PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before due due. DATE DUE DATE DUE DATE DUE T133 0 7 2994 ’- - ; ' ; r“ r .._____ ‘ DEE? fl 6 YUM ll MSU Is An Afflrmdtve AdiorVEquel Opportunity Inetltwon owns-9.1 THE EFFECT OF MARKET STRUCTURE, PROFITABILITY AND ORGANIZATION SIZE ON COMMERCIALIZATION, CLUTTER, CLEARANCE AND COMPLAINT PRACTICES USED BY COMMERCIAL TELEVISION STATIONS NATIONWIDE By: Jan LeBlanc Wicks A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY IN MASS MEDIA College of Communication Arts and Sciences 1989 ABSTRACT THE EFFECT OF MARKET STRUCTURE, PROFITABILITY AND ORGANIZATION SIZE ON COMMERCIALIZATION, CLUTTER, CLEARANCE AND COMPLAINT PRACTICES USED BY COMMERCIAL TELEVISION STATIONS NATIONWIDE By: Jan LeBlanc Wicks The last decade represented an era of unprecedented change in broadcast regulation and self-regulation. The National Association of Broadcasters abandoned its Television Code in 1982, and the Federal Communications Commission deregulated commercial television in 1984 and relaxed local station requirements for screening or clearing ads in 1985. These changes may affect the amount and accuracy of commercial information consumers receive. But what, if any, effect is this free marketplace approach having on the advertising practices of commercial TV stations? How well is the public interest served in the decision-making hands of commercial broadcasters as compared with FCC policies and self-regulatory bodies? A national mail survey of broadcast TV management was conducted to consider these questions and to determine if advertising practices changed after deregulation. The goal was to use economic theory, mass media policy research and organizational theory to predict performance differences in: 1) the level of commercialization, 2) the number and composition of commercial breaks, 3) the number and type of standards and sources a station uses or consults when clearing ads, and 4) individual station complaint practices. The idea that stations manipulate these practices to make their broadcast fare more attractive to viewers and commercial time more attractive to advertisers was also proposed and tested. The primary independent variables were market structure or concentration, station profitability, and organization size. Regressions were used to analyze these variables' relative effect on the dependent variables. T- tests were used to assess the usefulness of distinguishing different types of markets based on the level of concentration, profitability of stations and organization size. Study results suggest that some relationships between market structure, profitability and organization size, and commercialization, clutter, clearance and complaint practices exist. Results also support the notion that station managers may manipulate the dependent variables to make their airtime more attractive to viewers and advertisers. The implications of the marketplace approach are also discussed. Dedicated to Rob ACKNOWLEDGMENTS I credit my Mother for giving me the love of writing, and my Father for giving me the love of reading. My admiration, love and respect is theirs always. There are a number of friends who taught me how a kind word or deed can make a difference. Keith Adler, Ann Alchin, Liz Bain, Tom Baldwin, Martin Block, Trevor Brown, Janice Bukovac, John Clogston, Peter Golia, Allen Harris, T. Y. Lau, Barry Litman, Gary Pizante, Stan Soffin, Sandra Smith, Mike Steele, Bruce Vanden Bergh, and Dave Weaver each helped in some special way. My thanks to you. My dissertation committee deserves special thanks. Steve Lacy, Tom Muth and Todd Simon helped me to pursue my research area of choice. I'll always be grateful to them. I consider myself fortunate to have Fred Fico, my dissertation director, as a friend. He worked me to the bone, but always let me know that he was on my side. He allowed me to take risks, but made me justify every one. I wish every doctoral candidate could have a dissertation director like Fred. He's the finest. Most importantly, thanks to my husband, Rob. TABLE OF CONTENTS LISTOF TABLESOOOOOOOOOOOOOOOOOOO ..... OOOOOOOOOOOOOOOOOOOX mmER I: INTRODUCTION...OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOI The Study.............................................5 Implications of the Study............................10 Implications of the Study for Industry............11 Implications of the Study for Mass Media Economic Research.................................12 Implications of the Study for Mass Media Policy Research...................................14 Implications of the Study for Organizational Research..........................................16 Summary......................... ..... ................17 NOTES: CHAWERI OOOOOOOOOOOOOOOOOOO 0.0.0.0000000000019 CHAMER II: LITERATURE REVIEW.OOOOOOOOOOO0.0.0.00.0000026 Economic Theory......................................26 Economic Theory Applied to Commercial Television..........................33 Mass Media Policy Literature.........................36 ackground........................................36 Commercialization and Clutter.....................38 Advertising Clearance and Complaint Practices......................................45 Fairness Doctrine.................................53 Organizational Theory................................56 Summary of the Literature Review.....................59 NOTES: CHAPTERIIe ...... 0.00.0.0000...0.0.0.0000000061 vi CHAPTER III: PROPOSED THEORY AND HYPOTHESES............78 k/The Attractiveness of Advertising and Viewing Environments..............................82 Hypotheses...........................................92 Market Structure or Concentration.................92 Profitability.....................................95 Organization Size.................................97 The Joint Effect of Market Structure and Profitability..............................98 Research Questions..................................101 NOTES: CHAWER IIIOOOOOOOOOOOOOOOOOOOO0.00.00.00.00103 CHAHER Iv: METHODOIJOGYeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeelos Study Approach......................................105 Methods of Survey Research..........................106 Sampling Scheme.....................................107 Nonresponse in Mail Surveys.........................109 Questionnaire.......................................110 Operationalization of the Independent Variables.....111 Predictor Variables..............................111 Market Structure..............................111 Station Profitability.........................115 Organization Size.............................116 Control Variables................................117 Market Rank...............................J...117 Type of Advertising...........................119 Direct Broadcast Substitutes..................120 Market Quality................................121 Rates.........................................122 Former NAB Code Subscription..................123 Operationalization of the Dependent Variables.......123 Commercialization................................124 Clutter..........................................125 Formality of Clearance and Complaint Practices.....................................126 Statistical Techniques Employed.....................129 NOTES: CHAPTER IV...COOOO..0.0..0.0.0.0000000000000133 vii CHAPTER V: Violations RESULTS........................ ...... ......140 Data Examination and Scale Construction.............142 of Normality..........................142 multiCOIIinearity................................147 Hypothesis Hypothesis Hypothesis Hypothesis Hypotheses Hypothesis Hypothesis Hypothesis Hypothesis Hypotheses Hypothesis Hypothesis Hypotheses Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Reliability Of measures.............................159 Discussion of the Hypothesis Tests..................164 1.....................................165 2.....................................166 3.....................................170 4.....................................170 3 and 4...............................172 5.....................................174 6.....................................174 7.....................................175 8.....................................175 7 and 8...............................177 9.....................................179 10....................................182 9 and 10..............................183 11....................................185 12....................................186 13....................................186 14....................................187 13 and14.............................187 NOTES: Research Research Research Research Research Research Research Research Research Research Question Question Question Question # 1............................188 # 2............................191 # 3............................195 # 4............................198 Questions # 3 and # 4...................198 Question Question Question Question # 5............................l98 # 6............................201 # 7............................203 # 8............................206 questions#7and#8...................207 Summary of Hypothesis and Research Question Results..............................207 CHAPTER IV..................................211 viii CHANERVI: DISCUSSION.........O......................216 Implications of the Study...........................221 Implications for Industry........................221 Implications for Mass Media Economic Research.....................................224 Implications for Mass Media Policy Research......................................235 Implications for Organizational Research......................................250 Conclusion..........................................253 NOTES: CHAPTERVI..................................256 APPENDIXA: MAILINGS..................................258 APPENDIX B: SURVEY QUESTIONNAIRE INSTRUMENT...........261 APPENDIX C: PEARSON CORRELATIONS BETWEEN THE INDEPENDENT VARIABLES...................263 APPENDIX D: REGRESSIONS TO ASSESS MULTICOLLINEARITY.......................271 BIBLIOGMPHY.............. ..... ........................273 ix Table Table Table Table Table Table Table Table Table Table Table Table 10: LIST OF TABLES sampling SChemeOOOOOOOOOOO....;ODOOOOOOOOOO108 Survey Responses by Market Structure.......141 Skewed Independent and Dependent variables.I.0....0................000000144 Principal Components Analysis on the Independent Variables (Factors and variance Explained).....................153 Principal Components Analysis on the Independent Variables (Varimax Rotated Factor Matrix)............ ...... 154 Factor Analysis on the Independent Variables (Factors and Variance Explained).......O................0.0.0.157 Factor Analysis on the Independent Variables (Varimax Rotated FactormatriX)..........................158 Regressions With Untransformed variableso..D.......O...........OOO0.00.167 Regressions With Outliers Reduced to 3 Standard Deviations...................168 Transformed Regressions....................169 Mean and Standard Deviation for the Number of Formalized Complaint Policies by Market Structure............171 T-Tests Comparing Market Structure Categories With Complaint Formality (one-Tailed TestS)......................171 Table Table Table Table Table Table Table Table Table Table Table Table 11: 12: 13: 14: 15: 16: 17: 18: 19: 20: 21: Means and Standard Deviations for the Communication Form and Method Used Most by Market Structure................173 T-Tests Comparing Market Structure Categories With the Formality of Advertising Practices (One-Tailed Tests)......................173 Mean and Standard Deviation for the Number of Formalized Complaint Policies by PrOfitabilitYeeeeeeeeeeeeeeeeeeeeeeee176 T-Tests Comparing Profitability With Complaint Formality (One-Tailed Tests).............. ........ 176 Means and Standard Deviations for the Communication Form and Method Used Most by PrOfitabilitYO..0....0...............000177 T-Tests Comparing Profitability With the Formality of Advertising Practices (One-Tailed Tests).........................178 Means and Standard Deviations for the Number of Hours Devoted to Clearance, Number of Policy Areas, and Number of Policy Sources by Organization Size....................................180 T-Tests Comparing Organization Size With Clearance Formality (One-Tailed Tests)......................181 Mean and Standard Deviation for the Number of Formalized Complaint Policies by Organization Size...........182 T-Test Comparing Organization Size With Complaint Formality (One-Tailed Tests)................ ...... 183 Means and Standard Deviations for the Communication Form and Method Used Most by Organization Size....................184 T-Tests Comparing Organization Size With the Formality of Advertising Practices (one-Tailed TeStS)eeeeeeeeeeeeeeeeeeeeeelas xi Table Table Table Table Table Table Table Table Table Table Table 24: 25: 26: 27: 28: 29: 30: 31: 32: 33: Means and Standard Deviations for Commercial Time, Non-program Time, and Number of Program Length Commercials by Market Structure.....................189 T-Tests Comparing Market Structure Categories With Commercialization (one-Tailed TestS)eeeeeeeeeeeeeeeeeeeeee190 Means and Standard Deviations for the Number of Breaks and Consecutive Non-program Announcements by Market Structure.....................192 T-Tests Comparing Market Structure Categories With Clutter (One-Tailed Tests)......................193 Means and Standard Deviations for the Number of Hours Devoted to Clearance, Number of Policy Areas, and Number of Policy Sources by Market Structure..................196 T-Tests Comparing Market Structure Categories With Clearance Formality (One-Tailed Tests)......................197 Means and Standard Deviations for Commercial Time, Non-program Time, and the Number of Program Length Commercials by Profitability............l99 T-Tests Comparing Profitability With Commercialization (One-Tailed Tests)......................200 Means and Standard Deviations for the Number of Breaks and Consecutive Non-program Announcements by Profitability........................201 T-Tests Comparing Profitability With Clutter (One-Tailed Tests)..............202 Means and Standard Deviations for the Number of Hours Devoted to Clearance, Number of Policy Areas, and Number of Policy Sources by Profitability........................204 xii Table Table Table Table Table Table Table Table 34: 36: 37: 38: 39: 40: 41: T-Tests Comparing Profitability With Clearance Formality (one-Tailed TestS)......................205 Station Advertising Policies...............236 Station Policy Sources.............. ....... 237 Station Advertising Complaint POIiCieSeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee0238 T-Tests Comparing Performance of Stations Having/Not Having Advertising Policies (One-Tailed Tests)......................243 Employee Responsible for Clearance ReView..0.0000000000000..........0000000252 Pearson Correlations Between The Independent Variables...................263 Regressions to Test for Multicollinearity (Standardized Beta Weights).............271 xiii CHAPTER I INTRODUCTION Commercial television advertising is an important source of product information for American consumers. Most learn about products or brands from television advertising,1 and regard it as the most influential type.2 Changes in the content of such advertising affects the accuracy and quality of product information available to many Americans. And changes in the amount of advertising affect how much consumers can learn. Consumers only recall about four commercials at any one time.3 The deregulation of commercial television by the Federal Communications Commission in 19844 has important implications for the type and amount of commercial information available to consumers. Now local broadcasters decide for themselves how much commercial time and how many breaks to schedule per hour, and whether or not to air program length commercials in which the message is interweaved so closely with program content that the entire program must be considered commercial.5 A significant increase in the amount of commercialization and clutter will affect how much consumers can learn about new products and services from this important source. FCC actions may also affect what consumers learn. The FCC eliminated licensee obligations regarding deceptive advertising in 1985. Local broadcasters no longer need to exercise particular care when deciding to accept advertising which is the subject of an unadjudicated Federal Trade Commission complaint, or to review the reliability of every prospective advertiser.6 And stations of greater size and resources are no longer expected to make a correspondingly greater effort in this area.‘7 Consequently, these relaxations could result in a proliferation of inaccurate advertising on stations of all sizes and of all degrees of profitability. Further, the National Association of Broadcasters abandoned its Television Code in 1982.8 The NAB TV Code was the most important self-regulatory mechanism for local broadcasters, providing guidelines regarding commercialization, clutter, and clearance. The NAB Code Authority Board also cleared commercials for national and regional advertisers. Preclearance by the NAB was tantamount to network approval (although the networks still determined for themselves whether a commercial would be accepted).9 After the Code's demise, the networks assumed primary responsibility for clearance.10 However, concerns over clearance effectiveness were raised in 1987 when ABC, CBS and NBC cut back their Standards and Practices divisions as a cost-cutting move.11 As a result of the events of the past seven years, individual station management has more commercial decision-making freedom than ever before. Local broadcasters may now decide how many commercials to air and what procedures to use to determine if advertisements are deceptive. They also have more responsibility for clearing national and regional spot advertising. At the same time, station responsibility for determining how the local community views station policies was relaxed by the FCC. The new flexibility extends to ascertainment of public views regarding advertising policies. The FCC no longer concerns itself with ascertainment methodology, but only with how responsive stations are to community needs.12 So managers may decide what new policies to implement, and how to determine how the community feels about them. Licensee discretion is essentially "penalty-free," as the FCC also eliminated license challenges based on commercial time considerations.13 Violations of the deceptive advertising policy are now considered only in 4 which rarely result in "character" proceedings,1 meaningful action. The FCC considers ascertainment satisfactory if "programming presented by the licensee satisfies its obligation,"15 which seems vaguely worded and thus difficult to disprove. The FCC asserted that competitive market forces regulate the industry, precluding viewers from watching and advertisers from buying time on stations adding too many commercials.16 It noted a decline in public complaints about advertising as support.17 Advertisers disagreed, predicted clutter problems, and warned broadcasters to continue keeping logs to verify that advertising schedules aired.18 But what effect is the marketplace approach having on the advertising practices of TV stations? Will the quantity and quality of consumer information be affected? And is the public interest served as well in the decision- making hands of commercial interests as it was by FCC regulations? The potential for dramatic changes in advertising practices now exists. But public or FCC monitoring now is also more difficult, precisely when these new advertising practices might become a problem. The present study attempts to both illuminate what changes have occurred and provide a basis for evaluating their consequences. MM: A national mail survey of commercial television station management was conducted to determine if deregulation changed advertising practices. The goal was to use economic theory to predict performance differences in clutter and commercialization levels, and in clearance and complaint practices. The idea that stations manipulate these practices to make programming more attractive to viewers and commercial time more attractive to advertisers was also proposed and tested. Economic theory provides a means of distinguishing markets and making predictions based on those differences. The industrial organization paradigm states that an industry's market structure predicts its conduct and ultimately its performance.19 Market structure (or organization) is the sum of the economically significant features of a market affecting firm behavior in the industry supplying that market. These factors include the level of product differentiation and number of sellers and buyers.20 Market conduct consists of a firm's policies toward its product market and toward the moves made by its rivals in that market.21 Conduct includes pricing behavior, product strategy, and advertising.22 Conduct varies by market structure, and is considered "better" under some structures than others.23 Market performance is the appraisal of how equitable the economic results of an industry's conduct are to consumers and producers. Performance is assessed by considering whether an industry operates efficiently by earning enough to pay expenses and make a profit. Economists also consider whether an industry supports research, incorporates new technological developments in a timely manner, and employs all who desire jobs. Ideally, the industry, workers and consumers are all satisfied. The industry earns a reasonable profit, workers get a reasonable wage and are awarded for initiative, and consumers get a quality product at a reasonable price.24 Markets are distinguished by calculating Herfindahl- Hirschman Indices for each. The index takes the number and market shares of firms into account and indicates how concentrated a market is.25 Conduct differences in advertising practices are predicted based on market differences shown by the index. Performance is assessed by determining if ad practices are consistent with the FCC's public interest interpretation.26 The conduct studied was individual commercial television station commercialization, clutter, clearance and complaint practices. Commercial television stations are defined as those that depend on national and regional spot, and local advertising for profits, and may also receive funds from national commercial television networks. They may carry network programming, originate syndicated and paid programming, and may compete with other stations for a share of the viewing audience. Stations that depend primarily upon fund raising efforts or direct sales for revenues, such as certain religious or Home Shopping Network stations, are not included. Satellite stations are not included as they do not originate, but simply carry "mother" station programming. Commercial television stations are also distinguished as being "more" or "less" profitable. Broadcast band and network affiliation are commonly used as proxy variables for profitability,27 since individual station financial data is confidential. More profitable stations typically broadcast on channels 2 to 13 (VHF) and are affiliated with a national commercial television network. Less profitable stations typically broadcast on channels 14 to 83 (UHF)28 and are independent (or not affiliated with a network). A station's ranking in its home market was also used as a profitability indicator in this study, and these three variables were combined to create a profitability index.29 Commercialization includes the average amount of commercial and non-program time a station airs during early fringe and prime access times (3:00 to 7:00 p.m. or 4:00 to 8:00 p.m.), and the average number of program length commercials a station airs per month. Commercial time is defined as the amount of commercial matter30 a station schedules per hour of locally-originated programming, previously limited to 16 minutes by the FCC.31 Non-program time is the average amount of non- program time (including commercials, promotional announcements, and public service announcements), formerly limited to 16 minutes per hour by the NAB TV Code.32 The average number of program length commercials a station airs per month is also a measure of commercialization. Program length commercials were banned before deregulation. Findings are compared to the appropriate guideline to identify changes in practices. Clutter is the number of breaks, and consecutive announcements scheduled in breaks, in locally-originated programming, formerly limited by the NAB Code.33 Findings are compared to these guidelines to identify changes in commercial practices. The Code also banned multiple product announcements presenting two or more products or services in a single commercial less than 60 seconds, except where integrated to appear to the viewer as a single message. A multiple product announcement was considered integrated if the products or services in them were related, and if similar production characteristics were used throughout.34 New time units, like 15 and 45 second spots sold individually,35 were introduced when the multiple products standard was abandoned. Findings were examined to determine if stations have begun to accept these new units. Clearance and complaint practices are the formal station mechanisms for communicating with, and addressing the concerns of, advertisers and the public. Clearance consists of the practices a station utilizes to determine whether to accept an ad, such as having standards banning or restricting certain types of ads, and prescreening commercial copy for compliance with station policy, community standards, and legal requirements.36 It is based on FCC policy and decisions indicating that stations must have a program for preventing deceptive ads from airing. Clearance is important as it represents the initial point of interaction between advertisers and broadcasters. It is at this point where an ad or program length commercial is accepted, altered, or refused, based on formal or informal station advertising policies. Complaint resolution is the existence of procedures to deal with viewer, advertiser, interest and minority 10 group complaints about ads after they air. It is assumed that the more groups a station officially accepts complaints from, the more formalized these practices are, and the more a station is trying to keep abreast of public concerns. The formality of both clearance and complaint practices is also measured using management's usual form and method of communicating its policies to employees. Written forms of communication (such as a station manual including guidelines and policy statements) are assumed to be more formal than informing subordinates of new policies verbally. Impliggtigns 9f the Study: Television is being surveyed because it is the primary media source for commercial information. For example, 64 percent of respondents in a national survey reported they were most likely to learn about products or brands from television commercials,37 and in another study, 84 percent regarded television commercials the most influential type of advertising.38 Cable advertising is not included as it only accounts for a small percentage of spot and local advertising revenues.39 The study seeks to incorporate knowledge from a number of disciplines to more fully illuminate the 11 advertising performance of commercial television stations. It attempts to incorporate knowledge from the study of mass media by economists and policy researchers, and from the study of the sociology of organizations, to explain and predict the performance of commercial television stations using traditional explanatory variables from each. Findings may suggest the wisdom of combining theoretical areas in future research. Implications of the Study for Industry: The study is timely as broadcasters have had seven years to adapt to the NAB Code's absence, five to adapt to deregulation, and four to adapt to the relaxation of deceptive advertising policies. The study may provide useful suggestions for industry on dealing with commercialization, clutter, clearance and complaint practices because it is national in scope. For example, managers in larger markets may provide insight and advice on impending problems to their counterparts in smaller markets. 12 Implications of the Study for Mass Media Economic Research: No previous economic study was primarily concerned with local station commercialization and clearance practices. This study is apparently the first to examine the commercialization, clutter, clearance and complaint practices used by television stations nationwide at a time when greater discretion in individual decision-making exists. It is also the first to test if these practices are used to manipulate commercial time attractiveness. Previous research interpreted how economic theory applies to broadcasting.40 The national commercial television market was explained,41 and studies of conduct including the level of programming diversity,42 the market for theatrical movies,43 and the effect of divestiture of network owned and operated stations would have on public service programming performance,44 provided insight. A few studies examined the relationship between a local station's profitability and its performance in public service programming.45 Studies also assessed how competitive the local television news market 15,46 how stations set advertising rates,47 the effect of type of ownership on advertising rates,48 and the possible effects of deregulation.49 13 The study extends economic knowledge by testing whether the underlying television market structure varies enough to result in conduct differences in advertising practices. It may also suggest a more accurate method of calculating Herfindahl-Hirschman Indexes. Previously, shares from individual Arbitron ratings reports were used.50 Indexes used in this study were calculated from shares in the Arbitron ADI Viewing Allocation Report, which reflects the actual level of market concentration more accurately, and utilizes the same diary data used in the individual reports. The report is designed to help users compare cable to non-cable household viewing information. All home market stations, stations from outside the market (or Area of Dominant Influence), cable channels and superstations which achieve at least a .1 percent share51 are included, representing a more liberal standard than found in the individual market reports. This means that viewing is credited to more stations than can be reported in the local reports.52 14 Implications of the Study for Mass Media Policy Research: Policymakers can use study results to assess whether commercial deregulation is in the public interest, and if the marketplace effectively regulates advertising. For example, policymakers may conclude that deregulation is working if overall results indicate that broadcasters increased commercial time only slightly. Large increases might suggest that deregulation is appropriate only for some markets but not others or that the industry should be reregulated. The study provides an "independent" assessment of local station commercialization, as FCC or NAB staffers conducted the others prior to deregulation.53 An independent study examined network commercialization practices.54 Other studies have examined the effect of clutter on commercial effectiveness and viewer response to it.55 But no study reported the amount of commercial time and clutter on all types of stations in markets of differing sizes throughout the country. Advertising clearance is not widely studied, with most articles centering on "post-hoc regulation" done by "formal and informal self-regulatory bodies."56 Only one 15 study of FCC clearance policy was found, which noted that stations should be more vigilant when reviewing spots they conceived and produced.57 The proposed study thus represents the first attempt to assess the nature of local television clearance and complaint practices and how and why they differ. The goal is to provide insight into these practices, suggest new research areas, and provide a foundation for future studies assessing the long-term effects of deregulation. The study will also fill a gap by assessing broadcasters' self-regulation. Few studies have evaluated "the social performance of individual advertisers, advertising agencies, media, and the television networks,"58 and little information exists about local self-regulatory practices.59 Policy research is also extended by evaluating both standards and performance to reveal whether deregulation and relaxation of policies is having its intended effect. Standards and performance are evaluated first by determining existing advertising policies and how much staff time is devoted to them. Stations with and without policies are also compared to determine if the existence of a policy results in performance differences. Comparisons are then made with the old guidelines to 16 discover if stations are exercising the individual discretion intended to provide them "with the flexibility necessary to respond to the needs and demands of their communities."60 Implications of the Study for Organizational Research: Ideally, by combining traditional research from the mass media with organizational research, the study may lead to collaborations across traditional theoretical boundaries, resulting in more complete knowledge for all. Therefore, this study represents one attempt to incorporate, and hopefully encourage, interdisciplinary research on the mass media. The study extends organizational research into mass media research by applying the theory of bureaucracy to television stations to determine if organization size is a predictor of advertising policy formality. Organization size has long been a topic of interest in organizational research.61 The study suggests that organizational theory may be used to conceptualize and measure station differences in new ways. For example, larger and smaller stations differ in the number and type of formal advertising policies. Most stations may be expected to have policies in certain 17 areas, like political advertising.62 But larger stations may also have policies regarding specific advertising techniques, such as bait and switch advertising.63 The theory of bureaucracy suggests that as organization size increases, more types of tasks are carried out by more specialized staffs.64 Thus, organizational theory provides a way of explaining these, and perhaps other, differences at television stations of various sizes. And the study also utilizes a traditional measure of organization size and demonstrates its usefulness for mass communication researchers. Organization size may thus provide a way to assess the formality of other broadcast policies in other performance studies. Summarx= The last seven years have represented an era of unprecedented change in broadcast regulation and self- regulation. Guidelines which suggested how to meet public interest considerations for commercialization, clutter, clearance and complaint practices were eliminated. Local broadcasters were given a great deal of discretion in implementing advertising policies at their stations, and may respond as they wish. Consequently, the time has come to consider whether the commercial 18 marketplace is capable of regulating itself, and whether local broadcasters are capable of responsible voluntary regulation of advertising. The study introduces these four advertising performance areas to mass media policy research to attempt to answer this current public policy question. It attempts to explain and provide measures for them, using a number of theoretical approaches. Perhaps the most important contribution of the study is that it provides a foundation for examining advertising practices used at commercial television stations nationwide. The intent is to allow researchers to evaluate whether the public is able to obtain accurate and reasonable amounts of commercial information, both now and in the future. NOTES: CHAPTER I 1 ' t es owa e v s'o ' in a Time of Change, The Fourteenth Report in a Series by the Roper Organization,Inc. (New York: Television Information Office, 1985), p. 22. The remaining 4 percent of responses were to Don't Know/No Answer. 239w te Use Television (New York: Television Bureau of Advertising), p.6. Citing a study by R. H. Bruskin, "a company which asks different kinds of people about their media habits." 3Peter H. Webb and Michael L. Ray, "Effects of TV Clutter." 12urnal.2f_Adxerti§ing_Be§earsh 23 (February/March 1983), p. 9-11. 4In_ths_Mattsr_2f_Ihe_Bexisien_gf_£resramming_and o erc a i atio o. ic es sce 21’ P O',,-.l! 0' "'l-. ‘ ‘ S O,_ 'lLi‘, ,- e ‘V.?'! Stations. 98 FCC 20 1076 (1984). 5In_the_natter_2f_Eresram_Length_QQmmercials. 39 FCC 2d 1062 (1973). GElimination er Uaaeeessary Broadcast Begglatien, Fed. Reg. 5583 at 5589-90 (1985). 7Qenter_f2r_Lan_and_§osial_£olisx. 23 RR 2d 187 at 195 (1971). ayaited States v, National Assoeiation er Breageaetere, 536 F.Supp. 149 (D.C. Cir., 1982). 9"Agencies, Networks Battle Over Censor's Role," gggeek 14 November 1983, p. 52. 1OnThe TV Code is Dead," Marketing_§_uedia_nesi§ien§ 18 (December 1983), p. 65. See also Lynda M. Maddox and Eric J. Zanot, "Suspension of the NAB Code and Its Effect on Regulation of Advertising," Journalism Quarterly 61 (Spring 1984), p. 125-30, 156. 19 20 11L. J. Davis, "Looser, Yes, But Still The Deans of Discipline," gnaaaele (July/August 1987), p. 33-4. 1298 FCC 2d 1076 at 1101. 13Ibid., p. 1102. 1450 Fed. Reg. 5583 at 5590. 1598 FCC 2d 1076 at 1101. 1698 FCC Zd 1076 at 1076, 1105. 17Qommercialization an IV Statieas, 49 RR 2d 391 (1981). 18Diane Mermigas and Steven W. Colford, "Don't Burn Those Logs, Ad Units Tell TV Stations," Adyertisiag Age, 2 July 1984, p. 1, 42. 19F. M. Scherer. In_ustrial_uark_t_§trueture_and Economis_£erformanse (Chicago= Rand McNally. 1970). p- 3- 2oRichard Caves, ° , 5th ed., (Englewood Cliffs, N.J.: QQBQBQE1_EQEZQIEQDE§ Prentice-Hall, 1982), p. 16. 211bid., p. 48. 22Scherer, p. 4-5. 23Edwin Mansfield. Erinsiple§_2f_Misroesenomiss 3rd ed. (New York: Norton, 1980), p. 218. 24Caves, p. 66-7. 25Scherer, p. 51-2. 26The standard to which the FCC must conform in regulating broadcasting is that its actions must be "consistent with the public interest, convenience and necessity" (Communications Act, 47 U.S.C.A. Section 307). 27See Barry R. Litman, "Public Interest Programming and the Carroll Doctrine: A Re-examination," learaa1_et fireageaetiag 23 (Winter 1979),p. 56, for example. 21 28Frequency assignments and general information about broadcasting may be found in "The ABC's of the Fifth Estate," a e s o (Washington D.C.: Broadcasting Publications, 1988), p. A- 3, A-S. 29Traditionally, broadcast band, network affiliation and net weekly circulation are used as profitability indicators. However, rank was found to be an indicator of profitability in this study, and net weekly circulation an indicator of market size. See the data analysis section in Chapter 5 for a complete explanation. 30In 1963, the FCC noted that "Commercial material in prime time includes billboards, public service announcements, promotional announcements for other programs, as well as commercial copy...Commercial material within non-prime time does not include public service announcements, promotional announcements for other programs, and opening and closing 'billboards' which give program or sponsor identification." See AQ!§I§1§1DQ.QD v' ' s t o , 28 Fed. Reg. 5158 at 5160 (1963). Based on this early distinction, the FCC guideline is defined to include only commercial time for the purposes of this study as the daypart of concern is early fringe/prime access. Commercial matter has been defined as including both commercials and promotional announcements in W. 49 R 2d 391 at 392 (1981)., for example. See note 1, p. 293: "'Commercial matter' would be defined as all commercial announcements but would not include (a) station identification announcements, (b) public service announcements, (c) political advertising, whether paid or broadcast free of charge, (d) counter commercials, whether paid or broadcast free of charge, and (e) statements of editorial opinion by a station or responses to such statements." However, more recently in In the Matte r e: Bevisroa of Programming and W 0 ram e emen s o c a e ev s' o Stations, 2 FCC Rcd 6822 at 6824, 6828 (1987), when asking for comments regarding how many additional minutes of commercials children view "now that the guidelines have been deleted," the FCC specifically states in note 32, "In responding to these questions, we request that commenters restrict their remarks to the number of minutes per hour designated as commercials." Therefore, the interpretation used in this study is not inconsistent with FCC intent. 22 31 s on's Ru es-- 'ss'o O --W’t e e ' s ' ' Bureau, 43 FCC 2d 638 at 640 (1973). 328r2adsas11ng_Qablesseting_xearbeek_12§1 (Washington, D.C.: National Association of Broadcasters, 1982), p. D-18. This was the last version of the Code printed before it was dropped in 1982. Non-program guidelines were 9 1/2 minutes per hour during prime time, and 16 minutes per hour in other times for network affiliated stations. Non-program material could not exceed 12 minutes during children's programming airing on Monday through Friday, and 9 minutes and 30 seconds on Saturday and Sunday. Independent stations were limited to 8 minutes in a 30 minute period or multiples thereof during times other than prime time. 33NAB Television Code, p. D 17-8. 34Ibid. 35"ls-Second TV Commercials Resurfacing Once More, " I§l§_i§iQnLB§Q12_AQ§ 9 May 1983. p- 59- 60- 36The FCC has spelled out station responsibility for clearance in a number of deceptive advertising decisions, including: In_1he_Ma1ter_2f_KMEQI_Ths_§tatien_Qf_the Stare, 6 FCC 729 (1939).: FCC Public Notice, Lieeueee b' t W t es ect to h o d s M1s1sad1_g_2r_Dsgeet1xe_Ad_ertising 7 November 1961 RR 2d Current Service 11: 205 (1987).: and genter_fier_Lag_ang 592131.291121 23 RR 2d 137 (1971) 372ubl1s_A1111udes_I2_ard_Te15.1sion_and_91hsr_ue_1a in a Time at Qua age, p. 22. However, results must be considered in light of the fact that this study is sponsored by a television industry group. 38H2!_£2_H§§_I§l§!i§1227 p. 6- An industry 9rOUP also sponsored this study. 39Calculated from data in the Statietiea1_5betraet_gr tne_uniteg_§tatee;12§1, 107th Ed. (Washington, D.C.: U.S. Dept. of Commerce, Census Bureau, 1986), p. 538, Advertising-Estimated Expenditures by Medium. 23 40Benjamin J. Bates, "Economic Theory and Broadcasting," Presented to the Mass Communication Theory and Methodology Division of the Association for Education in Journalism and Mass Communication Convention, Memphis, Tennessee, August 1985, p. 5. 41Bruce M. Owen, Jack H. Beebe, and Willard G. Manning, Jr., "Chapter 4--The Behavior of the Networks," Telexieien_82202112§ (Lexington. Mass-: Heath. 1974). p. 91-116. 42Barry R. Litman, "The Television Networks, Competition and Program Diversity," EQQIDQ1_Q£ BreaQeaeting 23 (Fall 1979), p. 393-409. 43Barry R. Litman, "The Economics of the Television Market for Theatrical Movies," leurna1_er_ggmmuuieatieu 29 (Autumn 1979), p. 20- 33.: and Barry R. Litman, "Decision- Making in the Film Industry: The Influence of the TV Market " Journal_91_92mmunis_1120 32 (Summer 1982).p- 33- 52. 44Barry R. Litman, "Measuring Divestiture of Network Owned Television Stations: An Econometric Approach," The Antitru§1_flulle110 25 (Summer 1980). p- 363-76- 45Robert H. Prisuta, "The Impact of Media Concentration and Economic Factors on Broadcast Public Interest Programming," figurnai_efi_fireageaeting 21 (Summer 1977), p. 321-32.: and Litman, "Public Interest Programming," op. cit. 46Michael O. Wirth and James A. Wollert, "The Effects of Market Structure on Television News Pricing," Qeurual s ' 28 (Spring 1984), p. 215-24.: Barry R. Litman, "Market Share Instability in Local Television News," lgurne1_gf_areadea§1109 24 (Fall 1980). p- 499- 514.: and Robert Prisuta, "Local Television News as an Oligopolistic Industry: A Pilot Study," Qeurna1_efi Brgageaetiug 23 (Winter 1979), p.6l-8. 4‘7See Benjamin J. Bates, "Determining Television Advertising Rates," in Robert N. Bostrom, ed. semmun1eet1en_xearbeek_l (Beverly H1118. Ca-= Sage. 1983), p. 462-75, especially p. 462-3 for a discussion of previous research on rates. 24 48Michael O. Wirth, "The Effects of Crossmedia Ownership on Television and Newspaper 'Prices', Unpublished Doctoral Dissertation, Mass Media Ph.D. Program, Michigan State University, 1977. See p. 11-25 for a discussion of other ownership studies, as well as Compaine, op. cit. 49James A. Wollert and Michael O. Wirth, "Deregulation of Commercial TV in the USA," Telesommunigation§_£21121 6 (September 1982). p. 155-63- 50Wirth and Wollert, p. 217. 51Share is defined as the percentage of the total Households Using Television (HUT) and Persons Viewing Television (PVT) reached by a station during a specified time. So, any station with at least a .1 percent share of the persons viewing television at a given time is included in this report, for example. See the AD1_yieuing Allegatieh_3euert, cited below, for an explanation of terms and survey methodology. Share and other terms are defined in the Glossary of Arbitron terms on pp. xvi-xvii. A description of the viewing estimates reported, survey data, and survey limitations is found on pp. iii-xv. 52 o 'n s ev's' ° V w' oc ' n Beuert (New York: Arbitron Ratings Co., 1987), February 1987 (Sweeps Period from February 4 - March 3), p. iii. 53See In_1h__Matter_214Th__Bexi_1en_of_£regremming o n P s a . . 0. am ," ;e.ls Hm .i'..'.'.- TeleviSiOh fitatiens, 94 FCC 2d 678 at 699, 715-7 (1983): and 98 FCC 2d 1076 at 1103, for commercialization studies examining only one region of the country, and only VHF stations. See Nicholas Johnson, Breedsa_ting_1n_Ameriea1 1"1'01-1'0 O ' S 1‘ 0' Marketa, 42 FCC 2d 3 at 25- 30 (1973), for an examination of commercialization practices of network affiliates in the top 50 markets. This was a case study prepared by FCC Commissioner Johnson and his staff and seminar students. 54Barry R. Litman and Jan LeBlanc Wicks, "The Changing Advertising Market for the U. S. Television Networks," John D. Leckenby, Ed. The Preceedinge e: the H 01“1‘° -..-demociva#1-. (Austin, Texas: John D. Leckenby, College of Communication, The University of Texas at Austin, 1988), pp. RC27-33. 25 55Michael L. Ray and Peter H. Webb, "Three Prescriptions for Clutter," o v Beeeareh 26 (February/March 1986), p. 69-77.: Michael L. Ray and Peter H. Webb, dve is'n fective owde ev o v' nme t-- e o t o. 7 - (Cambridge, Mass.: Marketing Science Institute, September 1978), 38 pp.: Michael L. Ray and Peter H. Webb, erime ta Research n t e ffects of C utt r- i 'c e a v' o men -- e o - 192 (Cambridge, Mass.: Marketing Science Institute, April 1976), 50 pp.: and Marvin S. Mord and Edith Gilson, "Shorter Units: Risk-Responsibility-Reward, " Journalmef Ad_er11§1ng_Be§earsh 25 (August/September 1985). p 9-19- 56Eric J. Zanot, "Unseen But Effective Advertising Regulation: The Clearance Process," Jgurha1_ef_5gxert1§ihg 14 (1985), p. 44. 57Leon C. Smith, "Local Station Liability for Deceptive Advertising." leurnel_21_8roadsestins 15 (Winter l970-1),p. 107-12. 58Gordon E. Miracle and Terence Nevett, yeiuhtary Begulatien_gf_Adxertising (Lexington. Mass.: Heath. 1987). p. 200. 59Ibid., p. 231, Note 2. 6098 FCC 2d 1076 at 1114. 61Robert O. Slater, "Organization Size and Differentiation, " In Samuel B. Bacharach and Stephen M. Mitchell Eds.. Be§ear2h_1n_1he_S2cielesx_91_9rganizatien§ 4 (Greenwich, Conn.: JAI Press, 1985), p. 128. 62Bruce A. Linton, "Self-Regulation in Broadcasting Revisited, " Qeurhaii_h_Quarter1y 64 (Summer-Autumn 1987),p. 488. 6323 RR 2d 187 at 189-90. 64Robert Dubin, "Technical Characteristics of a Bureaucracy," In Robert Dubin, ed., fluuah_3e1atiehe_in s (New York: Prentice-Hall, 1951), p. 156. CHAPTER II LITERATURE REVIEW This study attempts to incorporate perspectives from a number of research areas to explain and predict commercial television performance. Economic theory and its application to the mass media is reviewed first to illuminate the underlying characteristics of television markets nationwide and what conduct may be expected. The basis for FCC policy is established and policies regarding commercialization, clutter, clearance and complaint practices are explained and interpreted. Organizational theory is then discussed to show how it can be used to explain and predict differences in advertising practices. 0 om' : The FCC deregulated television based on the assertion that competitive market forces now regulate the television industry.1 An analysis of television markets using economic theory allows us to determine their level of concentration and consider whether they are actually competitive. It also suggests what conduct is expected in TV markets and whether performance is "good." 26 27 Indeed, the fundamental assumption in economic theory is that "good" performance is what society wants. Performance is good when a firm can produce enough of a product at a fair price to satisfy consumers and earn a fair profit as well. Performance is also considered good if a company conducts research and incorporates technological innovations into production in a timely manner. Economists also look for a company to employ as many people as possible and pay a fair wage. The underlying ideal is that any outcome is fair to consumers, workers and firms. Good performance implies maximum satisfaction of all of these goals.2 Economists use the industrial organization paradigm to assess performance. The paradigm states that market structure predicts market conduct which predicts market performance. There are four structural models (perfect competition, monopolistic competition, oligopoly, and monopoly) which yield predictions about firm market behavior. "Actual markets are in fact located somewhere between the polar extremes of perfect competition and monopoly,"3 suggesting a continuum of market concentration. 28 Under perfect (or pure4) competition, there are many buyers and sellers, producers' products are homogeneous, and market entry is relatively easy. Since each firm's market share is small and products are similar, none can influence price significantly by increasing or reducing output. Perfect competition is favorable to consumers because prices are naturally kept close to actual producer costs (or marginal cost), and the market dictates price changes and production decisions.5 At the other extreme, monopolies are characterized by a single seller, a unique product, and high barriers to market entry. These firms decide how many units to produce and how much to charge for them. The result is reduced output at higher prices to earn higher profits, rather than producing more at a cost affordable to more consumers. Goods are more expensive than under perfect competition, so performance is not as good under monopoly.6 Closer to pure competition, monopolistic competition is characterized by many sellers who each sell a somewhat differentiated product. Differentiation is defined as any significant basis for distinguishing the goods or services of one seller from another, such as convenience of the seller's location to a group of consumers or creating 29 differences in consumers' minds through advertising.7 Sellers compete by varying product quality, trying different sales techniques, or developing creative advertising campaigns, rather than by price. Performance is not as good as under perfect competition, as the result is "too many" firms producing "too many" similar products, wasting society's resources.8 Both monopolistic and competitive elements are present in monopolistic competition. For example, each retailer is unique because of his store location, reputation, and product line, etc. This monopolistic aspect gives him some control over supply. Yet each retailer competes with others having different characteristics, reputations and locations. This is its competitive aspect.9 Oligopoly, closer to monopoly, is characterized by a few sellers who recognize their interdependence. Oligopolists consider how their competitors may react when making price and output decisions. Interdependence is actual or perceived, resulting in coordination and anticipation of the actions of competitors.10 However, because oligopolists' fortunes are not independent, simply being conscious of rivals and anticipating their moves does not necessarily represent "tacit agreement."11 30 Because oligopolists, like other firms, seek to adjust their prices and products to maximize their profits,12 the tendency is to consider both their direct and indirect influence on price or product changes. For example, if a seller assumes that his actions do not affect his rivals, he only considers his direct influence on price. But if he seeks to maximize profits, he must consider how his actions affect his rivals, and then ultimately himself again. He thus considers his total influence on price or product.13 But these actions are complicated by different selling and production costs, and by uncertainty. Oligopolists have different costs because their products "possess distinctive features and vary widely among themselves in size and quality."14 Uncertainty arises because a seller may not know how rivals may respond to a price or product change, how sophisticated or "far-seeing" they are, and ultimately how their responses will affect his own fortunes. This last factor is aggravated by imperfect knowledge about the extent of buyer preferences for his product.15 Since oligopolists tend to cooperate, they avoid price competition in favor of indirect competition through disguised price cuts (i.e., improving product quality, 31 providing free product-related services such as delivery) and through non-price competition (by differentiating products through advertising). Higher prices and profits for all are generally maintained, because the "winner" improves his market share for the short term, until the "losers" counteract with new ad campaigns of their own. Performance under oligopoly is not good because it often results in a transfer of income from consumers to producers, and a better opportunity to prevent entry of new firms (i.e., due to high advertising costs).16 However, collusion is prohibited by the Sherman Antitrust Act,17 and becomes more difficult as the number of firms increases, products become more heterogeneous, or the more firm cost structures differ. It is also difficult to achieve because by fixing prices, firms have a natural temptation to cheat and temporarily increase market share by selling below the agreed upon price.18 It is also difficult to tell when firms stop "cooperating" and start "competing." For example, monopoly ends and oligopoly begins when the number of sellers increases from one to two, but it is hard to specify when oligopoly becomes monopolistic competition-- at 5 firms or 10 firms? Generally, if the sellers are so 32 few that each believes his profits are influenced by his competitors actions then the market is considered an oligopoly.19 Yet oligopolistic behavior is applicable to larger groups of competitors, divided into subclasses. Products, like clothing, often fall into distinct quality or price classes, appealing to groups of consumers with differing incomes or tastes. A seller often competes with just a few class members, and avoids price competition since a price cut forces his closest competitors to follow suit, but not others.20 "Chain" linking of markets is another manifestation of oligopolistic behavior in "large" groups. Gas stations, for example, represent a market where each seller is linked by location to his nearest competitors. Degree of competition lessens with distance, and subclassing is illogical as stations on the borders of arbitrarily assigned areas compete more with competitors on the adjoining border than others in their own "area." So a price cut by one seller might influence his closest competitors only, or force them to meet it and in turn force others to do so indefinitely throughout the chain. A single seller may ultimately bring about a group change, making consideration of indirect influence a factor in "chain" markets.21 33 Economic Theory Applied to Commercial Television: Television markets are oligopolies22 as a result of FCC license allocation policies.23 Analysis of the networks suggests the behavior expected in three station markets. The networks may compete (or cooperate) by manipulating certain factors to gain a competitive edge (or achieve parity to limit it), including: the price of advertising time: the number of minutes of commercials per hour: the quality, type, and scheduling of programming offered: and payments to affiliates. The networks are evidently aware of their interdependence, manifested by "the jockeying over the fall schedule and midseason changes."24 The networks apparently cooperate in areas where cheating is easily detected and responded to, such as the price and quantity of advertising time, because the market is well organized and informed. Agreeing on program popularity and ratings25 is much more difficult, and competitors cannot respond quickly: programming schedules must be rearranged or new programs found to replace canceled ones.26 34 Before the NAB Code's demise, the networks scheduled three minutes of commercial time per half hour of prime time, enabling them to restrict the supply of time and the range of competition.27 Evidence suggests they still do, as all exceeded the old guideline a bit, with the number one network selling slightly more time than the others.28 The networks may survive other disruptions, such as changes in relative market shares, because the underlying market structure is unchanged.29 One might guess a transition period followed deregulation and the Code's demise, with equilibrium and cooperation eventually reinstated. Bates found that a station with higher ratings attracts more advertisers and probably charges more for commercial time.30 A few economic studies noted that audience size, market size, market concentration, the number of stations in a market, network affiliation, broadcast band, market quality and the presence of cable TV all affect station rates.31 Some of these same variables, such as network affiliation, broadcast band, audience size and the number of stations in a market, are also used as proxies for profitability.32 Type of ownership did not influence rates or profitability.33 35 Commercialization was included as a dependent variable in two studies. Prisuta suggested that market profits and network affiliation account for a small amount of variance in overcommercialization, with stations increasing commercial time to take advantage of surplus advertising revenue.34 However, these results were not replicated in a similar study by Litman.35 Profitability may also be an important influence on station advertising practices. For example, profitable stations can afford to reject some types of advertising that less profitable stations must accept. Clearance may represent a means of choosing between advertisers to maintain subclass membership. Large market stations, especially those in the top 50, evidently sell more of their commercial time inventory to national or regional "spot" advertisers, while smaller market stations sell more local ads.36 Some differences in clearance, complaint, commercialization, and clutter may therefore be a function of station size and the geographic focus of advertising it accepts. 36 Mase Media Eeiiey Literature: Prior to deregulation the FCC instructed local stations on how to meet community needs in policy statements, guidelines and rules for meeting the public interest. FCC enforcement of these practices served as a sort of "countervailing power37" against the oligopolistic tendencies it created in TV markets. Consideration of economic theory in tandem with policy literature broadens the basis for predicting how stations behave and perform, based on past FCC expectations. It also allows a comparison to study results to see if market forces truly represent a new countervailing force replacing the FCC. Background: The Federal Communications Commission's mandate to license and otherwise regulate broadcasters "as public convenience, interest, or necessity requires" originated in the Radio Act of 1927 and was retained in the Communications Act of 1934.38 Congress amended the 1927 act to ensure that stations were located across the country, thus establishing the localized nature of broadcasting.39 From the beginning, television licenses were allocated geographically so as many communities as possible could have at least one TV station.40 37 Interpretation of localism according to the public interest standard resulted in a principle of individual decision-making by broadcasters. The individual licensee is responsible for selecting and presenting broadcast matter, and "fulfillment of the public interest requires the free exercise of his independent judgment."41 As applied to advertising, licensees must "take all reasonable measures to eliminate deceptive advertising," and "avoid abuses with respect to the total amount of time devoted to advertising continuity as well as with which regular programs are interrupted for advertising messages."42 The FCC also decided that ascertainment of community programming needs was a principal ingredient of a licensee's public interest obligation. The concept of ascertainment originated in the 1960 Eh_hahe_£regramhihg Thguiry, where it was first noted that licensees should try to discover local broadcasting "tastes, needs and desires,"43 presumably to enhance their individual judgment. In 1961 the FCC rejected a license application where no effort was made to ascertain local needs, and which included the program schedule from other commonly- owned stations instead.44 Ascertainment was later expanded to include "commercial content and practices" as well as programming.45 38 Two Primere specified ascertainment methodology, including the number of interviews to be made, the type of leaders to contact, what survey methods might be utilized, and how to report survey results.46 Licensees are no longer required to follow these procedures, with the FCC now focusing on how responsive programming is rather than how a licensee arrives at programming decisions. The FCC thus identified two major areas of broadcaster concern regarding advertising: commercialization and clearance. However, given deregulation and the relaxation of relevant policies, the meaning of "the free exercise of independent judgment," and what constitutes "reasonable measures" regarding deceptive advertising and "abuses" of commercial time might vary significantly. And licensees may not consult community members on advertising concerns as readily since ascertainment methodology has been relaxed. Commercialization and Clutter: The FCC has maintained that commercialization is an important element in judging a licensee's overall program performance.47 Yet it has repeatedly declined to promulgate commercial time rules. In 1928, the Federal Radio Commission said that while advertising was necessary to support broadcasting,48 39 "stations are not given these great privileges...for the primary benefit of advertisers. Such benefit as is derived by advertisers must be incidental and entirely secondary to the interest of the public." It implied a commitment to this principle by requesting information about time sales in license applications from the beginning.49 And commercialization was a factor in license actions as early as 1928.50 Yet this strong language was not translated into meaningful action. Congress was evidently dissatisfied with the FRC's efforts, because in 1932 the Senate passed a resolution instructing it to survey and report on overcommercialization practices, including what plans might be adopted to limit advertising.51 But no commercial limits resulted. In 1946 the FCC again stated its concern with overcommercialization, defining it to include the amount of time devoted to advertising, the length of individual commercials, the number of commercials in a break, and the amount of programming time between spots. Yet it again seemed reluctant to take decisive action itself:52 "There is need, however, for a thorough review by the industry itself of current advertising practices, with a view towards the establishment and enforcement of sound standards by the industry itself." 40 It also noted that it considered the elimination of commercial advertising excesses a factor in determining whether a licensee's proposed program service was in the public interest when issuing and renewing licenses.53 In 1961 the FCC instructed its Broadcast Bureau to submit applications proposing more than 12 minutes of commercials per hour to the Commission for 54 Rulemaking resulted from the many consideration. applications with commercialization problems, supported by public complaints. The FCC lamented the failure of self- regulatory solutions but at the same time noted it preferred industry rather than FCC resolution.55 It did not adopt rules because of congressional and industry opposition,56 deciding to review commercialization in the overall evaluation of station performance on a case-by-case basis instead. It emphasized that ascertainment must include commercial content and practices.57 In the late 1960's and early 1970's the FCC began to note that some broadcasters were "subordinating programming performance in the interest of the public to programming in the interest of salability" by airing program length commercials.58 Minor fines and admonitions were leveled to deter the practice.59 The FCC finally 41 banned program length commercials in 1973,60 and reiterated its commitment to eliminating them in 1974.61 Broadcast Bureau delegations regarding commercial time were revised in 1973 to reflect industry standards.62 Applications proposing more than 16 minutes of commercial matter per hour would be forwarded to the Commission for review,63 a change that implicitly approved increasing commercial time from 12 to 16 minutes per hour. In 1981 the FCC again refused to institute a time rule due to declining viewer complaints and recent First Amendment protection afforded commercial speech.64 The FCC therefore has illuminated the principle it wants licensees to adhere to, but has taken no meaningful action to ensure compliance, expressing a desire for the industry to do so. A few studies examined commercial practices. In 1973, departing FCC Commissioner Nicholas Johnson prepared a report assessing the program performance of network affiliates in the top 50 markets, including commercialization. Despite his contrary opinions, his analysis suggested that stations rarely exceeded the 16 minute guideline.65 Further analysis of this data demonstrates that one third of stations violated the guideline at least once, with violators equally distributed among the networks. In 42 a fourth of the markets, two or more affiliates exceeded 16 minutes at least once,66 supporting the notion of cooperation in oligopolies. An FCC staff study found 99.67 percent of all broadcast hours within guidelines. Affiliates aired more commercial time than independents, averaging just under 11 minutes per hour.67 An NAB study found 99.84 percent of broadcast hours were within guidelines, with affiliates averaging 10.4 commercial minutes per hour and independents 9.6.68 These results suggest that more profitable stations (e.g., network affiliates) sell more time. Another NAB study found children's programs contained about 8 1/2 minutes of commercial time and about 13 minutes of non-program time.69 Questions may be raised about these studies. Reporting statistical averages may hide overcommercialization practices.70 A study of network affiliates in New York City provides support, as all three averaged 13.5 minutes of non-program material during prime time, when only 9.5 minutes is allowed.71 Most of the data for the FCC and NAB studies were collected prior to the code's demise. The FCC sample only included 60 stations from 3 southern states, while the NAB 43 study included only VHF stations. Neither was national in scope nor encompassing different types of stations, and findings are dated. Most stations have policies regarding time units (i.e., whether or not they accept 15 second spots or program length commercials), the number of commercial breaks per hour, the number of commercials per break, the number of commercial minutes per hour, product protection (i.e., guaranteeing a certain time will elapse between competing advertisements), and product acceptance (i.e., banned product categories such as contraceptive ads).72 A recent study suggests stations are less concerned with product protection and time standards, especially less profitable ones.73 Clutter research supports the notion that the more ads there are, the less attractive commercial time is to advertisers. As the amount of non-program material increases there is a consistent decrease in commercial effectiveness (i.e., correctly recalling the ad itself or the sponsor's brand name). Since viewers only recall about four commercials at any one time, the more ads there are in a break, the lower the probability that any one will be remembered.74 Another study provides support, finding that as television became more cluttered due to 44 the shift from sixty to thirty second spots, correct brand identification declined.75 Research also suggests that 15-second spots are less effective than 30's, especially when included in the same break.76 Viewers learn when to expect breaks and thus avoid commercials.77 They also notice when the number of spots in a break is increased, especially VCR owners, cable subscribers, and viewers between 18 and 34. Viewers believe the extra commercial time is taken away from programming, blame advertisers for it, and as a result become more negative toward ads.78 Studies about the environment that ads appear in (e.g., which program, its type, and how popular it is) suggest an increase in the number of commercials leaves viewer interest in a program unchanged.79 Researchers disagree as to whether commercials interrupting interesting programs are less effective,80 or more effective, especially if the commercial itself is interesting.81 However, a number of studies found that program type or environment affects commercial effectiveness.82 And a quality image or environment is important to advertisers, as they apparently pay extra to be associated with the number one network.83 45 Advertising Clearance and Complaint Practices: The history of an expressed principle accompanied by inaction is also found in FCC policy regarding deceptive advertising. As early as 1939 the FCC noted84 "...certainly stations should carefully investigate... enterprises to which they are requested to lend their facilities..." especially those the broadcaster knew to be of questionable character. Specifically,85 "It is enough that advertising continuity of such character be offered a station to cause it to carefully scrutinize the same, as well as to thoroughly investigate the particular scheme to be presented." The FCC originally attempted to regulate advertising content and noted techniques to avoid, including appeals based on patriotic feelings during times of war: "physiological" commercials regarding ailments, body odor, and "sluggish bile": propaganda in commercials (e.g., commercials advocating a point of view on an issue rather than a product or service): and the intermixture of programming and advertising, which made it difficult for listeners to know when the program ends and the ad begins.86 Although the FCC later relinquished responsibility for content regulation to the Federal Trade Commission, it retained some control. It addressed the intermixture of 46 programming and advertising in the ban of program length commercials,87 and addressed issue ads advocating a point of view in the fairness doctrine, which required fair presentation of both sides of a controversial issue.88 According to the FCC/FTC liaison, the FTC was to determine whether advertisements were deceptive,89 and the FCC would:90 "...continue to take into account pertinent considerations in this area in determining whether broadcast applications for license or renewal of license shall be granted or denied and in the discharge of other statutory responsibilities." In other words, deceptive advertising complaints were to be handled by the FCC only if they involved local matters.91 The FCC noted, however, that licensees did have obligations regarding deceptive advertising:92 "Broadcasting licensees must assume responsibility for all material which is broadcast through their facilities. This includes all programs and advertising material which they present to the public. With respect to advertising material, the licensee has the additional responsibility to take all reasonable measures to eliminate any false, misleading, or deceptive matter...This duty is personal to the licensee and may not be delegated."93 The FCC explained in 1961 that clearance responsibilities included not only reviewing copy, but taking reasonable steps to assure every prospective advertiser's "reliability and reputation" and "ability to 47 fulfill promises made to the public over the licensed facilities,"94 especially those of "questionable character."95 Licensees must also familiarize themselves with the charges when an ad is the subject of an FTC complaint, and "responsibly" determine whether to continue broadcasting the ad.96 The FCC's "reasonable diligence" standard requires licensees to keep abreast of FTC policy on deceptive advertising and other "obvious areas of concern," such as the reliability of local advertisers. When a manager had grounds for doubt or a significant complaint, he should call for substantiation of any factual claim.97 The FCC also expects larger and more profitable stations to make a "correspondingly greater effort" in this area, although all stations must have at least one staffer responsible for clearance.98 In 1973 the FCC instructed its Broadcast Bureau to submit for review license applications or renewals which were the subject of a final FTC cease and desist or consent order during the three-year period preceding the application.99 But review by the Commission did not guarantee action. 48 The FCC paid special attention to clearance practices when a station prepared copy itself, or advertised itself, expecting more diligence when licensees directly examine and/or prepare ad claims. It also acted when a station neglects to implement adequate clearance practices.100 In 1985 the FCC eliminated "unnecessary" broadcast regulations. As a result, stations are no longer required to have a deceptive advertising program, to exercise particular care when deciding whether to accept advertising which is the subject of an unadjudicated Federal Trade Commission complaint, or to review the reliability of every prospective advertiser.101 Violations of the deceptive advertising policy are now considered in character proceedings, where the FCC focuses on determining "truthfulness" and "reliability" by considering whether the licensee violated the FCC Act, or FCC rules or policies.102 It also considers "fraudulent statements to government agencies, certain criminal convictions, and violations of broadcast related anti- competitive and antitrust statutes."103 The FCC determines the weight to give to misconduct by considering its willfulness, frequency and recency. It also considers how serious the infraction was, the nature of the manager and owner involvement, and efforts made to 49 remedy the situation. An applicant's record of compliance with FCC rules and policies is normally considered as well.”4 Only a knowing presentation of a deceptive ad is included in a character qualifications matter. The licensee must either knowingly create a deceptive ad, or knowingly broadcast one which is the subject of a final FTC action. Or if licensee actions involving deceptive advertising result in an adjudicated violation similar to those discussed in relevant non-FCC misconduct, then the behavior may be considered.105 The FCC has thus taken the approach that the licensee must be proven guilty before any action is taken. This represents an implicit decision to ignore questionable advertising practices until the FTC or another government body finds them illegal, or they become so offensive that the public, advertising industry or competitors complain in such great numbers that they cannot be ignored. Such practices may therefore develop and go unregulated for years until legal proceedings are finalized. This is in direct contradiction to an FCC principle expressed in 1951: 50 "It is well known that once certain practices develop, it is exceedingly difficult in applying corrective measures to restore the situation to the same healthy conditions that would have prevailed had not the restrictive conditions been permitted to arise at all."106 The deceptive advertising policy was relaxed without hearings because notice and comment is not required when general statements of policy are eliminated.107 The FCC therefore eliminated it without apparently discerning whether stations were actually complying with it, or if it was truly burdensome. It is also unclear why the FCC decided it was burdensome for stations to have at least one person familiar with deceptive advertising policy. As many stations check into an advertiser's ability to pay, a brief background check does not seem burdensome either. Stations probably ask for substantiation of an ad when a "significant complaint" exists anyway simply to placate feuding advertisers. Keeping such requirements "on the books" only makes it easier for the FCC to identify violators, rather than imposing burdens. Also consider that the former FCC deceptive advertising policies were "narrow in scope."108 Yet the FCC narrowed the scope of character qualifications (the area in which deceptive advertising is now considered) because they were "overly broad," and to make the process 51 more "equitable and efficient."109 By explaining the types of infractions it considers in character qualifications, the FCC treats licensees consistently, eliminating different findings in cases with similar facts. Additionally, narrowing reduces the amount of time and resources spent in examining such questions.110 One must wonder why the FCC chose to take the "narrow" approach in the deceptive advertising area and replace it with the "narrow" approach in character qualifications. Retaining at least some guidance as to what the FCC considers "reasonable measures" appears to accomplish the same efficiency goals, and maintain consistency in the FCC's regulatory approach. Most stations apparently have mechanisms for handling "deceptive advertising." The sales and traffic department handle clearance and advertiser complaints, and the traffic manager is responsible for monitoring advertising content. The General Sales Manager oversees traffic, and the station General Manager makes the final decision on problem ads.111 Most stations have policies for political advertising, product and copy acceptability, issue advertising, product protection (i.e., separating competing ads), and time standards.112 Larger stations 52 may have policies for mail and direct selling accounts (i.e., buying products by calling an "800" number), bait and switch advertising, contests and games, demonstrations, medical products, free offers and guarantees.113 Stations might investigate advertisers by consulting the Better Business Bureau, reviewing its Code of Advertising or its National Advertising Division's Case Reports on national advertisers. The NAB and the networks can also provide advice.114 Larger stations may also contact advertisers or visit a store.115 Stations might review ads by perusing scripts, storyboards, a product sample or a label or package insert, before and/or after production. Claim substantiation or authentication of demonstrations or testimonials could be requested as well.116 Commercials may be challenged by competing advertisers, who must present supporting data. Outside technical expertise may be called in, and ads withdrawn only if the complaint is found valid. Viewer complaints may also be investigated, and ads possibly discontinued until complaints are resolved.117 53 Fairness Doctrine: The relaxation of the FCC's deceptive advertising policy becomes more ominous when one considers that the Fairness Doctrine no longer exists.118 Licensees have more discretion in deciding whether or not to accept controversial issue ads for broadcast, and could conceivably accept one without conducting a background check or selling time to an opposing camp. Ironically, another major concern was expressed by the FCC in 1946. It noted that the public interest required that "ample play for the free and fair competition of opposing views" was necessary "to all discussions of issues of importance to the public,"119 and supposed:120 "If time is also...to be sold for the presentation of a point of view, what precautions are necessary to insure that the most time shall not gravitate to the side prepared to spend the most money?" None, evidently. At first, the FCC did not consider licensee editorializing to be a proper part of the discussion of public issues,121 but later changed its position.122 Broadcasters were to adequately cover issues, reflect opposing views and pay for coverage if sponsors were unavailable.123 Editorializing was eventually recognized as a major component of meeting the public interest.124 54 The FCC explained in the 1964 £airneee_3euert how to decide whether to afford a reasonable opportunity for presenting alternate views on controversial issues. A licensee must make a reasonable judgment in good faith based on the situation, and may decide whether the issue is truly controversial, what viewpoints to present, the format, and spokespersons to present them.125 The Supreme Court upheld the doctrine's constitutionality in 1969.126 The FCC adopted the Zapple rule in 1970, which requires a licensee to sell time to supporters of opposing candidates during a campaign when it sells time to one camp. However, it does not require that free time be given to respond to paid ads.127 The 1974 Fairness Report provided guidance on how to determine if an issue constituted a matter of public controversy. Considerations included the level of public debate, whether and how much the issue appeared in the local news media, and the existence of an election question on the matter.128 Broadcasters had broad discretion in deciding how much time to devote to an issue and how to present it. In 1984 the Supreme Court declined to reconsider the fairness doctrine "without some signal from Congress or the FCC that technological developments have advanced so 55 far that some revision of the system of broadcast regulation may be required."129 The FCC responded in 1985 that the scarcity rationale for broadcast regulation was invalid as the explosive growth of information sources gave the public access to numerous viewpoints, making the fairness doctrine obsolete. However, it declined to act, asserting that Congress and the courts should do so.130 The fairness doctrine's demise was the result of an FCC decision that the Meredith Corp. station, WTVH in Syracuse, New York, violated the doctrine by broadcasting ads favoring the construction of a local nuclear power plant without presenting opposing points of view.131 Meredith appealed to the U. S. Court of Appeals, D. C. Circuit, arguing that the FCC did not adequately consider the petitioner's arguments regarding the Fairness Doctrine's constitutionality. The Court agreed and remanded the case for that purpose.132 After much prodding, the FCC considered alternative courses of action,133 and finally decided the fairness doctrine violated broadcasters' First Amendment rights and no longer served the public interest. It reiterated that the scarcity rationale was outmoded and that the same First Amendment principles should be applied to both the print and electronic media.134 56 W: The review of mass media economic and policy literature suggests the behavior and performance to expect of commercial television stations, as manifested in advertising practices. Adding organizational theory to this analysis illuminates how the nature of the organization itself alters these predictions. An organization constitutes "two or more people working together to achieve a common goal."135 "The defining criterion...is the existence of procedures for mobilizing and coordinating the efforts of various, usually specialized, subgroups in the pursuit of joint objectives."136 Bureaucracy refers to an organization's administrative component, which is responsible to top management for coordinating and executing policy and administrating funds.137 Bureaucrats are "the members of management who stand between top policy makers and workers."138 Sales and traffic managers in television stations fit this category, as the sales manager insures organization policy regarding advertising is carried out, and the traffic manager reviews advertising content. 57 In bureaucracies, a specialized and expert administrative staff maintains the organization and its lines of communication. Organizational duties are partitioned among employees in positions that are organized into a hierarchical authority structure. Rules and regulations guide their decisions and actions.139 Size is the most important cause of bureaucracy. As organization size increases, more types of tasks are carried out by more specialized staffs. Communication is more difficult140 since direct access to the top manager is limited by the time she devotes to her job duties and her knowledge of the organization's functions. Delegation of tasks to an intermediary becomes necessary,141 and a system of official rules is implemented to standardize operations and restrict direct supervision to unusual situations.142 Larger organizations probably use formal rules, including detailed descriptions of job duties, or keeping written records of work performance. Control by rules seems more efficient and less costly for larger organizations, because job routinization makes it easier to supervise more employees, especially when supervisors and subordinates are separated by several hierarchical levels.143 58 Smaller organizations probably use surveillance, such as closely supervising employees or inspecting the quality of work. Surveillance requires personal observation and contact between supervisors and employees, and seems more efficient and less costly for smaller organizations where both often work side by side.144 Organizational charts show that clearance and complaint formality may vary by station size. Large market stations employ more people than smaller ones,145 and station policy and communication formats vary. Stations reported codifying policies in a manual, having mostly written policies, using mostly memoranda, using the last NAB Code, or conveying policies verbally.146 Written policies may be rules of varying formality and verbal conveyance, surveillance. Staff policies are communicated either through existence made known, staff discussion, staff encouraged to read and staff required to read.147 Existence made known and staff discussion (less to more formal surveillance) is probably found in smaller markets and staff encouraged and required to read (less to more formal rules) is found in larger markets. Therefore the larger the station, the more traffic and sales employees there are, the more people involved in clearance and complaint practices, and the more formal the policies. The literature review suggests that changes in advertising practices utilized by commercial television stations nationwide may have occurred. Economic theory suggests that stations may fall into different classes and exhibit different conduct. Given the FCC's elimination or relaxation of all of its important policies regarding advertising, licensees now have broad discretion on how much commercial time to schedule, what ads to accept, and what clearance and complaint practices to utilize. Market forces may now be more decisive than in the past in determining these behaviors. Commercialization and clutter may have increased as a result of deregulation. Stations occasionally violated FCC and NAB guidelines before deregulation, so they may be more likely to do so now. It also appears that clearance and complaint practices may vary in formality because of deregulation and the demise of the NAB TV Code. But these are merely plausible outcomes that require empirical research to substantiate. Organizational theory also suggests that differences in the type and number of advertising policies, and in how those policies are communicated, may be expected based on organization size. Advertising policies may have changed 60 over the past few years, since managers may now feel free to adapt policies to their individual stations whose needs and constraints may vary because of organization size. Television now stands at a point where sweeping changes may occur in the practices licensees utilize. The events of the past decade represent a complete turnaround of FCC policies existing since regulation was instituted. And some policy changes occurred without any independent examination of whether and how well licensees complied with them. It seems prudent to examine them now to attempt to evaluate station performance before it becomes too difficult to eliminate any negative effects which may result. The conclusion from the literature review is that the amount and content of commercial information available to consumers may be changing. But how might these changes be expected to occur? A proposed theory of how stations utilize these practices will now be presented. NOTES: CHAPTER II 1In_1he_Ma11er_2f_The_Re_1e120_of_£regramm1ns_and Qumnereial izatien Eoiicies, Ascertainment Be eguiremente, -.10 PO -1,°° '--_1--m !_S _0 cu e o. ‘ V,— 0! eratiane 98 FCC 2d 1076 at 1077 (1984). 2F M. Scherer. Ind__1r111_Market_fitructure_and Eeene_ie_2errernanee (Chicago: Rand McNally 1970),p p. 3-4. 3Ernest Gellhorn. An111rue1_Lau_and_Esonomic§_1n_a Nuteheli 2nd ed. (St. Paul, Minn.: West, 1981), p. 51. 4Edward Chamberlin. The_Theerx_Qf_Menenelistie , (Cambridge, Mass.: Harvard University Press, 1933), p. 6-7. Chamberlin prefers the term "pure" to indicate an absence of monopoly elements, rather than "perfect," because the latter may imply perfection "in many other respects than in the absence of monopoly," including perfect knowledge of the future resulting in the absence of uncertainty, which is unlikely in reality. SEdwin Mansfield. Er1ncinle§_ef_nicr9esenomic§ 3rd ed. (New York: Norton, 1980), p. 217-39.: and Scherer, p. 9-10. 6Mansfield, p. 240-60.: and Scherer, p. 16. 7Chamberlin, op. cit., p. 56. 8Mansfield, p. 261-82. 9Chamberlin, p. 62-3. 10Mansfield, p. 261-82. 11Chamberlin, p. 31. 12Ibid., p. 169. 13Ibid., p. 31-2. 61 62 14Ibid., p. 149. 15Ibid., p. 101. 16Mansfield, p. 261-82. 17The Sherman Antitrust Act of 1890, 15 U.S.C.A. Sections 1-7. 18Mansfield, p. 272. 19Scherer, p. 10. 20Chamberlin, p. 102-3. 211b1d., p. 104. 22Benjamin J. Bates, "Economic Theory and Broadcasting," Presented to the Mass Communication Theory and Methodology Division of the Association for Education in Journalism and Mass Communication Convention, Memphis, Tennessee, August 1985, p. 13. 23The FCC set aside 12 very high frequency (VHF) and 70 ultra high frequency (UHF) channels for television. Yet no more than 7 VHF channels can be used at any location, because the remaining 5 must be unused to prevent interference from adjacent channels. Stations using the same channel must also be separated geographically. Coupled with the fact that the FCC assigned stations to as many communities as possible, the number available to any market was naturally limited, resulting in many 3 station markets. And although UHF allocations were made, many frequencies are unused because signal quality is inferior, making it hard for UHF stations to compete with VHF stations. See Roger G. Noll, Merton J. Peck and John J. McGowan, o c e Teieuieien_3eguiatien (Washington, D. C.: The Brookings Institution, 1973), p. 3-7,116.; Bruce M. Owen, Jack H. Beebe, and Willard G. Manning, Jr., e 's ' 5 (Lexington, Mass.: Heath, 1974), p. 6-7.: and Douglas H. Ginsburg, Regulation of Breadcasting (St. Paul, Minn.: West, 1979), p. 163-6. 63 24Owen et al., p. 103. See Chapter 4, p. 91-116: and Barry R. Litman, "The Economics of the Television Market for Theatrical Movies," geurnai_ej_§ennunieatien 29 (Autumn 1979), p. 20-3, for an explanation of the network cartel. 25A rating is defined as the estimated percent of all television households or persons tuned to a specific station. See Mew To Read Xour Arbitron Televieion Market gengrt, (New York: Arbitron Ratings Co., May 1987), p. 3., for an explanation of basic broadcast research terms and methodology. 26Owen et al., p. 101-7. 27Litman, "Theatrical Movies," p. 22. 28Barry R. Litman and Jan LeBlanc Wicks, "The Changing Advertising Market for the U. S. Television Networks," John D. Leckenby, Ed. The Breceegingeo of the C t e c dem o (Austin, Texas: John D. Leckenby, College of Communication, The University of Texas at Austin, 1988), p. 17. 29Barry R. Litman, "The Television Networks, Competition, and Program Diversity," leurna1_ej Breageaeting 23 (Fall 1979), p. 405-6. 30Benjamin J. Bates, "Determining Television Advertising Rates," in Robert N. Bostrom, ed. Qommunicatign_xearbeok_l (Beverly Hills, Ca-= Sage. 1983). p. 472-4. 31Ibid., p. 472-4.: Michael O. Wirth and James A. Wollert, "The Effects of Market Structure on Television News Pricing." qurnal_ef_nreadca§1109 28 (Spring 1984). p. 222-3.: and Michael O. Wirth, "The Effects of Crossmedia Ownership on Television and Newspaper 'Prices'," Unpublished Doctoral Dissertation, Mass Media Ph.D. Program, Michigan State University, 1977, p. 78. 32Barry R. Litman, "Measuring Divestiture of Network Owned Television Stations: An Econometric Approach, " The Antitru_t_fiull_tin 25 (Summer 1980). p- 368 371- 3- 64 33Wirth and Wollert, p. 223.: Barry R. Litman, "Public Interest Programming and the Carroll Doctrine: A Re-examination." Journal_of_firoadca§11ng 23 (Winter 1979). p. 57.: and Wirth, p. 11-25. 34Robert H. Prisuta, "The Impact of Media Concentration and Economic Factors on Broadcast Public Interest Programming," Qeurnai ef Broageasting 21 (Summer 1977), p. 325-6. 35Litman, "Public Interest Programming," p. 55. 36Wirth, p. 79-80. 37The concept of countervailing power states that large firms cannot capitalize on their market power because they must deal and negotiate with other large firms, large unions and government. The expectation is that the conduct of a large oligopolist becomes more like the conduct of a perfect competitor as a result. See John Kenneth Galbraith, Aneriean_§auita1i_n (Boston: Houghton Mifflin, 1952). A more basic explanation is provided by Mansfield, p. 278. 38William E. Francois. Maaa_Media_Lay_and_Begulation 4th ed. (New York: Wiley, 1986), p. 509. 39Edwin G. Krasnow, Lawrence D. Longley, and Herbert A- Terry. The_2211112a_of_Broadga§t_Begulat1on 3rd Ed- (New York: St. Martin's Press, 1982), p. 14.: and 2 F. R.C. Ann. Rep. 11-13 (1928). 40Basically, the number of stations that could be licensed depended on transmitter power, station location (including a consideration of local geographic characteristics), time of operation and antenna height. See Ginsburg, p. 163-6. The Commission created a Table of Assignments designed to: (1) provide at least one TV station to all parts of the country, (2) provide each community with at least one TV station, (3) provide a choice of at least two TV services to all parts of the country, (4) provide each community with at least two TV stations, and (5) assign any remaining channels to communities depending upon their size, geographical location, and the number of TV services available from TV stations located in other communities. 65 41 - . banc_£rogramming_lnguiry. 44 FCC 2303 at 2309 (1960)- 42Ibid., p. 2313. 4344 FCC 2303 at 2312. 44§uburban_nroadgaster§ 20 RR 951 (1961). affirmed in HQDIY_Y1_£Q§. 302 F.2d 191 (D.C. Cir., 1962), certoriari denied, 371 U.S. 821 (1962). 451n the Matt er gt Amendment ef Part 3 at the m ss 'o les and Re ula i s o , 1° ’1 . 1d 39 -!° ‘ -V S 01 3 0:! :- fitatiene, 36 FCC 45 at 49 (1964). 46 2r1mer_2n_Aa2er1a1nment_21.92mmnnity_£roblem§_by fireageaet_hnniieante, 27 FCC 2d 650 at 682-7 (1971).: and Aacerta1nmen1_2f_C2_mun11y_£robl_m__by_sommers1al Broadca§1_Applioante. 57 FCC 2d 418 at 441- 6 (1976) 4736 FCC 45 at 47. 48§1atemen1_made_py_1he_somm1§§1on_on_bugu§1_211 e t v eres ven e neeeeeity, 2 F.R.C. Ann. Rep. 166 at 168 (1928). Citation refers to the quote that follows in the text. 4931 FCC 45 at 46. 50See the case of station WCOT, 2 F.R.C. Ann. Rep. 152 (1928), where the Federal Radio Commission not to renew the station's license because it was used "as a means of direct advertising." Also see the case of WCRW, p. 155-6, where the station's power was reduced, among other reasons, because it existed "chiefly for the purpose of deriving an income from the sale of advertising." 51Eup112_Sery12e_Bespgnaibil1ty_of_hroadoa§1 Lieensees, Public Notice 95462, 2 July 1946, p. 41-2. More commonly known as the "Blue Book," which it is hereafter referred to. 52Blue Book, p. 43-7.: Cite that follows in the text, p. 47. 53Ibid., p. 12. 66 54In_1h__M_1ter__f_Ihe_Re_1.1on_of_£rogramm1ng_and mer a at on o ic s s nme 0- u 0° {‘21 ‘U‘l - 0 sun '2 ’ 7 °- 0! s_atigne, 94 FCC 2d 678 at 684- 5 (1983). 55N2112e_of_2r2poaed_Bulemaking1_Adyer11§1ng_on s1andard1_EM1_and_Ie1ey1e12n_firoadsaet_§tat1on§. 28 Fed.Reg. 5158 at 5159 (1963). 55In_Be_Beneyals_of_Br2adcaat_Licen§e§_for_Arkan§a§1 Leuieiana and Mississippi, 42 FCC 2d 3 at 26 (1973). This is the dissenting opinion of Commissioner Nicholas Johnson (prepared along with his staff and seminar students) entitled "Broadcasting in America: The Performance of Network Affiliates in the Top 50 Markets. 57Bep2r1_and_Q1der1_Commercial_Adxertising. 36 FCC 45 at 49 (1964). 58These programs were for products/services such as chinchilla ranching, real estate, speed reading courses, and gardening. See Tepper_§erpgratien 21 FCC 2d 148 (1968). Ame_12an_Br2adcaating_gompan1e§1_ln21 23 FCC 2d 132 (1970); Keep-12. Inc . 24 FCC 2d 149 (1970). Columbus Broadga§112g_921 25 FCC 2d 56 (1970): Multimedial_lnsl 25 FCC 2d 59 (1970); National Broadeas sting pg. 29 FCC 2d 67 (1971); Dena Pictures, Inc. 31 FCC 2d 206 (1971). Enani Incl 37 FCC 2d 748 (1972); EE111_1021 38 FCC 2d 411 (1972): Taft Breadcasting 99, 39 FCC 2d 1070 (1973): Eeigel_Broadga_ting_C_1 41 FCC 2d 370 (1973): and Rueh Broadea_ting_gorp_ 42 FCC 2d 483 (1973) 67 59A number of stations were asked to submit a statement as to their future policies regarding program length commercials and told that their performance in this area would be considered in their next license renewal. Some were fined, with amounts ranging from approximately $1,000 to $8, 000 for logging violations, such as failing to identify the sponsor of the program, and/or improperly indicating the start and end of the commercial matter in the program (essentially, all of it). See, for example, 8111.312111111109_12121. 42 FCC 2d 483. 486 (1973): Qhannel §eventeen, Ind,, 42 FCC 2d 529 (1973): Midland Television egrp., 42 FCC 2d 591 (1973); Mid_ueu_xprk firgadgagtihg_ggrpt, 42 FCC 2d 594, 597, 1088 (1973): Turner Broadeasting at North Careiina, 42 FCC 2d 622, 626 (1973): united Television gonpany pf Mew Manpehire, 42 FCC 2d 632, 636 (1973): WKQN-IV. Ind,, 42 FCC 2d 639, 642 (1973); Eyening Meus As seeiatio n. 49 FCC 2d 380 (1974); Eugene Televisien go , goos Bay, Qregen, 61 FCC 2d 1131 (1976): and HBQDQ£Q_§LQQQQQ§LiflQ. 62 FCC 2d 970 (1977). As shown here, violations dropped dramatically after the ban in 1973. Most of these cases involved programs sponsored by National Chinchilla, Inc., which were designed to promote chinchilla ranching in the home, along with associated products and services. 5°In_1he_Ma11er_of_21ogram_Leng1h_12mmercial§. 39 FCC 2d 1062 at 1062-3 (1973). It reiterated this ban in 1n_3e t f' 'on o Ka'se road as n o. O C on - 'no :--L-s - - Tele1ae1_o1_E1ogram_Leng1h_Com_ere1al_. 45 FCC 2d 344 (1974). c o m ss o c es n - Qumnergiaie, 44 FCC 2d 985 (1974). The Commission noted that is was concerned "when a licensee clearly broadcasts program matter which is designed primarily to promote the sale of a sponsor's product or services, rather than to serve the public by either entertaining or informing it. The primary test is whether the purportedly non-commercial segment is so interwoven with, and in essence auxiliary to, the sponsor's advertising (if in fact there is any formal advertising) to the point that the entire program constitutes a single promotion for the sponsor's products or services. This test will be construed strictly and the determination that a program is entirely commercial will be reached only when the facts clearly justify that conclusion.“ (See p. 986). A number of examples are also provided in question/answer form as illustrations. 68 62 - u 94 FCC 2d 678 at 684-5. 0 ' , 43 FCC 2d 638 (1973).: and 63During periods of high demand for political advertising, up to 20 minutes of commercials in 10 percent or less of a station's total weekly hours of operation was allowed. See 43 FCC 2d 638 at 640. 64Comme111aliza1ion_on_Ty_§1a1ion§. 49 RR 2d 391 (1981). 6542 FCC 2d 3 at 25-30. 66Ibid. 6794 FCC 2d 678 at 699, 715-7. o t e v's'o ' Qenmercialiaation Policies, Ascertainment Requirements, and Program Log Beguirements for Qomnereiai Televisign fitations, 98 FCC 2d 1076 at 1103 (1984). From Erwin G. Krasnow, Barry D. Umansky and William E. Kennard, dunnente er the Mationai Assoeiation of uroadcastere, MM Docket No. 83-670, 21 November 1983, p. 4. 69Edward E. Cohen, C ° e ' ' Commerc1al1za11o__fiurxey (Washington. Dv C-= National Association of Broadcasters), p. 1. 7098 FCC 2d 1076 at 1132. 71Michael L. Ray and Peter H. Webb, Adyertieing c 've s 'n a Crowde v's o nv'ro m -- e t Mo. 7§-ii3 (Cambridge, Mass.: Marketing Science Institute, September 1978), p. 5-7. 69 72William E. McCavitt and Peter K. Pringle, Ele11ron11_Media_Managemen1 (Boston: Focal Press. 1986). p. 154. 73Bruce A. Linton, "Self-Regulation in Broadcasting Revisited, " deurnaliem_Quarteriy 64 (Summer-Autumn 1987), p. 487. 74Peter H. Webb and Michael L. Ray, "Effects of TV Clutter." 12urnal_of_Adyer1ising_Besearch 19 (June 1979). p. 9-11. 75Leo Bogart and Charles Lehman, "The Case of the 30- Second Commercial. " J2u1nal_of_Adxer1111ng_Ee§earch 23 (February/March 1983), p. 15. 76Marvin S. Mord and Edith Gilson, "Shorter Units: Risk- Responsibility-Reward," dournai ef hdyertieing Beeeareh 25 (August/September 1985), p. 11. 77Michael L. Ray and Peter H. Webb, Experimental ;‘-’a 1 on ‘ ,7 e S O, , t‘ ' 9‘: :1- 1 a ' ' v' m -- - (Cambridge, Mass.: Marketing Science Institute, April 1976), p. 43. 78Mord and Gilson, p. 11-3, 19. 79Ibid., p. 11. 80Gary F. Soldow and Victor Principe, "Response to Commercials as a Function of Program Content, " deurnai pf Ad_sr1ising_Beseargh 21 (April 1981). p- 65- 81Herbert E. Krugman, "Television Program Interest and Commercial Interruption." 1211111_21_Adyer11sing Beeeareh 23 (February/March 1983), p. 23. 70 826. A. steiner. The_2eople_Look_a1_Ielexision (New York: Knopf, 1963).: "Video Scope: Evaluation of Television Advertising," Erinters ink 289 (1964), p. 40.: J. R. Kennedy, "How Program Environment Affects TV Commercials," deurnai of Advertising Research 11 (1971), p. 33-8.: and John H. Murphy, Isabella C. M. Cunningham and Gary B. Wilcox, "The Impact of Program Environment on Recall of Humorous Television Commercials," leurna1_er 81111111189 8 (Spring 1979). p- 21- 83Litman, "Program Diversity," p. 397. 84In_1he_Ma11er_o1_EMEC1_T11.11a1ion_21_1he_§11111 IDQ1 6 FCC 729 at 730 (1939). Refers to following quote in the text. 85Ibid. Refers to following quote in the text. 86Blue Book, p. 45-7. 8739 FCC 2d 1062. 88W. 99 Fcc 2d 1389 at 1398-1401 (1984). 71 89The Federal Trade Commission has been responsible for preventing "unfair methods of competition” in or affecting commerce since 1914. Originally, the FTC's power to control deceptive advertising was upheld in Sears Roebuck & Co v. FTC, 258 Fed. 307 (7th Cir., 1919). This power was steadily narrowed over the next twelve years, culminating in FTC v. Raladam Co., 283 U. S. 643 (1931), in which the Supreme Court held that the FTC must find that competition, not just consumers, were injured by a deceptive ad. The FTC's authority over deceptive advertising was confirmed with the passage of the Wheeler- Lea Act, 52 Stat. 111 (1938). It broadened the FTC's jurisdiction to include "unfair and deceptive acts in commerce," as well as "unfair methods of competition." Basically, in determining if an advertisement is deceptive, it must be determined whether the ad accurately describes the product and whether the consumer is likely to be mislead. The net impression which the advertisement makes is important and an ad which has the capacity to deceive the public, even though its facts are completely true, may be found deceptive. This capacity to deceive standard was established in Charles of the Ritz v. FTC 143 F.2d 676 (2nd Cir., 1944). Proof of a false advertising charge involves four questions. First, the audience of the commercial and/or their level of intelligence is considered. For example, an ad intended for children would be considered differently than one aimed at a well- educated target market. The FTC considers both the intended and unintended audience of the ad. Second, what the advertiser promised in the ad must be determined, including what the consumer actually understood it to mean. Literal truth is not a defense because the ad is viewed as a whole and is judged by the general, overall impression it creates. The third question, truth, incorporates a consideration of audience and promise. The issue here is whether the promise of the ad as comprehended by the audience is actually true or false and the basic question is whether the product or service performs as claimed. The final question to be determined is whether the deception is capable of affecting a purchase decision, called materiality. See Glen 0. Robinson, Ernest Gellhorn and Harold H. Bruff, The Administrative Process (St. Paul, Minn.: West, 1982), p. 432-84, for an explanation of deceptive advertising. See also Donald M. Gillmor and Jerome A. Barron, Maee memunigatign_Lau 4th Ed. (St. Paul, Minn.: West, 1984), p. 613-34. 72 9°PTQ, PCC in Liaison Agre ement on Ads, 27 April 1972, RR Current Service 2d 11: 212-4 (1987).: See Liaieen twee FCC and FTC e at to a s s 'n Ra ' and TV Advertising, 21 February 1957, RR Current Service 2d 11:201 (1987), which notes that the FTC will advise the FCC of questionable advertising broadcast over radio and television stations. It also notes that "licensees should not rely solely on the action or inaction of the Federal Trade Commission, nor should they suspend their own continuing efforts in determining the suitability of advertising material to be broadcast over their facilities." 91 a e Dece t ve dvertisin From v's'o Pet't'o er TQPE (Termination of Unfair Broadeasting Exeesses)], 32 FCC 2d 360 at 361 (1971). This interpretation predates the FCC/FTC liaison document in the previous note. 92 t e a ter o 1' in t'o o ss Preadeaet_keguiatien, 50 Fed. Reg. 5583 at 5584-5, 5589-90 (1985). 9344 FCC 2303 at 2313 (1960). 94L11en__e_Be_p2n_111111y_111h_Bespe_1_12_1he : 0-0 - S y S ‘.e .0 , e, 74 FCC 2d 623 at 624 (1961). 956 FCC 2d 729 at 730. 73 96 m ' t Qrdoxerl_herkeleyi_gal111. 29 FCC 2d 807 at 813 (1971)- Also known as the Chevron decision. See also 1n_Pe m lai ct o o r n' s T ev s o w Turk, M,x,, Qoneerning Teie vieio n Adv ertisemente, 32 FCC 26 412 at 412 3 (197l)-. In_B__Qomplain1.21_Alan_£1 Ne1kr111_and_L__rence_Bi_Qrdoyer. 37 FCC 2d 528 (1972)-. and t o s Westi house 40 FCC 2d 1045 (2973). The FCC declined to require broadcasters to publicize on their stations that the commercials in question might be deceptive. This request was made on the basis that fairness doctrine obligations were involved because the ads were a controversial issue of public importance because they sought to take advantage of people's fears regarding pollution. The basis for this argument is found in App11_a_1111y_91_1he_zairnes_ tr e ret e v rt s 9 FCC 2d 921 (1967), affirmed in Banzhar v. Edd, 405 F.2d 1082 (1968)., certoriari denied 396 U. S. 842. However, cigarette advertising was deemed a unique situation where fairness doctrine obligations were applicable to advertising. This interpretation was not extended to other product advertising, but it was noted that there might be cases where fairness obligations applied to advertising. Time was not granted for counter-commercials regarding the environment in E11end§_of_1he_Ear1h_x1_£CC. 24 FCC 2d 743 (1970), affirmed in 449 F.2d 1164 (D.C. Cir. 1971). The FCC examined these questions about the fairness doctrine in In_11e_Ma11e1_21_1he_aandl1ng_of_£ublic_lssues_Under the Pairness Qoetrine, 30 FCC 2d 25 (1971). In Tn_the , 4- .a . no . ' olic sues use 1‘ 1 .--s c s nda s o Qemmunigations Apt, 48 FCC 2d 1 (1974)., the FCC decided that the fairness doctrine no longer applied to product commercials. 97Cen1e1_f21_Lay_and_§21111_291111. 23 RR 2d 187 at 194 (197l)-. or see In_Re_1_mplain1_py_12n_umers m ° , 32 FCC 2d 400 at 406- 7 (1971), for the identical text under a different title. 9823 RR 2d 187 at 194-5. 990111911120.21.Au112111y. 43 FCC 2d 638 at 639 (1973). 74 1°°nAB_Lega1_Qu1de_t2_ECC_8rgean§t_Begulatigns 2nd ed. (Washington, D.C.: National Association of Broadcasters, 1984), p. IV 1-3. For more information see Leon C. Smith, "Local Station Liability for Deceptive Advertising." 1222n11_gf_areadsa§ting 15 (Winter 1970-1). p. 107-12. See also a case where broadcasting deceptive advertising on a radio station was part of the reason its license renewal was denied, in Re In____822112211229_2f united IeleVi§ign gg,. Ing., 55 FCC 2d 415 (1975)., and.1n Bs_ARR112et12ns_2f_Qn119d.121§21§12n_92111n21. 55 FCC 2d 431 (1973). 10150 Fed. Reg. 5583 at 5589-90. Reconsideration denied in the atte ecess atongcgst Regulation, 58 RR 2d 864 (1985)., affirmed in Gun 4 1 .018 "S C n0 ; :01 ‘4 ‘ , 800 F. 2d 1181 (D. C. Cir. 1986). 1°2E21122_Begerd1n9_Char1212:.99111f121§12n§_1n fitnnggn§t_Liggn§ing, 102 FCC 2d 1179 at 1208-11 (1986). 1°3Ibid., p. 1193-1203. 1°4Ibid., p. 1225-9. 1°5Ibid., p. 1212-4. 106In the Ma H: e: 9: ES 1 ablienm Mn 5‘ g unifgrm .20 H]: CH 0 e cens 1n 0 S gages in Connection With Violations by an Anniignnt 9f ans 9f the U.§, che; Inan IDQ ggnnnnigntigng Ant of 12;Ai_A__Angndgd, 42 FCC 2d 399 at 404 (1951). 10750 Fed. Reg. 5583- 10850 Fed. Reg. 5583 at 5589. 109102 FCC 2d 1179 at 1181, 1183. 11°Ibid., p. 1181. 111McCavitt and Pringle, p. 29, 146-7.: and 23 RR 2d 187 at 189-90. 112Linton, p. 488. 75 11323 RR 2d 187 at 189-90. 114Ibid: Enat_1e_a_§et_er_hugine§§_nureauz and 888 Cede_gf_8d_erti_1ng (Arlington. Va-= CBBB 1985).: HAD gn§§_3gnnzt, Vol. 13, No. 6, 15 July 1983, p. 20- 3. 11523 RR 2d 187 at 189-90. 116Eric J. Zanot, "Unseen But Effective Advertising Regulation: The Clearance Process," 14 (1985), p. 44- 6.: Capital Cities/ABC Inc., ngggttifiing Standard§_and_§uideline§. Department of Broadcast Standards and Practices, November 1986.; CBS Televigign Standarde January 1984-: and n8__Br_adga§t_§tandards_fer Iglgxiginn, "Advertising Standards," January 1986. 117Ibid.; Gordon B. Miracle and Terence Nevett, v s (Lexington, Mass.: Heath, 1987), p. 99.; Steven W. Colford, "Four A's Pushes Nets on Screening Ads," Agggttiging_ngg, 20 April 1987, p. 36.: Interview with Steve Donowski, General Sales Manager, WILx-TV (NBC-Ch. 10), Lansing, Michigan, 29 July 1987.; and Interview with Bill Shipley, Operations Manager, WSYM- TV (Independent-Ch. 47), Lansing, Michigan, July 1987. 118 ac ' v v 2 FCC Red 5043 at 5057-8 (1987). 119 t tte e ' Bzgndcasting Q9, 39. 4990, 3 F.R.C. Ann. Rep. 32 at 33 (1929). 120Blue Book, p. 39. Refers to following quote in the text. 121Maxflgner_8rgadsa§ting_921. 8 FCC 333 at 339-41 (1941). 122 13 FCC 1246 (1949). 123W. 395 0.8. 367 at 377-8 (1969). 12444 FCC 2303 at 2314. 76 12540 FCC 598 at 599. 126395 U.S. 367. 127W 23 Fee 26 707 at 707- 9 (1970). affirmed in In_Be_CQ_Rleint§_2f_Cgmmittee f_r_th__Fa1r_nr2ad2a_t1ng_Qf_92nfr_xer§121_1§§ue§. 25 FCC 2d 283 (1970). 1281B the H5539 I Q: IDQ HQDQJiDQ Qt EHDJjE 15 ”HQ ‘ -l°. 1‘ Standard__2f.the.§2mm_n1_at1gn__agf 48 FCC 2d 1 at 11- -12 (1974). 129£CQ v. nggne pf Women theps pf Cglifptnig, 468 U.S. 364 at 377, note 11 (1984). ”0me12; m ss'on's u d e ul io 5 Co e . ,‘S_ Doc . - .0 go. a... o 3 010‘ .‘7 Lig§n_gg§, 102 FCC 2d 143 at 246- -7 (1985). 131Syragu§e_Eea9e_C2un21l_x1.melexisign_§tatign_fliyn. 99 FCC 2d 1389 (1984). 132ugzggj§h Qgrp. v, Egg, 809 F.2d 863 (D.C. Cir., 1987). 133588 In IQ Syxjacnsg Epacp QQU UDQilV V, Elva; QIQQI Rggngsting Comment 2 FCC Red 794 (1987).; In tne Ma tte e: o n ir 'nto ection 7 .1910 0 th Comm ss 0 's u s d ulat'o s Concern'n Alte atives to the n ai ess Doctr'ne Ob i at'ons o roadcas i ense 8' Notice of Ingniry, 2 FCC Rcd 1532 (1987).; Report of the Commissipn, 2 FCC Rcd 5272 (1987).; and Mpnptnngnn_gpinipn gng_p;gg;, 3 FCC Rcd 2050 (1988). 1342 FCC Rcd 5043 at 5058. See also Syzngn_§_£gngg o ' v. e_ev'sio St. 5- ads: no .n- m 0- _., d Otdgt, 3 FCC Rcd 2035 (1988). 135Ricky W. Griffin and Gregory Moorhead thanignt ipna a; Bgnaviop (Boston: Houghton Mifflin, 1986), p. 20. 77 136Peter M. Blau. 9n_1he_Nature_gf_9rganizatien§ (New York: Wiley, 1974), p. 29. 137David Beetham, fintpnnptnpy (Minneapolis, Minn.: University of Minnesota Press, 1987), p. 12-13. 138Robert Dubin, Ine prlg pf Kong (Englewood Cliffs, N.J.: Prentice-Hall, 1958), p. 370. 139M. H. Gerth and C. Wright Mills, Eds., Ergm_ngx flppgrg Egsnyg in fipcioiogy (New York: Oxford University Press, 1946, Reprinted in 1975), p. 196-98. Contains a complete discussion of Weber's characteristics of bureaucracy. 14°Dubin, The szig pf Wogg, p. 365-70. 141Robert Dubin, "Technical Characteristics of a Bureaucracy," In Robert Dubin, ed., flnmnn_Rg1ntipn§_in A_nini§ttntipn (New York: Prentice-Hall,1951), p. 156. 143William A. Rushing, "Organizational Size, Rules and Surveillance," In Joseph A. Litterer, ed., Qrgan11at19ns1_§frugture_aud_fienaxigr 3rd ed. (New York: Wiley, 1980), p. 396-9, 404. 144Ibid. 14SMcCavitt and Pringle, p. 26-9. 146Linton, p. 488. 147Ibid., p. 489. CHAPTER III PROPOSED THEORY AND HYPOTHESES The literature reviewed in the previous chapter pointed out various strains of research that converge to explain and predict conduct in commercial television markets. Specifically, degree of market concentration and organization size may influence that conduct. These markets are oligopolies, where stations cooperate or compete to reap the highest profits for all. Again, the price of advertising time and the number of commercial minutes per hour are two factors which may be manipulated to make a station's commercial time more or less attractive to advertisers. Clutter and clearance and complaint practices may also be manipulated for this purpose. Therefore, local stations within a market may be expected to sell similar amounts of commercial time, schedule about the same number of breaks per hour, and run about the same number of consecutive announcements. Differences in the price of air time would thus be restricted to program popularity (or program ratings). 78 79 Stations offering popular programming could charge higher rates than those offering unpopular programming. Stations within a market might also decline to accept certain kinds of advertising, say for contraceptives, to make their commercial breaks similar in terms of content. Advertisers could feel confident that their ads would not appear next to "questionable" ones. Presumably, they seek out highly rated programs on stations restricting the amount of commercial time and clutter, and banning questionable advertising, to ensure that their ads run in a quality environment. And stations interested in maintaining the quality of this environment will be responsive to complaints. The review of organizational literature pointed out that differences in advertising policies may also be due to station size. Size may be a predictor of policy formality, how many different types of standards a station has, and the number of groups a station will officially respond to complaints from. The review of mass media economic research suggests that other variables besides market concentration and organization size also may affect the quality of a station's air time and hence the rates charged advertisers, including: 80 a) station profitability (traditionally measured using network affiliation, broadcast band and net weekly circulation), b) audience size or program quality (measured by a station's ranking in its market during a period when stations choose and schedule their own programming and advertising), c) the presence of substitutes (measured as the penetration of cable television and VCR's in a market), and d) market quality (or the buying power of the average viewer). These variables have traditionally been used as predictor variables in studies assessing various aspects of station performance. The type of advertising a station favors might also affect decisions about the quality of a station's air time. Large market stations sell more of their commercial time to spot advertisers, so it is included as a predictor variable to see if it explains any differences in advertising policies. The review of mass media policy literature provides further illumination of what might affect decisions about the quality of a station's air time. The FCC formerly expected larger and more profitable stations to make a correspondingly greater effort to restrict deceptive advertising (clearance). It also noted that it was 81 deregulating television to encourage individual decision- making. So, larger and more profitable stations may be expected to have more policies "on the books" before deregulation, thus accounting for some differences. And, because more profitable stations as a whole tended to schedule more commercial time before deregulation, they might also be expected to do so after. By considering the precedents from both economics and policy research, station profitability appears to be an important predictor of advertising policy differences. Economic theory helps to illuminate this further by suggesting that more and less profitable stations may be considered different "classes," with each having distinct price and quality differences. Stations with more popular programs can charge higher rates, and turn down questionable advertisements that less profitable stations cannot. Thus, differences in advertising policies may also be a result of how profitable a station is. Consequently, the primary independent variables of interest were market structure or concentration, organization size, and station profitability. Their influence on the dependent variables, advertising commercialization, clutter, clearance and complaint practices, is the focus of interest. 82 However, other variables noted above that have traditionally been used for assessing other types of performance were also included in the study. The study sought to test whether precedents from other performance studies could be extended to examine advertising performance. These other predictors of advertising policies include the quality of a station's programming or a station's overall ranking in its home market, the type of advertising, the availability of substitutes, market quality, and the price of advertising time or rates. Finally, stations were asked to indicate whether or not they formerly subscribed to the NAB TV Code to see if previous commitment to a self-regulatory organization represents a greater commitment to the public interest. The Attractiveness of Advertising and Viewing Environments The underlying assumption is that all of these predictor variables combine in a unique fashion in each market, and at each station within each market, to affect the relative attractiveness of a station's airtime to viewers and advertisers. The result is that the individual combination of the dependent variables, commercialization, clutter, clearance and complaint practices, are the manifestations of that attractiveness. Being able to predict the relative attractiveness of a 83 station's airtime may help to identify markets and stations that exhibit "good" or "bad" advertising performance. It also helps to explain why performance varies and suggests how to take steps to improve performance, if necessary. It is assumed that station managers consider (consciously or unconsciously) all of these predictor variables and how they interact to decide how much commercial time to schedule, how cluttered their breaks will be, which ads to accept and decline, and which complaints to respond to. Their decisions are assumed to affect how attractive their air time is to both viewers and advertisers. Advertisers like to be associated with a quality environment, and viewers resent it when time is taken away from programs and given to advertising. Thus, a manager must consider the effect such changes will have on the popularity of his broadcast fare. It is assumed that local stations strive to develop a quality image to make their commercial time attractive to advertisers.1 Commercial time is therefore the "product" a station sells,2 and a station manipulates a number of variables to make its broadcast fare more attractive to 84 viewers to increase ratings, and ultimately attract more advertisers. It is in that way that the viewing and advertising environments are interrelated. The viewing environment describes the way these manipulated variables are perceived from the viewer's perspective. Although the variables in the viewing and advertising environments are similar, they are not exactly the same. The viewing environment is how attractive a station's broadcast fare is to viewers. If a station's programming is not acceptable a viewer might turn to a substitute form of entertainment, such as a cable channel or a newspaper. If he perceives too much commercial time is being taken away from programming, and breaks have too many ads in them, this might also cause him to switch. If a station airs a number of offensive ads and fails to respond to his complaints, he may again turn to a substitute form of entertainment. The viewing environment is therefore the product a station "sells" to viewers. The attractiveness of a station's viewing environment is expressed as follows: v=f(x,x,c,c,o,0) l l 2 1 2 1 2 Where: V = The attractiveness of a station's viewing 1 environment to viewers. X = The quality of a station's programming (measured 1 by market ranking). X = The availability of entertainment substitutes 2 (Cable, VCR's, newspapers, radio. etc.). C = The perceived amount of commercial time aired on 1 the station. C = The perceived amount of clutter. 2 O = The perceived amount of offensive advertising 1 (formality of clearance practices). 0 = The perceived responsiveness of the station to 2 viewer and community concerns (the formality of complaint practices). A station increases the attractiveness of its viewing environment by improving program quality (X1), decreasing the amount of commercial time (C1) and/or clutter (C2), and/or increasing the formality of clearance (01) and/or complaint (02) practices. When a station improves the quality of its programming, it can expect to increase its audience, and therefore attract more advertisers. 86 When a station decreases the amount of commercial time and clutter, it becomes more attractive to both advertisers and viewers, as advertisers believe that their ads have a better chance of being recalled, and viewers perceive that less time is taken away from programming. When a station increases the formality of its clearance practices, it becomes more attractive to viewers because they perceive that the station airs less "offensive" advertising, and to advertisers because they believe that their ads will not air among others perceived as "offensive" by viewers. When a station increases the formality of complaint practices, it becomes more attractive to viewers and advertisers because both perceive it as being sensitive and responsive to their concerns. In other words, when a station improves the attractiveness of its viewing environment, it can expect to increase its audience. And when it increases its audience, it can expect to improve the attractiveness of its advertising environment. The assumption is that as the attractiveness of the viewing environment increases, the attractiveness of the advertising environment increases. The expectation is that the attractiveness of the viewing and advertising environments is positively related. 87 Therefore, the advertising environment is considered to be the "advertiser" counterpart to the viewing environment, and describes how the independent and control variables are perceived by the advertiser. The advertiser wants to obtain the greatest reach for each dollar spent, so audience size is important to him. (He also likes to be associated with the number one station in a market, because of the "prestige" image it conveys to viewers.) If the size and cost efficiency comparison is not cost effective, he may switch to a substitute form of advertising (rather than entertainment). If there is too much commercial time aired on a station, and its breaks are cluttered, the station's airtime is less attractive because his commercial has a poorer chance of being recalled. If a station airs offensive ads, he may be concerned about the environment his ads appear in, and how that environment affects viewer perceptions of his product. These problems may all be exacerbated if the station is unwilling or unable to respond to his concerns. Commercial time is therefore the "product" a station sells to advertisers, and the attractiveness of a station's advertising environment is expressed as follows: 88 The attractiveness of a station's advertising environment to advertisers. The size of the station's audience at any given time (or market ranking). The availability of advertising substitutes (cable, newspapers, radio. etc.). The amount of commercial time scheduled per hour. The amount of clutter per hour. The formality of clearance practices. The formality of complaint practices. The quality of the market the station is in (i.e., the buying power of the average viewer). The price of commercial time on the station. The attractiveness of a station's viewing environment (i.e., the average viewer's perception and/or image of the station). A station may increase the quality of programming (to increase audience size - 21), decrease commercial time (C1) and/or clutter (C2), and/or increase clearance formality (Fl) and/or complaint formality (F2) to improve the attractiveness of its advertising environment. Given that 21, C1, C2, F1, and F2 are manipulated to make a 89 station's advertising and viewing environments more or less attractive, these demand conditions are considered interdependent. Several of the independent variables (market concentration or structure, station profitability, the type of advertising accepted and/or organization size) may affect how a station manipulates the dependent variables (commercial time or C1, clutter or C2, clearance formality or F1, and complaint formality or F2). Therefore: 0 H H) 50 U) 2 H C = The amount of commercial time. R - The profits a station earns, or station 1 profitability. S = The structure of the market the station operates 1 in, or market concentration. 2 II The number of different types of advertisers 1 (i.e., national spot, local) which find the station's advertising environment attractive. I = Other intangible variables such as the quality 1 of the sales staff, management's views regarding fulfilling the public interest, etc. And: C Where: C R S N I And: F Where: 90 The amount of clutter. The profits a station earns, or station profitability. The structure of the market the station operates in, or market concentration. The number of different types of advertisers (i.e., national spot, local) which find the station's advertising environment attractive. Other intangible variables such as the quality of the sales staff, management's views regarding fulfilling the public interest, etc. The formality of clearance practices. The profits a station earns, or station profitability. The structure of the market the station operates in, or market concentration. Organization size, or the combined size of the sales and traffic departments. Other intangible variables such as the quality of the sales and traffic departments, management's views regarding fulfilling the public interest, etc. 91 And: F = f(R , S , D , I ) 2 l l l 1 Where: F2 = The formality of complaint practices. R = The profits a station earns, or station 1 profitability. S = The structure of the market the station operates 1 in, or market concentration. D = Organization size, or the combined size of the 1 sales and traffic departments. I = Other intangible variables such as the quality of the sales and traffic departments, management's views regarding fulfilling the public interest, etc. It is assumed that broadcast managers take the independent variables into consideration when determining advertising and viewing environment attractiveness. They are also assumed to safeguard advertising and viewing environment attractiveness by manipulating the dependent variables. For example, by limiting the amount of commercial time sold, limiting the number of consecutive announcements, and utilizing formal clearance and complaint practices, a station increases the attractiveness of its viewing and advertising environments. 92 W: Market Structure or Concentration: Given the structure of the television industry, it is assumed that stations consider how their counterparts vary commercialization, clutter, and clearance and complaint practices to differentiate their viewing and advertising environments (or make them more similar). Although this is expected of oligopolists, it is predicted that conduct varies by "type" of oligopolist, or market structure, as well. In other words, while the level of concentration may vary from market to market as a continuous variable, particular levels of that concentration may stand out for their particular effects on station conduct. Consider that market concentration varies. The number 2 Area of Dominant Influence (or market),3 Los Angeles, California, has 14 commercial television stations and is less concentrated than the number 213 ADI, Glendive, Montana, which has 1.4 Station conduct in these markets is expected to differ, as their concentration level differs. It should be more difficult for market members to cooperate in "loose oligopolies," or less concentrated markets. Because there are more stations with different overall ratings and profits, their reasons for deciding 93 how much time to sell and what spots to clear should vary. One might therefore expect to find greater "within-group" variance (or greater product differentiation from varied levels of commercialization and clutter) in less concentrated markets. Because station advertising and viewing environments are expected to vary the most, some stations are expected to adhere to old FCC and NAB guidelines, while others exceed them. Some stations accept variable length time units, schedule more breaks with more consecutive announcements in them, while others do not. Some may clear program length commercials and potentially offensive ads. And clearance practices vary, with some stations having more formal, written advertising codes, and others simply discussing problems as they arise. Markets with concentration levels greater than loose oligopolies, termed "oligopolies," are expected to exhibit less variance in clutter and commercialization levels. Clearance and complaint formality are expected to vary as well, but not as much as in loose oligopolies. Such practices are also expected to be less similar than in tight oligopolies. In more concentrated markets, or "tight oligopolies," stations are predicted to behave like the networks. The 94 old time and clutter guidelines will be exceeded slightly and to a similar extent, with the number one station scheduling slightly more. The formality of clearance and complaint practices is also expected to be similar. Station advertising and viewing environments are comparable, and competition restricted to programming to keep profits high. In monopoly power markets with one (or one significant) competitor, stations are expected to adhere to the old FCC guidelines. Since economic theory predicts that monopolists restrict output to raise prices, they may sell the least time of all, and be the least cluttered. They are expected to have the least formal clearance and complaint practices. Therefore, the following predictions are made regarding market structure: 1. As market concentration (81) increases, the level of commercialization (C1) decreases. 2. As market concentration (81) increases, the level of clutter (C2) decreases. 3. As market concentration (S1) increases, the formality of clearance practices (F1) decreases. 4. As market concentration (81) increases, the formality of complaint practices (F2) decreases. 95 A few studies examined advertising commercialization, clutter, clearance and complaint practices, but none explained how and why these practices differ. Results may provide new insight into station commercial practices, and illuminate how these practices differ according to market structure. By dividing stations into market structure types, predictions can be made regarding conduct based on economic theory, and these hypothesized conduct differences tested. Any differences the deregulation of commercial practices had in these market structure types can then be assessed and explained. If the notion that market structure type affects station conduct is supported, a new, useful means of operationalizing market structure would result which might explain differences on other performance variables. If not, economic theory may still provide insight by determining whether stations are behaving as oligopolists "in general." Profitability: But other predictors may affect the dependent variables. Managers must also, consciously or unconsciously, take into account how profitable their station is when making decisions about the attractiveness of their advertising and viewing environments. A manager 96 at a more profitable station might safeguard the attractiveness of its advertising and viewing environments. For example, by limiting the amount of commercial time sold, limiting the number of consecutive announcements, and utilizing formal clearance and complaint practices, a station increases the attractiveness of its viewing and advertising environments. Indeed, a manager might manipulate these variables in a certain way to inadvertently or knowingly create a "subclass" of stations. It is likely that different stations have different reasons for deciding how much commercial time to sell, how many breaks and spots to schedule, and what ads to clear. Stations with attractive viewing and advertising environments might schedule a little more commercial time because they can sell it, and because their quality programming abates its negative effects. Relative attractiveness is maintained by employing formal (and more strict) clearance practices. These profitable stations might therefore form a "subclass" in this fashion. The "damage" to relative attractiveness is minimized if all stations in a market increase commercial time to a similar extent or all decline to clear a potentially offensive ad (i.e., a contraceptive ad, or other product 97 which was the subject of a policy statement). However, stations with less attractive advertising and viewing environments may find it necessary to accept a potentially offensive ad. In that way, subclasses of stations might develop within markets. Consequently, the predictions based upon station profitability are: 5. The more profitable (R1) a station is, the greater its level of commercialization (Cl) will be. 6. The more profitable (R1) a station is, the more cluttered its commercial breaks (C2) will be. 7. The more profitable (R1) a station is, the more formal its clearance practices (F1) will be. 8. The more profitable a station is, the more formal its complaint practices (F2) will be. Findings can be used to reveal if stations adhere to FCC clearance and complaint expectations, and to discern what relationship exists between commercialization and profits. Organization Size: Organization size, as well as market concentration and profitability, may also affect advertising practices. Organizational theory suggests the following relationships between size and formality in organizations. The more formal a station's advertising practices are, the more likely its policies are written rather than verbal. 98 Further, the more formal such practices, the more complex they are (e.g., using more policy sources and dealing with more problems through the communication of policy standards). Therefore, the predictions based on organizational theory are: 9. The larger the organization (or the larger a station's sales and traffic departments (01)) are, the more formal its clearance practices (F1) will be. 10. The larger the organization (or the larger a station's sales and traffic departments (D1)) are, the more formal its complaint practices (F2) will be. Supporting results might suggest directions for other, in-depth organizational studies examining these and other performance variables. The Joint Effect of Market Structure and Profitability: Although the theoretical support is not as strong, it is predicted that market concentration and profitability will have a joint effect on the dependent variables. If we assume that a station with monopoly power restricts supply, clearance may provide a means of doing so. As the FCC does not expect as much from smaller stations, this might be a rationalization for keeping clearance costs down. Yet, if this station is quite profitable, its clearance procedures may be formal. So great differences 99 in clearance formality may exist in the monopoly power category. Greater "within-group" variance is therefore expected. Stations in tight oligopolies are expected to be aware of competitors' advertising and viewing environments, and cooperate to restrict competition to programming. They sell similar amounts of commercial time, have similar clutter levels, and similar clearance and complaint practices. Less profitable market members clear some ads their more profitable counterparts will not. This is mirrored by NBC and CBS traditionally allowing ABC to cheat,5 and by the less profitable cable networks and superstations clearing contraceptive ads while the networks refuse to do so.6 Therefore, the most profitable stations in tight oligopolies will have more formal clearance and complaint practices, sell slightly more time, and be slightly more cluttered than other market members. The most profitable stations in oligopolies are expected to be more commercialized and cluttered than those in tight oligopolies, but less than stations in loose oligopolies. The same intermediate behavior is expected for clearance and complaint formality. 100 It is assumed to be more difficult for stations to cooperate in loose oligopolies, so greater variance in commercial time and clutter is expected. Clearance is more formal because the FCC expected the most from profitable, larger market stations. Thus, the most profitable stations in loose oligopolies have the most formal clearance and complaint practices, sell the most commercial time, and are more cluttered overall. 11. As both market concentration (81) and profitability (R1) increase, the variance commercialization levels (C1) decreases. 12. As both market concentration (81) and profitability (R1) increase, the variance clutter levels (C2) decreases. 13. As both market concentration (81) and profitability (R1) increase, the variance clearance formality (F1) increases. 14. As both market concentration (31) and profitability (R1) increase, the variance complaint formality (F2) increases. Findings may support the proposition that commercial time and clutter, and clearance and complaint formality, are jointly affected by market structure and profitability. in in in in 101 law: The following research questions will be examined to determine if the differences between market structure categories are meaningful. Specifically: 1. Does commercialization vary by market structure type? 2. Does clutter vary by market structure type? 3. Do clearance practices vary by market structure type? 4. Do complaint practices vary by market structure type? In addition, the following research questions will be examined to determine if profitability classes exist and if differences between them are significant. Specifically: 5. Does commercialization vary by profitability class? 6. Does clutter vary by profitability class? 7. Do clearance practices vary by profitability class? 8. Do complaint practices vary by profitability class? Results may reveal that performance differences may be expected of stations in different profitability classes. 102 The analysis of the hypotheses and research questions should at least provide insight on advertising practices used by stations nationwide, and possibly explain why practices may differ. Whatever the outcome, findings will expand knowledge into some new areas of FCC policy. NOTES: CHAPTER III 1The idea for this theory developed from discussions with Dr. Barry Litman regarding his economic model of the network advertising market found in Barry R. Litman and Jan LeBlanc Wicks, "The Changing Advertising Market for the U.S. Television Networks," John D. Leckenby, Ed. Int Epoceedings of tng i288 Confergngg of the Ametignn Agagemy of Advertising, (Austin, Texas: John D. Leckenby, College of Communication, The University of Texas at Austin, 1988), p. RC27-33., whose influence is hereby gratefully acknowledged. The conceptualization and operationalization of this theory developed from discussions with Dr. Stephen Lacy, whose influence is also gratefully acknowledged. The author is solely responsible for its development and content. 2Benjamin J. Bates, "Economic Theory and Broadcasting," Presented to the Mass Communication Theory and Methodology Division of the Association for Education in Journalism and Mass Communication Convention, Memphis, Tennessee, August 1985, p. 5. 3The area of dominant influence is an exclusive geographic area which reflects those counties in which the dominant share of television viewing is to home market stations. The ADI is the industry's standard definition for allocating advertising dollars and often corresponds to the distribution and sales territories of many national and regional advertisers. From Rpn_Ip_ang_XQnt_Appitppn Iglevisipn napRgt Repprt (New York: Arbitron Ratings Co., 1987), p. 5. 4Televisipn g anig Engtppok, Stations Volume, No. 55, (Washington, D.C.: Television Digest, Inc., 1986), p. A1-4. 103 104 5Bruce M. Owen, Jack H. Beebe, and Willard G. Manning, Jr., Igipvision Eppnpnips (Lexington, Mass.: Heath, 1974), p. 108. 6Susan Morse, "Ads for Contraceptives Fail to Rile Viewers," ngnpgR, 21 January 1985, p. 38-9.: and Maurine Christopher, "Nets Stand Fast on Birth Control Ads," 10 November 1986, p. 36. CHAPTER IV METHODOLOGY fitndy Apptoach: Surveys are the most generalizable means of social research. The more a study occurs in a natural environment, making it more consistent with real life, the more valid it is also considered. More robust statistical tests like correlation and regression are less likely to miss a significant relationship where one actually exists.1 Survey research is often used by economists,2 and regression is a preferred statistical tool for economic analysis of media issues.3 A survey thus was conducted using regressions to analyze the relative effect of the independent variables on the dependent variables or the advertising practices of TV stations. T-tests were used to assess whether distinguishing market structure types, profitability classes, and organization size (larger and smaller) was useful. Means and standard deviations were used to show how differences were manifested and if practices varied as 105 106 predicted. Two way analyses of variance were also used to discern if variances were affected jointly by market structure and profitability. The study population was VHF and UHF commercial television stations in the top 213 Areas of Dominant Influence as reported by Arbitron. The unit of analysis was the commercial television station. The small markets edition with relevant data was not available for use, and those markets were omitted. The 147 stations excluded from the survey include small market stations, satellite stations which carry another station's signal rather than originating programming, commercial religious stations that rely primarily on fund raising rather than advertising, and Home Shopping Network stations.4 ods e es arch: The aim was to predict and explain differences among stations in different types of markets, so the survey method chosen had to be feasible for a national sample. A mail survey was the logical choice, as it could be conducted relatively quickly and inexpensively by one person. And because questions about time sales and clutter might be considered sensitive, the anonymity that mail responses provide encouraged candid responses.5 107 Although personal interviews and other field methods attain higher response rates and allow more probing,6 costs and research considerations precluded their use. The costs of visiting stations individually severely restricted sample size and generalizability, and was unacceptable given the aim of obtaining widely generalizable results. One major drawback of telephone interviews for a national survey was expense.7 Second, it was restricted by the necessity of calling sales managers at work, who are pressed for time on the job. A crucial concern was to make it easy for them to respond in about five minutes,8 as completing a tightly-constructed mail questionnaire at their leisure should do. W: The survey population was stratified to represent all 213 Arbitron Areas of Dominant Influence in one of the market structure categories. Strata were created by assigning all 870 of the stations into one of four market structure categories based on their ADI's Herfindahl- Hirschman Index (see Table 1), or measure of concentration described in the Operationalization section. 108 TABLE 1 Sampling Scheme STRATA l STRATA 2 STRATA 3 STRATA 4 Loose Tight Monopoly Oligopoly Oligopoly Oligopoly Power H-H Index H-H Index H-H Index H-H Index up to 1599 from 1600 from 1800 of 2000 to 1799 to 1999 and above 241 Stations 277 Stations 208 Stations 144 Stations All 870 stations in the universe were included so the expected response rate would be consistent with usual sample sizes in broadcast research. Sample sizes in broadcast economic research appear to be in the "300" range. One study included stations in only 13 markets,9 while others utilized secondary FCC data for 144,10 and 347 stations.11 Studies of rates used secondary data for 130 to 232.12 Stations in each strata were listed to utilize systematic sampling techniques for the pretests.13 After eliminating approximately 90 stations used in pretests, a planned-for response rate of 50 percent of 780 stations results in a sample size of 390. As about 10 independent variables were anticipated in the planning stages, the predicted response rate allowed for 30 cases each, well above the recommended minimum of 20.14 109 Hontpsponse in uni; §npygys= Nonresponse is a concern in mail surveys because nonrespondents may differ from respondents on the survey variables, resulting in biased estimates. To prevent this problem as much as possible, four pretests were conducted to ensure that the proper employee in the station was being surveyed, that questions were easy to answer and correctly understood, and that no question was considered offensive or too confidential, to ultimately increase the response rate. ’ Two follow-up mailings were also used to reduce nonresponse. Procedures and a mailing schedule demonstrated to be effective in minimizing nonresponse were used.15 The first mailing included an explanatory letter, questionnaire and return envelope.16 The first follow-up was a reminder postcard mailed bulk-rate one week after the first mailing.17 The second follow-up consisted of a revised letter, a replacement questionnaire and a return envelope mailed three weeks after the first mailing.18 (See Appendix A for mailings materials.) 110 W: Broadcast managers were consulted prior to, and after, questionnaire construction to provide a validity and reliability check. (See Appendix B for the final version of the questionnaire.) First, personal interviews were conducted with managers in one market before the questionnaire was written to confirm the meaning of proposed questions and their usefulness, and suggest others. Four pretests were conducted after the preliminary questionnaire was constructed. Respondents were again queried regarding question meaning and clarity, and nonrespondents asked why they did not respond. Suggestions were incorporated to improve the ease of completion by rewriting a few questions and eliminating others deemed redundant. However, question elimination and streamlining was primarily due to pretest subject indications that they were generally too busy to respond to questionnaires taking more than approximately five minutes to complete. Therefore some topics of interest had to be eliminated.19 The final questionnaires were numbered to indicate station identity for collecting appropriate secondary data (such as network affiliation and broadcast band) to 111 minimize item nonresponse (See Appendix B). This was explained in both accompanying letters, and station anonymity was preserved. on o t e de e V ’ : The primary independent variables of interest were market structure or concentration, station profitability, and organization size. Their influence on the dependent variables, advertising commercialization, clutter, clearance and complaint practices, was the focus of interest. The other independent variables were the quality of a station's programming or a station's overall ranking in its home market, the type of advertising, the availability of substitutes, market quality, the price of advertising time or rates, and former subscription to the NAB TV Code. These were initially included as predictors to extend previous research and to determine if they affect advertising performance. Rtgdictpt Entinbies: Market Structure: Concentration level is the most important determinant of market structure. The first step in measuring and proving concentration is to define the industry's relevant market. The criteria for including other local media is if they are considered close substitutes for TV 112 advertising.20 Direct broadcast substitutes, cable and VCR penetration, were included in the regression equation (see the section on the independent variable Direct Broadcast Substitutes, which follows later in this chapter, for further discussion). The next step is to decide upon a measure of market concentration. Several methods are commonly used in economics. One method is to count the number of market competitors. This measure neglects the audience share differences between stations, for example.21 The Lorenz Curve is inappropriate for this study as it masks concentration levels when there are few industry firms and evenly distributed market shares, which may be found in many TV markets. The concentration ratio, a measure of the cumulative market share of the top 4 or 8 firms in a market, may also mask inequalities. For example, two TV markets with identical four-firm concentration ratios of 60 can be very different: the market share distribution in one might be 45, 5, 5, 5 in one and 15, 15, 15, 15 in the other.22 The Herfindahl-Hirschman Index, calculated by summing the squared market shares of all firms, is preferred because it increases as the number of firms declines, and as inequality among firms rises. It equals one if the 113 market is a pure monopoly, and zero if perfectly competitive.23 The HHI is the Justice Department's preferred measure of market concentration and an index of 1000 (or .10) is considered the minimum level for raising antitrust concerns. It can be calculated using audience or revenue shares,24 and has been used as a measure of concentration in FCC proceedings.25 The ADI audience share of stations from 9 a.m. to Midnight has been used to calculate the H-H Index.26 Station shares for 6 a.m. to 2 a.m., Monday-Sunday, from the February 1987 Arbitron ADI Viewing Allocation Report were used in this study. (A share is the estimated percent of households using TV tuned to a specific station.)27 The report includes any station achieving at least a .1 share (including home market stations, stations from other ADI's, superstations and cable networks), a more liberal criteria than found in individual market reports. Both are calculated from the same diary responses.28 Using this report eliminated the disadvantage of using ADI shares from individual market reports, where households are assigned to one unique viewing area, and the overlap of broadcast signals is not accounted for. 114 Researchers have typically used the following rules of thumb for evaluating H-H Indexes. An H-H Index greater than .20 (or 2000) indicates a significantly concentrated industry, while one between .10 (or 1000) and .20 (or 2000) indicates a moderately concentrated industry. The indicator for significant concentration has also been placed at .18 (or 1800).29 Using the established rules used to guide station sampling, market structure for sampled stations was similarly operationalized as follows: 1. Stations in markets with H-H Indexes 1599 or below will be assigned to the "loose oligopoly" category. 2. Stations in markets with H-H Indexes between 1600 and 1799 will be assigned to the "oligopoly" category. 3. Stations in markets with H-H Indexes between 1800 and 1999 will be assigned to the "tight oligopoly" category. 4. Stations in markets with H-H Indexes of 2000 and above will be assigned to the "monopoly power" category. The index was entered as a continuous variable in the regression equations. Concentration level ranged from 902 to 3361 in the study, with the average being 1751. Markets in general were therefore at least moderately concentrated, supporting their traditional characterization as oligopolies. 115 Station Profitability: Proxy variables are used to measure station profitability, as such information is confidential.30 These proxies include broadcast band, network affiliation, the number of stations in the market, and market size (or potential audience size).31 It is generally acknowledged that broadcast band affects profitability to some extent. Broadcast band is measured simply by indicating whether a station is VHF or UHF, and including it as a categorical variable.32 Network affiliation has consistently been shown to be an indicator of higher profits. It is usually operationalized by designating which network a station is affiliated with (or its independence) as a categorical variable, to take network ratings variations into account.33 Affiliates may also be combined into one category, to represent an affiliate/independent dichotomy. Researchers may not use the number of stations in a market as a profitability indicator because it is highly correlated with and acts as a scale variable for market size.34 Net weekly circulation, an estimate of the number of unduplicated television households that watched a station for at least 5 minutes at least once during the week,35 is considered a better indicator of market size 116 and also represents a station's signal power. Signal power is considered a proxy for profitability because stations that reach more people have a larger audience and therefore earn larger revenues and profits.36 In one study, when the number of stations was deleted from the regression equation, net weekly circulation accounted for most of its variation.37 Thus, net weekly circulation is a good substitute because it also "measures" the number of stations in a market. The original intent was to use broadcast band, network affiliation and net weekly circulation as proxies for profitability. They are all reported in the WM 38 However. a profitability index incorporating broadcast band, network affiliation and market rank was ultimately used as the measure of profitability. (See the Data Analysis section in the next chapter for an explanation.) Organization Size: Organization size has been operationalized as assets, revenue, value of products or services, or the number of employees.39 The number of departmental employees is also a measure of size.40 Because the traffic and sales departments are responsible for the practices of interest, their combined size was used to measure organization size. Responses were included as a continuous variable. 117 W: Market Rank: The audience for a program is assumed to depend on the popularity of adjacent programs, as well as the nature of the program itself. Consequently, a station manager must pay attention to the popularity of individual as well as adjacent programs on her own station, as well as programming airing at the same time on competing stations. She may find it necessary to obtain new programs through syndication, and/or change the time a program airs, to counter a successful program. Although she may introduce a new type of programming to counter another station's fare, and thus start a programming trend, providing quality programming often takes the form of providing more of those program types which are popular on other stations or in other markets. Thus, program quality is measured by contribution to audience size.41 The more successful a station is at determining audience tastes, the larger its audience is. And, the larger its audience, the higher its ranking in the market. Therefore, market rank (or rank) was intended for use as a measure of program quality in this study, and entered as a continuous variable. 118 Station shares in early fringe (ADI Daypart Audience Estimate Summaries of shares for Monday-Friday, 3:00 p.m. to 6:30 p.m. or 4:00 p.m. to 7:30 p.m., depending on time zone) were used to determine ranking within a market. It is the longest period an affiliate is responsible for programming,42 and a time when independents compete directly with them. In a sense, these rankings represent how successful managers are at manipulating their schedules and/or counterprogramming. Shares from individual Arbitron TV reports were used because rankings within markets were desired, and these reports include both meter and diary estimates. The national news airs in this period, allowing the network factor to be considered. It also includes the early evening local newscast, an important source of competition, and a profit center for local stations. Local news is a way to create a station identity. Many managers believe that station image is greatly influenced by newscast quality, which stations control because they are locally-produced.43 In other words, this is when a station controls its advertising and viewing environments. Stations choose their own syndicated programming, decide in which order these programs will air, decide how many spots to 119 schedule, and produce an "image" program during this period. Although testing one daypart limits generalizability, the time period is a logical one for a preliminary test. The choice of this time period may be confounding. Managers are asked to estimate the average amount of commercial time that originates on their station during this period. To figure this out, network affiliates must eliminate the 30 minute network news program. However, pretesting suggested that this was not a problem. Data analysis suggested that rank was a better indicator of profitability, and it was incorporated into a profitability index with broadcast band and network affiliation. (See the Data Analysis section in the next chapter for an explanation.) Type of Advertising: The literature review pointed out that larger stations typically sell more of their advertising inventory to national or regional spot advertisers. Because the type of advertising may affect the dependent variables, stations were asked to indicate what proportion of their commercial time inventory was sold to spot and local advertisers. The proportion of local advertising 120 was entered as a continuous variable. (Only local was used for analysis because the two were highly correlated. See the Data Analysis section in the next chapter for a more detailed discussion.) Direct Broadcast Substitutes: Other local media including newspapers, magazines, radio, cable television, and other broadcast entertainment options like home VCRs can be considered substitutes for both viewers and advertisers for commercial television. Again, to measure and prove concentration, an industry's relevant market must be defined, and the criteria for including other local media is if they are considered close substitutes for TV advertising. Because advertising spending tends to be constant, and advertisers would probably switch out of TV if its price rises relative to other media, they can be considered substitutes. Presumably, a broadcast manager considers a price increase or advertising environment change very carefully if viable substitutes exist in his market. Yet newspaper and radio are not perfect substitutes,44 and researchers generally consider competition in one only medium because presently there is no accepted way of measuring interindustry competition.45 121 Therefore, the model was tested using "more perfect" broadcast substitutes, cable and VCR. Cable's influence on a station's audience is measured by entering an ADI's cable penetration as a continuous variable into a regression.46 Cable and VCR penetration figures are provided in Arbitron TV market reports, and were operationalized in this manner. They were later combined and divided by two to represent the average penetration of substitutes in a market. (See the Data Analysis section in the next chapter for more details.) Market Quality: Markets where consumers have more money to spend on advertised goods and services may be more attractive to national and regional spot advertisers. Greater wealth may also suggest a higher average level of education. Consequently, a manager must adapt her advertising and viewing environments to the composition of her audience. Market quality or prosperity is operationalized in one study using data from Sales & Marketing Management's Survey of Buying Power. Measures were the projected growth in households, and Effective Buying Income, which is roughly equivalent to total market disposable or after tax income, and represents audience ability to purchase advertised goods and services. It was operationalized as 122 a categorical variable, categorizing a market's median EBI as being above or below the national median by varying degrees, and proved to be a better measure of market quality than growth rate. However, a problem with using it is EBI market definitions do not always match Arbitron's.47 Consumer spendable income per household in each station's market is also a good measure of market quality.48 Average household after tax income is calculated by dividing total market EBI by the total number of households.49 Individual Arbitron reports include EBI, adjusted to Arbitron's definitions. Average household income was calculated for each market from this data, and included as a continuous variable. Rates: Previous research suggests that more profitable stations charge higher rates for advertising time,50 so rates may be an indicator of subclass membership. The price of advertising time may also affect the relative attractiveness of the advertising environment, perhaps forcing some advertisers to buy time on less profitable stations. 123 Spot prices have been operationalized as the highest rate quoted for a 30-second spot from the WW.“ Another study of news prices used rates from the fitpngntg_Rntg_nng_Qgtn W.52 The early fringe rate in the 8r2ad2ast1ngzgablesasfins_xearbook was entered as a continuous variable, as it covered most of the study period. The prime time rate was also included, as it might overlap with the study period in some cases. Former NAB Code Subscription: Stations were asked if they ever subscribed to the NAB Code, as previous involvement with a group involved in self-regulation may represent a predisposition to serving the public interest. Responses were included as a categorical variable, to see if former subscription affected performance on the dependent variables. QRerat12nal1zat12n_2f_the_DeRendent_Yariables Commercialization, clutter, and advertising clearance and complaint practices were the dependent variables. They were developed from precedents in past research. They were tested to determine if they embodied viable representations of station performance, in the hopes of extending broadcast policy research into the realm of advertising. 124 Commercialization: The old FCC 16 minute commercial time guideline, and the old NAB TV Code 16 minute non-program time guideline were used to assess commercialization level. The average number of program length commercials was also used to assess commercialization level. Station-originated programming was relevant for study because stations have complete control over scheduling in these programs. Because the NAB Code included other non-program material in its guideline, stations were asked how much commercial time they average per hour of station- originated programming plus the amount of non-program time (ads, promos, PSA's) they average. Stations were asked to calculate these times from either 3:00 to 7:00 p.m. or 4:00 to 8:00 p.m., depending on their time zone.53 Both were entered as continuous variables in the regression equation. Stations were also asked if they air program length commercials (and if so how many they air per week and how long they are) to see if lifting the ban had any effect. The average number aired per month was entered as a continuous variable in the regression equation. 125 One problem with measuring commercial time is that determining "excessive" amounts is somewhat arbitrary. The Code and FCC guidelines were based on industry standards. There is no "proof" that these standards are the actual line between just enough and too many commercials. These measures provide a convenient and logical point for comparison, but are considered in light of this fact. Clutter: The criteria for assessing clutter are from the NAB Code, as the FCC had no such guidelines. Generally, affiliated and independent stations were limited to 10 breaks per hour of station-originated programming: 8 breaks within programs and a break at the hour and half- hour.54 Five was the maximum number of non-program announcements to be scheduled consecutively in a break within an affiliate-originated program, and 4 in independent-originated programs. For affiliates, 4 or less of these announcements could be commercials, and 3 or less non-program material.55 The code also banned multiple product announcements, unless they were integrated to make them appear to be a single ad.56 126 Stations were asked to indicate the number of breaks they average per hour57 of station-originated programming, how many consecutive announcements they average in breaks, and whether, and how many, variable length units they accept, to indicate clutter level. Responses were entered as continuous variables in the regression, and/or compared to preexisting levels. Public service announcements were exempt from the consecutive announcement count, and newscasts from the interruption standards.58 The measure is confounded as stations exceeding the clutter standard may be airing PSA's, or interruption averages may reflect extra breaks in the news. Results were considered with this problem in mind. Formality of Clearance and Complaint Practices: Clearance and complaint formality were operationalized by asking which of the following policy formats were used, and which one was used most, and scored as follows: 1. Policies codified in a manual by the station were scored as first in level of formality (i.e., most formal operationalization of rules). 2. Policies which were mostly written were scored as second in level of formality. 127 3. Policies which were represented mostly in memoranda were scored as third in level of formality. 4. Policies which involve using another group's code (such as the NAB, BBB, etc.) were scored as fourth in level of formality. (This was assumed to be the least formal operationalization of rules, because station management did not develop them.) 5. Policies which were conveyed verbally were scored as fifth in level of formality. (Orally conveyed policies were assumed to be the least formal, and were representative of surveillance.) The original intent was to develop an index using all of these formats to represent their relative use and importance. Although the response options provided were based on previous research, an open-ended option was included to determine if other formats predominate, as a check. The policy form a station used most was ultimately used as a measure. (See the Data Analysis section in the next chapter for an explanation.) The method of communicating staff policies was operationalized using the surveillance/rules approach. 1. Requiring the staff to read communications regarding policy was scored as first in level of formality. (This was assumed to be the most formal means of communicating station rules.) 2. Encouraging the staff to read communications regarding policy was scored as second in level of formality. 128 3. Staff discussion of policy was scored as third in level of formality. (This represented the most formal operationalization of surveillance.) 4. Making the existence of policy known was scored as fourth in level of formality. (This was assumed to represent the least formal means of surveillance, and the least formal way of communicating policy overall). Again, the intent was to develop an index to represent the relative use and importance of these options. An open-ended option was again included to discern other significant methods of communicating policy. The communication method used most was finally used as a measure. (See the Data Analysis section in the next chapter for an explanation.) The questionnaire included a list of expected clearance policy areas and complaint groups suggested by the literature review. Stations were asked to indicate what standards they had for these variables. Presumably, the more official policies a station had, the more formal its practices. Responses were entered as continuous variables. 129 t' ' e ' es 0 The primary independent variables of interest were market structure or concentration, station profitability, and organization size. Their influence on the dependent variables, advertising commercialization, clutter, clearance and complaint practices, was the focus of interest. Regressions were used to test Hypotheses 1 through 10. The final list of independent variables included in the regressions were the primary predictors of interest and two control variables, the type of advertising and the average penetration of direct broadcast substitutes. As a result of the discovery of multicollinearity and non- normality, some variables were tested after transformation and bringing in outliers to three standard deviations, and others were dropped. Specifically, net weekly circulation and rates were dropped as they appeared to be contributing to multicollinearity. Market quality was dropped because the principal components analysis suggested it was measuring the same quality as market concentration. Former NAB code subscription was dropped because it did not contribute meaningfully to the regression equations. 130 T-tests were also used to assess Hypothesis 1 through 10, where appropriate. For example, organization size was divided at the mean and T-tests conducted to see if meaningful differences existed between smaller (mean and below) and larger (above the mean) organizations. T-tests were also used to test the relationship between the dependent variables and policy form and communication method used most, as these variables were not continuous and could not be tested using regression. Two way analyses of variance were used to test Hypotheses 11 through 14. The question was whether the variances for the dependent variables increased or decreased based upon the joint effect of market concentration and profitability. For example, as market structure and profitability increased, did the variance in commercialization decrease? The intent, for instance, was to consider whether a continuum existed with the more profitable stations in loose oligopolies selling the most commercial time, and the less profitable stations in monopoly power markets selling the least commercial time. Therefore, if the variances within groups were significantly greater than those between groups, it could be argued that market concentration and profitability had a joint effect on the dependent variables. If a 131 significant two way interaction was found, means and standard deviations were reported to illuminate the underlying finding. T-tests were used to determine whether the dependent variables varied by market structure categories and profitability classes. Market structure was categorized into four and sometimes two groups (less concentrated = loose oligopolies and oligopolies, more concentrated a tight oligopolies and monopoly power). The profitability index was also divided at the mean, so up to and including the mean represented more profitable stations, and above the mean represented less profitable stations. Tables with means and standard deviations were included to show how these differences were manifested. Chi-square tests were used to assess whether certain types of stations were more or less likely to accept certain time units, or have certain clearance standards. The data were examined to see if assumptions were met before testing the hypotheses. Multicollinearity was found in the independent variables, and some did not have normal distributions (see the Data Analysis section in the next chapter). T-tests using a separate variance estimate (rather than a pooled variance estimate) were used in some instances, because it could not be assumed that the 132 variances for the two groups were equal. A separate variance estimate T-test is used whenever probabilities for the F-test for equality of variance are small. The separate variance test was used when F-test significance levels were smaller than .10.59 The measures of the independent variables were regressed on each other to find the contributors to multicollinearity.60 A principal components analysis and a factor analysis (or principal factors extraction) were run to confirm the predicted theoretical relationships between independent variables, and to decide which ones to drop from the final regression equations. When relationships appear to exist among variables in a correlation matrix, as was the case here, these procedures determine whether the relationships within some subsets of variables are higher than those between the subsets.61 For example, were broadcast band, network affiliation and net weekly circulation actually predictors of profitability, and were spot and local advertising really interchangeable measures of the type of advertising? These procedures can suggest an answer, and both analyses were employed because finding similar factor solutions supports the notion that the final factors are stable and thus reliable.62 NOTES: CHAPTER IV 1Robert H. Prisuta, "Policy Research, " In Joseph R. Dominick and James E. Fletcher, eds., BIQAQQAELinQ R§_§n;pn_nptnpg_ (Boston: Allyn & Bacon, 1985), p. 171. 2Graham Kalton. Infr2dustion_to_§urxex_§amelins (Beverly Hills, Ca.: Sage, 1983), p. 5. 3Barry R. Litman, "Economic Methods of Broadcast Research, " In Joseph R. Dominick and James E. Fletcher, eds., Br2adQasf1ng_Eeseargh_uetnods (Boston: Allyn & Bacon, 1985), p. 107-15. 4The 1988 Br9ad2ast1ngLCablesasfing_Yearhook reports that there are 1017 commercial TV stations in the United States, and 870 were in the survey population. Therefore, 147 stations were omitted. It is not known how many of these are small market, religious, satellite, or Home Shopping Network stations, as the small markets edition was not available to the researcher. 5Earl R- Babbie. Tne_2rastiss_of_§22111_3esear§h. 3rd ed. (Belmont, Ca.: Wadsworth, 1983), p. 236-7. 61bid., p. 229. 7Ibid., p. 235. 8Nonrespondents in the four pretests indicated that they did not respond because they thought it would take too much time to fill out the questionnaire. Quite a few indicated in follow-up telephone calls that they would not complete any questionnaire that looked like it would take more than five minutes to complete. 9Robert Prisuta, "Local Television News as an Oligopolistic Industry: A Pilot Study," Qpnpnnl_pt atoadgasting 23 (Winter 1979), p. 63. 133 134 10Robert Prisuta, "The Impact of Media Concentration and Economic Factors on Broadcast Public Interest Programming. Journal_2f_8roadga§fing 21 (Summer 1977)." p- 56. 11Barry R. Litman, "Measuring Divestiture of Network Owned Television Stations: An Econometric Approach,," Th2 Antitrust_nulletin 25 (Summer 1980). p. 369- 12Benjamin J. Bates, "Determining Television Advertising Rates," in Robert N. Bostrom, ed. Communisafion_xearppok_l (Beverly Hills. Ca-= Sage. 1983). p. 465.: and Michael O. Wirth and James A. Wollert, "The Effects of Market Structure on Television News Pricing," Iournal_ofaroadsasting 28 (Spring 1984). p- 221-2- 13Stations in each market were listed by market rank, to implicitly stratify by station type. Because VHF/affiliated stations are often ranked in the top three, and UHF/independent stations afterwards, choosing a random start and selecting a fixed skip interval insured the selection of different station types. Periodicity, or the problem of drawing a biased sample because the list of elemetns is arranged in a cyclical pattern, it avoided because markets vary in the number of stations. The fixed skip interval therefore did not "land" on the same type of station each time. See Earl R. Babbie, fintxgy_3g§pntgn uptnpgp (Belmont, Ca.: Wadsworth, 1973), p. 93 for an explanation of periodicity. The sample list was not included as the explanatory letter promised that station identity would not be revealed in any published report. Information regarding the list may be obtained through personal requests to the author. 14Barbara G. Tabachnick and Linda S. Fidell, fining Multixariate_Statistip_ (New York: Harper & ROW. 1983).p p- 91-2. 15Don A- Dillman. Ma11_and_Telephone_§urxexs (New York: Wiley, 1978), p. 180-91. 135 16The explanatory letters were printed using a near letter-quality printer on two-color university letterhead, and signed individually with a blue ball-point pen. The questionnaire was reproduced on high-quality cotton bond, and prepaid, bulk-rate response envelopes were included with it. All were folded to comply with expert recommendations and inserted into stamped envelopes with individually typed addresses, to make each letter look as personalized as possible. See Dillman, p. 181. Review Chapter 5, "Implementing Mail Surveys," for excellent recommendations on how to conduct a mail survey. 17Both sides were hand-typed and copied onto Antique Beige card stock. Each was then individually signed with a blue ball point pen to make it look personalized. See Dillman, p. 183-6. 18The same personalization procedures from the first mailing were used. See Dillman, p. 186-8. 19For example, deeper probing on the methods used for clearance review at stations: whether a background check is conducted for prospective advertisers, or advertisers of questionable character: and whether and how stations preview ads before airing by inspecting package inserts, storyboards, etc. 20Litman, "Economic Methods," p. 116.: and F. M. Scherer, Industr1al_Market.§1rusture_and_nspnom12 Egtfptngngp (Chicago: Rand McNally, 1970), p. 53. 21Litman, "Economic Methods," p. 117. 22Scherer, p. 50-1. 23Ibid., p. 51-2. 24In_the_Matter_2f_Ame_dment_of_Seption_1111§§§ [fppne :11 Sgctions 73. 35. 73. 240 nng 73,636] pf the Qpnmission's Ruins Reinting to Mpltipl g anpts hip 92 AM, EM, and Teigvisipn Rroadcast Mtntipn 100 FCC 2d 17 at 42 (1984) See also n1_S1_Department_of_lu§fise_uerger §_igglinp§, issued June 14, 1982. 136 25Iloid. See also WW ns 73 7 . a 7 3 f 's ' o t' wne s ' o Ipigvision ppoadcast Stgtions, 95 FCC 2d 360 at 386 (1983) . 7 and WW -0011 ’ oms 701': 1° ,0! 'v o v 5 cat 0 o o un 't ntenna Telev's'o stem5° and i w es ec thereto to Eppnulate Regulatgpy Rpligy nng Rnignghing WM. 91 FCC 2d 46 at 51-2 (1982)- 26Wirth and Wollert, p. 217. 27 O 1- S ‘ ‘ 8 01'1' ‘WI' 1 0! Rgpptt, (New York: Arbitron Ratings Co., February 1987). 28Ibid. The report includes all stations with at least a .1 share, calculated from the same diary data. 29Litman, "Economic Methods," p. 118-20. The Justice Department uses .18 or 1800 to indicate a significantly concentrated industry. 30Barry R. Litman, "Public Interest Programming and the Carroll Doctrine: A Re-examination," lpntngl_gf Rtpgggnpting 23 (Winter 1979), p. 52, 59. See Note #5. 31Litman, "Measuring Divestiture," p. 368, 371-4. 32Bates, "Determining," p. 464.: Litman, "Measuring Divestiture," p. 371-2.: Litman, "Public Interest Programming," p. 56.: Wirth and Wollert, p. 219, 223.: and Michael O. Wirth, "The Effects of Crossmedia Ownership on Television and Newspaper 'Prices'," Unpublished Doctoral Dissertation, Mass Media Ph.D. Program, Michigan State University, 1977, p. 56-7. 33Bates, "Determining," p. 468.: Litman, "Measuring Divestiture," p. 372.: Litman, "Public Interest Programmingr" p. 56.: Wirth, p. 47, 56-7.: and Wirth and Wollert, p. 219. 34Litman, "Public Interest Programming," p. 55-6.: and Wirth, p. 48. 137 35 .t o S w Yo - et ' e s m , (New York: Arbitron Ratings Co., 1987), February 1987, Survey Period: February 4 to March 3, 1987., p. vi. This definition can be found in any Arbitron market sweeps book. 36Litman, "Measuring Divestiture," p. 372-3. 37Litman, "Public Interest Programming," p. 60, Note 13. 38(Washington, D.C.: Broadcasting Publications, Inc., 1987). 39Joseph A. Litterer, O s: d Rphgyip; 3rd ed. (New York: Wiley, 1980), p. 372. 40Eric J. Walton, "The Comparison of Measures of Organization Structures," Academy pf ngngggnpnt Rgvipn 6 (1981), p. 157. 41Bruce M. Owen, Jack H. Beebe, and Willard G. Manning, Jr., Igigvision Ecpnonigp (Lexington, Mass.: Heath, 1974), p. 96. 42William E. McCavitt and Peter K. Pringle, Elestrpnig_nedia_nanagement (Boston: Focal Press. 1986). p. 132-3. 43Barry R. Litman, "Market Share Instability in Local Television News," lournal_of_firpadpa§ting 24 (Fall 1980). p. 503.: and Wirth and Wollert, p. 215. 44Benjamin J. Bates, "Economic Theory and Broadcasting," Presented to the Mass Communication Theory and Methodology Division of the Association for Education in Journalism and Mass Communication Convention, Memphis, Tennessee, August 1985, p. 8-9. 45John Dimmick and Eric Rothenbuhler, "The Theory of the Niche: Quantifying Competition Among Media Industries," Journal_2f_92mmunipatipn 34 (Winter 1984). p. 104. 46Wirth and Wollert, p. 219-20, 223. 47Bates, "Determining," p. 465, 471-2. 138 43w1rth, 28-9, 56-7.; and Wirth and Wollert, p. 219- 20, 222. 49Sales & Marketing Management, Egypt, Part 1, (New York: Bill Communications, 1986) 137:2, p. A 29-30. soLitman, "Measuring Divestiture," p. 368, 371-3. 51Bates, "Determining," p. 464. 52Wirth and Wollert, p. 218-9. 53Pretesting confirmed that managers understood this to mean the "clock hour," including non-program material within as well as adjacent to the program. For example, if a program began at 3:00 p.m. and ended at 4:00 p.m., managers included commercials within the program and at the 3:28:30 and 3:58:30 break. Calculations were made for each hour this way, and managers said it was not confusing to calculate programming originating on their station. The "clock hour" idea seems a good way to insure that all parties are counting time the same way and comes from: Edward E. Cohen. Ch1ldrenls_Ielex1sion§ommersializatipn fintgpy (Washington, D.C.: National Association of Broadcasters), p. 2, 6.: and Erwin G. Krasnow, Barry D. Umansky and William E Kennard Commenf§_of_fhe_national Asso_1ation_gffiroadsa§ter§. MM Docket No. 83- -670 21 November 1983, p. 3. Thanks to Dr. Richard V. Ducey, Senior Vice President of Research and Planning at the National Association of Broadcasters, for supplying these materials. 54NAB Television Code, 0 t s igntppph_12§1 (Washington, D.C.: National Association of Broadcasters, Broadcast Publications, 1982), p. D-18. SSIbid. 561bid., p. 0-17. 57See note 57. 58NAB TV Code, p. 0-18. 59Marija J. Norusis. W (Chicago: spss, 1986), p. 203. 139 60Tabachnick and Fidell, p. 82-3. 61Jae-On Kim and Charles W. Mueller, Inttpgngtipn_tp Rngtpt Anaiysis (Beverly Hills, Ca.: Sage, 1978), p. 9. 62Tabachnick and Fidell, p. 394. CHAPTER‘V RESUDTS A 62.6 percent response rate (482 of 769 stations) was achieved. Six of these questionnaires were unusable.1 Stations used in pretests (90), stations no longer on the air (5), and undeliverable surveys (6) were subtracted to obtain the population size (870 - 90 - 5 - 6 = 769). The 1988 Br2ad2ast1ngLCablesasting_Xearpook reports that there are 1017 commercial television stations in the United States (541 VHF and 476 UHF).2 Responses thus represent 47.4 percent of all commercial television stations in this country. Respondents represent a cross section of stations. Responses approximated market structure's natural proportions, demonstrated by comparing each category's percent of usable responses to its survey and total sample percentages. At least 58 percent of stations responded in each category (see Table 2). 140 Percent of 141 Total Sample (N=870) Percent of Survey Sample (N=769) Percent of All Usable Responses (N=476) Number of Responses Returned Percent of Category Responding TABLE 2 Survey Responses by Market Structure (N=769) Loose Tight Oligopolies Oligopolies Oligopolies (n=211) (n=253) (n=182) 27.7% 31.8% 23.9% 27.4% 32.9% 23.7% 26.1% 35.1% 23.3% 124 167 111 58.7% 66.0% 60.9% Monopoly Power (n=123) 16.6% 16.0% 15.5% 74 60.1% Responses for broadcast band approximated the population's natural proportions, where VHF stations represent 53.2 percent of all stations and UHF 46.8 percent. Of responding stations, 56.5 percent were VHF and 43.5 percent UHF, representing a departure of only 3.3 percent. Affiliation status also mirrored nature. Independents represented 28.2 percent (134 of 476) of respondents, with network affiliates representing 71.8 142 percent (342). Independents represent about 32.2 percent of all commercial television stations (approximately 327 of 1017), and network affiliates represent about 67.8 percent (690 of 1017).3 Only a four percent difference existed. When network affiliation is partitioned, NBC stations accounted for 23.9 percent (114) of respondents, CBS stations 25.4 percent (121) and ABC stations 22.5 percent (107). 'o d s : Violations of Normality: Frequencies were run to determine if the assumptions of normality were valid before analyzing the joint effects of all of the independent variables using regression. Some of the independent and dependent variables were skewed (see Table 29), exhibiting non-symmetric distributions with more observations (or a tail) at one end. Positively skewed distributions have a pileup of cases with very low scores, and those with negative skewness very high scores.4 Some variables are naturally skewed. For example, since incomes are rarely less than zero, and some are much higher than average, a long tail to the right is expected.5 The same is true for some of the study 143 variables. Commercial and non-program time cannot be less than zero, and a few stations averaged much more commercial time than others. Consider also that responses to organization size are limited, as a station cannot have fewer than 1 employee in sales and traffic, and the maximum response was 55. The coefficient of skewness "summarizes the extent of asymmetry in a frequency distribution," and its value is zero if the distribution is exactly normal. A negative coefficient indicates a negatively skewed distribution, and the same for a positive coefficient.6 A variable is considered skewed if its coefficient is 1 or above.7 The positive coefficients for commercial time, non-program time, number of program length commercials, clearance hours, rank and organization size support the notion of a non-zero value with expected outliers with large values (see Table 3). Although moderate departures from normality do not seriously affect the validity of many procedures based on the normal distribution,8 transformations can change a variable's distribution to reduce skewness.9 A successful transformation may eliminate non-normality.10 Typically, logarithmic or square root transformations are used.11 144 TABLE 3 Skewed Independent and Dependent Variables DEPENDENT VARIABLES Commercial Time Non-program Time Accept Program Length Commercials? Number of Program Length Commercials Per Month Hours Per Week Devoted to Clearance 15 Second Time Unit 20 Second Time Unit 45 Second Time Unit Seen Ad You Declined On Other Station? INDEPENDENT VARIABLES NAB Code Subscription Sales/Traffic Department Size Rank Prime Time Rate Early Fringe Rate Net Weekly Circulation Penetration of Direct Broadcast Substitutes COEFFICIENT OF SKEWNESS Transformation Outliers Moved Before After to 3 St. Devs. 2.187 .646 .469 3.792 1.653 1.223 1.347 + 5.090 .453 2.226 3.020 .757 -3.683 + 1.294 + 1.022 + 1.968 + 2.433 *+ 1.377 .251 .586 1.639 ** 7.561 * 5.661 * 8.810 * 3.868 .784 2.284 *Variables which were not retained for the final regression model were not transformed. **Although Rank was moderately skewed, it was not transformed, as it was categorized and included in the profitability index. +Dichotomous variables were not transformed. 145 Both transformations were tried on the appropriate variables. The criteria for choosing between transformations were: a) if the sign on the coefficient of skewness remained the same, and b) if the transformation reduced skewness. Using these criteria, a natural log transformation was selected for non-program time, the hours devoted to clearance and the number of program length commercials per month. Square root transformations were selected for commercial time, organization size and the average penetration of broadcast substitutes.12 Market rank was not transformed as it was categorized for inclusion in the profitability index. Extreme cases may also be rescored or changed in such a way that their influence is reduced. Outliers were moved back to three standard deviations from the mean to retain their deviancy and prevent them from distorting the correlations.13 Regressions using natural data, transformed data, and data with the outliers brought in to three standard deviations were utilized for analysis. Decision rules for accepting or rejecting the hypotheses were therefore developed. Dependent variables with three measures (commercialization, tested in Hypotheses 1 and 5 and measured by the average amount of commercial and nonprogram time and by the average number 146 of program length commercials: and clearance formality, tested in Hypotheses 3, 7 and 9 and measured by the number of hours devoted to clearance, the number of policy areas and policy sources) required nine significance tests. That was because each one variable was tested using three regressions (natural data, transformed data and data with the outliers brought in). Therefore, if one to three of the tests were significant, the hypothesis was considered weakly supported. If four to six were significant the hypothesis was partially supported and if seven to nine were significant the hypothesis was considered strongly supported. If all regressions were insignificant, the hypothesis was not supported. A similar decision rule was developed for clutter, which had two measures (the average number of breaks and consecutive nonprogram announcements, tested in Hypotheses 2 and 6). There were six significance tests in this case (three for clutter and three for consecutive nonprogram announcements). If one or two of the tests were significant, the hypothesis was weakly supported. If three or four were significant it was partially supported, and if five or six were significant it was strongly supported. If all regressions were insignificant, the hypothesis was not supported. 147 Multicollinearity: Pearson correlations between the independent variables were examined for multicollinearity (see Appendix C), as previous research suggested it might be present.14 Multicollinearity is found by looking for a high correlation between two variables, and by regressing each independent variable on all others.15 The betas for each independent variable contributing to the "dependent" variable, and its contribution as a percentage of the variance explained, were examined to aid in this assessment. Variables appearing to be the cause of high multicollinearity were dropped.16 The preferred method of dealing with multicollinearity is to drop the offending variables. The choice of which variables to drop is guided by their theoretical relevance and ease of interpretation.17 Highly intercorrelated variables can be combined into a single indicator, if it makes conceptual sense.18 Net weekly circulation appeared to be a cause of multicollinearity, as it was highly correlated with other independent variables. (For example, .63 with organization size, .80 with prime time rate, .75 with early fringe rate, and .61 with VCR penetration). It measures differences in the signal power of each 148 individual station and reflects the number of intra- as well as inter-market viewers, and has been used as an indicator of market size (and a proxy for profitability).19 Net weekly circulation appears to be a measure of size rather than profitability in this study, given its non-significant and negative correlation with network affiliation (-.03, p=.45). Spot and local advertising, cable and VCR penetration, and network affiliation, market rank and broadcast band also appeared related, respectively.20 Regression analysis suggested that market structure, NAB code subscription, and local advertising were independent (See Appendix D). Net weekly circulation appeared to be a cause of high multicollinearity, as the other independent variables explained about 87 percent of its variance. It appeared to be a multiple indicator, which is consistent with theoretical precedents suggesting that it measures both market size and profitability.21 A principal components analysis and a factor analysis (or principal factors extraction) were run using varimax rotation22 to confirm the predicted theoretical relationships between independent variables, in order to decide which variables to drop or combine. For example, if broadcast band, network affiliation and net weekly 149 circulation all load on one factor in the analysis, the suggestion is that they are measuring profitability and may be combined into an index. Finding similar solutions in both procedures suggests that the factors are stable.23 Using both procedures thus provided a reliable way to discern if net weekly circulation was actually measuring profitability or market size in this study. Principal components analysis and factor analysis generally represent a set of variables in terms of a smaller number of hypothetical variables. In other.words, when there appear to be relationships among variables in a correlation matrix, these techniques tell us whether the relationships within some subsets of variables are higher than those between the subsets.24 Factor analysis is the more rigorous of the two techniques because the variance that is unique to each variable in the analysis, and the error variance, is excluded. Therefore, only the "pure" underlying relationships in the data are analyzed. Unique and error variances are included in a principal components analysis, which makes it easier to obtain factors with this technique.25 Therefore each technique is commonly used in certain situations. Principal components analysis is used to reduce a large number of variables to a smaller number. 150 It is often used in exploratory research where underlying theoretical relationships are to be clarified and is recommended as a first step before a more detailed factor analysis.26 Factor analysis is used to confirm whether the underlying structure of variables is as predicted by theory. Multicollinearity is not a problem in principal components analysis,27 and violations of normality and linearity assumptions are not a problem when principal components analysis and factor analysis are used simply to summarize relationships, as in this instance. Both procedures were used to confirm to a reasonable degree that the factor solutions were reliable. A relatively large sample should be used for factor analysis. Sample size is typically evaluated using the following scale: 50--very poor: 100-—poor: 200--fair: 300--good: 500--very good; and 1000--excellent.28 The sample size of 476 is therefore acceptable. Accepted guidelines exist for deciding whether a variable actually loads on a factor, and are as follows: loadings in excess of .71 (50% variance) are considered excellent, .63 (40%) very good, .55 (30%) good, .45 (20%) fair, and .32 (10% of variance) poor.29 Factor loadings above .30 are considered "salient," suggesting a relationship between variable and factor. 151 Items loading at .60 on one factor but above .30 on another should be deleted for this reason.30 This .30 difference criterion was used. The typical criterion for deciding to retain a factor is whether it has an eigenvalue greater than 1.0, which was also used.31 The principal components analysis and factor analysis were reviewed to discern if any of the independent variables was measuring the same construct, so alternate variables measuring the same quality could be dropped, and related variables could be combined into an index. The factors were also examined to see if variables that "should" load together actually did. For example, the literature review suggested that broadcast rand, network affiliation and net weekly circulation should all load together on a profitability factor. However, the Pearson correlations suggested that this may not be true. As a result, we might expect that net weekly circulation was measuring market size in this instance, and would load on the same factor with rates and other independent variables which might indicate market size, like organization size. Recall that the literature review suggested that larger stations earn higher profits32 and thus charge higher rates.33 152 But if net weekly circulation is not a measure of profitability in this instance, what might be? One would expect network affiliation and broadcast band to comprise a profitability factor. If true, one could construct an index from the two, as they are measuring the same underlying construct. Examining these independent variables in the context of a principal components and factor analysis also provides an opportunity to suggest how they might be interrelated to aid future research. The combination of the large sample size and operationalization of many variables traditionally used in mass media economic research provided an opportunity to confirm, explain, and possibly reveal interrelationships. For example, were the previous assumptions about profitability and market size supported? Were there other relationships among variables which may have been previously missed? And could certain variables be used interchangeably, with the theoretical interests of the study dictating which should be used? The principal components analysis resulted in four factors which were generally consistent with theoretical precedents discussed in the literature review (see Table 4.1 and 4.2). Results confirmed the suspicion that net 153 weekly circulation was actually a measure of market size. It did not load on the same factor as the other profitability indicators (see Table 4.2). TABLE 4.1 Principal Components Analysis on the Independent Variables (Factors and Variance Explained) Factor Eigenvalue Percent of Cumulative variance 1 Market Size 4.20443 30.0 30.0 2 Profitability 2.97026 21.2 51.2 3 Market 1.77785 12.7 63.9 Competitiveness 4 Type of 1.22745 8.8 72.7 Advertising 5 Factor .84913 6.1 78.8 6 Factor .76909 5.5 84.3 7 Factor .53890 3.8 88.1 8 Factor .49575 3.5 91.7 9 Factor .37057 2.6 94.3 10 Factor .26643 1.9 96.2 11 Factor .2235? 1.6 97.8 12 Factor .19322 1.4 99.2 13 Factor .0804? .6 99.8 14 Factor .03288 .2 100.0 154 TABLE 4.2 Principal Components Analysis on the Independent Variables (Varimax Rotated Factor Matrix) EQQLQI_1 E§E£2£_Z EQQLQI_1 EQQLQ£_£ Subscribe to NAB Code .03130 -.60641 -.22698 .06350 Spot .10430 .07193 .04681 -.92687 Local -.04114 -.05173 -.03843 .93015 Market Concentration .13013 -.02107 -.61285 .21539 Broadcast Band .24614 .76137 .00075 -.03040 Network Affiliation .04029 .85194 -.20865 .00832 Rank -.02688 -.82175 .33090 .09235 Prime Time Rate .87488 .14990 -.02601 -.10174 Early Fringe Rate .87437 .08054 .00867 -.l4688 Net Weekly Circulation .92033 -.01539 .18934 .09876 Market Quality .20029 -.03l93 .77026 -.01345 Cable Penetration .53168 -.23965 .69699 .14727 VCR Penetration .52709 -.21525 .68260 .17024 Organization Size (Sales & Traffic) .66773 .17398 .30527 -.08193 (NOTE: Rank's correlation is negative because of the original coding scheme. For example, independents were originally coded as "0" and affiliates as "1" for the variable Network Affiliation. Thus, as affiliation "decreased," rank "increased" (say, from first to third). Coding was changed when the profitability index was created to eliminate this problem.] 155 Factor 1 is described as market size, and includes prime time rate, early fringe rate, net weekly circulation and organization size. Because organization size was a predictor of primary interest, it was retained and the other variables were dropped. Factor 2 is called profitability, as the traditional theoretical proxies (network affiliation and broadcast band) achieve excellent loadings on it. Market rank also achieves an excellent loading here, suggesting that it should be considered as a profitability indicator in future studies. The combination seems intuitively correct when one recalls that VHF, network-affiliated stations usually have higher market rankings (and audiences) than independent, UHF stations, and are probably more profitable. Therefore, the decision was made to combine these variables into a profitability indicator. Factor 3 is called market competitiveness, as market concentration achieves a good loading, and market quality an excellent loading. Market concentration represents the level of competition between stations in a market. Market quality decreases as concentration increases, suggesting that the more stations there are, the more attractive the market is likely to be to advertisers, because of the wealth of the audience. Thus, the market is more 156 "competitive" for the spot advertising dollar than more concentrated markets. Since market structure is a predictor of primary theoretical interest, market quality was dropped from the regressions. Factor 4 clearly represents the type of advertising, with spot and local both achieving excellent loadings. Results support the notion that these are interchangeable measures. Local was retained, as the multicollinear regressions suggested that it was more independent of the other independent variables (see Appendix D). The factor analysis results suggested that three of the four factors were stable, as the factors for market size, profitability and type of advertising were found again (see Tables 5.1 and 5.2). Factor 1 again appears to indicate market size (see Table 5.2), with cable and VCR penetration also loading on it this time.34 The profitability factor structure appears stable, as Factor 2 is again composed of broadcast band, network affiliation and rank. Factor 3 (Type of Advertising) is also stable as spot and local again achieving excellent loadings. However, the market competitiveness factor was not replicated. 157 TABLE 5.1 Factor Analysis on the Independent Variables (Factors and Variance Explained) Factor Eigenvalue Percent of Cumulative variance 1 Market Size 3.97774 28.4 28.4 2 Profitability 2.65976 19.0 47.4 3 Type of 1.53035 10.9 58.3 Advertising 4 Factor .82419 5.9 64.2 Using both techniques aided in the decision on what size measures to retain. Although organization size, and cable and VCR penetration both loaded on Factor 1 in the factor analysis (see Table 5.2), all were retained as size measures. Cable and VCR penetration did not load on any factor in the principal components analysis (see Table 4.2). Although they had excellent loadings on market size in the factor analysis (see Table 5.2), their Pearson correlations with organization size were not quite as high (Cable = .45, sig =.000: VCR = .44, sig 8.000) as their correlation with each other (.95, sig = .000). It seemed they might be measuring different aspects of size, so cable and VCR penetration were summed and divided in two to represent the average penetration of substitutes in a market. 158 TABLE 5.2 Factor Analysis on the Independent Variables (Varimax Rotated Factor Matrix) E§£L2£_l £§E£Q£_2 EQQEQI_1 EEEEQI_A Subscribe to NAB Code -.05721 -.34118 -.02427 -.21044 Spot .13849 .38398 -.82058 -.00627 Local -.08000 -.34261 .78422 .00070 Market Concentration -.16894 .10815 .19543 -.15826 Broadcast Band .16813 .63172 .16172 .29483 Network Affiliation -.10111 .76557 .26431 .25936 Rank .16962 -.81973 -.20043 -.13371 Prime Time Rate .71735 .37684 .09626 -.34672 Early Fringe Rate .72883 .31479 .02703 -.32691 Net Weekly Circulation .89434 .11680 .20992 -.27261 Market Quality .4958? -.19957 -.11246 .27178 Cable Penetration .82967 -.38262 -.00397 .31947 VCR Penetration .80634 -.35894 .02466 .32044 Organization Size (Sales & Traffic) .65039 .19994 .00625 .05238 [NOTE: Rank's correlation is negative because of the original coding scheme. For example, independents were originally coded as "0" and affiliates as "1" for the variable Network Affiliation. Thus, as affiliation "decreased," rank "increased" (say, from first to third). Coding was changed when the profitability index was created to eliminate this problem.] 159 We: Multiple measures were operationalized for profitability, as previous research had used broadcast band, network affiliation and net weekly circulation as proxies. However, results of the factor analyses showed that broadcast band, network affiliation and market rank were actually profitability indicators in this study. A profitability index incorporating the three was developed.35 Lower scores indicated less profitable stations, and higher scores more profitable stations. The next step was to determine if the index was reliable. Reliability estimates based on the average correlation among items comprise internal consistency. Coefficient alpha is the commonly accepted way to determine reliability based on internal consistency,36 and one considers a measure's intended use to determine the satisfactory level of reliability. In the early stages of research on hypothesized measures of a construct, a modest reliability of .70 is satisfactory. For basic research, reliabilities of .80 or above are considered satisfactory. In applied research where important decisions are made based on scale scores (e.g., whether a child should be placed in a special class based on a low IQ score), a reliability should meet or exceed .90.37 160 Given that theoretical precedents existed for the index being created, .80 was selected as the reliability criteria. Coefficient alpha was .8543 for the profitability index, so it was retained for hypothesis testing. Multiple measures were also used for three of the dependent variables. For commercialization, the average amount of commercial and non-program time, and the number of program length commercials per month were used. For clutter, the average number of breaks, and the average number of consecutive non-program announcements in breaks were used as measures. For advertising clearance, the hours per week devoted to clearance, the number of policy areas a station had, and the number of policy sources usually consulted were used. The usual form and method by which policies were communicated were also used to measure the formality of both clearance and complaint practices (called advertising practices where appropriate). Attempts were made to create indices or scales for the dependent variables. Individual measures for commercialization, clutter, clearance and the usual form and method of communicating policies were retained because the study was exploratory. Their usefulness could therefore be assessed to determine if they should be retained for future advertising performance studies. 161 Summed scales were created as measures of clutter, and clearance and complaint formality. Specifically: a) the total number of time units accepted by a station was intended to be one measure of clutter, b) the total number of policy areas a station has was one measure of clearance formality, c) the total number of policy sources a station consults was a second measure of clearance formality, and d) the total number of groups a stations had complaint policies for was a measure of complaint formality. The .70 reliability criteria was used for these scales as they were hypothesized measures of a construct being tested in an exploratory study. The Kuder- Richardson (KR-20) test was used as the reliability measure. It is a special version of coefficient alpha that is used when dichotomous items form a scale,38 as was the case here. The total number of time units was not retained for analysis, as its reliability coefficient was too low (KR- 20 Alpha = .4571). Chi-square tests were used in Research Questions 2 and 6 to determine if market structure or profitability affected the likelihood of accepting or rejecting certain time units. The chi-square technique is used as discrete categories (whether or not a station accepts a certain time unit) were being analyzed.39 162 However, the sum of the number of policy areas (KR-20 Alpha = .8793), the sum of the number of policy sources (KR-20 Alpha = .7123), and the sum of number of complaint policies (KR-20 Alpha = .8918) each station had were retained for analysis. The attempts to create individual or summed scales for policy form and communication method were unsuccessful. Combining day-to-day forms and methods with the one used most for each were tried, as well as various combinations of the two variables. No measure was more reliable than simply using the form and method a station used most often as formality indicators. So, the form in which a station's clearance and complaint policies were communicated and the method used to communicate both clearance and complaint policies were discussed in special hypothesis sections. These special hypothesis sections, entitled Hypotheses 3 and 4 (for market structure), Hypotheses 7 and 8 (for profitability), and Hypotheses 9 and 10 (for organization size), were discussed this way because the form and method of communication were measures of formality for both clearance and complaint practices. In other words, when a station indicated its usual form and method of communication, the question was phrased to mean 163 all of its advertising policies. Stations were not asked to indicate the usual form and method of communication separately for clearance practices, then complaint practices. Consequently, these measures represented the usual form and method in which nil of a station's advertising policies were communicated. Multiple measures were used to represent clearance and complaint formality. That is why the formality of both clearance and complaint practices are considered jointly. For example, Hypothesis 3 discusses the relationship between market structure and the formality of clearance practices, measured as the number of policy areas a station has and the number of policy sources it usually consults. Hypothesis 4 discusses the relationship between market structure and formality of complaint practices, measured as the number of groups a station has established complaint policies for. A third section, entitled "Hypotheses 3 and 4," discusses the relationship between market structure and usual form and method used to communicate hpth clearance and complaint practices. The same format is used to discuss the relationship between profitability and the usual form and method used to communicate both clearance and complaint practices (the section named "Hypotheses 7 and 8"), and organization size 164 and the usual form and method used to communicate both clearance and complaint practices (the section named "Hypotheses 9 and 10"). One option in the policy form questions was if a station's policies were communicated in the last NAB Code (see question 12 in Appendix B). This option was suggested by a previous study. Examination of correlations suggested that the written and verbal components were related, but this option unrelated to both. It appeared that this option measured some distinct quality apart from the other four, so it was dropped. '3 ss 0 es's s: Hypotheses 1 through 10 were tested using regression analysis. The final list of independent variables included in the regressions were the predictors of primary concern (market structure, profitability and organization size) and two control variables (the average penetration of direct broadcast substitutes and local advertising). Regressions on the natural data, variables with outliers reduced to three standard deviations and transformed variables were used to test the hypotheses. Results were similar, and often identical, regardless of the regression analysis interpreted. However, all three are reported as some variables were significant after transformation or 165 bringing in outliers, but not before. Decision rules for nine significance tests were used for Hypotheses 1, 3, 5, 7 and 9, and decision rules for six significance tests were used for Hypotheses 2 and 6. Regression analysis was not used for complaint formality and the usual form and method of communication, as each of these variables had less than five categories. In those cases, T-tests were used. T-tests were also used to demonstrate differences between categories of the independent variables, where appropriate. Hypotheses 11 through 14 were tested using analysis of variance. Results were only reported when a significant two way interaction was found. In that instance, means and standard deviations were also presented to see if variances increased or decreased as predicted. T-tests were used to test Research Questions 1 through 8. If significant differences between market structure types or profitability classes were found, the argument could be made that distinguishing stations by type or class was useful. Hypothesis 1: Hypothesis 1 predicts that as market concentration increases, commercialization decreases. The average amount of commercial and non-program time scheduled per 166 hour of station-originated programming during early fringe and prime access (3:00 to 7:00 or 4:00 to 8:00 p.m.), and the average number of program length commercials scheduled per month were the measures of commercialization. Regression results are in the predicted direction but significant only for program length commercials using untransformed data (see Table 6). As market concentration increased, the number of program length commercials scheduled by stations decreased. Hypothesis 1 is weakly supported. Hypothesis 2: Hypothesis 2 predicted that as market concentration increases, the level of clutter decreases. Clutter was measured using the average number of breaks per hour, and the average number of consecutive non-program announcements per break. Results are significant for all three regressions for the number of commercial breaks as expected (see Tables 6, 7, 8), and the independent variables explained more than eight percent of the variance. No relationship was found for the number of consecutive non-program announcements. Hypothesis 2 is partially supported. 167 TABLE 6 Regressions Uith Untransformed Variables W Qgpgnggnt Market Profitability Organization Substitute yptinpigg Concentration Index Size Commercial Tine °.070 -.041 .122* Non-program Time Not significant Program Length -.111* -.289*** .009 Commercials Breaks -.129*‘ .285*** .006 Consecutive Non-program “at significant Announcements Clearance Hours .022 .029 .061 Policy Areas -.001 .225*** .241*** Policy Sources ~.008 .240*** .078 * p is less than .05 ** p is less than .01 *** p is less than .001 #476 (Hinim pairwise umber of cases I 418.) Penetration .074 .067 .356*** .022 .047 Local Adjusted O 027 .128** .080 -.025 -.035 .057 R Square .012* .115*** .086*** .125*** .122“‘ .055'** .1658 TABLE 7 Regressions With Outliers Reduced to 3 Standard Deviations nde Index .061 .103‘ - . 23‘... .285**‘ °.014 .218*** .240*** mm Market r bl Concentration Commercial Time -.062 Non-program Time -.045 Program Length -.032 Commercials Breaks -.129** Consecutive Non-program Not significant Announcements Clearance Hours .011 Policy Areas .000 Policy Sources -.008 * p is less than .05 ** p is less than .01 *‘* p is less than .001 n V l Profitability Organization Size .191*** -.072 .002 .038 .Z4S**' .080 ll=476 (Hiniuun pairwise nulber of cases = 418.) Substitute Penetration -.020 .108* .229*** -.000 .038 Local Adjusted .012 .019 -.006 -.029 .057 R Square . 03‘0*** .074*** .080“ .050‘** .123*** .055*** 169 TABLE 8 Transformed Regressions We; Qppgnggnt Market Profitability Organization yprjpbleg Concentration Index Size Commercial Time -.067 .054 .188*** Mon-program Time -.051 .135** .065 Program Length -.O77 -.247*** -.O39 Commercials Breaks -.129** .285*** .003 Consecutive Mon-program Mot significant Announcements Clearance Hours .031 .017 -.033 Policy Areas .002 .218*** .249*** Policy Sources -.008 .240*** .081 * p is less than .05 ** p is less than .01 *** p is less than .001 M=476 (Minimum pairwise number of cases 8 418.) Substitute Local Adjusted Penetration 3 Square -.070 -.009 .033*** .004 .052 .016** .205*** .018 .123*** ' . 072 080 . 0861“. .312*“ -.050 .095tee .031 -.022 .125*** .051 .057 .055*** 170 Hypothesis 3: Hypothesis 3 posited that as market concentration increases, the formality of clearance practices decreases. The formality of clearance practices was measured using the average number of hours devoted to clearance each week (clearance hours), the number of policy areas or standards a station has (policy areas) and the number of policy sources a station usually consults when making advertising policy decisions (policy sources). Advertising policy formality was also measured using the usual form and method both clearance and complaint policies were communicated in. These are discussed in a section entitled "Hypotheses 3 and 4" following Hypothesis 4. Hypothesis 3 is not supported. No significant regression results were found (see Tables 6, 7, 8). Hypothesis 4: Hypothesis 4 stated that as market concentration increases, the formality of complaint practices decreases. T-tests were used to test Hypothesis 4, as complaint formality was not a continuous variable. Market structure and complaint formality appear unrelated as stations in all categories have complaint policies for two groups (either viewers, advertisers, or interest or minority groups) (see Tables 9 and 10). No significant T-tests were found, so Hypothesis 4 is not supported. 171 TABLE 9 Mean and Standard Deviation for the Number of Formalized Complaint Policies by Market Structure (1=least formal, 4=most formal) Loose Tight Monopoly Oligopolies Oligopolies Oligopolies Power Mean 2.43 2.23 2.46 2.22 SD 1.71 1.76 1.64 1.68 TABLE 10 T-Tests Comparing Market Structure Categories With Complaint Formality (One-Tailed Tests) COMPARISONS T SIG DF m ' t a Loose Oligopolies/Oligopolies 0.98 .164 289.00 Loose Oligopolies/Tight Oligopolies -0.15 .440 233.00 Loose Oligopolies/Monopoly Power 0.82 .205 196.00 Oligopolies/Tight Oligopolies -l.12 .132 276.00 Oligopolies/Monopoly Power 0.02 .494 239.00 Tight Oligopolies/Monopoly Power 0.96 .169 183.00 *Denotes significance at the indicated level. +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. 172 Hypotheses 3 and 4: Results regarding the usual form and method used to communicate advertising practices (including both clearance and complaint practices) will now be discussed. Hypotheses 3 and 4 posited that as market concentration increases, the formality of advertising practices decreases. Verbal forms and methods of communication are less formal than written forms. T-tests are used to test these variables as they are not continuous. Most stations communicated policies verbally, in both form and method. The prediction that policies become more formal as market concentration decreases is not supported. Loose oligopolies had the most formal policy form, as expected (see Table 11). Oligopolies unexpectedly had the least formal policy form, with the implication that policy formality and number of clearance hours is related. Oligopoly stations spend the fewest hours clearing ads, and loose oligopolists spend the most. Significant differences between both categories are found for clearance hours and policy form used most (see Table 12). Although differences are not significant, the means for communication method used most tend to suggest a more concentrated versus less concentrated split. Stations in less concentrated markets may communicate more formally. 173 TABLE 11 Means and Standard Deviations for the Communication Form and Method Used Most by Market Structure (1=most formal/written, 4=least formal/verbal) Loose Tight Monopoly Oligopolies Oligopolies Oligopolies Power 0 'c s 05 Mean 2.74 2.99 2.86 2.80 SD 1.16 1.07 1.06 1.09 W Mean 2.64 2.61 2.70 2.85 SD 1.11 1.16 1.15 0.96 TABLE 12 T-Tests Comparing Market Structure Categories With the Formality of Advertising Practices (One-Tailed Tests) COMPARISONS Roiigy Eppn Used host Loose Oligopolies/Oligopolies Loose Oligopolies/Tight Oligopolies Loose Oligopolies/Monopoly Power Oligopolies/Tight Oligopolies Oligopolies/Monopoly Power Tight Oligopolies/Monopoly Power o 'o t s os Loose Oligopolies/Oligopolies Loose Oligopolies/Tight Oligopolies Loose Oligopolies/Monopoly Power Oligopolies/Tight Oligopolies Oligopolies/Monopoly Power Tight Oligopolies/Monopoly Power *Denotes significance at the +Denotes that a T-Test using was used since the variances Such tests are used whenever for equality of variance are indicated T -1.65 -0.74 -0.33 0.86 1.13 0.35 0.21 -0.32 -1.18 -0.54 -1.40 -0.84 level. SIG .050* .229 .372 .194 .130 .365 .418 .374 .119 .293 .082 .202 DF 222.00 180.00 152.00 218.00 190.00 148.00 216.00 165.00 141.00 209.00 185.00 134.00 a separate variance estimate within groups appears unequal. probabilities for the F-Test "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. 174 Hypothesis 5: Hypothesis 5 stated that the more profitable a station is, the greater its level of commercialization will be. As profitability increased, the amount of commercial and non-program time increased. Regression results were not significant for commercial time, but were significant for non-program time after the outliers were pulled in and transformations made (see Tables 6, 7, 8). The number of program length commercials decreased as profitability increased, contrary to expectations. Results were significant in all regression equations (see Tables 6, 7, 8). Less profitable stations air more program length commercials. So, Hypothesis 5 is weakly supported. Hypothesis 6: Hypothesis 6 posited that the more profitable a station is, the more cluttered its commercial breaks are. Clutter was measured using the average number of breaks per hour, and the average number of consecutive non- program announcements per break. Results were significant and as expected in all regressions for the average number of breaks (see Tables 6, 7, 8). As profitability increased, the number of commercial breaks increased. More profitable stations tend to be more cluttered. 175 However, results were not significant for the number of consecutive non-program announcements, so Hypothesis 6 is partially supported. Hypothesis 7: Hypothesis 7 stated that the more profitable a station is, the more formal its clearance practices will be. The number of clearance hours does not increase with profitability. Results suggest that less profitable stations spend more time clearing ads (see Tables 6, 7, 8). However, results are as expected and significant in all regressions for the number of policy areas and sources (see Tables 6, 7, 8). More profitable stations tend to have more policy areas and use more policy sources. Perhaps this is why they spend less time clearing ads. So, Hypothesis 7 is partially supported. Hypothesis 8: Hypothesis 8 posited that the more profitable a station is, the more formal its complaint practices will be. More profitable stations appear to have more formalized complaint policies (see Table 13), and significant t-tests support this conclusion (see Table 14). Hypothesis 8 is supported. 176 TABLE 13 Mean and Standard Deviation for the Number of Formalized Complaint Policies by Profitability (1=least formal, 4 = most formal) More Less Profitable Profitable Qonpigint Rpiigigs Mean ' 2.45 2.07 SD 1.69 1.72 TABLE 14 T-Tests Comparing Profitability With Complaint Formality (One-Tailed Tests) COMPARISONS T SIG DF 0 a'nt 0 'es More Profitable/Less Profitable 2.24 .012* 474.00 *Denotes significance at the indicated level. +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. 177 Hypotheses 7 and 8: We will now examine whether profitability affects the form and method stations usually use to communicate their advertising policies (both clearance and complaint practices). Both hypotheses stated that as profitability increases, the formality of advertising practices increases. Inspection of means shows that results are in the expected direction (see Table 15), but t-test results are not significant (see Table 16). The hypothesized relationship is not supported. TABLE 15 Means and Standard Deviations for the Communication Form and Method Used Most by Profitability (1=most formal, 4=least formal) More Less Profitable Profitable ol'c 0 se ost Mean 2.83 2.95 SD 1.13 1.02 Qpnnunigation nethog Usgd Most Mean 2.65 2.73 SD 1.11 1.12 178 TABLE 16 T-Tests Comparing Profitability With the Formality of Advertising Practices (One-Tailed Tests) COMPARISONS T SIG DF o ic orm se 05 More Profitable/Less Profitable -1.02 .153 372.00 u at‘o e od s d os More Profitable/Less Profitable -0.60 .274+ 192.87 *Denotes significance at the indicated level. +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. Policy form and the usual method of communication do not seem to be strongly related to profitability, as most stations communicate advertising policies verbally. The implication may be that stations as a whole have a casual approach to advertising clearance and complaint practices. It may also be that stations do not perceive many ads to be problematic, or the measures are inadequate. 179 Hypothesis 9: Hypothesis 9 stated that as organization size increases, the formality of clearance practices increases. Regression results generally supported the hypothesis, but results were significant only for the number of policy areas (see Tables 6, 7, 8). So Hypothesis 9 has only weak support. When organizations are categorized as larger (above the mean) and smaller (the mean and below), Hypothesis 9 is supported for all three measures of clearance formality. Larger stations have more formal clearance practices. T-test results suggest that larger stations spend more time clearing ads, have more policy areas, and consult more policy sources (see Tables 17 and 18). Larger stations may be more likely to promulgate new policies when issues arise. They were more likely to have a standard banning or restricting 900 phone numbers, with 66.7% (or 24 of 36) of stations having such a standard being larger (X2 = 9.96, df = 1, sig = .0016, Cramer's V/Phi = .15). Larger stations were also more likely to have an Issue-Advocacy standard, perhaps promulgated in response to the Fairness Doctrine's demise, with 60.7% (or 17 of 28) of stations having them being larger (X2 = 4.21, df = 1, sig = .04, Cramer's V/Phi = .10). 180 TABLE 17 Means and Standard Deviations for the Number of Hours Devoted to Clearance, Number of Policy Areas, and Number of Policy Sources by Organization Size Organization Size (Sales and Traffic Departments) Mean and Below W ot ea n e Mean 3.55 SD 5.81 u er o o ' 5 Mean 11.35 SD 4.59 W Mean 4.80 SD 2.68 Above Mean Results support the notion that larger organizations have formalized rules for dealing with concerns, even recent ones. Larger stations are probably asked to air more types of spot and local ads, and must therefore have more policies to deal with all types of problems. 181 TABLE 18 T-Tests Comparing Organization Size With Clearance Formality (One-Tailed Tests) COMPARISONS T SIG o s e te e e Smaller Organization/ Larger Organization -2.88 .002+* umb r eas Smaller Organization/ Larger Organization -5.98 .000+* W Smaller Organization/ Larger Organization -1.68 .047* *Denotes significance at the indicated level. DF 254.69 451.29 465.00 +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. 182 Hypothesis 10: Hypothesis 10 posited that as organization size increases, the formality of complaint practices increases. Larger stations averaged significantly more complaint policies (see Table 19 and 20). Results suggest that the relationship between organization size and policy formality should be considered in other broadcast performance studies. Hypothesis 10 is supported. TABLE 19 Mean and Standard Deviation for the Number of Formalized Complaint Policies by Organization Size (1=least formal, 4 = most formal) Organization Size (Sales and Traffic Departments) Mean and Below Above Mean WM Mean 2.14 2.63 SD 1.73 1.62 183 TABLE 20 T-Tests Comparing Organization Size With Complaint Formality (One-Tailed Tests) COMPARISONS T SIG DF Co ' ° ' s Smaller Organization/ Larger Organization -3.02 .001* 465.00 *Denotes significance at the indicated level. +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. Hypotheses 9 and 10: Hypotheses 9 and 10 posited that as organization size increases, the formality of advertising practices increases. There is support for a relationship between organization size and the usual form of communicating advertising policies. Examination of means supports the notion that larger organizations communicate more formally (see Table 21), although results were significant only for the form policies were normally communicated in (see Table 22). However, results were in the expected direction for communication method used most, and near significance. The hypothesis that communication formality increase along with organization size is partially supported. 184 All results for policy form and method suggest that it may be fruitful to test new or revised measures. Some relationship seems to exist and it may simply be that these measures need to be revised. However, it may also be that such policies do not differ meaningfully from station to station. TABLE 21 Means and Standard Deviations for the Communication Form and Method Used Most by Organization Size (1=most formal, 4=least formal) Organization Size (Sales and Traffic Departments) Mean and Below Above Mean 261w Mean 2.96 2.72 SD 1.06 1.13 memunicgtion Mgthod Mean 2.74 2.57 SD 1.10 1.13 185 TABLE 22 T-Tests Comparing Organization Size with the Formality of Advertising Practices (One-Tailed Tests) COMPARISONS T SIG DF 0 ' rm Use ost Smaller Organization/ Larger Organization 2.10 .018* 366.00 mu ' d s Smaller Organization/ Larger Organization 1.40 .081 346.00 *Denotes significance at the indicated level. +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. Hypothesis 11: Hypotheses 11 through 14 were tested using analysis of variance and results and tables are presented only when a significant two-way interaction is found. Hypothesis 11 posited that as market concentration and profitability increase, the variance in commercialization decreases. In other words, it was expected that the most profitable stations in the most concentrated markets would be the most commercialized, and the least profitable stations in the least concentrated markets the least commercialized. 186 Because more, and more different types of, stations may be found in less concentrated markets, these stations should vary more in how commercialized they are. No significant two way interactions were found for commercialization. Hypothesis 11 is not supported. HYpothesis 12: The predicted relationship of Hypothesis 12 was that as market concentration and profitability increase, the variance in clutter levels decreases. The underlying assumption was that the most profitable stations in the least concentrated markets would air the most breaks, and the least profitable stations in the most concentrated markets would air the least. No significant two way interactions were found for the number of breaks, or the number of consecutive non-program announcements. Hypothesis 12 is not supported. Hypothesis 13: Hypothesis 13 predicted that as market concentration and profitability increase, the variance in clearance formality increases. Recall that the FCC expected more of larger and more profitable stations in the deceptive advertising area. Therefore, in smaller markets with fewer competitors, differences were posited to be due to individual differences in management commitment to 187 clearance. This hypothesis was not supported as no significant two way interactions were found for clearance hours, the number of policy areas, or the number of policy sources. Hypothesis 14: The predicted relationship of Hypothesis 14 was that as both market concentration and profitability increase, the variance in complaint formality increases. The assumption was that greater differences in complaint practices would be found in more concentrated markets as differences in these policies would be based on individual management philosophy. Recall that the FCC originally expected more from larger and more profitable stations, which are often found in less concentrated markets. So, managers in more concentrated markets may be more likely to follow their own consciences. However this hypothesis was not supported as the two way interactions was not significant. Hypotheses 13 and 14: The posited relationship here is that variance in advertising policy formality (for both clearance and complaint practice communication) increases along with market structure and profitability. However, the two way interaction was not significant, and it seems that 188 stations vary little on the form and means of communicating policies. Results suggest that most stations simply communicate in the same fashion, or that new and better measures need to be developed. Research Question # 1: Does commercialization vary by market structure type? A very weak relationship may exist. Examination of the means in Table 23 suggests that commercialization decreases as market concentration increases. Some significant T differences were found (see Table 24), however commercialization was generally not significant in the overall regressions (except for the number of program length commercials using untransformed variables -— see Tables 6, 7 and 8). T-results may only be significant because other control variables were eliminated. New regressions using the final list of independent variables were run to determine if the T-test results represented true differences. Market structure was partitioned into the four market structure types and twelve different regressions (four using untransformed variables, four using variables with outliers reduced to three standard deviations, and four using transformed variables) were run on the three measures of commercialization. That way, it could be determined if commercialization varied by market structure type. 189 TABLE 23 Means and Standard Deviations for Commercial Time, Non-program Time, and Number of Program Length Commercials by Market Structure Loose Tight Monopoly Oligopolies Oligopolies Oligopolies Power o e i ° e Mean 12:21 12:31 12:17 11:12 SD 5:09 4:03 4:02 3:09 W Mean 15:17 14:25 14:09 13:36 SD 7:15 5:23 4:09 4:16 t a s Mean 7 6 4 3 SD 14.3 10.1 6.9 2.8 Only one regression, for loose oligopolies using untransformed data on the number of program length commercials, was significant. The suggestion is that stations in the least concentrated markets air more program length commercials. The relationship between commercialization and market structure is very weak, at best. Yet this weak relationship could grow stronger over time. Program length commercials were banned before deregulation so these results may represent a developing relationship. 190 TABLE 24 T-Tests Comparing Market Structure Categories With Commercialization (One-Tailed Tests) COMPARISONS W Loose Oligopolies/Oligopolies Loose Oligopolies/Tight Oligopolies Loose Oligopolies/Monopoly Power Oligopolies/Tight Oligopolies Oligopolies/Monopoly Power Tight Oligopolies/Monopoly Power Less Concentrated (LO & OL)/ More Concentrated (TO & MP) Oligopolies (LO, OL, TO)/ Monopoly Power upn-pppgtnn Time Loose Oligopolies/Oligopolies Loose Oligopolies/Tight Oligopolies Loose Oligopolies/Monopoly Power Oligopolies/Tight Oligopolies Oligopolies/Monopoly Power Tight Oligopolies/Monopoly Power Less/More Concentrated Oligopolies/Monopoly Power W Loose Oligopolies/Oligopolies Loose Oligopolies/Tight Oligopolies Loose Oligopolies/Monopoly Power Oligopolies/Tight Oligopolies Oligopolies/Monopoly Power Tight Oligopolies/Monopoly Power Less/More Concentrated Oligopolies/Monopoly Power *Denotes significance at the indicated level. 0.69 0.89 1.79 0.27 1.44 1.19 1.42 1.97 0.47 1.92 3.13 1.99 4.01 1.88 3.52 4.81 SIG .297+ .187 .005+* .324 .010+* .042+* .030+* .003+* .245+ .186+ .038+* .394+ .075+ .118 .078+ .025+* .319+ .029+* .001+* .024+* .000+* .031+* .000+* .000+* DF 221.29 214.00 174.41 245.00 161.47 161.45 395.79 115.18 195.32 187.21 176.70 237.90 158.98 147.13 422.91 119.85 204.14 176.24 135.03 261.77 205.37 150.07 424.15 402.10 +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the Such tests are used whenever probabilities for the F-Test for equality of variances within groups appears unequal. variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. 191 Market structure was collapsed into two categories representing less concentrated (loose oligopolies and oligopolies) and more concentrated (tight oligopolies and monopoly power) markets, and representing all oligopoly categories vs. monopoly power. Again, the significant T- results did not represent a relationship, as new regressions for these categorizations using the final list of independent variables were not significant. Research Question 4 2: Do clutter levels vary by market structure type? Examination of means in Table 25 suggests that as market concentration increases the number of breaks decreases. Differences were significant for all comparisons except loose oligopolies and oligopolies (see Table 26). However, results were not significant for the number of consecutive non-program announcements. It appears that the number of breaks varies by market structure type. 192 TABLE 25 Means and Standard Deviations for the Number of Breaks and Consecutive Non-program Announcements by Market Structure Loose Tight Monopoly Oligopolies Oligopolies Oligopolies Power Mean 6.81 6.70 6.07 6.15 SD 1.90 1.88 2.14 1.89 se 8 Mean 3.62 3.69 3.63 3.70 SD 2.32 1.48 1.18 3.70 Results suggest that market structure may be categorized into two groups: less concentrated (loose oligopolies and oligopolies) and more concentrated (tight oligopolies and monopoly power markets), with increased concentration accompanied by fewer breaks. A significant T-Test provides support (see Table 26). Concentration differences may be more meaningful between two categories rather than four. Given the results for commercialization and clutter, both dichotomous categorizations should be considered in future studies to determine which is the most useful. 193 TABLE 26 T-Tests Comparing Market Structure Categories With Clutter (One-Tailed Tests) COMPARISONS Bumper_of_Breaks Loose Oligopolies/Oligopolies Loose Oligopolies/Tight Oligopolies Loose Oligopolies/Monopoly Power Oligopolies/Tight Oligopolies Oligopolies/Monopoly Power Tight Oligopolies/Monopoly Power Less Concentrated (LO & OL)/ More Concentrated (TO & MP) Oligopolies (LO, OL, TO)/ Monopoly Power o ' o - o o Loose Oligopolies/Oligopolies Loose Oligopolies/Tight Oligopolies Loose Oligopolies/Monopoly Power Oligopolies/Tight Oligopolies Oligopolies/Monopoly Power Tight Oligopolies/Monopoly Power Oligopolies (L0,0L,TO)/ Monopoly Power *Denotes significance at the indicated -0.28 -0.05 -0.27 0.33 -0.06 -0.28 -0.46 level. SIG .319 .003* .013* .007* .025* .395+ .000* .059 .392+ .481+ .395+ .370+ .476 .390+ .324 DF 264.00 221.00 183.00 255.00 161.36 174.00 440.00 97.63 174.36 170.20 154.11 218.89 211.00 96.96 72.87 +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. 194 Acceptance of new time units instituted since the code's demise may suggest the experimentation that the FCC intended, but also a more cluttered environment since the more time units a station accepts, the more cluttered its breaks may appear. Examining individual time units separately with market concentration supports the notion that as market concentration increases, clutter decreases. Less concentrated market stations seem more likely to accept new units instituted since the NAB Code's demise. The great majority of responding stations accept 15 second spots (447 of 476, or 93.9 percent). When market structure is dichotomized as more and less concentrated and compared with acceptance of 15 second spots (Yes or No), significance is achieved (X2 - 4.2, df - 1, sig - .039, Cramer's V/Phi = .10, Kendall's Tau-C - -.048, sig = .012). As market concentration increases, stations are more likely to decline the 15 second spot. Therefore, more concentrated markets may be less cluttered, and clutter practices seem to differ between more and less concentrated markets. Most stations did not accept the 45 second time unit (346 of 476, or 72.7 percent). Market structure in four categories is related to acceptance (X2 8 10.48, sig - .015, Cramer's V = .148, Kendall's Tau-C = -.107, 195 sig = .009). As market concentration increases, proportionately fewer stations accept 45 second spots. Clutter practices seem to differ between categories. Research Question I 3: Do clearance practices vary by market structure type? Results suggest the answer is no. Examination of means suggests that market structure is weakly related to clearance hours, but not as predicted (see Table 27). Loose oligopoly stations averaged significantly more clearance hours than oligopoly and monopoly power stations, but not tight oligopoly stations (see Table 28) In markets where there are many competitors, and very few competitors, clearance plays a role. Larger stations have full time broadcast standards personnel which accounts to some extent for the mean in loose oligopolies. Tight oligopoly stations may need to spend more time clearing to insure that their advertising and viewing environments are competitive. The large standard deviation suggests that only certain stations do so, perhaps representing a difference between profitability classes. The large standard deviation for loose oligopolies may also represent clearance differences between stations in different classes, or the fact that more stations are found in these markets overall. 196 Market structure did not appear to be related to the number of policy areas a station has, or the number of policy sources a station uses. Examination of means suggested that differences were due only to randomness (see Table 27). T-tests confirm these suspicions (see Table 28). No significant differences were found in any comparison between market structure and the number of policy areas used, nor the number of policy sources consulted. TABLE 27 Means and Standard Deviations for the Number of Hours Devoted to Clearance, Number of Policy Areas, and Number of Policy Sources by Market Structure Loose Tight Monopoly Oligopolies Oligopolies Oligopolies Power W Mean 5.68 3.84 4.87 3.72 SD 8.52 6.12 9.16 6.90 W Mean 12.30 12.31 12.50 11.54 SD 4.23 4.47 4.42 4.79 W Mean 5.05 4.96 4.81 5.08 SD 2.43 2.86 2.97 3.00 197 TABLE 28 T-Tests Comparing Market Structure Categories With Clearance Formality (One-Tailed Tests) COMPARISONS HQEI§_D§¥2£§Q_£Q_91§QIQDE§ Loose Oligopolies/Oligopolies Loose Oligopolies/Tight Oligopolies Loose Oligopolies/Monopoly Power Oligopolies/Tight Oligopolies Oligopolies/Monopoly Power Tight Oligopolies/Monopoly Power W Loose Oligopolies/Oligopolies Loose Oligopolies/Tight Oligopolies Loose Oligopolies/Monopoly Power Oligopolies/Tight Oligopolies Oligopolies/Monopoly Power Tight Oligopolies/Monopoly Power e O 0 Loose Oligopolies/Oligopolies Loose Oligopolies/Tight Oligopolies Loose Oligopolies/Monopoly Power Oligopolies/Tight Oligopolies Oligopolies/Monopoly Power Tight Oligopolies/Monopoly Power *Denotes significance at the indicated T 1.99 0.67 1.72 -0.98 0.13 0.92 -0.02 -0.35 1.17 -0.34 1.22 1.40 0.30 0.66 -0.06 0.41 -0.29 -0.58 level. SIG .024+* .251 .043+* .164+ .450 .178+ .492 .363 .121 .366 .112 .081 .383+ .254+ .476+ .342 .386 .280 DF 203.58 213.00 168.82 150.18 222.00 164.68 289.00 233.00 196.00 276.00 239.00 183.00 283.61 213.12 129.53 276.00 239.00 183.00 +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. 198 Research Question # 4: Do complaint practices vary by market structure type? Results suggest that the answer is no. Complaint practices do not vary by market structure type. Most stations have complaint policies for dealing with two groups (usually for viewers and advertisers, and less often for minorities or special interest groups -- see Tables 9 and 10). Research Questions # 3 and t 4: Research Questions 3 and 4 posit that the formality of advertising practices, as measured by the form and method of communication of both clearance and complaint practices, varies by market structure type. Results again suggest that the answer is no. Clearance and complaint formality (measured as the form and method of communication) do not vary by market structure type (see Tables 11 and 12). However, these results may be due to inadequate measures. Better measures may ultimately reveal a relationship between market structure and advertising policy formality. Research Question 9 5: Does commercialization vary by profitability class? The answer to this research question appears to be yes. Results are in the expected direction for commercial time, 199 but not significant (see Tables 29 and 30). More profitable stations do schedule more non-program time and differences are significant (see Tables 29 and 30). The result that less profitable stations air significantly more program length commercials is interesting, as it suggests that more and less profitable stations have different advertising and viewing environments. Indeed, type of commercialization seems to represent a means of classifying stations. TABLE 29 Means and Standard Deviations for Commercial Time, Non-program Time, and Number of Program Length Commercials by Profitability More Less gpnnptgig1_1ing Profitable Profitable Mean 12:38 11:26 SD 4:11 5:21 Wimp Mean 14:59 13:52 SD 5:02 5:42 mb 0 W Mean 3.5 10.1 SD 4.9 16.3 200 Consider that certain types of advertisements may be expected on more and less profitable stations. More profitable stations air more commercial and non-program time in general but fewer program length commercials. Less profitable stations air less commercial and non- program time but more program length commercials. Advertisers learn which types of commercials stations, possibly exacerbating environment differences over time. TABLE 30 T-Tests Comparing Profitability With Commercialization (One-Tailed Tests) COMPARISONS T SIG DF 0 r ' ' e More Profitable/Less Profitable 1.53 .064+ 215.57 Wine More Profitable/Less Profitable 1.96 .025* 409.00 Number_2§_2§ograms More Profitable/Less Profitable -4.58 .000+* 142.34 *Denotes significance at the indicated level. +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F—Test Significance Levels were .10 or smaller. 201 Research Question 2 6: Does clutter vary by profitability class? Results suggest that the more profitable a station is, the more cluttered its commercial breaks are when clutter is operationalized as the number of breaks per hour. More profitable stations schedule significantly more breaks than less profitable stations (see Tables 31 and 32). However, results for the number of consecutive announcements are not significant. TABLE 31 Means and Standard Deviations for the Number of Breaks and Consecutive Non-Program Announcements by Profitability More Less m aks Profitable Profitable Mean 6.88 5.59 SD 1.80 2.07 W W Mean 3.67 3.47 SD 1.32 1.59 202 TABLE 32 T-Tests Comparing Profitability With Clutter (One-Tailed Tests) COMPARISONS T SIG DF umb o ks More Profitable/Less Profitable 6.22 .000+* 222.36 0 o c 'v o A e s More Profitable/Less Profitable 1.18 .119+ 189.18 *Denotes significance at the indicated level. +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. Although the T-Test for the number of consecutive announcements was not significant, Chi-Square results were. When categorizing the number of consecutive non- program announcements dichotomously as the mean and below, and above the mean, less profitable stations were more likely to exceeded the mean (X2 = 10.80, df = 1, sig = .001, Cramer's V/Phi = .169). Despite the fact that less profitable stations tend to schedule fewer breaks, some may schedule quite a few more consecutive announcements in them. This may be how some less profitable stations compensate for clutter. 203 Again, it appears that more and less profitable stations may have different advertising and viewing environments. Considering results for individual time units suggests that the advertising and viewing environments of more and less profitable stations differ in break composition. Although most stations accept 15 second spots, more profitable stations appear more likely to accept them (X2 = 4.2, df = 1, sig = .039, Cramer's V/Phi = .10). In contrast, 20 second spots may be found more often on less profitable stations (X2 = 9.96, df = 1, sig = .0016, Cramer's V/Phi = .15). Less profitable stations also seem more likely to accept the 30 minute (X2 = 11.20, df = 1, sig = .0008, Cramer's V/Phi = .159) and sixty minute (X2 = 8.31, df = 1, sig = .0039, Cramer's V/Phi = .138) time units. Research Question 1 7: Do clearance practices vary by profitability class? When clearance formality is operationalized as the hours per week spent on clearance, no differences are found (See Tables 33 and 34). Examination of means suggests that less profitable stations tend to spend more time clearing ads than more profitable stations, a finding contrary to expectations. 204 Differences in profitability classes are suggested when formality of clearance is operationalized as the number of formal policy areas a station has, and the number of sources a station consults when making advertising policy decisions. T-tests for both were significant (see Table 34). Findings suggest that managers at more profitable stations, having more established policy areas and using more policy sources, as a result tend to spend less time clearing ads than their counterparts at less profitable stations (see Table 33). TABLE 33 Means and Standard Deviations for the Number of Hours Devoted to Clearance, Number of Policy Areas, and Number of Policy Sources by Profitability More Less Profitable Profitable Ql§Q£§D2§_HQBI§ Mean 4.25 5.25 SD 7.40 8.31 o o s Mean 12.94 10.55 SD 4.09 4.80 W Mean 5.35 4.05 SD 2.81 2.56 205 TABLE 34 T-Tests Comparing Profitability With Clearance Formality (One-Tailed Tests) COMPARISONS T SIG DF §l§§I§D§§_HQEE§ More Profitable/Less Profitable -1.24 .107 437.00 W More Profitable/Less Profitable 5.17 .000+* 229.66 W More Profitable/Less Profitable 4.72 .000* 474.00 *Denotes significance at the indicated level. +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. Results also suggest that more and less profitable stations differ in their approach to clearance. More profitable stations appear more likely to stay abreast of FTC policy as required by the FCC before deregulation. More profitable stations were more likely to consult FTC publications when making advertising policy decisions (X2 = 8.04, df = 1, sig = .004, Cramer's V/Phi = .135). More and less profitable stations also appear to differ on clearance formality. Because more profitable 206 stations have more formal policies their clearance time is reduced. Less profitable stations may ban fewer types of ads, and decide whether or not to accept them on a case- by-case basis. Consequently, lower profits may force a station to consider accepting ads they may not otherwise. Again, the suggestion is that advertising and viewing environments may differ. The idea that more and less profitable stations may be clearing different types of ads is supported by responses to an open-ended question asking what other policy areas stations have. Less profitable stations seemed more likely to have policies banning or restricting 900 phone number ads (X2 = 6.73, df = 1, sig = .009, Cramer's V/Phi = .127), suggesting that they may receive more requests to air them. Standards banning or restricting issue-advocacy advertising seemed the domain of more profitable stations (X2 8 4.18, df = 1, sig = .04, Cramer's V/Phi = .103). Research Question 2 8: Do complaint practices vary by profitability class? Significant results support the notion that more profitable and less profitable stations differ on how formal their complaint policies are (see Tables 13 and 14). Again, the suggestion is that different classes of stations exist, based on profitability. 207 Research Questions 2 7 and f 8: Clearance and complaint formality (as measured by the usual form and method used to communicate advertising policies) do not vary by profitability class. Results were in the expected direction but not significant (see Tables 15 and 16). Summary of Hypothesis and Research Question Results: Overall, study results suggest that some relationships between market structure, profitability and organization size, and commercialization, clutter, clearance and complaint practices do exist. These measures of performance therefore deserve future study and clarification. Results also support the notion that station managers manipulate their advertising and viewing environments in response to various concerns. Preliminary evidence suggests that differences may exist between commercial television market structure categories. Support was also found for the notion that subclasses of stations exist based on profitability and their advertising and viewing environments may differ. Future research on these questions will be discussed in the final chapter. Market structure or concentration was a predictor for some of the dependent variables. Hypothesis 1 was weakly 208 supported and Hypothesis 2 was partially supported, as the level of commercialization tended to decrease as market concentration increased, and the number of breaks decreased as market concentration increased. Both the individual and combined Hypotheses 3 and 4 were not supported, as advertising policy formality did not decrease as market concentration increased. Neither clearance formality, complaint formality, nor the usual form and method of communication appeared related to market structure. Profitability was also found to affect some of the dependent variables. Hypothesis 5 was weakly supported and Hypothesis 6 was partially supported as the amount of non-program time and the number of breaks increased along with profitability. Hypotheses 7 was partially supported and Hypothesis 8 was fully supported when tested individually. The number of policy areas and sources, and complaint formality, increased as profitability did. The combined Hypotheses 7 and 8 were not supported as communication formality did not increase with profitability. Hypothesis 9 was weakly supported and Hypothesis 10 was fully supported when tested individually. Larger organizations seemed to have more formal advertising 209 policies. Larger stations spent more time clearing ads, had more policy areas, consulted more policy sources, and had more formal complaint policies. The combined Hypotheses 9 and 10 were partially supported as the policy form used most by larger organizations was more formal but only in the expected direction for the method used most. Hypotheses 11 through 14 were not supported, as market concentration and profitability did not have a joint effect on the dependent variables. It appears that commercialization and clutter may vary by market structure, but clearance and complaint practices do not. Commercialization differences seem to exist between profitability classes. More profitable stations air more commercial and non-program time, and less profitable stations air more program length commercials. More profitable stations are more cluttered as they air more breaks than less profitable stations. The notion that the different profitability classes have different advertising and viewing environments was also supported. More profitable stations are more likely to accept 15 second spots, and less profitable stations are more likely to accept 20 second spots, and 30 and 60 minute program length commercials. 210 Less profitable stations tend to spend more time clearing ads, contrary to expectations. More profitable stations have more policy areas and consult more policy sources which may explain why they spend less time clearing ads. Results also suggest that the profitability classes may air different types of ads. More profitable stations tended to have more formal complaint policies. Although results were not significant, more profitable stations also used more formal means for communicating advertising policies. NOTES: CHAPTER V 1More than half of the responses on these questionnaires were left blank, so they were excluded. The great majority of respondents who returned incomplete questionnaires left only one or two questions blank. These six therefore represented "unusual cases. 2"A Short Course in Broadcasting, 1988," Br2ad2ast1ngACaplesasting_xearbeok (Washington. D.C.: Broadcast Publications, 1988), p. A-2. 31bid. 4Barbara G. Tabachnick and Linda S. Fidell, fining Multi_ariete_§tatistis_ (New York: Harper 8 ROW. 1983). p- 5Marija J. Norusis, e S u t 8 (Chicago: SPSS, 1986), p. 176-7. 6John Neter, William Wasserman and G. A. Whitmore, Applipg_§tnti§tip§ (Boston: Allyn & Bacon, 1978), p. 66. 7M- G- Bulmer. Principles_of_§tatistiss (Cambridge. Mass.: The MIT Press, 1967). 8Francis J. Wall. Stat1st1eal_Data_Analxsis_Handbeek (New York: McGraw-Hill, 1986), p. 16.6. 9Barbara G. Tabachnick and Linda S. Fidell, fining Multimeriate.§tatistics (New York: Harper & Row, 1983), p. 84. 10A. C. Atkinson, o s o ' s Rpgtpppipn (Oxford: Clarendon Press, 1985), p. 81. 11See Atkinson, p. 80-1.; Norusis, p. 355-7.: and Tabachnick and Fidell, p. 84-5.: for guidance on tranformations. 211 212 12Both transformations normally resulted in quite similar regression equations, with only a few percentage points of difference in variance explained. For this reason, the transformation which mirrored the natural direction of the skew in nature was retained. Most of the transformations did not significantly change the variance explained in the regression equation, and in most cases, the same variables were included in both transformed regression equations. 13Tabachnick and Fidell, p. 76, 92. 14Barry R. Litman, "Public Interest Programming and the Carroll Doctrine: A Re-examination," fiphtnnl_pt Rppgggggting 23 (Winter 1979), p.55. 15Tabachnick and Fidell, p. 82-3. 16Elazar J. Pedhazur, Multiple_Begress12n_1n fighg_iptg1_3p§ggtgh (New York: Holt, Rinehart & Winston, 1982), p. 246. 17Tabachnick and Fidell, p. 83. 18Michael 8- Lewis-Beck. Applied_Besrea§12n1_An Inttpgnptipn (Beverly Hills, Ca.: Sage, 1980), p. 61. 19Litman, "Public Interest Programming," p. 55. 20Rank's correlation is negative because of its "reverse" nature. Originally, independents were coded as 0 for affiliates 1 for network affiliation for the pearson correlations. (These codings were later changed when the profitability index was developed.) Thus, as affiliation "decreased" rank "increased," say from first to third. 21Litman, "Public Interest Programming," p. 55. Specifically, "Net weekly circulation measures the reach of each television station according to the number of majority as well as minority viewers and is thus a superior measure of market size..." It is used as a proxy measure for profitability because "...stations located in large markets earn higher profits..." 213 22See Tabachnick and Fidell, p. 399. Varimax rotation is the most commonly used orthogonal rotation solution (e.g., the factors are uncorrelated with each other). Varimax rotation forms the simplest factors possible by maximizing the variance of the loadings across variables within factors. Interpretation is easier with this method as loadings tend to be higher for variables that have a high correlation with a factor, and lower for the others. 231bid., p. 380-1, 394-404. 24Jae-On Kim and Charles W. Mueller, Rngtpt_Annlypi§ (Beverly Hills, Ca.: Sage, 1978), p. 9. 25The basic difference between the two techniques is the contents of the positive diagonal in the correlation matrix, which contains the correlation of a variable with itself. In both, the variance that is analyzed is the sum of the values in the positive diagonal. In principal components analysis, "1"s are present in the diagonal, so the variance to be redistributed among components is the same as the number of observed variables. Each variable contributes a unit of variance (or a "1") and all of the variance is redistributed, including the variance unique to each variable and error variance. In factor analysis, only the variance that each variable shares with the other variables is analyzed. Here the common variance is estimated by inserting communalities instead of "1"s in the positive diagonal. Communality estimates are usually the squared multiple correlations of each variable with the others, or the highest absolute correlation in a row. Unique and error variance is omitted in this technique because of the belief that they confuse the structural picture that emerges from an analysis of underlying processes. See Tabachnick and Fidell, p. 395-6, and Kim and Mueller, cited in the next note, p. 21, for further explanation. 26Tabachnick & Fidell, p. 397. 27Tabachnick & Fidell, p. 373, 380. 28Andrew L- Comrey. A_F1rst_Course_1n_£astor_8nalxsis (New York: Academic Press, 1973), p. 200. 214 29Ihid., p. 226. 30R. L. Gorsuch, £ggtpz_Ann1y§i§ (Philadelphia: W. B. Saunders, 1974), pp. 186-94. See also Robert H. Wicks, : ° II e‘l',, - :‘S 9 "l e! - '1' a it '7.- ' .‘ I‘ - s on ews . oi- e e- e; s: 1 --u-; — ' 09‘ RQSed on Media Gratitications, Unpublished Doctoral Dissertation, Mass Media Ph. D. Program at Michigan State University, August 1987, p. 84. 31Eigenvalues represent variance, and since the variance that each variable contributes is 1 or less, any factor with an eigenvalue less than 1 is not as important as a variable. Therefore, the number of factors with eigenvalues greater than 1 suggests the maximun number of factors. Factor loadings represent the multiple correlation between a variable and a factor, and variables with high factor loadings are examined to discover the basis for a factor. Factor loadings must exceed .30 to be interpreted, since this indicates at least a 9 percent overlap in variance between a variable and factor. The greater the overlap, the more the variable is a pure measure of the factor. See Tabachnick and Fidell, p. 406, Comrey, p. 22L and Jun C. Nunally, prghpnpttignlhgpty, 2nd ed. (New York: McGraw-Hill,1978), p. 425, for more information. 32Litman, "Public Interest Programming," p.55. 33Benjamin J. Bates, "Determining Television Advertising Rates," in Robert N. Bostrom, ed. Communication_xearbook_l (Beverly Hills. Ca-= Sage. 1983), p. 472-4. 34Prime time rate, early fringe rate, not weekly circulation, and cable and VCR penetration have excellent loadings on Factor 1 (Market Size). Organization size achieved a very good loading. Net weekly circulation is dropped as an indicator because it is skewed, contributes to multicollinearity, and measures size instead of profitability. 215 35For broadcast band, VHF was scored as 2, and UHF as 1. For network affiliation, affiliation with a national network was scored as 2, and non-affiliation (independent) was scored as 1. For market rank, a ranking of first to third was scored as 2, and fourth or higher as 1. Each stations scores were summed to create the index. The lower sums, 3 and 4, represented less profitable stations (since 4 was the mean). The higher sums, 5 and 6, represented the more profitable stations. A number of variations were tried, and this method was by far the most reliable. Scores were also set up this way so increased profitability would be represented by a positive sign in the regression equations. 35Jum c. Nunally, R§22n2m§;:12_1hsgrx 2nd ed. (New York: McGraw- Hill, 1978), p. 229-30. 37Ibid., p. 245-6. 33Ibid., p. 229-30. 39Hubert M. Blalock, Jr. figgi§1_§;g§i§§ig§ 2nd Ed. (New York: McGraw- Hill, 1979), p. 279. CHAPTER VI DISCUSSION The Federal Communications Commission deregulated commercial television to encourage licensee experimentation. It is unclear if the FCC directly considered whether this would affect the quality and quantity of commercial information available to consumers. It is also unclear whether the FCC really considered if sales managers, who must meet sales performance goals to keep their jobs, are truly concerned with the public interest when deciding how many ads to schedule, and which ads to accept. By eliminating log-keeping and ascertainment, the FCC has made it difficult to determine what the effect on the consumer information marketplace might be. Indeed, the present study relies on self-reports of station policies and conduct. The quality of the results is directly related to the veracity of the average respondent. How did managers use their new freedoms with this underlying profit maximizing reality in mind? The average amount of commercial time aired per hour was 12:15, below 216 217 both the NAB Code and FCC guideline, supporting the argument that excessive commercialization would not occur after deregulation. Findings here mirror those existing before deregulation, that only a small minority exceed the guidelines (18 of 426 responding, or 4.2 percent).1 Most stations do not wish to alienate viewers, advertisers, or the FCC. However, some managers are apparently unconcerned with the effects overcommercialization may have on the quality of consumer information. The same appears true for the old NAB Code guideline for non-program time. The average amount of non-program time aired per hour is 14:25, about one and a half minutes under the old standard. Yet 17.7 percent exceeded the old guideline (73 of 411 responding), again suggesting that a some managers consider profits before the public interest. The average station in the sample airs 5 program length commercials per month. That translates into anywhere from two-and-a-half to five hours per month, or roughly an extra hour of commercials per week, or an extra 8 or 9 minutes per day, or 45 seconds per hour (in a 20 hour broadcast day). Viewed in this fashion, it seems that stations are trying out the new time unit fairly responsibly as a whole. By adding this to the non-program 218 time average, it translates to a little more than 15 minutes of non-program time per hour, still below the old NAB Code guideline. However, 13 of 453 stations responding (or 2.8 percent) aired 35 or more program length commercials per month. Viewed another way, 3.7 percent of the stations which actually accept program length commercials (or 13 of 347) air 35 or more per month. Some of these programs may be for baldness cures, real estate deals or other investment schemes. This finding again suggests that a minority of broadcasters may not be experimenting in a responsible fashion. Stations are scheduling fewer breaks than suggested by the former NAB Code, with the average station airing about seven per hour, and the most common response being eight. The average break includes four spots. The average station accepts seven time units, including spots and program length commercials. Although the length and composition of those four-spot breaks may be quite different, it appears that the average station is trying to keep its airwaves from becoming too cluttered. However, about 50 managers left only the response to the number of consecutive non-program announcements blank. Pretesting indicated that this question was not difficult 219 to understand or complete. Perhaps some managers found this difficult to estimate, although over 400 did not, and others were forthright in providing answers which clearly exceeded old standards. At any rate, clutter is, or is becoming, a concern, and the problem may be worse than results indicate. There is clearly a need to tape programming off the air and count the number of commercials per break, and the number of breaks per hour, independently. Evidence suggests that most stations are willing to accept new time units. Most stations accepted program length commercials and 15 second spots, while most declined 45 second spots. Although stations tend to respond the same way, a significant minority went against the tide in each instance. Either individual decision making is occurring, or some stations are not asked to air the new time units. Results reveal that the average station has 12 policy areas or standards, consults about five sources when making advertising policy decisions, has complaint policies for at least two groups, spends about five hours a week on clearance review, and conveys advertising policies and decisions verbally to employees. Stations apparently make some effort to screen deceptive ads. 220 Most stations do not have formalized advertising policies. Regarding the form most policies were in, only 66 stations reported having policies codified in a manual, 56 had mostly written policies, 113 mostly used memos, and 139 conveyed policies verbally. Regarding the communication method used most often, 87 stations required their employees to read them, 34 encouraged them to, 139 communicated them through staff discussions, and at 94 stations the supervisor told each employee individually. Thus, the average station "stores" and "communicates" its policies verbally. For most stations, the sales manager is responsible for deciding whether or not to accept advertisements for broadcast on a day-to—day basis. Very few had a broadcast standards director responsible for clearance (13): only very large ones had such a director. The largest stations are still apparently concerned with the higher expectations the FCC once had, as they have not eliminated this position. It appears that most stations still follow the old commercialization and clutter guidelines and have kept at least some of their clearance and complaint practices on the books. However, there are a number of other findings which have implications for professionals and scholars in 221 various fields. Let us now consider how study results may be useful to the industry, mass media economic researchers, mass media policy researchers, and organizational researchers. 1 t' h tu : Implications for Industry: The study provides an overview of what practices are typically used by stations throughout the country. Managers can compare results to the practices and policies used at their own stations to discern their own level of commitment to the public interest. Over 200 respondents requested a copy of the results to do so, as this idea was suggested in the explanatory letter. The finding that larger stations appear more likely to promulgate new policies may suggest to managers in smaller markets what concerns they may eventually be dealing with. A few stations reported having policies for 900 telephone number "social phone line" ads, late night strip club ads, contraceptive advertising, alcohol or lite alcohol ads, financial or investment advertising, hair loss advertising, weight loss advertising, adult products and services advertising, and competitive advertising (e.g. ads for programming or features in other media). Managers at smaller market stations may wish to consider 222 establishing policies for, or simply thinking about, these types of ads and what restrictions or guidelines they might impose. The study also provides a gauge for other agencies and groups to see how useful their counsel is to broadcasters. One organization contacted the researcher for permission to use the results to demonstrate that it is an important commercial adviser to the business sector.2 Results may also be used in the same fashion by other groups, or to suggest new policy sources for broadcasters to contact when making policy decisions. An important finding for industry is that stations having more policy areas tended to spend less time clearing ads. Busy managers should consider taking the time to establish and formalize policies, as it may make clearance decisions more efficient over the long term. Communicating these policies to all employees involved in clearance may make overall station operation more time efficient as well. Broadcasters need to consider what the long term effects of deregulation may be, and what the collective results of their experimentation will be on the quality and quantity of commercial information. A few 223 broadcasters appear to be responding irresponsibly. What actions need to be taken by industry members to correct this problem? Study results may at least be used as an example to demonstrate to some managers that their conduct is unlike the majority's. Finally, the commercial broadcasting industry must consider whether it is in the public interest for managers to remain responsible for clearance. Again, this is not intended as a comment on the ethics of the typical sales manager. But it is a reality that sales managers must meet performance goals to keep their jobs and earn raises. This represents a genuine conflict of interest that would be difficult for anyone to reconcile. Any resolution with a realistic chance to succeed must be spearheaded and supported by management. Recognition of and changes implemented to correct this concern would show that the industry was truly concerned with the public interest. More importantly, meaningful action without prodding from the government and the public would clearly demonstrate genuine industry concern for the content of commercial information. Action of this type is strongly recommended, before public criticism makes a response appear to be in the industry's self-interest. 224 Implications for Mass Media Economic Research: The study also has important implications for economic research. It expands the analysis of commercial television performance to include advertising variables like commercialization, clutter, clearance and complaint practices. The national scope of the study reinforces that these practices can be measured and provide useful results. There are important findings for economic researchers to consider. The first is that the underlying market structure of commercial television apparently varies enough to result in some performance differences. Market structure may also affect other performance variables as well. It should therefore be included as an independent variable in economic assessments of broadcast performance in the future. Ideally, researchers should consider what the best categorization for market structure may be, for explanatory and possibly regulatory purposes. Should market structure be divided into two or four categories? It was useful to categorize it both ways in this study. In fact, in many of the "more/less" concentrated dichotomies, near significance was achieved for the categorization by four categories. 225 It seemed that the four structure categorization was useful because the combination of deregulation and the greater degree of competition in some markets seemed to result in experimentation over the past few years. It also suggests that more change may occur as competition intensifies, suggesting that the four category structure may ultimately be the most useful. Indeed, starting with the dichotomous categorization, finding significance there, and then testing four categories seems a prudent way of approaching tests of market structure in the future. However, future research should continue to consider new categorizations. It seems advisable to calculate the Herfindahl- Hirschman Index using the An1_yigwing_Allggatign_Egpgrt if the intent is to divide market structure into categories. Concentration levels varied 902 to 3361, with the average index 1751 (or in the oligopoly category), and slightly below the indicator for significant concentration (1800). It may be difficult to obtain this degree of concentration difference if outside signals are not included. Future research should compare indices calculated from this report and from individual market reports to see if results differ. 226 Another important finding is that theoretical precedents previously used in broadcast economic research are apparently correct. Factors proposed to represent market size (prime time rate, early fringe rate, net weekly circulation, cable penetration, VCR penetration and organization size): and profitability (broadcast band, network affiliation and rank), suggest that the traditional theoretical conceptualizations of these variables are correct, and that distinct qualities are being measured. The choice of which indicator to use is simply dictated by theoretical considerations. Yet it may also be argued that the implication of the principal components and factor analyses is that the measures of market size and profitability need to be reassessed. Given the changing nature of the broadcast marketplace, traditional operationalizations may need to be reconsidered. For example, could the size variables actually represent profitability, and the profitability variables actually represent product differentiation? The importance of new broadcast substitutes like cable and VCRs may make differences in signal quality and program offerings on commercial stations more noticeable to viewers. Future research should consider if these traditional operationalizations are valid. 227 Results support the proposal that stations manipulate the study variables to develop a quality image to make their commercial time more attractive to advertisers and viewers. More profitable stations air more commercial time, and more non-program time to remind viewers to watch their quality programming. More profitable stations also schedule more breaks, but have more formal clearance practices to compensate. Complaint practices are more formalized, perhaps to insure that stations stay abreast of how viewers and advertisers perceive their environments. The notion that stations are aware of their own, and their competitors, advertising and viewing environments was supported by the finding that about 84 percent of responding managers saw an ad they declined air on another market station. This represented an attempt to operationalize knowledge of competitive environments. The measure suggests one approach to confirming the existence of knowledge, and ultimately cooperation, among oligopolists. Future research is needed to replicate these findings and examine what other aspects of the advertising and viewing environments may be manipulated. Research might examine if environments vary by program type 228 (action/adventure, situation comedies, etc.), or time period (early fringe vs. prime time, morning vs. afternoon soaps, etc.). It also seems fruitful to discern if and how the networks manipulate their environments, and if those findings are similar to environments found in local markets. Studies could also compare the national commercial television networks with cable networks to see if findings mirror those for more and less profitable stations at the local market level. More and less profitable stations may manipulate their environments differently to make them more attractive to different segments of advertisers and viewers. More profitable stations as a class generally air more commercial and non-program time, but air relatively few program length commercials. They average more breaks per hour than their less profitable counterparts, but accept fewer time units, so their breaks probably appear less cluttered. Specifically, more profitable stations are more likely to accept 15 second spots, and less profitable stations are more likely to accept 20 second spots and 30 and 60 minute time units. Future research might also examine if different program length advertisers appear on more and less profitable stations. Since more profitable stations air 229 fewer program length commercials, they may decline some types that less profitable stations accept. Perhaps more profitable stations decline program length commercials for baldness cures while less profitable stations cannot. This may explain one difference between classes. More profitable stations have more formal clearance and complaint policies, and seem more likely to stay abreast of FTC policies. They consult more policy sources when making advertising policy decisions, have more standards, are more likely to express those standards more formally, and seem more likely to promulgate new standards when new issues arise. It seems that managers at less profitable stations must spend more time clearing ads as a consequence of their less formal policies. Future studies should examine whether this is true. New advertising problem areas also appear to vary by class. More profitable stations were more likely to have standards banning or restricting issue-advocacy advertising, and less profitable stations were more likely to have a standard for 900 phone number ads. Future research should confirm this hypothesis and examine if other types of advertising vary by subclass. Findings might illuminate that segments of viewers obtain commercial information of higher or lower quality. 230 More and less profitable stations may therefore be thought of as fairly distinct quality and price classes, with product differentiation manifested as differences in commercialization and clutter, and clearance and complaint practices. More profitable stations probably compete with each other for a segment of viewers and advertisers rather than with less profitable stations, and vice versa. Station competition may be more intense among class members rather than between them. Future research should seek to confirm this, and discern how competition differs. Interestingly, with similar viewing and advertising environments within classes, price competition can be avoided. Distinct price differences between subclasses may be found, and a price war in the "more profitable" class could conceivably temporarily steal business from less profitable stations. It would be interesting to see if such an effect actually occurs. It would also be interesting to see if a significant number of different advertisers buy time on stations in different classes. If more advertisers of a "questionable" nature buy time mostly on one class, and a certain segment of viewers were linked to that class, future research might have important implications for the quality and quantity of commercial information for some 231 groups. For example, if people in lower socio-economic classes watched programming like "The Morton Downey, Jr." show on less profitable stations which carried more program length commercials for baldness cures and real estate deals, 900 phone number ads, and late night strip ads, how might the quality of the commercial information they receive compare to other viewers preferring programming on more profitable stations? If this knowledge gap were supported and if image differences were found, they could be defined more clearly, and their effect on profitability studied. Classes could then be studied over time to see if the presence of a certain type of advertising or viewing environment leads to financial failure or station closings. For example, if less profitable stations attract viewers of a lower socio-economic status and questionable advertisers, will it be harder for such a station to survive? Viewers with little expendable income are not relatively attractive to advertisers. If the advertising and viewing environments between classes become more distinct over time, it may worsen the image of less profitable stations, making it more difficult to return to a profitable state. 232 Consider this assertion along with the suggestion that local stations may eventually suffer the same fate as local newspapers, with new technologies taking business away from them. Competition from cable, for example, may make being the number three (or lower) station in any market dangerous.3 This danger may be exacerbated by an unattractive advertising and viewing environment, making failure more likely. Future research should test this thesis. Results of this study demonstrate that more complete assessments of station performance should be conducted. These measures of advertising should be considered with traditional programming or public interest performance variables in the future to provide a more comprehensive picture of station performance. Is superior performance in programming correlated with superior performance in advertising? Do stations that program more public interest programming also air fewer commercials? Do stations that air more "trash television" air more program length commercials? More importantly, do stations manipulate program types and advertising to create a more comprehensive "programming environment?" Is the advertising environment simply one aspect of the total environment a station 233 presents to viewers and advertisers? Because of this study, the basis now exists in broadcast economic research to begin testing this hypothesis. Future research should also discern if class membership is a function of the type of advertising. Given the suggestion that class membership is based on profitability, other classes might also exist. The spot market may be considered a classification of stations nationwide, with attractiveness varying by individual market quality and station profitability. Future research should consider this assertion, and consider if other ways of classifying stations might be found. Factor analysis results also suggest that the type of advertising is easily included in future studies. Recall that both Spot and Local advertising loaded on one factor. The type of advertising could be easily included in comprehensive performance studies as a control. Results of the factor analysis also suggest that one theoretical precedent needs to be reconsidered. Net Weekly Circulation apparently should be used as an indicator of market size, not profitability, as the factor analysis suggested it may be the best size measure. Future research should examine this assertion. 234 Results of the principal components analysis suggest that market concentration alone may not be a suitable measure of how competitive a market is. It may be useful to develop an index incorporating market quality and/or other related variables, as well as market concentration, to create a more complete picture of market competition. Future research should examine what other variables or indices might be useful for portraying broadcast markets more accurately. The study also suggests that Rank should be used as a profitability indicator in the future. Station shares were converted into a ranking and entered as a continuous variable in the regression equations. This same measure should be tested again in future studies. Whenever possible, researchers having a large sample and multiple predictors might run a factor analysis to confirm or refute the underlying theoretical structure discussed in this study. It might help to establish certain variables as the preferred measures for certain research situations. It may also help to discern whether the size variables are actually better indicators of profitability. Researchers should also inspect their data for violations of normality. Although results were not meaningfully altered by transforming the skewed variables 235 or bringing in outliers for regression analysis, such a problem could conceivably happen in future studies. Researchers should at least be aware of such a possibility and guard against it. Implications for Mass Media Policy Research: The study also has important implications for mass media policy researchers. Results suggest that stations are to some extent fulfilling the FCC's intent to encourage experimentation and individual decision-making. Stations are indeed scheduling program length commercials differently, for example. However, within this realm of experimentation, the FCC also requires licensees to take all reasonable measures to eliminate deceptive advertising, and avoid abuses with respect to the total amount of time devoted to advertising continuity as well as with which regular programs are interrupted for advertising messages. Given these guidelines, let us reconsider licensee performance. Regarding commercialization and clutter, most stations are not exceeding the commercial time, non- program time, and clutter guidelines. The conclusion is that the quantity of information in the commercial marketplace has not increased to unacceptable levels thus far. 236 Most stations indeed have advertising policies, with the average station having 12 formalized advertising policies. The breakdown of policy areas that stations have standards for is (see Table 35): TABLE 35 Station Advertising Policies (N = 476) 221192.129. W Wing Llievo 29115;: Political Advertising 96.6% 460 Product Protection 85.1% 405 Contests & Games 84.7% 403 Movie Trailers 82.8% 394 Mail Order/Direct Selling 80.0% 381 Product Acceptance 79.0% 376 Issue Advertising 78.6% 374 Copy Acceptance 78.4% 373 Contraceptive Advertising 77.9% 371 Children's Advertising 69.1% 329 Bait & Switch 67.4% 321 Time Standard 67.4% 321 Interruption 62.8% 299 Guarantees 49.6% 236 Free Offers 48.9% 233 Medical Products 46.4% 221 Demonstrations 25.8% 123 The average station consults about five sources when making advertising policy decisions. The breakdown of policy sources that stations consult is (see Table 36): 237 TABLE 36 Station Policy Sources (N = 476) W WW mums Consulting Station Policies 75.6% 360 FCC Publications 66.0% 314 Former NAB Television Code 61.6% 293 National Association of Broadcasters 55.5% 264 Group Owner 42.0% 200 FTC Publications 39.5% 188 Local or State Consumer Agency 34.0% 162 Better Business Bureau 31.3% 149 Network 26.5% 126 Network Code 19.3% 92 Group Code 17.9% 85 BBB Code of Advertising 14.3% 68 BBB Ad Review Committee 6.9% 33 NAD/NARB Case Reports 6.7% 32 Given the different number and frequency of policies and sources, the preliminary suggestion is that clearance formality differs and experimentation may have occurred. The same may be said for complaint policies, and most stations have complaint policies for at least two groups. The breakdown for complaint policies is (see Table 37): 238 TABLE 37 Station Advertising Complaint Policies (N = 476) W WW £21.19); 2.21.1911 Viewers 66.4% 316 Advertisers 59% 281 Interest Groups 55% 260 Minority Groups 54% 257 At most stations, the sales manager is responsible for clearance review (or deciding whether or not a station will accept an ad for broadcast), and he/she averages about five hours per week doing so. About 84 percent of responding managers saw an ad they declined to accept for broadcast air on another market station. Most managers also convey advertising policy decisions verbally. Now consider clearance review in the sense of determining an advertiser's (and/or an ad's) reliability and reputation. Although the average is about 5 hours a week, at 283 stations (or 64.5 percent) 2 hours or less is spent on clearance; at 67 (or 15.2 percent) between 3 and 5 hours is spent; and only at 89 stations (or 20.3 percent) is over 5 hours spent. The modal response was one hour. What does the sales manager do to determine reliability and reputation? He views the ad, calls the 239 advertiser with questions, thinks about his decision for a few minutes, and perhaps views the ad again with the general manager and chats with her about it before making his final decision. If there is only one questionable ad per week, then 2 hours or less might be all a reasonable consideration takes. The common feeling among respondents was that most ads are not a problem, and do not require special consideration. Telephone conversations during the pretest period, and a few notations in open-ended responses support this. In addition, advertisers or their ad agencies often alert them to questionable ads. It may be true that most ads are truthful. However, busy sales managers may send a subconscious message to subordinates: do not send me an ad for review unless its really necessary. Thus, many questionable ads go unscreened. Finally, it may be that the definition of a questionable ad is not the same for a sales manager and the average viewer. And in another sense, it can be argued that deciding upon an advertiser's reputation or an ad's veracity in two hours or less is a rather superficial examination. The average sales manager obviously must make these decisions rather quickly.4 How extensive a background check can you 240 conduct in 2 hours or less? Can you do a comprehensive job of investigating or substantiating a claim in two hours? One might reasonably argue, "no." Therefore, when considering experimentation in the realm of taking reasonable steps to eliminate deceptive advertising, performance appears to be fair or poor. However, as the FCC did not seriously enforce this policy before deregulation, results may not indicate a deterioration in performance. And if the belief that most ads are accurate is true, performance may actually be good. Future research should examine clearance practices more thoroughly to determine if they indeed screen deceptive ads. Practices could be compared with the type of ads that actually air on a station. If deceptive ads, or a significant number of questionable ads (i.e., baldness cures) are found on a certain station, perhaps their clearance practices might be found to be less formal. The FCC once required stations to have at least one person responsible for keeping abreast of FTC policies on deceptive advertising. Although not a direct test of this requirement, 39.7 percent (188 of 473) of responding stations usually consult FTC publications when making 241 advertising policy decisions, while 60.3 percent do not. It appears that stations are no longer making a serious effort, if any, to stay informed about deceptive advertising, assuming that they did so before deregulation. The finding is ominous when one considers that stations were required to familiarize themselves with the charges when an ad is the subject of an FTC complaint, and responsibly decide whether to air it. If these stations "usually" do not consult FTC publications "for advice when making advertising policy decisions,"5 one must wonder if they know when an ad is the subject of a complaint. A significant minority of stations did not consult FCC publications when making advertising policy decisions. Of the 473 stations responding, 33.4 percent (or 158) did not consult FCC publications, while 66.6 percent did. The implication is that: a) a third know FCC policies "by heart," b) a third simply telephone the FCC when they have a question, c) a third no longer consider (or never considered) relevant FCC policies when experimenting, or d) a third do not find FCC policies useful or relevant. Future research should discern if these or other reasons explain why so many stations do not consult FCC publications when making advertising policy decisions. If 242 a third of stations have truly abandoned FCC guidance since deregulation, one must consider what the effect of this may be, and whether it is in the public interest. One must also consider how useful FCC guidance really is. Are these stations ignoring the FCC because they usually find better advice elsewhere? An underlying assumption of the study is that the presence of a policy results in superior performance in that area. This assumption can be tested. For example, 321 stations (or 67.4 percent) reported having a time standard, and 155 (or 32.6 percent) did not. Did those 321 stations actually schedule less commercial time? Were the differences significant? Stations having a time standard average more commercial (14:36 vs. 13:22) and non-program (14:42 vs. 14:22) time, but fewer program length commercials (5.0 vs. 6.5). These differences were not significant (see Table 38). The presence of a policy appears to represent slightly poorer performance on the traditional measures of commercialization and better performance for program length commercials. However, as the commercial average is not above FCC guidelines, the policy may actually improve performance in the sense that it prevents a profitable station that could sell more time from doing so. 243 TABLE 38 T-Tests Comparing Performance of Stations Having/Not Having Advertising Policies (One-Tailed Tests) COMPARISONS Mam Commercial Time Does Not Have/Does Have Non-program Time Does Not Have/Does Have Number of Program Length Commercials Does Not Have/Does Have WW Number of Breaks Does Not Have/Does Have Number of Consecutive Announcements Does Not Have/Does Have 0 t c a d Number of Clearance Hours Does Not Have/Does Have 8 st tio Number of Clearance Hours Does Not Have/Does Have *Denotes significance at the indicated T -0.92 -1.20 -1.19 -0.57 -1.91 -1.47 level. SIG .179+ .115+ .072 .116 .286 .029+* .071+ DF 362.96 335.52 451.00 440.00 194.96 184.46 329.45 +Denotes that a T-Test using a separate variance estimate (rather than pooled variance estimate) was used since the variances within groups appears unequal. Such tests are used whenever probabilities for the F-Test for equality of variance are "small". Consequently, the Separate Variance Estimate T-Test was used whenever F-Test Significance Levels were .10 or smaller. 244 Clutter performance may also be assessed. Interruption standards existed at 299 stations (or 62.8 percent), but 177 (or 37.2 percent) did not have them. Stations with standards averaged more breaks (6.64 vs. 6.34) and more consecutive non-program announcements (3.70 vs. 3.58), but differences were not significant (see Table 38). Results again suggest that performance is slightly worse on stations having policies. Yet these policies may prevent true violations, as the actual number of breaks and consecutive non-program announcements is below or meets old Code recommendations, respectively. The FCC also suggested that stations should review prospective copy. Most stations (78 percent, or 373) had a copy acceptance standard, while only about 22 percent (or 103) did not. Stations having a standard spent significantly more time clearing ads (4.86 hours) than stations that do not (3.38 hours) suggesting that the presence of a standard results in superior performance (see Table 38). The FCC also noted that where there are grounds for a significant complaint, a station should call for substantiation of any factual claim. About 43 percent (or 205) of respondents have a substantiation policy, while about 57 percent (or 270) do not. Stations having a 245 policy tend to spend more time clearing ads (5.19 vs. 4.06 hours), although differences are not significant (see Table 38). It appears that the presence of policies can be beneficial, and possibly prevent abuses. Earlier findings suggest that stations with policies also save themselves clearance time, so the adoption of standards seems useful. The question of responsiveness to community concerns since the demise of ascertainment may be indirectly examined through existing station advertising complaint policies. Viewer complaint policies were found at 66.4 percent of responding stations (or 316), but not at 33.6 percent (or 160). Advertiser complaint policies exist at 59 percent (or 281 stations), but not at 41 percent (or 195). About 55 percent (or 260) had interest group complaint policies, while 45 percent (or 216) did not. And 54 percent (or 257) had minority group complaint policies, while 46 percent (or 219) did not. More than half of responding stations had policies for each group, suggesting that many are concerned with complaints. Station commitment most likely varies by the actual procedures used, and how serious stations actually are in implementing responses to those complaints. Future research should examine the characteristics of those policies, and how they differ by group. 246 Finally, one must consider whether deregulation is in the public interest, and whether the marketplace regulates advertising effectively. In the sense that clutter and commercialization have not increased significantly since deregulation, the conclusion is yes. However, the study has raised a number of new questions about advertising clearance and complaint practices. More profitable stations seem to be performing better in these areas. Are different practices used by more and less profitable stations? Are different procedures actually the cause of these differences? Will this superior performance continue in the future? Should these stations be regulated differently? In addition, related questions were dropped because pretesting indicated the questionnaire had to be shortened to receive an adequate response rate. Important questions to consider in future studies include: a) Do stations investigate prospective advertisers, especially those whose reputations they question? How do they conduct such an investigation? b) Do stations take reasonable steps to assure an advertiser's reliability, reputation, and ability to fulfill promises made to the public, especially those of questionable character? If so, what steps do they take? c) Do stations have any special safeguards or 247 procedures they use when preparing copy in-house, and/or promoting themselves? Do most stations have safeguards? Are they similar? and, d) Are stations aware of what present deceptive advertising requirements are, and how those violations are prosecuted? Are these procedures effective? If not, what others are? Another important finding that deserves future consideration is that market structure may affect conduct and thus performance on some variables. Although it is impossible to confirm, these differences may have developed as a result of deregulation, suggesting that they might continue to change as time passes. The question thus becomes: should all markets be regulated alike? Should steps be taken to promote competition in more concentrated markets, or simply reimpose guidelines there and not in less concentrated markets? Study results show that some differences do exist, but overall performance is acceptable, given the old guidelines. At present, no compelling reason for regulating markets differently exists. However, this question should be considered periodically in future research, to discern if performance worsens over time. Results of this study may be used as a benchmark. 248 Indeed, the FCC might be able to promote competition by reconsidering how it allocates frequencies rather than "re-regulating" television. Frequencies could be assigned by ADI with the appropriate concentration level taken into account. It might be useful to look at individual ADIs and consider the number of home and imported signals a community receives. As distant signals are imported into most markets, the concept of looking at localities in a vacuum seems outdated. Another question which needs to be considered is whether an industry that relies on advertiser support for its survival, no matter how ethical, can be expected to consider the public interest above its own economic considerations when deciding how much commercial time to schedule? And is it truly in the "public interest" to yield decision-making in this area to industry? Preliminary results do not indicate that local broadcasters are performing badly as a whole. However, results do suggest that some minority of managers appears unconcerned with the public interest. Keeping requirements on the books would have simply made it easier to prosecute those violators, and perhaps would have prevented the violations in the first place. 249 A damning suggestion is that FCC policies themselves may lead to poor performance. The airwaves may be worsening because less profitable stations cannot afford to turn down some questionable advertisements, especially since the deceptive advertising requirements were relaxed. Thus, by setting them up at a disadvantage initially through an inferior signal, and later by relaxing standards, the FCC may be encouraging poor performance by less profitable stations. Future research is needed to consider the veracity of this assertion. FCC policy may also insure that these stations remain less profitable, or fail. As they accept more and more questionable advertising, and more and more varied time units, their advertising and viewing environments may progressively worsen, and their viewing and advertising environments may become increasingly unattractive. Certain segments of viewers may become less likely to watch, and certain advertisers less likely to buy time on these stations, making them even less profitable. They may ultimately enter a downward spiral as newspapers do when circulation declines begin.6 Future research is needed to see if this assertion is true, and what steps might be taken to prevent this occurrence. 250 Implications for Organizational Research: The study also has important implications for organizational research. It demonstrates that traditional measures from organizational research can be applied in mass communication and economic research. The main finding was that as organization size increases, the formality of advertising practices increases. The notion of bureaucracy as a function of size is supported in this study. Larger organizations spend more time clearing ads than smaller organizations, have more policy areas, and consult more policy sources. The bureaucratic notion that as organization size increases, more types of tasks are carried out by more specialized staffs is supported. This suggests that organization size should be considered in future studies of all aspects of broadcast performance. One might expect that the number and type of tasks would vary on other performance measures as well. For example, can we expect that larger and more profitable stations will carry more types of public service programming because they have more employees to conceptualize and produce such programs? In that sense, organization size would be a useful control variable for performance studies. 251 It also appears that size necessitates the formalization of advertising policies so they may be communicated to employees. Larger organizations are more likely to promulgate new standards when issues arise. It might be useful to utilize qualitative methods, such as the "participant/observer" technique, to discern how and why standards are promulgated. However, organization size does not seem to have as great an effect on the form those policies take, or how those policies are communicated. Perhaps this is because all responding stations might be considered small organizations, given the range of 1 to 55 employees. On a day-to-day basis, formal communication may not be necessary. Yet, qualitative research may yield new, more effective, measures. Responses confirm previous findings that the sales and traffic departments are primarily responsible for clearance, or deciding whether or not to accept ads on a day-to-day basis. Remember that a subordinate of the sales manager must preview ads first, and then submit questionable ads to him or her for acceptance. Also recall that the traffic manager is supervised by the sales manager. Responses suggest that using the combined size of sales and traffic as a measure of organization size for this study was acceptable (see Table 39). 252 TABLE 39 Employee Responsible for Clearance Review (N = 476) Bergen; Emma: Title WW Sales Manager 52.1% 248 General Manager 15.6% 74 Program Director 7.8% 37 Operations Manager 5.9% 28 Station Manager 5.7% 27 Traffic Manager 3.4% 16 Broadcast Standards 2.7% 13 None .5% 3 Other 6.3% 30 An idea for measuring size is also provided by the study. For example, in programming performance research, one might measure size as the number of employees in the programming, public affairs, and/or operations departments. Pretests could be used, as in this study, to confirm what the appropriate departments are. Results suggest that clearance review can be made more efficient by formalizing policies so managers can spend less time clearing ads. Future research should confirm this and study station organization in more depth to determine ways of improving management techniques. Studies in other types of organization may provide insight and be readily applicable to commercial TV stations. 253 Interested researchers might also examine how other organizations handle ethical decisions that are related to performance. Remember that the sales manager is responsible for clearance review at most stations. He probably has to increase sales by a certain percentage each year as well. Do other organizations separate managers from tasks which affect assessments of their job performance? If so, how do they do so? And can their techniques be adapted for use at commercial TV stations? Miss: The study of commercialization, clutter, clearance and complaint practices at commercial television stations nationwide has expanded knowledge in a number of important ways. It introduced a new area, advertising, to broadcast performance research and proposed new variables as measures of performance. It also provided viable ways to measure those variables. Variables used in previous economic studies were also incorporated and demonstrated to be useful for advertising research. The study introduced a new theoretical area, organizational research, to broadcast performance analysis. It combined a number of theoretical areas in a new way to explain how and why commercial television stations vary advertising practices, described as the 254 advertising and viewing environments. The study extended this idea further and proposed that stations might be separated into distinct price and quality classes, using classical economic theory. Economic methods were also extended by the study. A new method of calculating the Herfindahl-Hirschman Index was introduced to solve the problem of excluding outside signals, cable channels, and superstations from the measure of market concentration. A factor analysis demonstrating the underlying relationships among variables traditionally used in broadcast economic research was also discussed. The study extended policy analysis by using new ways of assessing broadcast performance. Standards which existed at stations were compared to measures of performance to discover if they actually resulted in superior performance. The effect of FCC policy on station performance was also considered. The implications for the commercial marketplace are clear. Advertising policies may be manifested in different ways on different types of stations. Different audiences may therefore be receiving different commercial information of varying quantities and quality. The question has been raised, and future research must discern, whether these assertions are true. 255 Finally, the study was conceived and proposed to assess the changes the author suspected were occurring at stations nationwide. Is the quantity of advertising increasing due to the widespread acceptance of program length commercials? The answer is apparently, "yes." Could this result in information of lower quality for viewers? Again, the answer may be yes, given the questionable nature of some of these ads, and some others which new standards were created for. An important finding suggested by this study is that a few broadcasters appear to be disregarding the public interest. Hopefully, this is not an indication of a coming trend. The content and amount of commercial information available to some viewers may suffer as a result. The suggestion of this study is that profitability, market structure and organization size may be key predictors of the quantity and quality of commercial information. The task of future scientific research is to refine our ability to make these predictions. But there is also the continuing task of both our academic and political institutions to assess the consequences of these forces in order to make policy choices in line with our values. Hopefully the result will truly be in the public interest. NOTES: CHAPTER VI 1Assuming that none of the 50 respondents who declined to answer the commercial time question, and/or distinguish non-program time from commercial time, exceeded the guidelines. 2Letter from James E. Baumhart, President and Chief Executive Officer, Better Business Bureau of Chicago and Northern Illinois, 2 November 1988. He indicated that he would like to "give some publicity through our membership publication as well as properly utilize this source of information in an upcoming presentation to the Newspaper Publishers Association." 3Greg Mitchell, "The Battle for Boston," Channels 2 (January/February 1983), p. 39. 4Consider again that the questionnaire had to be trimmed from 6 to 2 pages, because pretesting indicated that sales managers would not respond unless it took five minutes or less to complete. 5This is the exact wording from the questionnaire. 256 257 6James N. Rosse with James N. Dertouzos, "The Evolution of One Newspaper Cities," Engceegings of Eh: W (Washington. D. C.: Bureau of Competition, Federal Trade Commission, December 14-15, 1978), p. 456-9. The downward spiral is based on the interdependence of a newspaper's two markets: advertising and subscriptions. The underlying notion is that the smaller paper in a two newspaper market is more vulnerable to going out of business. Assume as part of a controlled experiment that this smaller newspaper experienced a 10% decline in advertising demand which means that the advertising space of the paper has declined by 10%. If advertising publication drops then circulation drops 10%. If circulation declines, then ad sales will decline even more. This decline in advertising will again cause a decline in circulation, and so on. Consider also that the unit costs of the enterprise will increase as this is going on. Therefore the paper will have to charge higher ad prices, provide a lower quality product, or lose profits. Because a firm can only absorb so much of a loss, eventually the spiral will cause higher prices and/or lower quality. This, in turn, also compounds the downward spiral effect. Rosse states, however, that the effect may work in reverse as well. APPENDIX A MAILINGS (Explanatory Letter, Follow-up Postcard and Follow-up Letter) 25f! SCHOOL OF JOURNALISM INDIANA UNIVERSITY Ernie 9er Ha" Bloomington. Indiana 47405 (812) 855-9247 May 17. 1988 Mr. John Doe Gen. Sales Manager, KTTV-TV P. O. Box 100 Bloomington, IN 47401 Dear Mr. Doe: With the demise of the NAB Code and deregulation, broadcasters now have the freedom to tailor advertising practices to meet the needs of their own markets and stations. However, many may not be sure of how the ”average“ station is adapting, and may not be aware of practices that other stations are implementing that might be useful in their own markets. You are one of a select group of broadcast managers being asked to provide information about the advertising practices you use at your station. Your name was drawn in a random sample of commercial television stations nationwide. In order that the results will truly represent practices used by stations of all sizes and types, it is important that you complete and return the questionnaire, which takes about five minutes to fill out. Returning it represents voluntary participation on your part. You may be assured of complete confidentiality. The identification number on the questionnaire serves to prevent me from sending a follow-up questionnaire to those who have already responded. It also enables me to access public data about your station (from the fixgnggnnning_x;nxhnnk) so I can tabulate the results by statistical groups (e.g., UHF or VHF stations, affiliates or independents). Under no circumstances will you or your station be identified in any report, published or unpublished. Ideally, the study will provide a “benchmark“ of typical practices employed in large, medium and small markets. It will enable managers to compare how advertising practices differ across market types, and possibly use that information to analyze practices in use at their own stations. If you'd like a summary of the results. please write “Copy of Results Requested" on the back of the return envelope, and print your name and address below it. Please gn_nn; put this information on the questionnaire itself. The study is not connected with any commercial enterprise. It is purely academic and results will be used in a doctoral dissertation and scholarly journals. If you have any questions, please feel free to call me at (812) 335-1725. Thank you for your assistance. S}ncerely, r Lap/L/[l/‘(a/LQ Vgln Wicks Lecturer 259 Ad Practices Survey NON PROFIT ORG. Indiana University BULK RATE School of Journalism U.S. POSTAGE PAID 200 Ernie Pyle Hall Bloomington, IN 47405 Bloomington, IN 47405 PERMIT No. 2 Mr. Joe Sales General Sales Manager WTTV-TV P. 0. Box 0000 Anytown, USA 00000 May 24. 1988 Last week a questionnaire seeking your expertise about advertising practices implemented since the NAB Code's demise and deregulation was mailed to you. Your name was drawn in a random sample of sales managers employed at commercial TV stations nationwide. If you have already completed and returned the question- naire, please accept my sincere thanks. If not, please do so today. It is extremely important that you complete and return it, so practices used at stations of all types and sizes will be represented. If by some chance you did not receive the questionnaire, or it got misplaced, please call me right now, collect (812-335-9240), and I will get another one in the mail to you today. incerely, Mao/«J Jan Wicks Lecturer 26“) SC 00 OF OURNALISM INDIANA UNIVERSITY Em: 9er Hal]! Bloomington, Indiana 47405 (812) 855-9247 June 8, 1988 Mr. John Doe General Sales Manager. KTTV-TV P. 0. Box 100 Bloomington, IN 47401 Dear Mr. Doe: About three weeks ago I wrote to you seeking information about the advertising practices used at your station. As of today I have not yet received your completed questionniare. I have undertaken this study because I believe that research on important media policy issues should include input from industry professionals. And professional input is especially important for this study, as it seeks to illuminate how broadcasters have adapted to the freedom resulting from deregulation to tailor advertising practices to the needs of their own markets and stations. The intent is to provide a benchmark report of typical practices employed by VHF, UHF, affiliated and independent stations in small, medium and large markets. I am writing to you again because of the significance each questionnaire has to the usefulness of the study. In order for the results to truly represent stations of all types in different kinds of markets, it is essential that you return your questionnaire. Again, you may be assured of complete confidentiality. The identification number on the questionnaire simply enables me to access public data about your station (from the fignnggngtjng_xgnghnnk) so I can tabulate the results by statistical groups (e.g., UHF or VHF stations, affiliates or independents). Under no circumstances will you or your station be identified in any report, published or unpublished. If you'd like a summary of the results, please write “Copy of Results Requested“ on the back of the return envelope, and print your name and address below it. In the event that your questionnaire has been misplaced, a replacement is enclosed. Please feel free to call me at (812) 335-1721 if you have any questions or concerns. Your cooperation is greatly appreciated. Sincerely, I 254/[Meio- Jan Wicks Lecturer APPENDIX B SURVEY QUESTIONNAIRE INSTRUMENT 10. 261. NATIONAL SURVEY OF ADVERTISING PRACTICES Did your station ever subscribe to the former NAB Code? ves_ eo__ sou-t um:— Please ghggk_gll the commercial time units your station accepts: ___Josuae ___srmuuo ____sauumo ___Jssuao .__J58flnfl .__Jaauau> ___30snme .__Jmsune ‘__JNRI Does your station accept program length commercials? as an «aura» a. If Yes, how many usually air per month? b. If Yes, what is their usual length? Please gagimngg the proportion of spot and local advertising that airs on your station? 9m' 1 men 1 On average, how many minutes and seconds of gnmmgzgigls air pg: n93: in programming which originates on your station during early fringe and prime access (3:00-7:00 or 4:00-8:00 p.m.)? On average, how many minutes and seconds of gsner_ngnnzggzam maggxial (PSA's, promos, etc.) are aired ngr_ngnr in programs originating on your station during early fringe and prime access? On average, how many agents are scheduled within and between programs n§z_ngnz of programming originating on your station during early fringe and prime access? on average. how many W (“5. PSA's, promos, etc.) are scheduled consecutively within breaks in programming originating on your station during early fringe and prime access? What is the total number of employees in: mm: sues Which gng of the following employees at your station is primarily responsible for clearing (or deciding whether or not to accept) ads for broadcast on a day-to-day basis? (gnegk_gnlx_gne) ,__Jmumunnwmn ___mesmuuml _____srArlou MANAGER _____SAL£SPERSONIACCDUI1 EXECUTIVE _omumous was: _noemn omens _opemlous Assnsmn _nocuu ”smut _m\mc was: _mnou m: _____IIAFFIC ASSISIAII .____rsnouc1tou ASSISIAII _£ualu££n|uc mace: __tucmmus ASSISTANT OI HER 10--a. How many hours per week do you estimate this employee usually devotes to advertising clearance? 262 11. Please indicate all of the areas in which your station has advertising policies by placing a check next to each one: ._____Iu:sasuvvtou staunaaos _____IIIE stannaans ___flmmxrmmumneu ___ymmc:ummnmm _POLIUCAL ADVERIISIRG _lsse mnsme _uuu. cents/oust: scum: _w: a sum _cout£srs : ms _nu onus _oeueus1unous _murt£s ___cvvumwnmm __nnmoupmmmu _____tuxtna£u-s ADVERIISIIG _____aavnt TRAILERS ._____OOWIRACEPIIVE ADVERTISING SUBSIAIIIAIIOI VIEWER COMPLAINTS ADVERTISER COMPLAINTS IWIERESI caauv COMPLAIIIS aluoaxrv csnuo CORPLAIITS ___mme __Jflfll 12. Please ghe§k_311 of the forms in which your station's advertising policies are communicated: _coom£o II A m _us£ um m an: ___uostu wanna: _omwtv mum mu ___awnvmunemmnoa ___;om ___Jnua 12--a. What form is used most of the time? 13. Please gnggk_311 of the methods used to communicate advertising policies at your station: tucouaAtEO to lean POLICY STAFF otscussxau REQUIRED IO READ POLICY SUPERVISOR IELLE EMPLOVEEE INDIVIDUALLY ROME OTHER 13--a. How are policies communicated most of the time? 14. Please ghggk_all of the sources your station usually consults for advice when making advertising policy decisions. LOCAL/Stat! consume: AGENCY _____IAIIOIAL assocIAtnou or IROIDCASIERS scans: Ian ratsvxsuou can: _____;:Atnou Iotnclzs ___Jflumr .__Jenmu¢mx ___Jmascumn ___pmubdm£ _n£n£a muss: mm _m cm: 0' mamas _vcc summons ____us mvemmm uvxeu commas _____rtc PUOLICAIIOIS _____yao/uaas CASE assent: ___mme ___pnm: 14--a. Which gun is consulted most of the time 15. Which types of ads, if any, are not accepted for broadcast on your station (e.g., X-rated movie trailers, etc.)? 15--a. Have you ever seen an ad you've declined air on another TV station in your market? as so umwxmu 16. Are there any new ad practices I've missed? MIL 10: Ad Sum, [Misha School of WHO, Ernie "to an, Ilouinutm, II 47405. APPENDIX C PEARSON CORRELATIONS BETWEEN THE INDEPENDENT VARIABLES APPENDIX C PEARSON CORRELATIONS BETWEEN THE INDEPENDENT VARIABLES TABLE 40 Pearson Correlations Between The Independent Variables (2-Tailed Significance - Divide in Half for 1-Tailed Sig.) Sub. To NAB Code Spot Local Org. Size Subscribe to 1.000 -.104 .100 -.099 NAB Code p=.027 p=.034 p=.035 Spot -.104 1.000 -.791 .164 p=.027 p=.ooo p=.ooo Local .100 -.791 1.000 -.111 p=.034 p=.ooo p=.017 Organization -.099 .164 '-.111 1.000 Size p=.035 p=.000 p=.017 Market .014 -.127 .129 -.084 Concentration p=.755 p=.006 p=.006 p=.072 Broadcast -.279 .135 -.109 .296 Band p=.ooo p=.003 p=.018 p=.ooo Network -.310 .061 -.042 .064 Affiliation p=.000 p=.186 =.356 p=.165 263 sub. Rank Prime Time Rate Early Fringe Rate Net Weekly Circulation Market Quality Cable Penetration VCR Penetration TABLE 40 (cont'd.) to NAB Code .320 p=.000 -0120 p=.049 -.111 p=.070 -.000 p=.996 -.054 p=.247 .019 p=.679 .013 p=.771 264 Spot -.128 p=.006 .168 p=.005 .206 p=.001 -.014 p=.754 .062 p=.183 -.021 p=.641 -.035 p=.451 Local .111 p=.017 -.109 p=.067 -.132 p=.027 .040 p=.390 -.036 p=.431 .057 p=.222 .069 p=.l39 Org. Size -.071 p=.124 .480 p=.000 .524 p=.000 .632 p=.000 .348 p=.000 .453 p=.000 .445 p=.000 Subscribe to NAB Code Spot Local Organization Size (Sales and Traffic) Market Concentration Broadcast Band Network Affiliation TABLE 40 (cont'd.) Market Concentration .014 p=.755 -.127 p=.006 .129 p=.006 -.084 p=.072 1.000 .105 p=.023 .079 p=.086 265 Broadcast Band -.279 p=.000 .135 p=.003 -.109 p=.018 .296 p=.000 .105 p=.023 1.000 .591 p=.000 Network Affiliation -.310 p=.000 .061 p=.186 -.042 p=.356 .064 p=.l65 .079 p=.086 .591 p=.000 1.000 Rank .320 =.000 -.128 p=.006 .111 p=.017 -.071 p=.124 -.101 p=.028 -.548 p=.000 -.737 p=.000 266 TABLE 40 (cont'd.) Market Broadcast Network Rank Concentration Band Affiliation Rank -.101 -.548 -.737 1.000 p=.028 =.000 p=.000 Prime Time -.014 .249 .196 -.131 Rate p=.805 p=.000 p=.001 p=.029 Early Fringe -.026 .245 .080 -.069 Rate p=.661 p=.000 p=.179 p=.248 Net Weekly -.O74 .150 -.034 -.007 Circulation p=.110 p=.001 p=.459 p=.879 Market -.313 .008 -.169 .251 Quality p=.000 p=.853 p=.000 p=.000 Cable -.202 -.000 -.287 .420 Penetration p=.000 p=.984 p=.000 p=.000 VCR -.l77 .019 -.256 .392 Penetration p=.000 p=.678 p=.000 p=.000 267 TABLE 40 (cont'd.) Prime Time Early Fringe Net weekly Market Rate Rate Circulation Quality Subscribe to -.120 -.111 -.000 -.054 NAB Code p=.049 p=.070 p=.996 p=.247 Spot .168 .206 -.014 .062 p=.005 p=.001 p=.754 p=.183 Local -.109 -.132 .040 -.036 p=.067 p=.027 p=.39o p=.431 Organization .480 .524 .632 .348 Size (Sales p=.000 p=.000 p=.000 p=.000 and Traffic) Market -.014 -.026 -.074 -.313 Concentration p=.805 p=.661 p=.110 p=.000 Broadcast .249 .245 .150 .008 Band p=.000 p=.000 p=.001 p=.853 Network .196 .080 -.034 -.169 Affiliation p=.001 p=.179 p=.459 p=.000 268 TABLE 40 (cont'd.) Prime Time Early Fringe Net Weekly Market Rate Rate Circulation Quality Rank -.131 -.069 -.007 .251 p=.029 p=.248 p=.879 p=.000 Prime Time 1.000 .782 .804 .200 Rate p=.000 p=.000 p=.001 Early Fringe .782 1.000 .756 .198 Rate p=.000 p=.000 p=.001 Net Weekly .804 .756 1.000 .312 Circulation p=.000 p=.000 p=.000 Market .200 .198 .312 1.000 Quality p=.001 p=.001 p=.000 Cable .351 .380 .599 .543 Penetration p=.000 p=.000 p=.000 p=.000 VCR .315 .368 .617 .513 Penetration p=.000 p=.000 p=.000 p=.000 269 TABLE 40 (cont'd.) Cable VCR Penetration Penetration Subscribe to .019 .013 NAB Code p=.679 p=.771 Spot -.021 -.035 p=.641 p=.451 Local .057 .069 p=.222 p=.139 Organization .453 .445 Size (Sales p=.000 ~ p=.000 and Traffic) Market -.202 -.l77 Concentration p=.000 p=.000 Broadcast -.000 .019 Band p=.984 p=.678 Network -.287 -.256 Affiliation p=.000 p=.000 Rank Prime Time Rate Early Fringe Rate Net Weekly Circulation Market Quality Cable Penetration VCR Penetration 270 TABLE 40 (cont'd.) Cable Penetration .420 p=.000 .351 p=.000 .380 p=.000 .599 p=.000 .543 p=.000 1.000 .952 p=.000 VCR Penetration .392 p=.000 .315 p=.000 .368 p=.000 .617 p=.000 .513 p=.000 .952 p=.000 1.000 APPENDIX D REGRESSIONS TO ASSESS MULTICOLLINEARITY 2771. A—oo. can» 666. m_ a... “we. ceeu 666. 6_ ac. “mo. case 666. we may CC. ...6ss. 666. ~66. 666. a66.. 666.. n~6.- 666. 6.6. 666.. ~66. Ct! IKE. 666. 63. 6.66. 36. 6M6. ~66. .26. - .66. 666. - 86. CC. ...~.6. 6k6.- 666.. 6.6.. 6e6.. a66.. 6.6.. 666. 6m6.. ~66.. m~s. sum. to Gt. 0. Ci. ...666. 666.- 6N6. .m.. an». 6A6. nn~.- ~66. map. 6.6. 666.- ~66.. to lei 3.8, . 6~6. - 68. . R... R6. 666. . 6.6.6. - E6. 26. . ~66. so»: :6. .66.6 «“66 6666 666666 a .66k .666 ..666 .s.x: 666_ta 66.» .6666 .cea pou6:_p< 66> 6.666 .36: 66: >.t66 agate .6666 6666 .636: 3:66 .66: m656<_e<> c26626666z_ .eu;m_om euom poumoeeoceumc >u_ceoc__.oowu_:z ace was» an «commmOeoou we mao hzwozwawo 2172 ceamuo. eseopfi. ssswcn. essmoo. sssnhc. sssmNO. stepmo- ssLOPF. seesaw a 66.66_6< ... n~o.- Num- one.- ... h~0.- one. .coa au> ... CI. ... pan.- o~o.. ... one. Ono. amo.- .coa ocbau ..66. 666. 666. 6. ~mo.. one. ~—o. omo. ... com.- ..666 .66: ... m6~. 666.- CR. 66..- .66. 66... ~66.- ... ms.- .0. 6~6. .0. CR. 66.. 66- 6~6. ~66. CC. C 66~.- 66.. .66.. m~6. .ocmu can“ ...s) 66c... 662 >.eeu ... .-.- 6.6.. ... 66.. 6.6. .~6.- 6~6.- ... 666. ..6.- ... 6~6. n~6. .66. 6~6. . ... 66.. 6-- . 666. .6.- can“ 66.. 2:6 .663 n.n.uc06v pg m4m .266266662. aces “pa. cccu mum. mm ass “ma. Coca «Po. -o.. ... nw~.- «No.- one. «no. w~o. .UCOU .uscx 666. 6. ace 0. nmc.- m.o.- ... one. -o. 66o. Ono. ... mno. poo. npo.. coo.- ..i no..- coo.- moo. moo.- omo. mne- oeo. neo- ...< sea“ .66: coo.- ~oo.. moo. ome- poc.- ~oo. pro. 6666 .666 Mme.- P60. Ono. ... —~p.u one. opp. woo. o~_m .meo n—o.. Noo. nno.. spo. 0.9. o~o. co_6etuocoa uu> no.6etuocea o_beu >u..656 66x66: co_ue.aue.u >.xooa 66: eyes ooc_ea >.eeu ease 66.. 66.66 .6666 6666 COmHQLuCOUCOU .6166: mmam<~¢<> pzwozuawo BI BLIOGRAPHY BIBLIOGRAPHY Advertising-Estimated Expenditures by Medium. SeeEieeieei enstreee e: gne Unined fitQEQS-l952- 107th Ed. (Washington, D.C.: U.S. Dept. of Commerce, Census Bureau, 1986). v ' sio d a §§e§iene. 28 Fed. Reg. 5158 (1963). "Advertising Standards," 0 s a s Teleyieien. (New York: National Broadcasting Co., January 1986). "Agencies, Networks Battle Over Censor's Role." Agneek 14 November 1983, p. 52. Amerisan_Breadeastins.§emeaniesi_lnei 23 FCC 2d 132 (1970). Anello, Douglas A. and Cahill, Robert V. Qennenne_efi_ene O a sso 'at' . Docket No. 15083. 30 September 1963. Appel, Valentine. "Editorial Environment and Advertising Effectiveness." Jenrnei of Advertising Research 27 (August/September 1987), p. 11-16. t a tte Advertising. 9 FCC 2d 921 (1967). Affirmed in Banzhaf v, E g. 405 F.2d 1082 (1968). Certoriari denied 396 U. S. 842. v's o ' V ° c Bene;_. (New York: Arbitron Ratings Co., February 1987). Ascerneinnent 9f Qenmuniey Eroniems ny gemmezeigi Breadeast_heelisant§- 57 FCC 2d 418 (1976)- 273 274 Atkinson. A. e. 2l2ts1_Iraasfermatiees1_and_8esressien. (Oxford: Clarendon Press, 1985). Babbie. Earl R. Surxsx_8ssearsh_uetheds- (Belmont. Ca.: Wadsworth, 1973). - The_Braeties_ef_§esial_Beseare_. 3rd ed. (Belmont, Ca.: Wadsworth, 1983). Bates, Benjamin J. "Determining Television Advertising Rates." in Robert N. Bostrom, ed. gennnnieeeien Xeenneek_1 (Beverly Hills, Ca.: Sage, 1983), p. 462-75. ."Economic Theory and Broadcasting." Presented to the Mass Communication Theory and Methodology Division of the Association for Education in Journalism and Mass Communication Convention, Memphis, Tennessee, August 1985. Beales, Howard. "Television Program Quality and Restrictions on the Number of Commercials." ETC Wenking Pane; Ne. 39. (Washington, D.C.: Federal Trade Commission, Bureau of Economics, 1980). Beetham, David. finneeneneey. (Minneapolis, Minn.: University of Minnesota Press, 1987). B e v s‘ . (Arlington, Va.: Council of Better Business Bureaus, 1985). Blalock, Hubert M., Jr. Seeiei_§;eeie§iee. 2nd ed. (New York: McGraw-Hill, 1979). Blan- Peter M. Qn_the_Nature_sf_9rsanizatiea§- (New York: Wiley, 1974). Bogart, Leo and Lehman, Charles. "The Case of the 30- Second Commercial." u v ° ' s are 23 (February/March 1983), p. 11-19. Broedceseinngabieeesfiing Yeanbook i98§. (Washington D.C.: Broadcasting Publications, 1988). Bzeedceseing eriecasEing Yeazneek iggi. (Washington, D.C.: National Association of Broadcasters, 1982). 275 Brosterhous, Patricia. "United States v. National Association of Broadcasters: The Deregulation of Self-Regulation." Eederal_semmunieetiens_Las genznei 35 (1983), p. 313-46. Bulmer- M. G- £rinsieles.ef_§tati§tise- (Cambridge. Mass.: The MIT Press, 1967). Camebell_fieue_semeanxi_zti_811 77 FTC 664 (1970)- Capital CitieS/ABC InC-. Ad2ertiains.§tandarde_and enigeiinee, (New York: Department of Broadcast Standards and Practices, November 1986). Caves, Richard. ' ' t Eezfieznenee. 5th ed. (Englewood Cliffs, N.J.: Prentice-Hall,1982). v' 'o a . (New York: Columbia Broadcast System, January 1984). Ceater_fer_Las_end_Seeial_£elisx- 23 RR 2d 187 (1971)- Chamberlin, Edward. Th§_IDEQIY_QI_MQDQEQli§LiQ Cemnenieien. (Cambridge, Mass.: Harvard University Press, 1933). Channel_§exentsen1_1nei 42 FCC Zd 529 (1973)- Cherlss_ef_the_81tz_21_£IC- 143 F.26 676 (2nd Cir-. 1944)- Christopher, Maurine. "Nets Stand Fast on Birth Control Ads." Advereising Age 10 November 1986, p. 36. Cohen. Edward E. Childrsa1s_Telexisien.§emmereialiaatien finnyey. (Washington, D.C.: National Association of Broadcasters, 1987). Colford, Steven W. "Four A's Pushes Nets on Screening Ads." ngyeneieing_nge. 20 April 1987, p. 36. QQlehB§_Bzgsfi§s§LinQ_§Qi 25 FCC 2d 55 (1970)- Qenme;eie1_ngye;e1eing. 36 FCC 45 at 49 (1964). Cemmsrsialiaatien_en_Ty_§tatiens- 49 RR Zd 391 (1981)- 276 Communications Act of 1934. 47 U.S.C.A. Compaine, Benjamin M. "The Expanding Base of Media Competition." gouznai of annnnieetien 35 (Summer 1985), p. 81-96. Comrey, Andrew L. ' st 0 s s s. (New York: Academic Press, 1973). Davis, L. J. "Looser, Yes, But Still The Deans of Discipline." Qnennele (July/August 1987), p. 32-5, 38. Dene Eietnres. Ins. 31 FCC 2d 206 (1971). Dillman. Don A. uail_and_1e1eehene_§urxexs. (New York: Wiley, 1978). Dimmick, John and Rothenbuhler, Eric. "The Theory of the Niche: Quantifying Competition Among Media Industries." leurnal.ef_Cemmunieati_n- 34 (Winter 1984), p. 103-19. Dominick, Joseph R. and Fletcher, James E., eds., Breadsastinsifieseersh_uetheds- (Boston: Allyn & Bacon, 1985). Dubin. Robert. ed. Hsman_Be1atiene_in_edm1nistratiea- (New York: Prentice-Hall, 1951). . Ih§_NQ£lQ_Q£_N2Ik- (Englewood Cliffs, N.J.: Prentice-Hall, 1958). Ehrlich, Isaac and Fisher, Lawrence. "The Derived Demand for Advertising: A Theoretical and Empirical Investigation " American.§se_emis_zexieu 72 (June 1982),p p. 366-88. 'n t' s s o . 50 Fed. Reg. 5583 (1985). Reconsideration denied. 58 RR 2d 864 (1985)- See also Tsle22mmunisatien§_8essareh_and Actien.§enter_!1_fig_- 800 F. 2d 1181 (D. C. Cir. 1986). Eugene Television Qe,, gees Bey, Oregon. 61 FCC 2d 1131 (1976). Eyening Hens Asseeiegien. 49 FCC 2d 380 (1974). 277 ECQ_x1_Leagus_ef_Wemen_Yeters_ef_Califernia- 468 U.S. 364 (1984). ETC1_ECC_1n_Liaisen_agrsement_sn_8d_- 27 April 1972. RR Current Service 2d 11:212-4 (1987). ETC_21_Cslgats:£almelixs_921 380 U-S- 374 (1965)- ETQ_21_Beladam_921 283 U- 8. 643 (1931)- ETC_y1_§eerrx_1_flstshinsen_gei 405 U-S- 233 (1972). "IS-Second TV Commercials Resurfacing Once More." Telexisienzfladie_ege- 9 May 1983. p- 59-60. 101-2. Francois. William E- Mass_Media_Las_and_Bsgslatien. 4th ed. (New York: Wiley, 1986). . 24 FCC 2d 743 (1970). See also 449 F.2d 1164 (D.C. Cir. 1971). Galbraith, John Kenneth. Ame;ieen_geni§eliem. (Boston: Houghton Mifflin, 1952). Gellhorn. Ernest. Antitrust_Law_and_Esenemiss_in_a unnenell. 2nd ed. (St. Paul, Minn.: West, 1981). Gerth, H. H. and Mills, C. Wright, eds. EIQm_M§E_fl§2§IL Eeeeye_in_§eeielegy. (New York: Oxford University Press, 1946, Reprinted in 1975). Gillmor, Donald M. and Barron, Jerome A. Meee gemnnnieenien_Le_. 4th Ed. (St. Paul, Minn.: West, 1984). Ginsburg. Douglas H. Besslatien_ef_nreadeastingo (St. Paul, Minn.: West, 1979). Gorsuch, R. L. £ee;e;_nnelyeie. (Philadelphia: W. B. Saunders, 1974). Qreat_Lake§_Breadea§tins_921 3 F-R-C- Ann- Rep- 32 (1929)- Griffin, Ricky W. and Moorhead, Gregory. engenizenienei Beneyi_;. (Boston: Houghton Mifflin, 1986). 278 Heller, Mary Ann. "An Argument for Elimination of Ascertainment RequirementS-" Issraal.sf_firsadsastieg 25 (Winter 1981), p. 71-80. Hew To Read You; Arbinren Ieievieien Henge; Renezn. (New York: Arbitron Ratings Co., May 1987). ow 'o . (New York: Television Bureau of Advertising). Babbard_nreadsastiag- 62 FCC 2d 970 (1977)- 55 FCC 2d 416 (1975). 55 FCC 2d 431 (1973). 1‘ 100 ° 6 Ions - W-s 010-8‘ F 0:1 -_ ego - 4O FCC 2d 1045 (1973). ,1 .-.U9,17! 9 1 .01 ‘0, I oer-1's ‘ 'V "-1 C 0 0! ‘1! 1° ‘ ‘V - 0! :9 ‘,, s‘ae! S. 32 FCC 2d 412 (1971). ' . ' w QreeseLI_8erkele11_Califi 29 FCC 2d 807 (1971)- Qxflgyex. 37 FCC 2d 528 (1972). g ;- sue ..g egg-” i;;' e! e7 I : ’- eginmnie. 32 FCC 2d 400 (1971). n e on.- -. l 0.1111 0. 8 1° 0 Centreyersial_1e§ue_ 25 FCC 2d 283 (1970). I °e No1 Tcatfo o .7 - = . . 4- 'n- . o.ki. 2 Con n Re s Or so a o 1' 0 1' ‘ 0 ' 9° -.u :1- ! ouu‘ Us. 45 FCC 2d 344 (1974). 1‘ a .0 ' ‘ 0! ‘ 1'10 9‘ 190.. as o Cemm1ss1ea_Ee11e1ss_en_ErogramzLength_Cemmercials. 44 FCC 2d 985 (1974). 279 " {‘1‘Wz ; ° 3 °e° .- ‘19‘: 1 21::1 Leu1s1eaa_end_uississieei- 42 FCC 2d 3 (1973)- Ia_Be_Beguest.hx.flishelas_Zaeele- 23 FCC 2d 707 (1970)- In Lhe Matter 9f AdOELIQD e; Stendezgs Qeeigneg to Eliminane Receptive Adveznising Egon Ieievisign 'et' 101 °, [.31 ‘QNO1at°° ° 1 a.” i 01! -S 7 ° Egeesse§)|. 32 FCC 2d 360 (1971). n .1‘ 1: ,e1 ° 11‘12u‘. ° '-1_ 9 ° 1‘ 7°1I7—S.O ' Ruies--Cennission nganizaeion--flinn Reen ct to De1sgat1ens_ef_Auther1tY_te_the_ghiefi_hreadsast eggegn. 43 FCC 2d 638 (1973). n 1- 1. - - ' -1-me . r... . 1- °wu7—S.°1'S Rules and Reguletions nigh Reeneet Le ngyertieing en Staadard1_EM1_aad_Is1ezisien_8readsast_§tetisns- 36 FCC 45 (1964). 1 1‘ y: e, ° Ll 1° ‘1 '-1 - . -°9:m_ °, 1‘ Cemm1ss12e1s_Ba1___and_Bsg_1etiees_aelatixe_te pivezsification ef Control e: gonnnnity Antenne Ielevisien Systems; eng Ingniny with Res sneet tnere;e L9 Rormuiete Regnlatozy Reiiey eng RuiemaRing endlen Legisletive 2:0 995 e1 . 91 FCC 2d 46 (1982). ,1 1e 1a t‘_ ° 1N‘1°u‘1 ° _‘ °1 , 7°3u‘ Sections 73.35, 73.250 eng Z§.§3§] ef the mm'ss'on's u s ela n to u t' e ° . 100 FCC 2d 17 (1984). t1‘ 1- ° .1‘ e ‘1 ° ‘ ,°1S 7 " a d 6 o t e Co iss on' s no es ' ' e ers an §§a§ions. 95 FCC 2d 360 (1983). I1 he 1a ‘1 O . ,e° ._1u‘1 ° 2 1 ,°su '° .° 3‘ wed 'ce ’ ' s on ses ' o ne t' Wit V 0 ° 5 b is La S 0 he - 0 a1 1e °uu 1' . .°1S Q£_l2§fiL_A__Amengeg. 42 FCC 2d 399 at 404 (1951). 280 In_the_Matter_ef_Ings1r1_inte_§estien_1111212_ef_the ss 0 l s e 'ons ' . .‘S‘ Re 1‘ Raffletre! O 3 are. .‘_t Li_en_e_§.102 FCC 2d 143 at 246- -7 (1985). -1 ‘ Va '. ° I’l- Q I ' ’ ' 7 ° ' ' I‘ 0 'ss' '5 u es e 'o s once at'v s t e ' n ss oc b ' i s s ' 8 5° 'c Ingniny. 2 FCC Rcd 1532 (1987). Renorn of the Commission. 2 FCC Red 5272 (1987). Memetandum Qainien_and_grder- 3 FCC Red 2050 (1988)- 1 ,.,. 1° ° = 11‘ 1e Fairnes__pestrine 30 FCC 2d 25 (1971) _1 1‘ y: ‘ ° 1‘ 1:11 1° ° 219_ or-‘= 12‘ e ess o 'n 'c r s ef.the.§emmunisatien§_Ast- 48 FCC 2d 1 (1974)- atte o ' '0 mm o e c t on i ' c Begu1rs_sats1_aa__2t2gram_Leg_Beguiremente_fer Cemmer2191.1elexisien_§tatiens- 94 FCC 2d 678 (1983)- r o ev's o a 'a o o 'c'es so in nt Regniren nns, and Ezegnem Leg Regnizenen;s fie; Cemmersiel.Tslsxis1en.§tatiens. 98 FCC 2d 1076 (1984). ev' n omme cial at’on Policies Ascer ainment e ments nd am 'reme C2mmercial_Telexi_1ea_§tatiens- 2 FCC Rod 6822 (1987L ITT_Cent1nenta1_Bak1ng_Cemean21_Insii_Et_Ali 79 FTC 248 (1971). IIT_Cent1nental_Bak1ng_Cemeanxi_1neii_Et_All 83 FTC 865 (1973). Johnson, Aaron C., Jr., Johnson, Marvin B. and Buse, Reuben C- Eeenemetries1_hasis_s_82211ed- (New York: Macmillan, 1987). 281 Johnson. Nicholas. Breadsastins_1n_8msrieai_1he nenReee. 42 FCC 2d 3 (1973). Kalton, Graham. (Beverly Hills, Ca.: Sage, 1983). KQQEZIYL_IDQL 24 FCC 2d 149 (1970). Kennedy, J. R. "How Program Environment Affects TV CommercialS." Ieurnal_ef_Adxertis1ng_Bessareh 11 (1971), p. 33-8. Kim, Jae-On and Mueller, Charles W. IDLIQQBQLIQD_LQ_E§§LQ£ Anelyeie. (Beverly Hills, Ca.: Sage, 1978). KMEC1_The_§tatien_ef_the_§tats. 6 FCC 729 (1939)- Krasnow, Edwin G., Longley, Lawrence D. and Terry, Herbert A. s 3rd Ed. (New York: St. Martin's Press, 1982). Krasnow, Erwin G., Umansky, Barry D. and Kennard, William E. Cemments_2f_the_Natienal_Asseeiatien_ef fizeegeeenene. MM Docket No. 83-670, 21 November 1983. Krugman, Herbert E. "Television Program Interest and Commercial Interruption." Reeeenen 23 (February/March 1983), p. 21-3. LaBarbera, Priscilla A. "Advertising Self-Regulation: An Evaluation." M§n_Busines§_Tssiss 28 (1980). p- 55-63- . Analyzing and Advancing the State of the Art of Advertising Self-Regulation." QQQIDQI_QI_AQ¥§ILIEIDQ 9 (1980), p. 27-38. . "The Antitrust Shadow Over Advertising Self- Regulation-" Curreat_Iss_e__and_Bsseareh_1a Advertising (1981), p. 57-70. . "The Diffusion of Trade Association Advertising Self-Regulation." QQBIDQI_QI_M§IKQLIDQ 47 (Winter 1983), p. 58-67. 282 Lewis-Beck. Michael 8- Asslisd_Bsgreesien1_Aa In;;e§ne§i_n. (Beverly Hills, Ca.: Sage, 1980). ewe S Mis1eadiag_Badie.and_Ty_Adxsrtis1ng- 21 February 1957, RR Current Service 2d 11:201 (1987). I 'c n e- -s--1-'b' ' 1 :es-- to .a. .s . Raise, Misleading o: Deeen§iye Adv ernieing. 74 FCC 2d 623 (1961). Issued 7 November 1961. RR 2d Current Service 11:205 (1987). Linton, Bruce A. "Self-Regulation in Broadcasting Revisited." IQBIDQIIEE_QBQILEIIY 64 (Summer-Autumn 1987), p. 483-90. Litman, Barry R. "Decision-Making in the Film Industry: The Influence of the TV Market." lennnei_e£ gennnnieeeien 32 (Summer 1982), p. 33-52. . "Market Share Instability in Local Television News-" Journal_ef_8readeasting 24 (Fall 1980). p. 499-514. . "Measuring Divestiture of Network Owned Television Stations: An Econometric Approach." The_An_it£u§t_Bulletin 25 (Summer 1980). P- 363-76- . "Public Interest Programming and the Carroll Doctrine: A Re-examination." lenznei_efi_fineegeeening 23 (Winter 1979), p. 56. . "The Economics of the Television Market for Theatrical Movies." leurnal.sf.§emmunieatien 29 (Autumn 1979), p. 20-33. "The Television Networks, Competition and Program DiversitY-" Journal_ef_8readeasting 23 (Fall 1979), p. 393-409. Litman, Barry R.and LeBlanc Wicks, Jan. "The Changing Advertising Market for the U. S. Television Networks." John D. Leckenby, Ed. The Ereeeedings ef ghe eiggg e d m f v (Austin, Texas: John D. Leckenby, College of Communication, The University of Texas at Austin, 1988), pp. RC27—33. 283 Litterer. Joseph A. Qrgaa11at1ens1_§truetsre_and_8ehaxier. 3rd ed. (New York: Wiley, 1980). Maddox, Lynda M. Maddox and Zanot, Eric J. "Suspension of the NAB Code and Its Effect on Regulation of Advertising " Journalism_Querterlx 61 (Spring 1984). p. 125- 30, 156. Mansfield. Edwin. Erinsislss_of_uisrcesenemie§- 3rd ed- (New York: Norton, 1980). Maxfleser_hreadsa§tins_sei 8 FCC 333 (1941). McCavitt, William E. and Pringle, Peter K. E1estreaie_uedia_uaeesement- (Boston: Focal PresS. 1986). McGregor, Michael A. "The FCC's Use of Informal Comments in Rule-Making ProceedingS-" Journal_ef_8reedsasting 30 (Fall 1986), p. 413-25. RenegiEh Cenn, v, Egg. 809 F.2d 863 (D.C. Cir., 1987). Mermigas, Diane and Colford, Steven W. "Don't Burn Those Logs, Ad Units Tell TV Stations." hgyereieing_hge, 2 July 1984, p. 1, 42. N1d_Nss_Yerk_Breadeasting_Cerel 42 FCC 2d 594 (1973)- hidland_Tslexisisn_22121 42 FCC 2d 591 (1973)- Miracle, Gordon E. and Nevett, Terence. Yeluntarx e 1 'o v 's' . (Lexington, Mass.: Heath, 1987). Mitchell, Greg. "The Battle for Boston." ghenneie 2 (January/February 1983), p. 39-41, 44, 58. Mord, Marvin S. and Gilson, Edith. "Shorter Units: Risk- Responsibility-Reward- " Journal_ef_Adxertisiag Reeeeneh 25 (August/September 1985), p. 9-19. Morse, Susan. "Ads for Contraceptives Fail to Rile Viewers." hgneeh, 21 January 1985, p. 38-9. 284 Murphy, John H., Cunningham, Isabella C. M. and Wilcox, Gary B. "The Impact of Program Environment on Recall of Humorous Television Commercials," lehxngl_efi Advernising 8 (Spring 1979), p. 17-21. Multimedial_lnsi 25 FCC 2d 59 (1970)- s ' . 2nd ed. (Washington, D.C.: National Association of Broadcasters, 1984). Natienal_flreadsastins_ssi 29 FCC 2d 67 (1971)- Neter, John, Wasserman, William and Whitmore, G. A. Aeslied.§tatisties- (Boston: Allyn & Bacon. 1978) Noll, Roger G., Peck, Merton J. and McGowan, John J. (Washington, D.C.: The Brookings Institution, 1973). Norusis. Marija J- Ths_SBS§_§uids_te_nata_Analeis- (Chicago: SPSS, 1986). Nunally, Jum C. Reyehemennie_1heezy. 2nd ed. (New York: McGraw-Hill, 1978). Qeean_§nraY_Cranherriesi_1nsii_fit_811 80 FTC 975 (1972). Owen, Bruce M., Beebe, Jack H., and Manning, Willard G., Jr. Telexisien_zsenemiee. (Lexington. Mass.: Heath. 1974). Pedhazur, Elazar J. ' o v Reeeexeh. (New York: Holt, Rinehart & Winston, 1982). Pfizer_1nei 81 FTC 23 (1972)- . .- — 1 - . . ,- . x-"._.. I. It 1 - ..1 °. .1- ! ° '2’. 1‘ Lieeneing. 102 FCC 2d 1179 (1986). Preston, Ivan L. "A Review of the Literature on Advertising Regulation." Current_1ssues_and_8esearsh In_Adxertising 2 (1983). P- 1-37- ' '1‘ ° °u1!_1 ' °.° ‘11: .° 3 °-_°. a: Appiigene_. ‘27 FCC 2d 650 (1971L 285 Prisuta, Robert. "Local Television News as an Oligopolistic Industry: A Pilot Study." Journal_ef_flreadsaeting 23 (Winter 1979). p-ol-B- . "The Impact of Media Concentration and Economic Factors on Broadcast Public Interest Programming." Journal_ef_areadeasting 21 (Summer 1977), p. 321-32. Pregram_Lensth_Csmmsrsials- 39 FCC 2d 1062 (1973)- a“: 1t ' U°‘S ,OW=1° ,‘ ‘ ; °1 it! 0 1‘ 1‘! a 1 ' e ha e t ' o ' ' . (New York: Television Information Office, 1985). O O o_ ‘ - - 1 - --— \ r..! . ' '2'. 2: Ir 0 Public NOtice 95462, 2 July 1946. F. Ray, Michael L. and Webb, Peter H. Advertising s ow ed v -- BQEQIL_N21_ZB:113. (Cambridge, Mass.: Marketing Science Institute, September 1978). Ray, Michael L. and Webb, Peter H. Expenimentel_Reeeezeh e E ects TV u ' D u ' v' -- e (Cambridge, Mass.: Marketing Science Institute, April 1976). Ray, Michael L. and Webb, Peter H. "Three Prescriptions for Clutter." Ieuraa1_ef_Adxertising_Bessarsh 26 (February/March 1986), p. 69-77. Bed_L1en_Breadseating_2211_lnsi_xi_ECC- 395 U.S- 367 (1969). {‘9° ' °uu —: °1 ‘1 921 £r_gramming_1nguirx 44 FCC 2303 (1989) . 13 FCC 1246 (1949). Robinson, Glen 0., Gellhorn, Ernest and Bruff, Harold H. The_Administratixs_2reees_. (st. Paul. Hinn.: West. 1982). 286 Rosse, James N. with Dertouzos, James N. "The Evolution of One Newspaper Cities," Ezgcgegimgs g: the Sympgsium 9m Media Concentration; Volume 11. (Washington, D.C.: Bureau of Competition, Federal Trade Commission, December 14-15, 1978). Bu§b_hrgsdsasting_sernl 42 FCC 2d 483 (1973)- Rushing, William A. "Organizational Size, Rules and Surveillance." In Joseph A. Litterer, ed., ' ° . 3rd ed. ' s: (New York: Wiley, 1980), p. 396-404. e Sales 8 Marketing Management. (New York: Bill Communications, 1986). Part 1. Scherer, F. 11. WW Egmfigmmgmgg. (Chicago: Rand McNally, 1970). figggs Bogpmgg Q 99, 2, E19. 258 Fed. 307 (7th Cir., 1919). "Self-Regulation of National Advertising." NAQ_gg§g_nggmt 13, 15 July 1983, p. 20-3. Sherman Antitrust Act of 1890. 15 U.S.C.A. Slater, Robert 0. "Organization Size and Differentiation," In Samuel B. Bacharach and Stephen M. Mitchell, Eds. S. 4 s a o (Greenwich, Conn.: JAI Press, 1985), p. 128. Smith, Leon C. "Local Station Liability for Deceptive Advertising." Jougnal Q: Bzgaggastimg 15 (Winter 1970-1),p. 107-12. Soldow, Gary F. and Principe, Victor. "Response to Commercials as a Function of Program Content." 12urnal.2f_Adzerti§ing_Besear2h 21 (April 1981). p. 59-65. . 6151 121‘ .0 1‘ 011- S 01 01,1-°L-- a. Belatize_t9_2ubli2_Intere§t1_senxeniensei_gr Nggggsity. 2 F.R.C. Ann. Rep. 166 (1928). STE Compomatiom et a1. 87 FTC 56 (1976). 237 steiner. 6- A. Ihe_E99219_Look_et_lelexision- (New York: Knopf, 1963). Suburben_firozgge§ter§ 20 RR 951 (1961)- See also flenrx_21 nee. 302 F.2d 191 (D.C. Cir., 1962). Certoriari denied. 371 U.S. 821 (1962). eyggcgse Eeeee Conneii 2, 3:23. 99 FCC 2d 1389 (1934). Sxre22§e_£eege_gouneil_11_fllyn 2 FCC Red 5043 (1987)- cse e e o- ' mu. 0 .- ;--.1.-r ,. 011-11‘1 2 FCC Red 794 (1987). I r‘ - . l A“‘! uem2rendum_921n12n_2nd_grder 3 FCC Red 2035 (1988). Tabachnick, Barbara G. and Fidell, Linda S. Heine Multixeriete_§teti§tie_- (New York: Harper& ROW. 1983). Iain anoadceseing Ce, 39 FCC 2d 1070 (1973). IEdexision_§_ssble_regtbook- Stations Volume- No. 55. (Washington, D.C.: Television Digest, Inc., 1986). "The TV Code is Dead" ugrketin9_§_nedie_nesi§ions 18 (December 1983), p. 64-66, 135, 13a. Toensr_goreer§tion- 21 FCC 2d 148 (1968)- T9rner_ar2ed2asting_2f_nQrth_Ceroline- 42 FCC 2d 622 (1973). Tyrie, Carl and Clift, Charles III. "Broadcast Public Files: FCC Requirements, Station Performance and Public Use-" 1991211_9£_n:2992991199 28 (Summer 1984), p. 305-13. S. Denangmene e: Justice Merge; enigelines. Issued June 14,1982. n'ted S tes v t' s o 536 F.Supp. 149 (D.C. Cir., 1982). Hnited_Telezisi9n_Compenx_9f_Neu_nemnshire. 42 FCC 26 632 (1973). 288 "Video Scope: Evaluation of Television Advertising," EELD£§I§_IDB 289 (1954): Po 40- Wall. Francis J. Statistigal_Data_Analxsis_nandbgoh- (New York: McGraw-Hill, 1986). Walton, Eric J. "The Comparison of Measures of Organization Structure." e o e v' w 6 (1981), p. 155-60. Earner:Lambert.§ompanx_21_£TC. 562 F.2d 749 (1977). Webb, Peter H. and Ray, Michael L. "Effects of TV Clutter." l2urnal_9f_Adxertisin9_Besearsh 23 (February/March 1983), p. 7-12. Heigel_firoadsasting_§21 41 FCC 2d 370 (1973)- WEIL1_In21 38 FCC 2d 411 (1972). When is e EQtEQI Beeinese EUIEQBZ (Arlington, VA.: Council of Better Business Bureaus, 1985). Wheeler-Lea Act of 1938. 52 Stat. 111. Wicks, Robert H. Using "Benefit Based Segmentation" Strategies to Define Television News Audience Segments: A Segmentation Model Based on Media Gratifications." Unpublished Doctoral Dissertation. Mass Media Ph.D. Program, Michigan State University, 1987. Wirth, Michael O. "The Effects of Crossmedia Ownership on Television and Newspaper 'Prices'." Unpublished Doctoral Dissertation. Mass Media Ph.D. Program, Michigan State University, 1977. Wirth, Michael O. and Wollert, James A. "The Effects of Market Structure on Television News Pricing." Journal_2f_nroadoasting 28 (Spring 1984). p- 215-24- Wollert, James A. and Wirth, Michael O. "Deregulation of Commercial TV in the Osa." Tel:sommunieations_roliox 6 (September 1982), p. 155-63. 289 m,me,__1;_r_(e_L 37 FCC 2d 748 (1972). exon-Iy. Ins. 42 FCC 2d 639 (1973). Zanot, Eric J. "The National Advertising Review Board, 1971- -1976-" lournalism_uonogranh_ 59 (February 1979). p. 1- 70. . "Unseen But Effective Advertising Regulation: The Clearance Process." l2urnal.9f.hd_erti_ing 14 (1985),p. 44.