify the employe concerned, and the Guild, of any demand on the Employer for such surrender or disclosure or authentication. If the employe is proceeded against under law on account of his refusal to surrender or disclose or authenticate, the Employer shall move to join as a party to such proceedings; [the Employer] shall meet all expenses incurred by the employe, including fees and expenses of legal counsel retained by the employe; and [the Employer] shall indemnify such employe against any monetary loss including but not limited to fines, damages or loss of pay. In no case shall an employe suffer loss of wages, employe status or benefits under this contract as a result of his refusal to surrender or disclose or authenticate. APPENDIX B SUPPLEMENTARY TABLES 173 TABLE B-l.--Percentage of contracts containing newsroom democracy provisions 7 Type of Language N Restrictive Model Model-plus Total coverage V-1 185 3.2 29.7 26.5 59.4 V-2 185 1.6 0.0 0.0 1.6 v-3 185 0.0 0.0 0.0 ' 0.0 I-l 189 2.1 2.1 0.5 4.7 I-2 189 9.0 4.8 0.0 13.8 I-3 189 12.7 55.0 0.5 68.2 I-4 189 18.0 2.1 0.0 20.1 I-5 189 1.1 6.3 2.6 10.0 I-6 189 7.9 6.3 0.0 14.2 I-7 189 3.7 0.5 0.0 4.2 P-l 189 14.8 1.1 0.0 15.9 P-2 189 6.9 l l 0.0 8.0 P-3 189 6.3 1.6 0.0 7.9 P-4 189 20.1 5.8 0.0 25.9 P-5 189 5.8 2.6 0.0 8.4 P-6 189 23.3 3.2 0.0 26.5 P-7 189 23.8 3.2 0.0 27.0 P-8 189 2.6 1.6 0.0 4.2 TABLE B—2.--Percentage of contracts containing Voice, Integrity and/or Privilege by date of first contract First contract Voice Integrity Privilege 12/33 - 8/45 56.8 60.6 63.6 8/45 - 8/55 18.5 17.4 18.2 9/55 - 8/65 21.0 17.4 11.4 9/65 - 8/76 3.7 4.6 6.8 N 81 109 44 Degrees freedom 3 3 3 Chi-square 3.64 3.45 2.10 Probability .30 .40 .60 TABLE B-3.--Percentage of contracts containing Voice, Integrity and/or Privilege by circulation Circulation Voice Integrity Privilege (50,000 24.1 24.1 25.0 50,000- 249,999 34.2 39.3 38.6 250,000+ 41.8 36.6 36.4 N 79 112 44 Degrees freedom 2 2 2 Chi-square 2.24 1.11 0.19 Probability .40 .60 .95 175 TABLE B-4.--Percentage of contracts containing Voice, Integrity and/or Privilege by affiliation Affiliation Voice Integrity Privilege v—— Group 39.3 51.5 54.5 Independent 60.7 48.5 45.5 N 107 136 55 Degrees freedom 1 1 1 Chi-square 8.20 2.46 1.25 Probability .00 .12 .26 .monFaEm mecouwum muu .mcmnmmpnaa name 176 .mmmsmcmp pumcucou Pmooz acmucoqmmcgou use mucmsmpmum swam on aux com I xwucmqa< oa cmwmmm a mom. mpn.~ cam. mom.p mmo.—u me. u mpm.m omm.m own mm mmno> moF.P voo._ mmo._ moo.~ Nwm. u ¢m~.~ mmm.m nmm.p ocm ow N¢¢.— mmo.~ —om.~ mmo.~ FmF. mn©.~ «mo.m mon.— uwm ow N Nm¢.P oqw. ONN._ opm. NPN. mm—.N mom.m mmm._ Rem ow ~ wem.F mom. «cm._ 0mm. mno. omo.N Pmo.m mmm._ nvm om mpnu> oxw. NFN.F Nmm. wom.P mvp.~u omp. Nvm.m oom.m new mm <~um> mm as; mm as; mm can mm n2; 0mm n23; mEmpH muco_cm> .>mu .vum mmmczwxm cum: 2 mgwmccowpmmzc oz» cw com: macaw muvo> mpqwocwga Pmcmcmm co mmxopasm pmwcou_um can mcocmwpnsa Low menu zucmzamcmuu.mum w4m . .>mc .vum mmwczmxm cum: 2 1 ~ . mgwmccowumwac on» cw umm: mew»? zuwcmmch mpnwucpca chmcmm co mmxopaem powgouwnm can mcmcmwpnzg com upon xucwzcmcu11.m1m m4m .>m_u .Uum mmesov—m ...me z mgwmccowummza use cm now: mama? mampw>wca mpnwocwca chmcmm co mmaopasm Popcouwuo can msmgmwpnaa com aunt zucmzcmcm11.N1m m4m<~ 179 .mmxopasm mecoywum u umu .mcmcmanaa u can; .mmczmcmp uumcpcou Pmuoz mcwucoammccou new mpcwsmpmum Emu? op Am; coc : xwucmgg< op memmm Nun. wmm.p wa. mop.P 5mm. 1 mes. 1 N_m.m mwm.m cam mm mm1u> oom.~ wmm. mmo.p mom. mue. 1 o—o._ ~m¢.m mmn._ Nam mu ~_m.~ mom. w¢P.P Nmm. emm. emm._ eem.~ mmo.p Fem on N N_¢._ amm. mmF._ Rom. oem. cmm.~ mmw.~ mum.~ Pam om ~<~1u> mmm._ NPm.~ mmF.p “mm.F Now. 1 map. mmm.m mum.~ mum ox mp1u> mvo._ 0mm.~ cmo.— -—.P wow. 1 ans. ~m~.m mmo.~ men as <~1u> mm can mm as; mm nag mm nan gum nan; memuH mucmwcm> .>mu .upm mmmczoxm com: 2 mgwnccowpmmza mg» cw cum: mama? mowo> Pmauumcpcou co mmxongm vacoupum ccm mcmcmwpnaa Low mumv zucmzcmca11.m1m m4m .>mu .upm mmmczmxm cam: z mcrmccowummsa mcp cw now: new»? xuwgmmucH Pozuoocpcou co mwzoFasm megoupcm ucm mcmnmwpnan Low upon aucmacmcu11.m1m m4m .>mu .uum mmmszmxm sew: z memessoeummsc ms» sw use: new»? mow—w>wss Pesuuesesoo so mmaopsso Pewsouwum use msmsmwpsss sou eueu ausmsamss11.o_1m m4m.os 185 TABLE B-l4.--Inter-item correlations for publishers and editorial employes on general principle Privilege items and indices Items Correlation Publishers Editorial Employes PG-lA - PG-lBa .1362*** .3182* PG-lA - PG-l .7816* .7289* PG-lB - PG-l .7041* .8791* PG-2A - PG-ZB .3478* .2573* PG-2A - PG-2 .8473* .7914* PG-ZB - PG-2 .7754* .7760* PG-3A - PG-3B .6628* .3358* PG-3A - PG-3 .8984* .7172* PG-3B - PG-3 .9242* .8942* PG-4A1 - PG-4A2 .4847* .3505* PG-4A1 - PG-4 .7578* .8403* PG-4A2 — PG-4 .8676* .7989* PG-5A - PG-SB .3463* .5204* PG-5A — PG-S .7729* .8440* PG-SB - PG-5 .8579* .8960* PG-6A - PG-6B .2883** .3448* PG-6A - PG-6 .7932* .7556* PG-GB - PG-6 .8037* .8726* PG-7A - PG-7B .3583* .3991* PG-7A - PG-7 .8324* .8141* PG-7B - PG—7 .7941* .8557* PG-8A - PG-8B .1453*** .2685* PG-8A - PG-8 .7360* .8697* PG-8B - PG-8 .7393* .7049* aRefer to Appendix H for key to item statements and corre- sponding Model Contract language. bPG-l = index sum of PG-lA + PG-lB; PG-2 = index sum of PG-2A + PG-ZB, etc. *p.5 .001 **p 5_.01 ***p > .05 186 TABLE B-15.--Inter-item correlations for publishers and editorial employes on contractual Privilege items and indices Items Correlation Publishers Editorial Employes PC-lA - PC-lBa —.2357** .2690* PC-lA - PC-l b .7055* .7293* PC-lB - PC-l .5070* .8372* PC-ZA - PC-ZB -.3654* .1662** PC-2A - PC-2 .2682** .7715* PC-ZB - PC-2 .7957* .7432* PC-3A - PC-3B -.3268** .2227* PC-3A - PC-3 .7818* .6679* PC-BB - PC-3 .2920** .8482* PC-4Al - PC-4A2 .6712* .4893* PC-4Al - PC-4 .9226* .9027* PC-4A2 - P —4 .9000* .8092* PC-SA - PC-SB -.3454** .3714* PC-5A - PC- .6612* .7815* PC-SB - PC-5 .4528* .8663* PC-6A - PC-6B -.3417** .2088* PC-6A - PC-6 .6607* .6904* PC-GB - PC-6 .4387* .8470* PC-7A - PC-7B -.4l65* .2880* PC—7A - PC-7 .6379* .7571* PC-7B - PC-7 .2713** .8107* PC-8A - PC-BB -.l436*** .l689* PC-8A - PC-8 .5583* .7941* PC-BB - PC-8 .6450* .7235* aRefer to Appendix H for key to item statements and corre- sponding Model Contract language. bPC-l = index sum of PC-lA + PC-lB; PC-2 = index sum of PC-2A + PC-ZB, etc. *9 **p ***p .001 1/\ ( .05 ) .05 APPENDIX C PROCEDURES FOR DRAWING EDITORIAL EMPLOYE AND PUBLISHER SAMPLES APPENDIX C PROCEDURES FOR DRAWING EDITORIAL EMPLOYE AND, PUBLISHER SAMPLES The editorial employe sample was drawn from two sources: (1) The Newspaper Guild's Guild Reporter mailing list and (2) the New York local dues list. The Guild Reporter mailing list contains the names and home addresses of all Guild members. The list is broken down by local and then by units within the local (for example, the Detroit Free Press unit within the Detroit local). The list was in the process of being updated when the sample was drawn in April 1977, and most of the names and addresses had been corrected. Within the Guild Reporter mailing list some units consist of only editorial employes. In other cases the unit may be divided according to department; and in a third category, the unit may con- sist of more than one department but members are not listed according to department. The Guild Reporter mailing list does not contain names and addresses for any members in the New York local. Therefore, all names drawn from units in that local had to be drawn from dues lists, which provide names only. Questionnaires sent to all persons sampled from the New York local had to be addressed to their places of employment. A final sample size of 355 was sought (standard error = 5.4 per cent at the 95 per cent confidence level). Anticipating a 60 per cent return on the target sample (which is partially unknown but which was estimated), the target sample was 592. From existing TNG data, the following were known: (1) 6,495 editorial employes could be identified from TNG lists; that is, these employes could be extracted from composite lists of the appropriate units; (2) 16,590 general TNG employes could be identified from the remaining units. The assumption was made that half of this number would be editorial employes (basing the assumption on the fact that 187 188 approximately half of the total Guild membership is editorial). The ratio of editorial to general employes in units where editorial could be extracted was equal to 6,495 9 13,379, or 49 per cent. The general membership part of the target sample was there- fore doubled. To arrive at the total sample, a constant sampling fraction of .04 was applied to all units. This gave an effective sample of 260 names from units where names could be drawn exclusively from the editorial department, and 664 (or 2 x 332) names from units where editorial employes could not be singled out. (The sampling fraction .04 = 6,495 + 16,590.) A final sample of 916 names was drawn. Excluded from the sample were those units too small to be included when the sampling fraction of .04 was applied. (Thirteen was the minimum size for a unit to be included in the sample: .04 x 13 = 0.52, rounded off to 1.) Random numbers between 1 and 13 were used to select the first name to be drawn from each unit. There- after, every ith name was chosen to complete the sampling fraction for that unit. The population from which the publishers' sample was to be drawn was less than 200, and a poor return was anticipated. Therefore, the only publishers omitted from the sample were those for whom names and addresses could not be obtained or who did not have contracts covering editorial employes. A total of 190 publishers were sent questionnaires. APPENDIX D PUBLISHER AND EDITORIAL EMPLOYE QUESTIONNAIRES OmOZumm Ow0200w OmOZuum OmOZmum Ow0200m OmOZmum 20_m_>0m0 hom. .022 Emcee; H>wv. 3:080 0.303 .0305: 05> 05 memeo 30E E SE 802. .sm>0.0Em 05 *0 5.8.80 20 ee 00 0.30% 0:02, 2: so .8500 5:00 05 3 236. 9 0>0.0Ee :e >0 >30: .0 Eat as... .823 3.30.3290 use 208 m: 50.53 3:0E:m.$e 0.00.5 $05 >.:0 8:00.00 0.30% m>0.0Eo :< .:0:0:030 00.. e... E 00020:. 3:30 305 >.:0 E0350 9 02.300: 00 0.305. m>o_0Em :< .:0_$s0m.0 MEmEmmeceE we 00 0.30% «E. “.020 so mE.>0 one so 33 ms... .:0_~e0..030 0:0. .00 >000 e... E 0.2: 83:20 5.2: *0 00:22:. 00 0.305. m>0_0Em :d. 8:063:08. 3.2 050.0 03 EEmSe :e “305.2, 0:e 202 >:e .0 :0.to§0 305:5 08:08.0 3 0.30:... >000 was: _.< $109 qSQOkRHw IRS; qqflQ Mkzmsxmk «QM k0 130th kmmxm 1.1:: Omoz<.m:o:m u H>wv. (<20 302800 e E 0,003.2: 80>) "SEES... 05 t ....esm: .me>0.0Ew 3:03.00 05... 55.330 e 50350 Hoezcoo e E EmEBeS e :03 mE03.0E 80000 :0 :02: 0.302, 30> £0.53 9 02300 05 .090 05 *0 00:. 0:23:30 2: :O .m. 0:e .EeESeem 05 5.2, .2053 E 0933.0 :0 meme 30> £0.53 3 00500 05 .030 05 so 003 0:923. 05 :O .: “3:253: 3.30:0. 9: .0 :08 .0 3000.8 02: :0 306.00 50> 0520:. 88.0 30> 0.302, .530. 8e000s00e 9.: mEBEu >m m20:.0.0Ew .Eoose30: 05 E 3::0200 0:0E0meseE *0 :0000. .0... 3 5:0 8302:0000: @333 0...“ 5.3 0.3300 0.305 33.0.5 0:... 300.820 0:0 >0..00 030: 3.0830: 0:00.000 ..e 8.0:: 03 £3... 05 E80. 0.305 00.95 0:... ..s 0: 000 2003302 >2 : 30>: 03 0 00:10 2000200 03020.: «F... ...0 .3000 tags 0E .0>0.0Em 05 5.3 5.02.3200 s05... >.:0 005.330 03 0.30% 0:03 e.0>0_0E0 :e *0 30.00030: .0 80:00:00 00.300800 02 m... E 00020:. 00: 0:0 ~05 2:08:930 ~30 >:eo 00 0>0.0E0 :e 0:300: >eE 88: 00 03.080 0:... .:0.00030 m.0>0.0E0 05 :03 00.....3 00 00: 0.30:“... 0E. 0.00.0 so 0:._>3 6. 0003.330 00 .000 .0082: e... E 000E 830.3 5.0.: mE0sem0s 00.0.00 :0 5.3 23200 00: 000: s0>o.0E0 05. 005.330 e. H. w. .020. 005 00 >.m300:00.3E.m 0:008: 00 Em: 05 022. 0.30%. 5:00 05 00 .020. e 00 308 05 e. 0.53 03:3 0>0.0E0 :< 00.0020 0.3..0: L30. 03800000 0003:7000 00:3 0:.Es0000 0.30% 0:20 s0>0_0E0 02.? .0022: 3:08.28 :0 00.2 0800.0 00 00:300. 00 0.30...." 0>0.0E0 309.00 02 .0003 e.0>0.0E0 :0 *0 20.60: .0. .3 95:00:00 E00 00 850:3 0:.Es0e00 0.30% 0:20 0:0Eomeces. 003swucou11mmH0.080 05 >3 0.00 0: 0.305 000.38 ...: .003. 0. 3.3.0. .0 «.300. 0 00 0>0.080 :0 >: 00.50:. 083. >:< 1 80508.08. .0 00.38 0.: 03.8.0 0. .0: 03.080 :0 .0 53.. 05 3.8.0.00 0.305 .0>0.080 0:... .8050... 0.: .000. 0. 9.03.0. .0. :0000 «.300 0000. 03.0 -80 :0 :0.:3 :. 800 0 0. 3.00 0 0.0 :3. 0. 0.6: .0: 0.305 03.080 0.... 80.80 :08 80.. «.300. >08 00:. 0.83. .0.0:0E. .0 330 .030. .0. 8.3.3003. :0 08380 0.305 3008.08. .0 00.30» 8.: .000. 0. .0: 000.000 0:3 03.080 :< ..0..0.08 00.300 0.: 0.00 ...:0530 0. .0: 0.6.080 :0 .0 0:2. 05 3.8.0.00 0.30:“ 03.080 0: h 0:083 .05 .0 :08: 0:0 .0>0.080 0.: 8.0.... 0.305 3508.08. .0 00.38 0.: 80.8.0 .0 3508.08. 80:08:» 9 :003 00:00 03.080 :< .88 .:00 0.03.080 05 305.3 0.050 0. :0.~08.0.E .05 .000. .0: 0.305 3008.08. .0 00.38 0.: 00.000. A.0: 03.080 :0 80:3 0. .0>0.080 :< .soz 08.08. .0 00.300 0.: 80.8.0 9 .0: E9. 05 0.0: 0.305 0>0.080 :< ..mmotbOm 98m... k0 Wtbm. 0 1.0%. Q1202 >.th200 QEWEWK (km. to $000.0 1.551.. mu: ..0::oe.00 >303 ..008 800.830: .0 3.30.8 05 0. .23 08.3300 0: 0.30:... m0>0_08w ..0::om.00 80803308 .0 >.:0 0800800 0: 0.305 00.00: .0..0..0m 800.0080 800.030: 0:0 88:00 030: .0 0.0208 .0>0.080 05 5.3 3.8383 .0. 0.300090 005.8800 0 0>0: 0.30:0 m0>0.08w 003:.ucou11mm.:0 0.30:. 0:3 96.0 .80 :0 0. :0..3...00. 00.08 0. 00.00.30 0: .0: 0.30:0 .0>0.080 :< .000. .000. ..0 >00 0.30:0 .0>0.080 0:. .000.300 0.: .0030. 0. 0803.0. .0. 0>0.080 :0 08000 :0..00 ..300 :_ 08.800 .0:. .0 :08: 0.0>0.0 .80 0:. 0:0 0>0.080 0:. 8.0.:. 0.30:0 :0..08.0.:. .0 00.300 0.0>0.0 -80 :0 03.8.0 .0 5:08.08. .00:0..30 0. :003 00:00 .0>0.080 :< .:0..08.0.:. .0 00.300 0.0>0.080 :0 .0 >...0..:00..:00 0:. £08.08 0. 02.300. 0: .0: 0.30:0 .0>0.080 0:... 00.0. 0. 000.000 0>0.080 0:. 8.80 .0 00.300 0:. .0 000.0.000. 000.300 0.: 00200.0 0. .003.0. 0 .0 ..300. 0 00 00:30:. 00000. ..0 0:0 >:0 .0:.000 0>0.080 :0 80.0.0 0.30:0 .0>0.080 :< 000.300 0.: 00200.0 0. .003.0. 0.0>0.080 :0 0:. .208. 0000 0 :. 000. .000. >:0 >00 0. 08.300. 0: .0: 0.30:0 .0>0.080 :< ..0..0.08 00.300 0.: 0.00..:0:.30 0. 003.0. 0. 20.. 0:. 030: 0.30:0 0>0.080 :< .0000 0:. 0. >..00 0 00 :.0_ 0.30:0 .0>0.0 .80 0:. .000.300 0.: .0030. 0. 0503.0. .0. :0..00 ..300 0000. 0>0.080 :0 .. 003:.ucoouumm.0 .30> 000000 30> 0.303 30... 0003.080 .0..0..00 08.0300 80.800 0 00: >:00800 .30> 00: 0.00> >58 30... 02 00> 280:0. 030.0 0 .0 .03808 0 :0..00..330 .30> 0. ..0.0 .0:.00008 00000030: >..00 08:.08. :0..00..330 .0 00> .. :0..0.30..0 0083800 0.:0..00..330 .30> 0.... 50> 529.5%... ...(mmew wmwz... 02—mw3m2< >m w¢. 0.50.5 00358815222353”. mmzm. 30:0 194 OmOZuum OmOZuuw 8020mm OmOmem OmOZuuw OmOZuum 205—>05. ho<¢bzoo 000000 20:92 000000 :o.:.00 0: .0>0. .0>0. >.0:0..0 H>mv. Om Z 0. mm ..0>0.080 0:. .0 :0..0.00.0 0.:. .0 03 0.30:0 ...03 0.: .0 .00....0 .0..00 0:. 0. 0.0..0. 0. 0>0.080 :0 >3 >.00. .0 .5: 0: .. ..00.£0 >..0:0.000.0.0 0.0 0.00. 0: :02? 0808:9000 ...03 000:. >.:0 8.0..00 0.30:0 0>0.080 :< 88.8.0000 30. 0.: 8 0003.08 00..30 000:. >.:0 8.0..00 0. 008300. 03 0.30:0 0>0.080 :< .:0..0.00.0 0..:08000:08 .0 03 0.30:0 0:.. ..00.0 .0 0:..>3 0:. .0 003 0:... .:0..00..330 0.0. .03 >000 0.: 8 0008 30:05 .808 .0 0088.8 03 0.30:0 0>0.080 :< .0:0.000.08. 00.0. 0.00.0 0. .080..0 :0 59...! 0:0 0.00. >:0 .0 50.980 30:83 008805 03 0.30:0 >000 0.50: ._< $100 ..SQQKRE 1.2.: .3me MKZWSWK «Rm ....0 130.00 50...... ml... Omoz<.0:o..0 :0.:.00 0: 00.00 20:93 Ow O 2 < (m 00.000... 00.00 H>wv. ...00..:00 0 8 0003.08 0.03 8080.20 0:. .. ....0:0: 080003 0.303 .0308... 3.0.. 0:. 00000 .008 8 .0:. 0.02. 003.080 .0..0..00 0:0 05:33 0 503.03 .0283 0 8 8080.80 0 :030 0803.08 000000 .0 .90. 0.30.5 30> :08? 0. 00.000 0:. .0000 0:. .0 00.0 0:0:..:0.. 0:. :0 .0. 0:0 ..:080.0.0 0:. :...s .0.0:00 8 00.000.0 .0 00.00 30> :0.:.s 0. 00.000 0:. .0000 0:. .0 00.0 0:0:-..0. 0:. :0 ... 020500000 20000302 00.0220..00=o 0.0.020 00.00..00 ”08080.80 0838.0. 0:. .0 :000 .0 0.00000 03. :0 0:08.00 .30> 0.00.08 0000.0 30> 0.303 ..0..0. 0.0..00.000 0:. 08.2.0 >0 195 00.003 Om O z M. 0m .0..0..00 :0 0200 0. 0080. ..0:. 80.. 00>..0.:000.00. .020 0.30:0 00>0.08w 0m 0 z < (.0 800.030: 0:. 8 3:80.00 .:08000:08 .0 :0..00. Om O 2 0. 0m -00 0. .0..0 00>..0.:000.00. 0.8083 0:. :...s ..30:00 0.30:0 .0>0.080 0: ... 0m 0 z < (.0 008.020 0:0 >0..00 Om O z 0 mm 0.50: 080.000. 0:80.000 ..0 00.08 0. 2.0.. 0:. 80.0. 0.30:0 .0>0.080 0:... em 0 z < (0 S0. .0.. =00tm§m2 2. 53.02. W>qu~0.080 0:. 5.! 5.8.3800 .0.... Om O z “. um >.:0 005.330 03 0.30:0 ...03 0.0>0.080 :0 .0 88.0030. .0 0:880:00 am a z < (m 88.8.0000 30. 0.: 8 0003.08 .0: 0.0 .0:. Om O z “. um 0808:8000 .30 >..00 0. 0>0.080 :0 08300. >08 008.. .0 .0>0.080 0: ... 0.0. 0 Z < (m Om O z H. mm 88.0030 0.0>0.080 0:. .26 0003 03 .0: 0.30:0 0:.. ..00.0 .0 0:..>3 < Ow D z < (m .:0..00..330 0. .0..0 .0..0.08 0.: 8 0008 Om O z 0. mm 20:05 .808 080.000. 0>0.080 :0 :...s ..30:00 .0: 000: .0>0.080 0: ... Om o z < (w 005.33 0. .. .. .0..0. .0:. 0. >.0300:0..38.0 080000. 0. .5: 0:. Ow O z u. um 0>0: 0.30:0 .0..00 0:. 0. .0..0. 0 .0 0300. 0:. 0. 3.03 000:! 0>0.080 :< am a z < (w 08.020 80.3: Ow 0 Z .... 0m .30.. 0.30.0080 00.3...0:00 .0:.5 088.0.00 0.30:0 0:0.0 .0>0.080 0: ... 0m 0 z < (m ..0..0.08 Om O z 0 “.m 08000.08 .0 00.0. 0.000.0 0. 008300. 03 0.30:0 0>0.080 .0..0..00 02 0m 0 z < (m ....03 0.0>0.080 :0 .0 88.00.. Om 0 z u. um .0. .3 25.80.50 8:0 0. .0508... 8.8.0.00 0.30:0 0:0.0 8080080.). Dm o z < (m 003:..counumm.0302u 30_080 0:. >0 0.00 0.. 0.3050 000.300 0... .008. 0. 0803.0. .0 ..300. 0 00 0>0.080 :0 >0 00:30:. 00000. >:< .:0..08.0.:. .0 00.300 0... 00200.0 0. .0: 0>0.080 :0 .0 20.. 0:. 088.200 0.390 .0>0.080 0...... 000.300 0... .025. 0. 0803.0. .0. :0..00 ..300 0000. 0>o.n -80 :0 30...! 8 0000 0 0. >..00 0 00 80. 0. 000.. .0: 0.3030 .0>0.080 0;... .:0..00 £030 80.. :30. >08 .0... 00000. .0.0:0:.. .0 0.000 .000. .0. 3580:0000. ..0 083000 0.30:0 :0..08.0.8 .0 00.300 0.: .026. 0. .0: 000.000 0...? 0>0.080 :< .0..0.08 00.300 0.... 0.00 ...:0530 0. .0: 0>0.080 :0 .0 .00.. 0:. 088.800 0.390 .0>0.080 0:... .0:0800 .0:. .0 :08: 0:0 .0>0.080 0.: 8.0.8 0.30:0 :0..08.0.8 .0 00.300 0.: 00200.0 .0 :0..08.0.8 .00:0..30 0. :003 00:00 0>0.080 :< ....00 .:00 0.03.080 0:. .305... 0.00.0 0. :0..08.0.:. .0:. .00>0. .0: 0.3050 :0..08.0.8 .0 00.300 0... 00.026. 00... 0>0.080 :0 8085 o. .0>0.080 :< .00.. 08.0.8 .0 00.300 0... 00200.0 0. .0: ....0.. 0:. 0.00.. 0.30.0 0>0.080 :< ..MWUQDOM. gm? KO 0.2.30.0 40.4.9202 2......0200 ”52.339. «Rm k0 0.39:0 412.... m2.» ..0::O0.00 >50... ..0030 800.030: .0 :0..00.00 0... 0. .0.... 00:30:00 0.. 0.30:0 00>o.08m ..0::00.00 808000008 .0 >.:0 000.3800 00 0.30:0 00.000 .0..0..0w 008.00... 800.030: 0:0 .:0.:00 0.50: .0 0.8.08 .0>0.080 0:. .....s 0800300.0 .0. 0.30000... 03.8800 0 000: 0.30.0 00>0.08w 003: ...:ounaumzzzomamzo m>o..02m ..Smotom. Owo owe Omo own. Omn— omo own. own. own. omo Dwo z<:0 0.30:. 0...: 0>o.0 .80 :0 0. 5.58.00. 0.08 0. 00.00.30 0.. .0: 0.3050 .0>0.080 :< .000. .000. ..0 >00 0.30:0 .0>0.080 0... 000.300 0... .0000. 0. 0803.0. .0. 0>0.080 :0 .08000 :0..00 ..300 :. 000800 .0... .0 :08: “.0>0.0 .80 0.... 0:0 0>0.080 0... 8.0.8 0.30:0 :0..08.0.8 .0 00.300 0.0>o.0 .80 :0 00800.0 .0 :0..08.38 .00:0..30 0. :003 00:00 .0>0.080 :< 88.08.88 .0 00.300 0.0>0.080 :0 .0 >...0..:00..:00 0... 80.808 0. 008300. 0.. .0: 0.30.0 .0>0.080 0...... .030. 0. 000.000 0>0.080 0... :0..00 .0 00.300 0... .0 000.0.000. 000.300 0... 00800.0 0. .003.0. 0 .0 ..300. 0 00 00:30:. 00000. ..0 0:0 >:0 .0:.000 0>0.080 :0 .00.0.0 0.39.0 .0>0.080 :< 000.300 0.... 03.8.0 0. .003.0. 0.03.080 :0 08 .308. 0000 0 8 000. .000. >:0 >00 0. 0083.00. 0.. .0: 0.30.0 .0>0.080 :< .0..0.08 00.300 0... 0.00..:0:.30 0. 003.0. 0. .00.. 0... 000.. 0.30.0 0>0.080 :< .0000 0... 0. >..00 0 00.80. 0.30:0 .0>0.0 .80 0... .000.300 0... .026. 0. 0803.0. .0. :0..00 ..300 0000. 0>0.080 :0 .. 003:..c00uummHo.02m .<.mop~om omoz<_..0003m ||||0>_..o0030 0:0 0>_.0< III .0050 00:30 000002.02 0.; ......) 800.0030 .30.. 000000 30> 0.302, 2.0... .000 1800 III 0000 Illl .:0..00xm 1.080.530. ._0>o_080 50> 5.2. 0880020. 0:02. 30> 000000 30> 0:302. 261 0:00.00030: 0 00 000.52. 30> 0>0: 0.00> >:0E >>oI ...0>o_0E0 80:30 .30.. .0. 00.5.5 30> 0>0.. 0.00> >:08 2.0... 02 00> ...:08..0000 _0..o..00 0... >0 .00....00 0:30 0:. .00:3 00.0.60 30> 0.< :0..0..0000 no. .0 0.... 30> ..mZO....mm.DO 4>m2< >m mm_.00 mm 0.30; 003: ...:ou-..m.m.