"IllllquIn-Jlllllllullhfll. UN IVERSITY LIB BRARIES 7 k |11111111|1111111111111111111 | 1|1111l111111| 3 1293 00910 8667 This is to certify that the thesis entitled MUSIC PREFERENCE AND PERFORMING ARTS TELEVISION VIEWING: AN ANALYSIS presented by Peter Alan Frahm has been accepted towards fulfillment of the requirements for M . A. degree in Te lecommunicat ions WV Major professor Date February 16, 1990 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution f‘ ' LiBRARY Michigan State L__ University PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. DATE DUE DATE DUE DATE DUE 11 _1L_ fl —‘—1 MSU Is An Affirmative Action/Equal Opportunity Institution cmkcmpma-pd "MUSIC PREFERENCE AND PERFORMING ARTS TELEVISION VIEWING: AN ANALYSIS." BY Peter Alan Frahm A THESIS Submitted to Michigan State University in partial fullfillment of the requirements for the degree of MASTER OF ARTS Department of Telecommunication 1990 ABSTRACT MUSIC PREFERENCE AND PERFORMING ARTS TELEVISION VIEWING: AN ANALYSIS BY Peter Alan Frahm Music has established itself as a viable television programming tool within the last ten years. It is hypotheti- cally possible to develop music programming for television that appeals to any type of musical interest. When develop- ing such programming, one must first determine if a rela- tionship exists among music preference, music consumption and fine arts television viewing, and identify any potential market niches. Two random samples were drawn from the Lansing, Michi- gan television market, consisting of persons who either did or did not financially support public television. A tele- phone survey instrument was used to gather data on the samples' attitudes regarding different musical styles, current musical programming and different media used to enjoy music. Analysis of variance, correlations and factor analysis revealed significant potential market niches where music programming could acquire an audience. Few statisti- cally significant differences between the two groups exist- ed; demographics had a larger impact in determining the possible niches. Copyright by Peter Alan Frahm 1990 ACKNOWLEDGEMENTS I would like to take this opportunity to thank those persons who made it possible for me to survive the rigors of graduate education in general, and this thesis in particular: First and foremost, I would like to thank my parents, Gerald and Marianne Frahm, for all of their love, support and patience during my undergraduate and graduate career. Little did I know all of the things they taught me before I ever made it to kindergarten, much less college, would be so instrumental to my successes at Michigan State University. Thanks for showing me the importance of discipline, pride, and getting an education. Thanks also for showing me the value of reading ... and listening. Professors Blaine McKinley and Beulah Monaghan, (American Thought and Language), and Professor Albert Drake (English) taught me the ins and outs of writing at various stages of my undergraduate days at Michigan State. They not only taught me how to write, but also how to critique my work. (Not an easy task, I might add.) The lessons learned in their classes helped make writing this little beastie a iv wee bit easier. Don Kemp, the TC department's chief engineer, also deserves mention in this space. While I never took a class from him, he did manage to teach me quite a bit. The most important thing I learned from Don was, no matter how small or unseemly the project, don't skimp. I will never forget the impish grin on his face when Eileen Mullin and I came up less than three feet short of phone cable when trying to "help" Don install the phone in the new MSU Telecaster office. All he said was, "You guys skimped on the cable, didn't you?" I'll never live that one down, yet I learned a lot from that little fiasco. My thesis committee in particular deserves all the thanks I can bestow upon them, and then some. I was fortunate enough to work with a really super bunch of folks. Dr. Linda Kohl, formerly of the MSU Department of Telecommu- nication, was instrumental in helping me with the research design of this project. She managed to keep me focused upon the problem at hand; consequently, little time was wasted in theoretical "backtracking." The research design section of a thesis is, in my humble opinion, the most difficult stum- bling block to overcome. Dr. Kohl's guidance kept me from tripping all over my own feet, so to speak. As far as I am concerned, she went above and beyond the call of duty while helping me out. Gary Reid, the TC department's audio specialist, was also a tremendous help during this endeavor. Gary played the role of devil's advocate in several bull sessions in the audio labs over the course of this project. Gary's approach during these sessions was simple and direct; consequently, good ideas didn't get lost in the esoteric shuffle. Gary's ability to get straight to the heart of the matter at hand was invaluable. Last, but not least, is Bob Albers, my thesis supervisor. His ability to evaluate a work as a whole without losing any of the details contained therein were invaluable during the conceptualization and writing stages. These same qualities are also what makes him an outstanding video instructor. He is the person who introduced me to the kind of television programming I enjoy working on the most: fine arts/musical performance. This type of programming was the inspiration for this thesis. Finally, I would like to thank those persons without whom this thesis would not be possible: Steve Dick's TC 300 class (who served as phone operators for this project); the Integrated Telecommunication Services Lab staff (who were most helpful and patient throughout); Indra DeSilva, for his insights into IBM computers and phone surveys; PhD. candi— date Jeff Brand, for brainstorming sessions with the "M-M.PhD. Chalkboard;" MSU Performing Arts Marketing Direc- tor'iIimmy Hilburger and last, but not least, MSU Telecaster menIbers Paul Buszka and Wendy Scarborough for taking time out car their busy schedules to help run the phones. vi Table of Contents Listing Of Tables.0.0...00............OOOOOOOOOOOOX Chapter I INTRODUCTION............................1 RESEARCH PROPOSAL............................2 FINE ARTS TELEVISION.........................3 MUSIC, MEDIA AND MASS APPEAL.................3 GOAISOOO.......OOOOOOOOOOOOOOO0.00.00.00.00004 Chapter II REVIEW OF LITERATURE ................... 5 THEORY..... ....................... ... ..... ...6 MEASUREMENT..................................9 MEDIA USE....................................13 MUSIC AND TELEVISION.............. ..... ......20 POTENTIAL IMPACTS FOR PTV....................26 TV AS AN ARTS MEDIUM.........................28 SWRYOOOOOOOOOOOOCOOO......OOOOIOOO... ..... 30 Chapter III RESEARCH DESIGN.......................32 METHODOLOGY..................................32 SAMPLING.....................................33 PROCEDURE....................................35 INSTRUMENTOOOOOOOOOO......OOOOOOOOOOOOOOOOO0.35 vii WSIC CONSUMPTION.‘......COCOOOOOOOOOOOOOO00.37 TELEVISION VIEWING PATTERNS..................39 PRETEST...OO0.0............OOOOOOOOOOOOOOO00.40 Chapter IV RESULTS AND ANALYSIS...................43 DEMOGRAPHICS.................................43 MUSIC PREFERENCE.............................48 DEMOGRAPHIC INFLUENCES ON MUSIC PREFERENCE...52 Preference and Age......................52 Preference and Income...................54 Preference and Occupation...............56 Preference and Education................59 Preference and Gender...................59 Preference and WKAR Membership..........63 Overall Preference Structure............64 MUSIC CONSUMPTION............................68 MUSIC ON TELEVISION..........................77 MEDIA PREFERENCES FOR ENJOYING MUSIC.........84 FINDINGANAUDIENCE.........OOOOOOOO00.......88 Chapter V CONCLUSIONS.............................94 REVIEW.......................................94 CONCLUSIONS..................................95 Market Segments.........................96 OUTSIDE INFLUENCES...........................99 Influence of Radio Programming..........99 Radio Listening Patterns......... ...... 101 viii Radio as a Preference Predictor........101 OTHER IMPACTS................ ..... . ..... ....102 Demographics...........................102 Michigan State University..............103 PTV Preconceptions.....................104 Television Program Variables...........104 Television Programming Factors.........106 SUMRY0000000000000000000000000 ....... 00000107 Program Development....................108 Recommendations for WKAR-TV............109 APPENDIX A SURVEY INSTRUMENT.....................112 APPENDIX B CORRELATIONS OF MUSIC STYLE PREFERENCES WITH TELEVISION PROGRAMTITLESOOOOOOOOOOOOO0.00000000.128 APPENDIX C CORRELATIONS OF MUSIC STYLE PREFERENCE WITH MEDIA PREFERENCES................133 APPENDIX D CORRELATIONS OF MUSIC STYLE PREFERENCE WITHMEDIAUSEOOOOOOOOOOCOOOOOCOOO0.0.138 GENERAL REFERENCE AND BIBLIOGRAPHY...............143 ix Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table Table 10 11 12 13 14 15 16 17 18 LISTING OF TABLES sample ParameterSOOOOO00.0.0000000000000.044 Sample Music Style Preference Means.......49 Additional Music Preferences..............50 Favorite Musicians and Artists............51 Music Music Music Music Music Style Means Style Means Style Means Style Means Style Means by Age Group............53 by Income Level.........55 by Occupation...........57 by Education..... ....... 60 by Gender...............61 Music Style Means by WKAR Membership.....63 FactorAnaIYSiSOOOOOOOCOOOO0.00000000000065 Local concerts Attended...............00.69 Radio Station Reach (Spring '89 Sweeps)..71 Television Viewership....................74 Cumulative Responses of "Occasionally" orMoreOOOOOOIOOOOOOOOO0.0.0.0...0000000076 Television Viewing.......................78 Local Music Programming Ratings..........79 Media Preferences........................86 Chapter I INTRODUCTION Most people would turn on their radio to hear music before they would turn on their television. This should not be a surprise, as music is one of the programming items that radio stations use to attract listeners. Music has also demonstrated its use as a programming tool for television as well. Several programming examples come to mind immediately: American gandstand, Soul Train, Nightflight, and, of course, Music Television. All of these programs utilize music as a hook to attract viewers. Each of the aforementioned programs caters to a specific audience: in these cases, the rock and dance music crowds. It would stand to reason that the audi- ences that tune into these shows are interested in the type of music that is being played. Hypothetically speaking, it should be possible to develop a television program with an emphasis on music for nearly any type of musical interest. All one need do is find out what your intended audience wants to see - and hear- and then deliver it to them. A commercial UHF station in Marlboro, MA., did just that; the station airs commercial music videos 24 hours a day. The station's playlist is determined via programming research gathered from their viewer request line (Weinstein, 1985). The concept has worked; the station registered in the ratings books within six weeks, instead of the usual 24. Current and up-to-date audience information is espe- cially crucial to non-profit programmers as well, as they have less margin for error in determining audience needs than do their advertising-supported counterparts. Advertis- ing based programmers can easily outspend non-profit pro- grammers in terms of audience research and promotions: thus, they have a distinct advantage in the marketplace. The non- profit status of Public Television (PTV), which relies on grants and donations, is a case in point. PTV has carved a niche for itself as a Fine Arts programmer on the national level; it should be hypothetically possible to carve a market niche for a public station at the local level as well. RESEARCH PROPOSAL: The author's intention is to find an answer to the question: Is there a relationship among Music Preference, Music Consumption, and Musical Television Viewing? If a relationship exists, then performing arts producers would find information concerning Music Preference and Music Consumption Habits invaluable, as it would assist them in determining their target audience's viewing preferences. A producer would then be able to customize programming at the local level, which, in turn, should lead to more viewers as well as increased viewer loyalty. This research should be particularly useful to PBS affiliate WKAR Television, which airs many performing arts programs of national and local origination, examples of which include Great Performances and segments of Michigan Skyline. FINE ARTS TELEVISION Performing arts programs have two major components: music and the accompanying visuals. It is necessary to find out what an audience would like in both areas for the pro- posed program to be successful. There are other factors that influence program success and failure, such as time slot, program length, station, et cetera. For the purposes of this thesis, only the creative component of musical style will be analyzed in extensive detail. MUSIC, MEDIA AND MASS APPEAL The key to this entire exercise is developing a reli- able yet effective method for measuring music preference. Academic journals such as Popular Musig and Society or The Journal of Music Therapy serve as starting points for the disclosure of how music preference is influenced, measured and structured, both for the individual and for mass audi- ences. Also of interest to the researcher are those media which are used most often by the consumer to enjoy music. Relationships may exist whereby certain audience segments enjoy their preferred music on different media. These rela— tionships can be used to develop or predict market segments that are inclined to make use of performing arts television. If the market segments that reveal themselves are indeed viable, and not a statistical accident, then relationships between music preference and musical programs that cater to specific tastes should exist as well. In effect, if the individual's music preference is known, then it should be possible to predict whether s/he will watch a television program that features a particular type of music. Also, knowing the aggregate music preference of a given population would assist in predicting the success or failure a particu- lar type of music programming. GOALS The ultimate goal of this study is to see if a rela- tionship exists among music preference, music consumption and fine arts television viewing patterns such that once the aggregate music preference of a population is known, televi- sion programming can be developed that will attract certain audience segments with hopes of success. While this study was performed at the local level, the methods used should be applicable in other markets regardless of size or location. Furthermore, the information garnered from this type of study may also be useful in terms of grant applications for the arts and letters, radio station programming and espe- cially advertising/promotion campaign planning for radio stations, record stores or concert halls. In any case, the information need is the same: what do people like to see and hear? Chapter II REVIEW OF LITERATURE One of the more recent successful programming phenome- na has been the music video, which was made into a household word by the Music Television Network. When examining the decision-making process behind the development of MTV, and other commercial successes, a pattern arises. The most prominent commercial triumphs came about as the result of solid market research. The starting point of the researchers was to determine audience music/band preferences (Desinoff, 1987 and Weinstein, 1985). A great deal of research has been done on music prefer- ence. However, much of it has to do with the music prefer- ence of children. Research has been performed on the types of musical attributes (things such as tempo and instrumenta- tion vs. voice) that appeal to children (LeBlanc and Cote, 1983). Hair (1981) studied the capabilities of children to consistently identify certain musical concepts. Environmen— tal effects such as how repetition affects the liking of music have also been investigated (Hargreaves, 1984). Howev- er, not much research has been done on the appearance of musical material on television. Most music and television studies involve the social and cultural impacts of music videos (Oglesbee, 1987; Walker, 1987; Wells, 1984 and Jones, 1988). THEORY LeBlanc (1982) developed an interactive theory of musical preference that takes several factors into account. This theory is of interest because it examines several of the same production and audience variables that are of interest to the television producer and the television programmer. LeBlanc maintains that several different physi- cal, psychological and sociological factors interact to form a person's musical likes and dislikes. These are divided into multiple (eight) levels that begin at Level 8, stimulus inputs, and continue through to the final decision of like or dislike (Level 1). These factors are considered by Le- Blanc to be potential influences on music preference because of previous studies or the inherent ease of study. Of par- ticular interest to this study are those variables that are of use to both the music educator and the fine arts televi- sion producer: the music stimulus and the socioeconomic status of the listener. Level eight, the music stimulus as originally presented to the listener, has two components. The first is the actual music itself, which is comprised of physical properties (e.g. timbre), complexity, referential (or intended) mean- ing, and the quality of performance. These variables are all under the control of the composer, arranger, and performer, and together make up a music genre. The second component is comprised of environmental influences on the listener. These include peer groups, the listener's family, educators and authority figures and incidental conditioning (random previ- ous experiences, good or bad). All of these variables assist the listener in developing music "values" as to what is or is not worthy of their listening time. Both of these components interact with the medium upon which the music is presented. These would include media that could successfully reproduce music performances: vinyl records, cassette tapes, compact discs, and video. This area is where the producer, either audio or video, would have a great deal of influence on whether the music in question would receive a favorable review from the listener. Further- more, the decision as to on which medium a musical piece should be released can and does have a major impact on the commercial and critical success of a song and/or artist. Levels seven, six and five of LeBlanc's model deal with what the music must "pass through" in order to make it to the brain for processing. (With each subsequent level, the stimulus gets "closer" to the listener so that a decision to like or dislike can be made.) Level seven is comprised of "physiological enabling conditions," or the body's ability to detect the stimulus in the first place. Level six refers to whether the subject is indeed paying attention to the music stimulus. Level five is what LeBlanc describes as the "mood" of the listener; this level acts more like a filter ‘than does level six, and is capable of interacting with levels six, seven and eight. The fourth level of LeBlanc's model deals with the personal characteristics of the listener. One set of factors for this level deal with personality. This includes in- creased ability to perceive differences in sounds (auditory sensitivity), musical ability, musical training, the listen- er's memory and the base personality of the individual. The second set of factors refer to the listener's social back- ground. This includes gender, ethnic group, socioeconomic status, and age/maturity. These latter factors are very useful in terms of media planning and programming, as they permit the matching of audience segments and programs (Eastman, Head and Klein, 1989). Levels three, two and one are representative of the decision-making process of the listener. Level three is comprised of the processing activities of the listener's brain. Level two consists of repeated listening and sampling activities, and the actual preference decision. Last, level one has two components. The first is acceptance of the music stimulus, along with repetition and heightened attention when the preferred music is heard again. The second is the rejection of the music in question. LeBlanc's model taken as a whole is very helpful in determining musical preference for a given audience segment. The model is comprehensive in nature and easily used as a part of the framework for this study. Levels four and eight will have the most impact on this study, as these contain many production and social variables that would influence who would watch a fine arts television program. MEASUREMENT The accurate measurement of these variables is critical for the success of a fine arts television show. Dixon (1980); Bruner (1979); Fink, Robinson and Dowden (1985); and Diehl, Schnieder and Pentress (1983) all made efforts to develop various methods to measure music preference and/or relate it to music consumption (behavior). Dixon (1980) used a five point semantic differential to measure 17 different musical genres. The response categories, coded 1-5 respec- tively, included "dislike very much," "dislike somewhat," "don't know/no opinion," "like somewhat," and "like very much." Dixon concluded that music dislike appears to have both direction and strength: 2 to 33% of all respondents "dislike(d) very much" one or more of the genres, and 5 to 25% "dislike(d) somewhat" one or more genre. However, no one respondent claimed to dislike all of the genres used in the study. The expected mean for each musical category was a three; however, 14 out of the 17 genres scored means above 3 (indicating favorable reactions from those surveyed). The three genres that did not exceed the expected mean were Spiritual (2.88), Hard Rock (2.85), and Punk music (2.5). Dixon surmised that since virtually every type of music was represented and that the majority of the music types received average scored above the expected mean, the breadth 10 of musical acceptance was rather high for the sample. One interesting facet of the Dixon study was the wide range of "don't know/no opinion" responses: 8 to 54% of all those surveyed. Dixon attributes this to a failure to measure familiarity with each of the categories. Schuessler (1948) had found that the attractiveness of a musical piece varied directly with familiarity; Dixon (by his own admission) did not control for familiarity. The two categories that had the lowest response rate for "DK/NO" were Disco (8.3%) and Hard Rock (8.6%), meaning that most of those surveyed had some sort of opinion about these types of music. Dixon attributes this phenomenon to something he calls a "musical period" that each person has in their lives. In short, a person would hypothetically give a high preference rating to a style of music that was domi- nant during the time when they were most likely to become a heavy music user (late preteens to early adulthood) and that there may be a correlation between preference and age. Fink, Robinson, and Dowden (1985) made use of multi- dimensional scaling to obtain a measure of musical prefer- ence in conjunction with behavior (consumption). Multi- dimensional scaling was selected because MDS would allow for "distance" in measuring among musical genres, and permit the measurement of several differing styles simultaneously. The data was taken from a 1982 study for the National Endowment for the Arts (ARTS '82). The actual data were dichotomous. Subjects were asked if they "like(d)" or 11 "dislike(d)" listening to various kinds of music, and were asked whether they attended any concerts featuring the kinds of music being studied (yes or no). The musical categories were classical, opera, Broadway/musical, jazz, soul/rhythm & blues, big band, country/western, bluegrass, rock, mood/easy listening, folk, barbershop, hymns/gospel, other, and "all types." These responses were cross-tabulated and translated into distances that represented how close the music types were to each other. Dimensions, or axes, were then derived. The first seven dimensions for musical preference accounted for 94% of the variance in the distances; of these, the first two axes account for about 58%. These two are defined as "complexity" and "geographic base" (urban vs. rural) respectively. When combined with the data for attendance, it is shown that attendance (consumption) varies positively with liking (preference). This is a key finding, because it indicates that knowledge of preference may be used to predict consump- tion with some degree of accuracy. Deihl, Schneider, and Pentress (1983) used seven point bipolar adjective scales to measure people's attitudes towards 10 music categories (classical, country/western, jazz, rock, punk rock, big band, soul, folk, beautiful music and opera) deemed by the authors to be the "most important" to explore. The data were drawn from a larger study regard- ing media style, orientation towards new technologies, and subject background characteristics. The survey itself was 12 conducted by telephone during various dayparts by undergrad- uate and graduate students enrolled in a survey methods course. The numbers to call were selected on a purely ran- dom, as opposed to representative, basis, and yielded 607 usable interviews. A factor analysis on the subjects' ratings of the various musical categories was used to determine any under- lying structures that might have existed. The three factors derived from the Varimax rotation were dubbed "traditional/highbrow," "contemporary/progressive" and "middlebrow/traditional" by the authors. The music styles loaded into the factors as follows: "high brow" includes classical & opera, and to a lesser extent, big band, folk, and jazz. Deihl et al. reasoned that these loaded together because of the relative complexity of the styles involved. The "contemporary" factor included rock, punk rock, soul, and jazz. The authors surmise that jazz loaded into this group because of existing innovative elements in the genre. The third factor included (in order) big band, country, beautiful music, and folk music. The authors suspect that folk loaded here as well because it contains a simpler music structure than classical or opera. Deihl et al. put forth that music preferences seem to group themselves along the lines of the aforementioned three factors as opposed to specific categories. For example, those that enjoy classical music will also tend to like opera or jazz more than those persons who dislike classical 13 music. The authors suggest that if the grouping tendency of music preference does fall along these lines, then current radio station practices that lean toward specific music styles -as opposed to broader groupings-may not be best utilizing the potential of the medium. The same may hold true for performing arts television programming. MEDIA USE Fink et a1. established a potential positive relation- ship between music preference and music consumption (1980). In the Fink study, the relationship between preference and concert attendance was mapped out using MDS. Assuming that an individual will indeed make use of a single medium that features the same type of music that they have a preference for, it stands to reason that the same individual would make use of several media that feature their preferred music style. Bruner (1979) conducted a study to determine the degree of association between recording purchases and the use of other musically oriented media. The study was de- signed to be of use to the recording industry; the research questions were aimed at discerning consumer behavior and demographic patterns of recording purchasers. Bruner looked at concert attendance, radio use, recordings use and amount of money invested in stereo equipment (among others), but not music preference across genre. The media habits of 258 college students were recorded and analyzed. The sample was divided into two groups accord- 14 ing to how many recordings (LPs and cassettes) they pur- chased each year. Those that purchased 12 or more recordings per year were considered to be heavy users. Moderate users were defined as those who purchased five to eight recordings per year, and light users were those who purchased four recordings or less per year. Of the men, 16% were considered heavy users, and 39% light users. About 54% of the women sampled were light users, and only 6% were heavy users. Heavy recording users apparently make more use of musically oriented media than do light users. About 45% of heavy users attend concerts four or more times a year, as compared to 10% of light users. Just under 76% of heavy users listen to the radio 15+ hours per week on the average, while only 39% of the light users do the same. About 55% of heavy users listen to their recordings 15 hours a week or more, as opposed to 12% of the light users. Furthermore, 9% of heavy users spend five hours a week or less listening to recordings, contrasted with 57% of light users. The rela- tionship between heavy use and hours spent listening to recordings appears to be linear. A noticeable linear rela- tionship also appeared to exist between amount of use and the amount of money invested in a stereo system: the more records purchased in a year, the more likely the purchaser 'was to own an expensive stereo. Less than 1% of light users put less than $1,000.00 into their stereo systems, and 48% of light users spent $100.00 or less on their present stereo system. 15 Bruner did not analyze two items that are relevant to the research question being studied: music preference across "heavy" and "light" users, and the use of television as a means of musical enjoyment. Statistics compiled by the British Phonographic Industry show that media use does vary with music preference (at least in England); for example, the majority of purchasers of "pop/rock" materials purchase cassettes (British Eggnggrgphig Iggustry rearhggk, 1987). Dixon (1980) analyzed media use along with music pref- erence. The media studied were: concerts (live music, no dancing), radio (weekday use), television (weekday use), recordings (weekday use), time spent actively listening (reported as a percent), total number of recordings pur- chased in the year prior to the study, and then weekend use of radio, television and recordings. In all, there were nine media use variables being measured. Dixon distinguished between weekday and weekend use because he reasoned that respondents would have difficulty in accurately recalling their exact media use in terms of "hours per week" in addi- tion to the type of media they did use. (Dixon also measured music preference across genre, as cited earlier.) The data were collected from 396 undergraduate stu- dents enrolled in 19 classes at the University of North Carolina-Wilmington. Averages were calculated for each of the media use variables. This score is referred to as "involvement." Then, each subject was assigned a "1" if their individual score was above the mean for the medium in 16 question. These standardized scores were then summed to obtain a musical involvement score ranging from 0-9. Partial Pearson correlations, with age and education partialed out, were then run on the media variables. Some of the apparent relationships were of interest. The variables seemed to group themselves into two groups: active and passive. Active items such as concert attendance, weekend use of recordings, recording purchases and active listening correlated highly with the other variables. Dixon surmised that this was because concerts and special program- ming are selective events that merit closer attention than a "background" medium such as radio. Since recording selection and purchase is a very active process, it naturally corre- lated highly with the other active variables. Radio and TV use (passive activities) had fewer significant correlations with the other variables, especially with the active ones. One set of relationships, in particular, was deemed inter- esting: recording purchases, extent of recording use and concert attendance correlated highly with each other. Dixon suggests that this may be an indicator that radio and tele- vision programming may have less of an impact on recording sales than previously thought, and that more study on this particular phenomenon was needed. Dixon then used factor analysis to determine if the media use were prone to active and passive groupings. The analysis did yield three factors. Factor one consisted of concerts, weekday and weekend use of recordings, active 17 listening, and total purchases. Factor two consisted of weekday and weekend radio use. Factor three was comprised of weekday and weekend television use. Dixon considered factors two and three to be subscales of music involvement, labeling factor one "active," factor two "passive radio" and factor three "passive television." The factors explained similar amounts (26, 16 and 16% respectively) of variance, strongly indicating that music involvement was multidimensional in nature. In an effort to test the applicability of the musical involvement scale, partial Pearson correlation coefficients, with age and education partialed out, were calculated among the 17 music genres studied and the total music involvement scale. On the "active" subscale, concerts had the highest number of significant correlations (13) out of all of the variables, followed by weekend recording use (8), active listening (7), weekday recording use (6) and total purchases (4). The radio "passive" subscale had six significant corre- lations, and television eight. What was of particular inter- est was how the different music genres loaded onto the media variables and the active/inactive subscales. Certain styles were significantly related to the active subscale and others the passive. Some styles, such as Hard Rock, had significant correlations with several of the active subscales, but not with the passive. Dixon suggests that such styles have hardcore fans that will go out of their way to find and enjoy that type of music. The radio 18 market also may have had an influence of which styles were related to which scales. Folk was significantly related to concert attendance, but not to radio; however, Folk received almost no airplay in the area being studied. Furthermore, the survey took place in a collegiate town, which may or may not have a wider variety of music formats than the norm. These factors may have had an impact upon the results. The way the music was used by the survey participants also may have influenced the results. Disco and Beautiful Music were negatively associated with the active scales, but were positively related to the passive scales. Dixon sur- mises that these styles are used more as background material or, in Disco's case, for dancing, hence the given relation- ships. The way the music is performed may also have an impact on the various relationships. For example, Classical music features long pieces of music played straight through, while popular music makes use of shorter (3-5 minute) cuts. This may explain why Classical music had significant nega- tive correlations with radio use, while Soul had positive correlational significance with the passive radio subscale in Dixon's study. The shorter cuts in pop lend themselves more to casual listening than do the Classical cuts. Dixon concluded that music-related behavior, i.e. consumption, is multidimensional, given the number of sig- nificant correlations found among the variables. He also concluded that the measurement scales developed for genre taste and involvement were good starting points, but that 19 more research was needed before the phenomenon of music preference could be fully understood. Dixon's treatise does have a few potential limitations. One of these is the actual sample utilized in the study. Dixon drew his sample from the student population of UNC- Wilmington. Bruner (1979) also used a student population in his report. While the results in both cases were statisti- cally valid, they may not be generalizable beyond the con- fines of the respective samples. Since both groups consisted of college students, the variance in age, socioeconomic background, attitudes and opinions are going to be much narrower than for a sample drawn from a city with a popula- tion in excess of 75,000 people. These will affect genre popularity. Another factor to consider is regional populari- ty. For example, Dixieland may be more popular down in the south than in the Midwest. Consequently, a particular form of music may receive a lot of airplay in one region, and none whatsoever in another. If the measures developed by Dixon are valid, the genre preference and music involvement results should change by region. Furthermore, if the locale of the study were to be moved, different genres should load on different involvement variables, e.g. the passive televi- sion subscale mentioned earlier. Another item to consider is that of music trends. As shifts in demographics occur within a given population, music styles and radio formats will undergo shifts in popu- larity. New radio formats such as Classic Rock and Roll or 20 New Wave are designed to attract certain demographic niches that may be ignored within a market. "Disco" enjoyed a brief heyday in the late 1970's, but its popularity has since waned. MUSIC AND TELEVISION An additional factor that merits examination in light of Dixon and Bruner is that of Music Television (MTV). Both Bruner's and Dixon's monographs were published before the debut of MTV. MTV is a basic cable service that airs promo- tional videos that have visual images set to music. It was originally conceived to appeal to those people that regular- ly purchased sound recordings -12-34 year olds -and was designed to be used in a manner similar to the way people listened to the radio, and the visuals are there if the person(s) listening cares to watch. Furthermore, MTV was also intended to be marketed towards the music industry as a potential outlet to promote music, either through the air- play of the music videos themselves, or through commercial advertising of new releases, by either new or already estab- lished artists (Wolfe, 1981). Some videos show the bands actually playing their latest releases, while others feature a visual interpretation(s) of the music and lyrics. Of particular interest are the general use patterns of music videos as an entertainment medium. According to a study by Melton and Galician (1987), radio was the most preferred medium for listening to music for 319 out of 414 college students (77%). Personal home collections were the 21 next most preferred, with 13%. Music videos were the medium of choice for musical enjoyment for about only 10% of the sample. Another listening/viewing habit they discovered was that, in general, more people watched music videos for shorter periods of time. While only 12.8% of the subjects listened to the radio one hour or less per week, 55.3% watched music videos 1 hour or less per week. About one third listened to the radio 1-2 hours per week, but only 18% watched music videos 1-2 hours per week. At the other end of the continuum, about 6% listened to the radio 7 or more hours per week, but none watched videos for that amount of time. The sample group did report watching more music videos on the weekends. Additionally, radio had a stronger impact on music purchases than did music videos. Those that gave radio as their first choice for enjoy- ing music said that "watching music videos limited interpre- tation of song meaning for them." This is an especially interesting statement when examined in light of the cancel- lation of ABC Television's lg anggrrr In Concert was a live performance music program that aired from 1972-75. It featured live performances (with stereo simulcasting) of popular rock groups. While the program drew good reviews from critics for its technical quality and originality, its format was considered to be "boring" after repeated viewing (Wolfe, 1981). Wolfe speculated that this boredom may have been the result of the type of blocking used for the pro- gram. It seems that the producers of In gogggrr used 22 proscenium style (lateral) blocking in an attempt to more accurately duplicate a stage show. MTV, on the other hand, features videos that provide the viewer with a set visual interpretation of the music. However, as Wolfe notes, tele- vision is a more effective visual medium when blocked for depth. The question is, does/did a program like 1n Qgggerr draw a steady repeat audience, or did the audience draw depend upon the featured act? This point is a crucial one for a fine arts producer. The show stayed with one kind of music-rock and roll-but changed bands, having several during the course of one evening. The program had sonic, but not visual variety, and apparently this helped lead to its downfall. Tucker and Hartman (1987) examined the music video viewing habits of 18-35 year olds. They found that one third of those surveyed reported watching music videos regularly. It seems that the primary use of music videos is for enter- tainment purposes and that men are more likely to watch than women. They also discovered that music video use decreased as education increased. The relationship between income and time spent viewing is an interesting one. About 20% of those who earned between SIG-14,000 annually reported watching 5-15 minutes of music videos daily; 15% of the same income bracket reported watch- ing 16-30 minutes daily. Of those who earned $15-19,999 a year, 12% watched for 31-60 minutes. 73% of those who earned $20-24,999 per year reported that they usually did not watch 23 videos at all. Yet, 65% of those who earned $25,000 plus per year agreed with the statement, "I consider myself to be a music video fan." This was a higher percentage than all of the other income groups. Tucker and Hartman offered no explanation as to why this occurred; however, the statistics bear out their hypothesis that viewership decreases with income. Perhaps their agreement did not translate into actual viewing. One of the reasons behind the success of Music Televi- sion was their in-depth research conducted before the debut of the channel in August, 1981. The research was supervised by Marshall Cohen, Senior Vice President of Research and Corporate Services for Music Television Networks. Cohen conducted a study of 500 randomly selected households that had cable television services (Desinoff, 1987). The reason- ing behind the research and the actual results attained by Cohen are most interesting. The initial research was aimed at 12-40 year olds, as Cohen felt that as age increased, interest in rock music would begin to trail off. Each person in the sample was read a prepared statement that described in principle what was to become known as MTV. Then those sampled were asked if they would be interested in receiving such a channel. People were also asked questions about how the as yet hypothetical service should be packaged in terms of station breaks, news and VJ's. Cohen also made inquiries concerning people's attitudes and beliefs (psychographics). 24 Part of the data gathering involved finding out peo- ple's reactions to different artists and musical styles. Cohen was attempting to find out what the music preference picture looked like by searching for audience segments that had similar preferences. Essentially, Cohen was proceeding along the same lines as Bruner, Dixon, and others mentioned previously: measure music preference, then see what type of structures lie underneath it. Once the structures are found, programming may be developed to fit those structures. Howev- er, Cohen performed his testing in a slightly different way: instead of using a list of musical categories and getting people to respond to those, he used the names of the bands, so as to get -in his opinion -the most accurate measure of the potential audience' music preferences. He felt that using style labels only created confusion in the eyes of the respondent and added noise to the research process. Cohen's point is an interesting one. Breakdowns in communication between interviewer and subject could create a significant amount of noise in the communication process, resulting in muddled data. A means of insuring internal validity and reliability will have to be developed in order to insure that any instrument used to measure musical pref- erence is indeed getting the best data possible. The research work by Cohen seems to have paid off. MTV registered in the Nielsen books in April, 1983, -about a year and a half after its debut - with a 1.2 rating (Desinoff, 1987). The financial and ratings success of MTV 25 led to a bevy of imitators on broadcast and cable, including Radio 1990 (USA), Friday Night Videos (NBC) and the Nash- ville Network (TNN) for country fans. The nature of cable television allowed networks to narrowcast towards specific audiences -based on psychographics and demographics -and thus could achieve profitability though small audiences (Sissors and Surmanek, 1982). Also, cable television serv- ices could be interconnected with music preference as per Dixon (1980) in a way that would show up in a statistical analysis, especially given MTV's increases in viewership on weekends (Melton and Galacian, 1987). The same may hold true for TNN or the other cable music video programs. One MTV imitator is not a program nor a national net- work. Rather, it is WVJV-TV ("V-66") a local UHF station located in Marlboro, Massachusetts. WVJV airs music videos 24 hours a day, seven days a week, and, like MTV, attempts to be a radio station that happens to offer visuals. The only difference is that WVJV is a broadcast station. Since WVJV is offered over the air for free and is picked up by several cable operators, the station has the potential to reach 100% of the audience in the New England area (Wein- stein, 1985). WVJV was an (relatively speaking) instant success. It took about six weeks (from the debut of the station) for V-66 to register in the Nielsen books. (Ordinarily it can take anywhere from six months to two years for a UHF station to make it that far.) According to Weinstein, V-66 and MTV 26 are very close in the ratings race for all adults. Part of the secret of V-66's success was their use of research. The station has four request lines, and all requests are kept track of on a computer. The telephone operators would per— form on the spot market research and inquire as to the viewer's demographics and socioeconomic status. All of this information is used to program the station. Given the rapid rise into the ratings books, the method appears to work. Again, the pattern holds true: the request lines measure music preference, then the additional market research infor- mation reveals underlying structure(s) that the managers can use to program the station. POTENTIAL IMPACTS FOR PTV What is really noteworthy about V-66 is that all of this takes place at a local level. It is also a prime exam- ple of the marketing concept at work: find out what the market wants, then deliver it at a profit (Kotler, 1985). While public television stations do not earn a profit per se, their benefit from engaging in market research would be viewers, as measured in ratings. Granted, non-profit enti- ties such as public television will, in general, have fewer resources to perform original research than will most com- mercial stations. Even so, precise use of those resources is crucial, such that PTV may serve the public interest. Some studies have been conducted on the viewing habits of the PTV audience. Lindsay, LeRoy, and Novak (1978) 27 investigated the viewing habits of 225 households in the state of Pennsylvania. PTV households were drawn for the sample in a three to one ratio to non-PTV households, so as to have a large enough sample of PTV households to compare socioeconomic status and other relevant data. A PTV house- hold was described as one that viewed a PTV station for at least two quarter hours during sweep week in November 1976. The findings were most interesting. The viewers were divided into three groups: PTV viewers, non-PTV viewers in PTV households, and non-PTV viewers in non-PTV households. Of the three groups, the PTV viewers watched the most tele- vision, both commercial and PTV. On the average, PTV viewers watched seven more quarter hours of commercial television than non-PTV viewers in non-PTV households (p 5.001). Fur- thermore, PTV viewers watch ten hours of commercial televi- sion for each hour of PTV. Children and adults 54+ use PTV in greater numbers than do teens and adults 18-54. Lindsay, LeRoy, and Novak think that in this instance that PTV view- ers are watching PTV in addition to their commercial view- ing, as opposed to using PTV as a commercial television substitute. They claim that PTV users are not selective users of television in general as much as they are selective users of PTV. So, while PTV may have been intended as a commercial television alternative, it is not being used as such. This point should be kept in mind, as it puts PTV on the same plane in terms of usage as commercial television, even though the two have different financial structures. 28 Again, this makes audience research even more crucial for PTV. TV AS AN ARTS MEDIUM Lou Harris and Associates (1988) is inclined to think that television and the VCR will have an impact on the performing arts in the future. The number of people who reported seeing a "symphony orchestra concert" on television has increased from 36% to 40% since 1984. This is an in- crease of 11%, and closely corresponds to a "loss of 9% in the number of people attending such concerts" in person. Those persons that reported watching modern dance or ballet on television increased 7% since 1984. Viewership of pop/rock concerts on television has declined, however, with attendance. Attendance has dropped 5%, and TV concert view- ing has dropped 8%. Opera has not fared well on television either. The number of persons who have reported seeing an opera on television has declined some 13% since 1984. With the exception of musical comedy, symphony orchestra and dance performances, there has been a 5% decline in the reported viewing of the arts on television. (This includes Oscar caliber (cinematic) movies as well.) According to Harris, of those persons polled, 22% would like to see more pop concerts, 13% would like to see more ballet and dance, 12% would like to see more symphony con- certs, and 7% would like to see more operas. Furthermore, when asked if they would watch more arts programs if 29 offered, 59% indicated they would "probably" view them, and a further 14% said they would "certainly" view them. When asked where they could find this type of programming, 57% said public television, 30% said cable, and 4% said commer- cial television. In short, Harris concludes that, all told, live per- formance attendance is keeping up with television in terms of arts consumption. Harris sees this as a potential criti- cism of television because television does not require as much effort on the part of the consumer as does actually attending an event. Television may have a market opportunity in terms of carrying more arts programming. Harris goes so far as to suggest that commercial television may want to take advantage of this potential niche to boost overall audience numbers. Harris also took a look at VCR use. In 1988, 49 million families reported having VCRs, about 55%. This was a 234% increase from 1984. The median number of rentals during 1988 was 20.5; estimated rental revenue is about $2.6 billion dollars (1.3 million tapes @ $1.99 per rental). VCR use is eating into the television audience; 44% of VCR owners are using their VCR to view tapes instead of watching regularly scheduled programs. In terms of the types of programming desired for VCR use, 77% of VCR owners said it was "certain" or "probable" that they would buy or rent a musical comedy or musical theater program. About 58% are interested in buying or 30 renting "the best pop concerts just after they have taken place;" 45% claimed that they would be "attracted" to pro- grams that featured "recent performances of top symphony orchestras and top performing artists." About 39% are inter- ested in programming that features performances by the "best ballet and modern dance companies in the world" and 23% expressed an interest in operatic performances. Assuming that one quarter of VCR rentals and purchases are from the arts, the potential aggregate number of sales or rentals for pop, classical, musical theater, dance and opera videos would come to about 200 million per year. Harris indicates that this is a conservative estimate. Harris maintains that the implications for these num- bers are that Videocassettes are a potential source of revenue (and thus funding) for the arts, and that VCR owners should be included in the aggregate arts audience. Harris estimates that VCR use could equal up to one third of live concert attendance and arts television viewing, offsetting current declines in these areas. Given these estimates, Harris feels that the VCR is indeed a promising means of delivering the arts to the mass audience. SUMMARY Music preference may be measured using a variety of methods. The music style or the artist serve as the inde- pendent variable. Likert scales, semantic differential scales and multiple dimensional scaling are all valid means 31 of attaining a measure of music style preference for a given sample. Preference groupings may be obtained by using factor analysis: these groupings may help define music preference segments within a population. Patterns exist among music style preferences and pre- sentation media consumption; these may be measured by multi- ple dimensional scaling or correlations. Self-reported recording purchases, concert attendance, music publication use and radio/television use are all legitimate means of measuring music related media consumption. Such patterns have been applied to commercial media programming with great success. The relationship between music style preference and the television medium will be of particular interest as it would be helpful in determining if a relationship exists among music preference, music consumption and fine arts television viewing patterns. Chapter III RESEARCH DESIGN The first step in determining what type of music pro- gramming will succeed in the Lansing market is to find out the music preferences in the market, and then discover if underlying structures exist that may be of use to a fine arts producer. If a relationship among music preference, music consumption, and fine arts television preferences exists at all, then some sort of underlying structure should be revealed that may indicate if one type of music is pre— ferred over another and, if so, if the audience has an interest in watching it on television. The information needed for this analysis falls into two broad categories: music preference (by genre) and media use; that is, which media the population in question uses for musical enjoyment. METHODOLOGY The original intent of this study was to determine the musical viewing preferences, if any, of the Lansing market in an attempt to see if WKAR-TV could better service the area. The best method for ascertaining these needs was the telephone survey because it has a reasonably high response rate and is not too time-consuming or expensive when com- pared to focus group studies or mail surveys. The principal benefit of a telephone survey is the rapidity with which the data is gathered (Kerlinger, 1986). 32 33 SAMPLING The population for this study was the Greater Lansing area, which includes Lansing, East Lansing, Okemos, DeWitt, Charlotte, Mason, and Grand Ledge. The population was divid- ed into two groups for later analysis: members of WKAR-TV and non-members. The population was divided because there may be a difference in music preference, and thus, fine arts television viewing, between members and non-members, even though both groups may (or may not) watch WKAR on a regular basis. A household was considered to be a member of WKAR if they were listed in the membership files as of the February, 1989 pledge drive. A non-member was anyone who was not listed in WKAR files during the same time frame. The WKAR-TV membership list, as kept by Jill Westmore- land-Ayers, was used as the sampling frame for members, and the Ameritech white pages for the Lansing area was used as the sampling frame for non-members. The non-member sample was checked against the member sample for possible overlap- ping, and more names were drawn as necessary to eliminate the overlap. Social science research criteria were taken into ac- count when determining the minimum sample size needed. It was desired to have a 95% or better confidence interval for any statistics calculated, as this is the standard for the social sciences (Blalock, 1979). The problem came in figur- ing out what the minimum sample size for both groups would be so that this standard could be met. According to Blalock, the sample size is a function of the desired confidence 34 the sample size is a function of the desired confidence interval (C.I.), precision level (P.L.), and the standard deviation (S.D.) of the population such that: square root (sample size n)* PL= CI*SD Since the music preference and media use of the popula- tion had not yet been determined, there was no standard deviation for the groups in question; therefore, a standard deviation from an existing study was used in its place for the purposes of estimating the needed minimum sample size. The standard deviation of the sample means in Dixon's (1980) measurement of music preference across genre was used in place of the standard deviation of the population. In this case SD=0.56 (Dixon, 1980). Thus: 0.05* square root (n)= 1.96* (0.56) where 5% is the level of precision, and 1.96 deviations from the mean is the confidence interval. In this instance, in order for the study results to be statistically significant, the sample size n for each group had to be at least 481, making for a total sample size of 962. However, getting that many completed telephone surveys would have been difficult to accomplish, so the precision level was reduced to 10%, and the sample size recalculated: 0.10 * square root (n)= 1.96 * 0.56 This yielded a minimum sample size of 120 persons per group. This made the total sample for both groups together 240. While the larger sample would have had less sampling error present, it would have also presented more difficulties for 35 PROCEDURE Systematic random sampling was used to generate the lists of telephone numbers (and associated names for cross checking between groups) needed for the telephone survey. Each telephone number had an equal chance of being selected from its respective frame. Random sampling was utilized so as to make sure that the samples drawn were representative of their respective populations. Each of the telephone numbers drawn from the Ameritech white pages was cross- checked against the master list of WKAR-TV members for duplication. Two hundred numbers from each frame were drawn to allow for non-participating households. Both the member and the non-member samples were drawn in the same fashion. Starting with page one, fifth entry, of the respective frames, every tenth name and accompanying telephone number was drawn and recorded. When the end of a page was reached, the persons recording the names skipped ahead 15 pages and repeated the process until the first 200 names were drawn; it was expected that this process would have to be repeated to gather enough telephone numbers for the sample. The WKAR member sample was drawn first, then the non-member sample, to allow for the prevention of duplicate names between lists. INSTRUMENT A 71-item questionnaire (see Appendix A) was adminis- tered in an attempt to determine the music preference, music 36 INSTRUMENT A 71-item questionnaire (see Appendix A) was adminis- tered in an attempt to determine the music preference, music consumption and performing arts television viewing habits of the drawn sample. With the exception of qualifier questions, e.g. "Do you own a stereo system," all of the questions asked were designed to yield interval level data or higher. Music preference was measured using two different methods: aided and unaided recall. Respondents were asked to give their reactions to 18 musical categories: Heavy Metal, Soft Rock, Jazz, Jazz Fusion, Progressive Rock, Country Western, Classical, Electronic/ New Age, Folk, Rhythm and Blues, Dance/ Disco, Reggae, Spiritual, Beautiful Music, Big Band and Rap. All of these category variables taken together make up the construct musical preference. A ten point seman- tic differential scale was used to gauge the respondent's reaction to the differing genres. Respondents were asked to rate each style from one to ten, with one meaning that the respondent did not like the style at all and ten meaning that they liked it a lot. (Each respondent was also asked to indicate if they were not familiar with the style in ques- tion. In this way it was possible to eliminate "noise" from the data gathering process.) This ten point scale allows for more flexibility in terms of data analysis, since it is interval in nature. The music categories used in this study are not iden- tical to those used by Dixon. These changes reflect the 37 changes in the musical spectrum since 1979, e.g. emergence of rap as a musical form. Unaided recall of musical style was also used to meas- ure musical preference; this time the respondent was allowed to define the variable used to help build the construct. The subject was asked to name their three favorite bands or artists. These were categorized according to where the artist(s) in question were found in music publication sales charts, e.g. "Billboard," and later totaled. Finally, the respondent was also asked if they had a favorite style of music not mentioned in the survey. These responses were later tallied and totaled. MUSIC CONSUMPTION‘ Music consumption was measured in a variety of ways as well. This construct is made up of activities in which people engaged to enjoy music. These consisted of concert attendance, radio station ratings, stereo use, television use and VCR use. These activities were measured in several different ways. Concert attendance was recorded for each individual respondent. They were asked to name up to five concerts (if any) that they had attended within the last calendar year. These were then cross-tabulated across music genre, again using sources such as "Billboard" for categorization guid- ance. Radio use was measured in terms of time spent using the 38 These were then cross-tabulated across music genre, again using sources such as "Billboard" for categorization guid- ance. Radio use was measured in terms of time spent using the medium. Respondents were asked how many days per week and hours per day they spent listening to the radio. Radio station ratings for the Lansing market were also examined from the May, 1989 sweeps period, to see which music format(s) were the most popular across age and demographic categories. Television use was examined as well. Again, respondents were asked how many days per week and hours per day they watched television. Furthermore, respondents were asked on which day/evening they watched the most television. Respondents were also asked how many music-oriented programs they rented in the last year. Home stereo use was also measured. Respondents were asked if they actually owned (or had access to) a stereo system. The amount of time (in hours) spent listening to prerecorded (i.e. non-radio) music on their stereo during a typical week was recorded. Respondents were asked to report on how many prerecorded LP's, cassettes, and compact discs they owned, and how many total units for all three media they purchased in the last year. A five point Likert scale was used to measure the respondents' overall medium of preference for musical enjoy- ment. Each subject was read a series of statements that favored one medium over another as a means of enjoying 39 level data during analysis, as the response structure strongly resembled a semantic differential scale. The media compared were live concerts, radio concert broadcasts, televised concerts, concerts/performances on videocassette, recordings and performance vs. "concept" videos. Once the series of statements was completed, an ordered list of media preference for enjoying music could be determined both for the individual subject and the sample. Furthermore, the data gathered are interval level, allowing for correlations to be run across music genre. TELEVISION VIEWING PATTERNS Performing arts television viewing habits were also operationalized and measured. The respondents were read a list of music-oriented television programs that were cur- rently available in the Lansing market on broadcast and/or cable. They were then asked to indicate how often (if at all) they watched these programs. A five point scale was used, the points being "often," "sometimes," "occasionally," "rarely" and "never," with "don't know" as a last option. Since not all of the programs listed were available on broadcast television, a screening question on cable televi- sion hookup was administered to "factor out" these programs for the appropriate respondents. Additionally, each person polled was asked to name any other favorite program not mentioned in the survey. These were later coded and tallied with the others. The Lansing market ratings for each of the programs on the list were recorded from the appropriate 40 for the appropriate respondents. Additionally, each person polled was asked to name any other favorite program not mentioned in the survey. These were later coded and tallied with the others. The Lansing market ratings for each of the programs on the list were recorded from the appropriate Nielsen sweeps periods as well. (Multiple sweeps periods were necessary because some of the programs chosen did not air on a weekly basis, e.g. Great Performances.) Additional ratings were found for other musically oriented programs not covered in the survey so as to supplement the information on the programs in the survey. The ratings acted as a check on the responses of the sample. Interval level data was sought on the respondents' socioeconomic background. This information was to be used for grouping purposes to determine if any demographic char— acteristics could be used to predict music preference or media use for music enjoyment. PRETEST The survey instrument was pretested on ten subjects. These persons were all residents in the Lansing area. No real problems occurred in the pretest save one. The style of music known as "Beautiful Music" or "Easy Listening" was referred to in the original draft as "Beautiful Music." None of the persons queried understood what "Beautiful Music" referred to: however, when subsequently asked for a reaction to "Easy Listening" the subject recognized this as a musical 41 cable television or a videocassette recorder, which made the survey proceed faster than in the actual data gathering sessions. (Skips were designed into the survey so those that did not own VCRs or subscribe to CATV did not have to answer questions that were not applicable to them.) The survey was then placed into a Survey System (ver- sion 4.0) file so that computer assisted telephone inter- viewing could be utilized. Each telephone operator was able to directly input data for the survey onto a floppy disk while speaking with the respondent. The computer managed the skips for the operator, ensuring proper question sequence. Each night the gathered data were added to a master data file. The phone interviews were conducted out of the Depart- ment of Telecommunication's Integrated Telecommunication Services (ITS) laboratory over a two week period. Volunteers from TC 300 and MSU Telecasters served as phone operators (along with the author) for the data collection. Each tele- phone operator was given a one page instruction sheet that detailed how the survey was to be conducted. They were also given an oportunity to review the survey and ask questions before calling out. The author supervised the data gathering for the duration of the collection period. After the collec- tion period was over, the file was then converted to the American Standard Code for Information Interchange (ASCII) for analysis in the Statistical Package for the Social Sciences (SPSS). 42 for the duration of the collection period. After the collec- tion period was over, the file was then converted to the American Standard Code for Information Interchange (ASCII) for analysis in the Statistical Package for the Social Sciences (SPSS). Chapter IV RESULTS AND ANALYSIS DEMOGRAPHICS Two hundred and twenty two surveys were completed out of 713 phone calls, resulting in a response rate of 31.14%. It was necessary to obtain more telephone numbers from the sampling frames to gather enough data. Of the completed surveys, five were thrown out because of several bad codings within the individual cases, rendering them suspect for the purposes of this study. This left a final sample size of 217. While this is somewhat short of the goal of 240, the loss of precision was equal to the estimated standard devia- tion divided by the square root of 217, or 0.038. Therefore, the final number of valid cases should be statistically significant. It was decided not to press on for all 240 completed surveys because of a shortage of time and re- sources. The data were then analyzed in the ITS Laboratory using SPSS PC+, version 3.0. Table 1 shows the sample breakdown according to stand— ard demographic groupings (gender, age, income, education, occupation and race) as well as cable television subscrip- tions and VCR ownership. These are all cross tabulated with WKAR-TV membership. (The final n for each cross tabulation does not equal 217 because some respondents elected not to complete the demographic section of the questionnaire.) The sample skewed slightly towards females with 42.5% 43 44 Table 1 Sample Parameters gender Tetel E__B Memeer Hen Member Male 90 (42.5%) 33 (35.9%) 57 (47.5%) Female 122.151I531 52 1511111 £1 lillfifil Total 212 (100%) 92 120 Age 18-34 82 (38.5%) 23 (24.7%) 59 (49.2%) 35-49 65 (30.5%) 34 (36.6%) 31 (25.8%) 50-65 40 (18.8%) 21 (22.6%) 19 (15.8%) 55: 25 1124211 IQ 1151111 11 124231 Total 213 (100%) 93 120 Ineeme $0-10K 16 (8.1%) 6 (7.1%) 10 (8.9%) 10-20K 26 (13.2%) 6 (7.1%) 20 (17.9%) 20-30K 37 (18.8%) 14 (16.5%) 23 (20.5%) 30-40K 32 (16.2%) 11 (12.9%) 21 (18.8%) 40-50K 28 (14.2%) 14 (16.5%) 14 (12.5%) 505+ pg (22.4%) 14 (40%) 25 (21.4%) Total 197 (100%) 85 112 Education Total WKAR Member Non Member H.S. 28 (13.5%) 13 (14.3%) 15 (12.9%) Some Call. 66 (31.9%) 22 (24.2%) 44 (37.9%) College 52 (25.1%) 24 (26.4%) 28 (24.1%) Post. c611. 25 (12.1%) 12 (13.2%) 13 (11.2%) ASE; DEQEQE éé Illsifil. 29 12231 15 1121851 Total 207 (100%) 91 116 Table 1 cont'd. chuparipn Part Hrly. Full Hrly. Mgt./Prof. Other Sal. Unemployed Retired Technical Stndent 20 (9.7%) 29 (14%) 56 (27.7%) 31 (15%) 12 (5.8%) 35 (16.9%) 8 (3.9%) 19.111131 Total 207 (100%) EQCQ White Afro-Amer. Asian-Amer. 203 (97.1%) 5 (2.4%) l (0.5%) Total 209 (100%) cnry Subs. Yes N9 Total ygg Owned Yes He Total Iotnl 133 (63.3%) 11 1361131 210 162 (76.4%) 52 1221:30. 212 45 10 (11.1%) 8 (8.9%) 28 (31.1%) 13 (14.4%) 5 (5.6%) 20 (22.2%) 3 (3.3%) g (3.3%) 90 89 (97.8%) 2 (2.2%) Q 1231 91 WKAR Member 62 (67.4%) ;n (32.6%) 92 76 (81.7%) ;1 (18.3%) 10 (7.9%) 21 (16.5%) 28 (22%) 18 (14.2%) 7 (5.5%) 15 (11.8%) 5 (3.9%) I; (1o.2§) 127 114 (96.6%) 0 (0%) l (0.8%) 115 Mpn Member 71 (60.2%) A1 (32.8%) 118 86 (72.3%) 33 (27.7%) 119 46 of the sample being men, and 57.5% women; 43% of the sample contributed financially to WKAR-TV. The average age of the sample was approximately 43 years: the reported ages were also demographically grouped. Approximately 38.5% of the sample were between the ages of 18 and 34, 30.5% between 35 and 49, 18.8% between 50 and 65, and 12.2% were older than 65 years of age. A Chi square test indicated that a signifi- cant relationship exists between the age of the respondent and whether they are a member of WKAR-TV (p<.01). An initial glance at Table 1 suggests that perhaps a significant number of 18-34 year olds from this sample may not be members of WKAR-TV. In terms of income, 29.4% reported a total household income of $50,000 per year or more. This was the largest group in the sample. Of those persons surveyed, 8.1% report- ed earning less than $10,000 per year, 13.2% between $10- 20,000 per year, 18.8% between $20,000-30,000, 16.2% between $30-40,000 per year, and 14.2% earned from $40-50,000 per year. The majority of the sample (86.5%) have had some part of a college education or more. Specifically, 13.5% have attended or completed high school, 31.9% have attended some college, 25.1% are college graduates, 12.1% have engaged in post-graduate studies, and 17.4% have completed an advanced degree. The majority of the sample classified their occupation as being management/professional (27.1%). Interestingly 47 enough, the proportion of management/professionals was the same for both WKAR members and non-members (13%). Approxi- mately 9.7% of the sample reported that they worked part time, and 14% reported that they worked full time on an hourly basis, while 15% said they worked in a non-management salaried position. Additionally: 5.8% reported being unem- ployed; 16.9% were retired and 7.7% said they were students. A small portion (3.9%) classified themselves as having a technical job, e.g. equipment repair. Most of the sample said they were Caucasian (97.1%), and 2.4% classified themselves as being Afro-Americans. One respondent (.5%) said they were an Asian-Pacific American. Nearly two thirds (63%) of those surveyed subscribed to cable television, and three fourths reported owning at least one videocassette recorder. Both of these are higher than the 1989 national estimates for CATV subscribers and VCR ownership (56% and 75% respectively) (McVoy, 1989). The Phi coefficient between CATV and VCRs was 0.2475 where p<.001, indicating a possible positive relationship. This would appear to fit in with national VCR and CATV trends. The sample's average time spent watching television was 2.70 hours per day. The average number of hours spent lis- tening to the radio each day was 3.5. The average number of records owned per person in the sample was 98.97, and the average number of cassette tapes owned per person was 31.58. No significant differences existed between WKAR members and non members in terms of media use or ownership. 48 MUSIC PREFERENCE Table 2 lists the music style preference ratings for those styles that were tested. Each respondent was asked to rate the styles from 1 to 10, with one being the worst, and 10 being the best. Classical music received the highest rating, with a mean of 6.38 on the 10 point scale. Easy Listening was next with a mean of 6.16, followed by Big Band with 5.97, then Rhythm and Blues (5.95). All the rest of the styles finished with means below 5.5, - the expected mean - with Heavy Metal (2.06), Rap (3.11), and Spiritual (4.15) rounding out the bottom of the list. Respondents were also asked during the course of the survey if they liked any other musical styles that did not appear in the questionnaire. Table 3 shows the most frequent styles cited. Out of 16 responses, Bluegrass was mentioned 6 times, or 37.5%. Broadway tunes and Polka were each men- tioned twice (12.5%). Since only nine different styles were mentioned, this may be taken as an indication that the styles list was somewhat comprehensive. The respondents were also asked to name up to three of their favorite artists. It is reasonable to assume that the artists named would correspond somewhat with the style preferences mentioned above. Table 4 shows the 10 artists most frequently mentioned. The most frequently mentioned artists were the Beatles with 16 mentions. Kenny Rogers was cited 14 times, and Barbara Streisand 13 times. Of the top ten, three were Rock and Roll artists, four were Country 49 Inble a Sample Muslg sryle Preference Means Style Mean—1 5M; .11 Heavy Metal 2.06 2.13 204 Soft Rock 5.24 2.84 209 Jazz 5.47 2.70 210 Jazz Fusion 4.59 3.09 115 Progressive Rock 4.45 2.90 168 Country Western 4.50 3.00 210 Classical 6.38 2.91 209 Electronic/New Age 4.23 3.02 169 Folk 5.10 2.59 207 Rhythm & Blues 5.95 2.53 211 Dance/Disco 4.29 2.62 206 Reggae 4.24 2.62 168 Spiritual 4.15 3.04 209 Easy Listening 6.16 2.80 207 Big Band 5.97 2.84 207 Rap 3.11 2.67 184 1 Scale: 1=don't like at all, 10= like a lot 50 fable ; Addlrionnl Musig Ereferenges Style Bluegrass Broadway Showtunes Polka Opera Folk Rock Swing Cajun Ragtime Ethnic/Folk Munber pr Mentions 51 Inple A reverite Musieians A Artists Artist Mumber pr Mentions l) The Beatles 16 2) Kenny Rogers 14 3) Barbara Streisand 13 4) Alabama 12 5) Oak Ridge Boys 9 6) Neil Diamond 7 7) Mozart 7 8) Randy Travis 7 9) The Judds 6 10) The Rolling Stones 6 Five Groups with 5 Nine Groups with 4 12 Groups with 3 29 Groups with 2 172 Groups with 1 52 Western artists, one could be classified as Easy Listening or Soft Rock (Streisand) and one as Classical (Mozart). Kenny Rogers has achieved crossover status between Country and Soft Rock/Easy Listening. In all, 237 different artists were mentioned. It is interesting to note that even though Classical music had the highest average preference, only one of the top ten artists with the most mentions was actually a Classical composer/performer. One possible explanation is that the artist in question is deceased. Another possibility are the potential marketing differences between popular and Classical music in the United States. Another possibility is the influence of commercial radio and its role in promoting recorded music. According to Eastman, Head and Klein (1989) rock and roll music is more widely listened to on the radio than any other style. DEMOGRAPHIC INFLUENCES 0N PREFERENCE Preference and Age In order to determine if their is a relationship among music preference, consumption and fine arts television viewing, the music preference spectrum, if you will, needs to be refined in a way such that market segments can be identified. Segmentation may make finding a relationship among the constructs easier. The music style preferences of the sample appear to be strongly influenced by age. Table 5 Shows the music style means by age group. Of the 16 differ- ent styles tested, only two (Rhythm & Blues and Dance/Disco) 53 Table 5 Musle firyle Meens py Age Gronp stile l§_3__- 41 15:32 51-55 .6_6_~: 5.19112 Metal 3.00 1.85 1.05 1.45 0.00*** Softrock 6.20 6.35 3.93 1.88 0.00*** Jazz 5.97 5.73 4.80 4.75 0.05* Fusion 5.58 5.09 2.39 3.00 0.00*** Prog.Rock 6.17 5.02 1.97 1.79 0.00*** Country 3.78 4.16 5.93 5.14 0.00** Classical 5.21 6.60 7.76 7.00 0.00*** New Age 4.79 4.96 3.13 2.25 o.00*** Folk 4.16 5.77 5.58 5.26 0.00** R&B 5.79 6.29 5.78 6.04 0.63 Dance 4.11 4.05 5.05 4.29 0.23 Reggae 4.85 4.08 3.28 3.76 0.05* Spiritual 3.44 3.50 5.25 5.68 0.00*** Easy Lis. 5.21 6.12 7.92 6.36 0.00*** Big Band 4.75 5.66 7.55 7.71 0.00*** Rap 4.15 2.63 1.83 2.60 0.00*** 1 Scale: 1= don't like at all, 10= like a lot. 3 Analysis of variance, df=3 * = p 5.05 ** = p 5,01 *** = p 5.001 54 were npr influenced by age category. This is not really much of a surprise, given radio's performance as a demographical- ly specific medium (Sissors and Surmanek, 1986 and Eastman, Head and Klein, 1989). Soft Rock, Jazz, Jazz Fusion, and Progressive Rock all had means above 5.0 with those persons below 50 years of age, and means below 5.0 for those 50 years or older. Heavy Metal, Electronic/New Age, Reggae and Rap appealed more to those persons under 50 as well. Country Western and Spiritual both had means greater than 5.0 with those persons over 50 years of age. Folk and Big Band mostly appealed to those persons over 34 years of age. Classical and Easy Listening both had means greater than 5.0 across all age groups, but the group with the lowest means was the 18-34 year olds. It would seem that these results parallel those seen in Dixon (1980); he found that music style preference and age were related as well. Preference and Income Table 6 shows music style means across income catego- ries. Out of the sixteen different styles, there were only three statistically significant differences among the means. Generally speaking, the mean liking of Soft Rock increased with the respondent's level of income. Those persons that earned less than $10,000 per year gave Soft Rock a mean score of 4.19, and this rating gradually increased with income; those earning $50,000 or more per year gave it a mean of 6.07. The only exception to this is a slight drop in the $20-30,000 group. This trend appears to be somewhat 55 Table 6 Music Style Means by Income Level 1.90 5.68 5.71 5.60 5.22 5.91 3.91 6.48 5.78 2.68 Style so-lox 810-20K 820-30K 830-40K Heavy Metal 1.38 3.35 2.17 Soft Rock 4.19 4.56 4.29 Jazz 4.38 5.73 5.89 Jazz Fusion 3.29 5.31 4.04 Progressive Rock 4.27 4.43 3.88 Country Western 5.56 4.73 4.38 Classical 7.69 5.58 6.46 Electronic/ New Age 2.64 4.19 3.90 Folk 5.06 5.46 4.58 Rhythm and Blues 5.38 6.46 6.38 Dance/ Disco 3.38 5.04 4.47 Reggae 4.27 5.33 4.13 Spiritual 5.38 4.73 5.03 Easy Listening 6.00 5.80 6.11 Big Band 5.50 6.50 6.43 Rap 3.38 3.17 2.81 Scale: l-don't like at all, 10- like a lot Analysis of variance: df-S *** p<.001 9* p<.01 * p<.05 840-50K 5.50 4.04 3.33 6.86 6.30 2.71 $50K+ 4.08 3.02 5.95 5.72 3.40 .02* .01** .45 .47 .48 .72 .22 .18 .75 .55 .26 .60 ,ooae: .71 .68 .78 56 reinforced by a positive Pearson's correlation coefficient of 0.2572 between Soft Rock and income (p=.001). Conversely, the mean liking of Spiritual music de- creased as income increased. Those persons earning less than $10,000 per year gave Spiritual a mean rating of 5.38. The mean score gradually decreased as income increased; those earning in excess of $50,000 per year gave Spiritual a mean rating of 3.02. This observation is also somewhat reinforced by a Pearson's correlation coefficient of -.2781 between Spiritual and income (p=.001). The last statistically significant difference between means of the income groups was that Heavy Metal was signifi- cantly preferred more by those in the $10-20,000 per year income category (3.35) as compared to the other groups. The correlation between Heavy Metal and income was not statisti- cally significant, so there does not appear to be any type of trend between these two variables. Preference and Occupation When looking at the differences among music style means across occupation categories, there are several sta- tistically significant differences (see Table 7). However, these primarily involve three groups: the unemployed, re- tired people and students. The unemployed as a group showed a lligher preference for Country Western (6.58) and Spiritual (7-w42) music. Conversely, this same group had lesser prefer- ences for Heavy Metal (1.0), Jazz Fusion (1.67) , and Rap dogma a zen—n «nape zones 6% Donovan—o: LS ankna pawn able menu dale zan.\onon. Onsan may. cooIoHOKoo ”cannon dense—new mneaean :aack zen-u ”.mo ~.aa ~.ow p.a. p.oo p.wo ~.qm ~.qu noun non» u.qm o.aa m.~q a.wa a.~u u..~ a.pu v.90 «can u.ou o.oq a.oo ..ou ..oc u.~o a.uu m.o~ noun weapon ~.ao u..w u..~ s.uq ~.mq n.44 u.qu n.4u onooneaupc¢ mb.N mb.m mw.v mm.~ vo.m mm.v ov.m vm.m mo.v v~.b Nm.m NH.m o~.m o¢.m Nm.v vo.~ .HHOU umom mo.va a 66.66 .6 Hoo.va «ea vnmo ”oucmauu> mo nunaumc4 666 m 6666 "66 .666 66 6666 6.666n6 “66606 66.6 66.6 66.6 666 66.6 66.6 66.6 6666 666 66.6 66.6 66.6 666666666 6666 66.6 66.6 66.6 ”66666666 66.6 66.6 66.6 666666 66.6 66.6 66.6 oumaoxmuemo 66.6 66.6 66.6 66666 6 960666 66.6 66.6 66.6 6666 66.6 66.6 66.6 664 >62 66.6 66.6 66.6 660666666 66.6 66.6 66.6 suuesoo 66.6 66.6 66.6 6066 .6666 66.6 66.6 66.6 666666 6666 66.6 66.6 66.6 nnm6 66.6 66.6 66.6 6066 6666 66.6 66.6 66.6 6666: 6>mu= 6666660 .606 :66: 66666 pmu0.HHoo Ho>oq coHumuoom he acme: uflxum Unuoz m manna Table 9 Music Style Means by Gender etxle Meleel Females 51922 Heavy Metal 2.61 1.66 0.00** Soft Rock 5.51 5.07 0.27 Jazz 5.70 5.33 0.32 Jazz Fusion 5.05 4.18 0.13 Prog. Rock 4.93 4.07 0.05* Country 4.58 4.43 0.72 Classic 5.95 6.69 0.07 New Age 4.12 4.33 0.65 Folk 5.00 5.17 0.64 Rhythm & Blues 5.94 5.96 0.64 Dance/Disco 3.75 4.68 0.01* Reggae 3.99 4.49 0.21 Spiritual 3.89 4.32 0.31 Easy Listening 5.76 6.46 0.08 Big Band 5.44 6.36 0.02* Rap 3.07 3.14 0.87 1 Scale 1= don't like at all, 10= like a lot. 3 Analysis of variance: df=1 * = p50.05 **= p50.01 62 women (2.61 and 1.66 respectively). On the other hand, women seemed to have stronger preferences for Dance/Disco (4.68) and Big Band (6.36) than did men. It would seem that gender is not as large a factor in predicting music preference as is age. Preference and WKAR Membership Some statistically significant differences between WKAR-TV members and non-members appeared within the data set as well. Table 10 shows that WKAR members had somewhat stronger preferences for Classical (7.01) and Folk (5.56) music than did non-members. On the other hand, non-members showed distinctly stronger preferences for Heavy Metal (2.47), Progressive Rock (4.98) and Reggae (4.74) than members. While these are interesting differences, WKAR membership is still not as strong a predictor for music preference as is age. The WKAR influence might not have been that strong. Classical music had means above five for all of the demographic groupings indicated, including those persons who were not members of WKAR-TV. The difference between the member and non-members was apparently that of liking vs. strongly liking, as opposed to like vs. dislike. 63 Ieple lg Musle Style Means py WKAR Mempership firyle Members Men Members gigrz Heavy Metal 1.53 2.47 .00** Soft Rock 5.11 5.34 .56 Jazz 5.30 5.60 .43 Jazz Fusion 4.04 5.00 .10 Prog. 3.66 4.98 .00** Country 4.30 4.66 .39 Classical 7.01 5.88 .01** New Age 4.02 4.38 .45 Folk 5.56 4.66 .01* R & B 5.75 6.10 .32 Dance/Disco 4.12 4.41 .43 Reggae 3.56 4.74 .00** Spiritual 4.18 4.14 .92 Easy Listening 6.29 6.07 .58 Big Band 6.21 5.78 .29 Rap 2.72 3.40 .09 1 1=don't like at all, 10=like a lot 3 Analysis of variance: df=1 ** p5.01 * pg.05 64 Overall Preference Structure A principle component factor analysis was run on the 16 different musical style preference ratings to determine if any underlying structures existed that could shed light on the music preferences of the sample. The computer was not given a specific number of factors to produce. Varimax rotation was utilized so as to achieve a clearer picture as to the overall structure of the resulting analysis output. Table 11 shows a reproduction of the SPSS output for the rotated factor matrix. Five factors were found among the 16 different styles. Jazz Fusion, Jazz, Electronic/New Age, and Rhythm & Blues all loaded properly onto Factor 1, with factor loadings in excess of 0.50. Progressive Rock and Reggae also loaded onto Factor 1 in excess of 0.50. However, Progressive Rock also loaded onto Factor 2 (.44996), and Reggae onto Factor 4 (.55601), making these variables sus- pect for this particular analysis. Factor 2 is something of a puzzle. Classical, Heavy Metal, Soft Rock, Big Band and Folk all loaded onto Factor 2 properly. However, Soft Rock, Big Band and Folk also loaded elsewhere. Soft Rock loaded onto Factor 3 (.44140) as did Big Band (.53799) and Folk loaded onto Factor 5 (.49534), making them all suspect variables. This leaves Classical and Heavy Metal as the only two remaining variables on Factor 2. What is of particular interest is that the loading for Heavy Metal is negative (as was the loading for Soft Rock). This would appear to imply that there is an inverse relationship O: 01 Table H Factor Analysis FACTOR /VARIABLES metal to rap /missing pairwise /analysis metal to rap /print rotation /rotation varimax /format sort. 31800 ( This FACTOR analysis requires 31.1K) BYTES of memory Page 14 SPSS/PC+ F A C T O R A N A L Y S I S Analysis Number 1 Pairwise deletion of cases with missing values Extraction 1 for Analysis 1, Principal-Components Analysis (PC) PC Extracted 5 factors. 1/22/90 Varimax Rotation l, Extraction 1, Analysis 1 - Kaiser Normalization. Varimax converged in 8 iterations. Rotated Factor Matrix: FACTOR l FACTOR 2 FACTOR 3 FACTOR 4 FACTOR 5 FUSION .88631 .00123 .02564 .03520 -.07360 JAZZ .81740 .11812 .15166 -.03750 -.01056 PROGROCK .67909 -.44996 -.16158 .08633 -.02093 NEWAGE .61990 .04765 -.05875 .19050 -.08978 REGGAE .56667 .17320 -.27342 .55601 .06681 BLUES .53152 .32472 .10773 .21712 .35466 Page 15 SPSS/PC+ 1/22/90 - - - - F A C T O R A N A L Y S I S - - - - FACTOR 1 FACTOR 2 FACTOR 3 FACTOR FACTOR 5 CLASSIC .23417 .77729 .05314 .00689 -.05710 METAL .11440 -.59110 -.20854 .05078 -.07017 SOFTROCK .36997 -.54486 .44140 -.07556 .05360 FOLK .32877 .54383 -.05754 -.07034 .49534 EASYLIS -.15110 .13500 .86768 .10913 .08087 BIGBAND .20900 .53799 .58972 .12053 .01698 RAP .22474 -.23865 -.00400 .75575 -.l7055 DANCE .16328 .00020 .40185 .54172 .18626 SPIRIT -.29164 .25578 .26445 .51578 .19003 COUNTRY -.19329 -.05405 .12341 .04543 .89789 Page 16 SPSS/PC+ 1/22/90 - - - - F A C T 0 R A N A L y s I s - - - - Factor Transformation Matrix: FACTOR 1 FACTOR 2 FACTOR 3 FACTOR 4 FACTOR FACTOR 1 .92038 .18993 .10342 .31814 .07017 FACTOR 2 -.27375 .75091 .47953 .11176 .34461 FACTOR 3 -.l4066 -.57552 .59267 .52540 .14723 66 (at least within the confines of this particular sample) between the preferences for Classical music and Heavy Metal. The correlation between the variables Heavy Metal and Clas- sical music is -.3754 (p<.001), indicating that those per- sons that rated Classical highly rated Heavy Metal somewhat lower, and vice versa. Factor 3 had only one variable (Easy Listening) load onto it properly (.86768). Factor 4 had another interesting set of groupings. Rap, Dance/Disco and Spiritual all loaded properly with no other significant (within 0.25) loadings elsewhere in the factor matrix. The sample preference means for all three styles were among the lowest of those styles tested; this grouping may merely be a reflection of a lack of popularity for these three variables with this particular sample as compared to the others. Country Western (.89789) was the only variable that loaded onto Factor 5 properly. With this particular factor analysis, only 65.2% of the variance is accounted for by the five factors. The main problem appears to be the growing diversity of rock and roll as a musical style, which is seen in the variety of radio formats based around rock music. It appears that Heavy Metal, Soft Rock, and Progressive Rock have appealed to distinct demographic groups within the sample, so it is possible to see how musical style could be used to attract a particular audience in the mass media. The music style is really only one variable used in the development of a radio format: news, on-air personalities 67 and other variables are included as well. In the end, a radio programmer has pulled together an entire package that appeals to a particular listener, i.e. the format. This format is what ultimately attracts the listener, not the music (Eastman, Head and Klein, 1989). At some point the listener may potentially develop a confusion between the music style and the radio format. (There were a few in- stances of persons listing "oldies" or "classic rock" as their favorite music style. These are also names of existing radio formats in the Lansing area. It may be that the radio formats and musical styles were indeed mixed up by the sample. This may explain why certain styles loaded onto multiple factors.) Another problem that might cause multiple loadings on the different factors is the respondent's unfamiliarity with the particular musical styles. This could be influenced by almost anything. Another look at LeBlanc's (1982) interac- tive theory of music preferenCe shows that certain circum- stances in a person's background may influence what types of music to which they have actually been exposed. While this factor analysis is somewhat limited, it does show that certain groupings do exist. Jazz, Jazz Fusion, Rhythm & Blues and Electronic/New Age all seemed to group together. The latter three all have elements of traditional jazz in them (e.g. rhythms, chord structures, etc.) so this grouping does actually make some sense. Also, it would appear that Easy Listening is a stand alone style, and, 68 making allowances for the multiple loadings in the factors, Country Western, Classical, and Heavy Metal are all stand alone styles as well. These results parallel Deihl, et al. (1983) insofar as that definite groupings could be culled out of the music style preference spectrum. Most likely, the reason the factors from this study are not identical to those seen in Deihl et al. (1983) is because different populations from different regions were being studied. Interestingly enough, Classical music loaded highly onto one factor in both studies. MUSIC CONSUMPTION The next question to examine is whether this sample's music preferences are paralleled by their music consumption habits. An examination of the Lansing area radio station ratings, local concert attendance, and fine arts television viewing was undertaken to determine if this is actually the case. Those persons responding to the questionnaire were asked to name the three most recent concerts that they had attended. Table 12 shows the 12 concerts mentioned most often. The top three most often mentioned were concerts performed by the Lansing Symphony Orchestra (13), the MSU School of Music (9), concerts sponsored by the Wharton Center (7) and the Smokey Robinson concert at the Michigan Festival in August 1989 (7). Of the 12 concerts most fre- quently mentioned, five were at the Michigan Festival. In 69 maple lg chal Concerts Artended o ce 8 1) Lansing Symphony Orchestra 13 2) MSU School of Music 9 3) Wharton Series 7 4) Smokey Robinson 7 5) Bangles 4 6) New Kids on the Block 4 7) Alabama 3 8) Chenille Sisters 3 9) Juilliard String Quartet 3 10) The Nylons 3 11) Spyro Gyra 3 12) The Who 3 11 with 2 72 with 1 * Denotes Michigan Festival Event 70 all, 95 different concerts were cited by the 217 respond- ents. (The Michigan Festival was noted as the concert spon- sor on 30 different occasions.) Correlation coefficients were calculated between each of the music styles and the number of concerts each respondent had attended in the last year. The r between Classical music and the number of concerts attended was .2870 (p<.001), indicating a possible positive relation- ship between the two variables. It would appear that the more a respondent liked Classical music, the more likely the respondent was to attend concerts. This parallels findings by Fink, Robinson, and Dowden (1985). Other significant correlations occurred between concerts and Soft Rock (-.1716, p<.01), Progressive Rock (-.1292, p<.01), Country (-.1575 p<.05) and Dance (-.1136, p<.05). All of these correlations indicated a potential negative relationship with concerts. One possible explanation is that these types of music may not be offered in the Lansing area as much as Classical music. Also, these types of music may be found in other settings (such as local bars) where it is used as background for other activities, e.g. dancing. This would appear to be supported by Dixon's 1980 study. The Lansing area Spring sweeps as compiled by the Radio Research Consortium were analyzed to see if any apparent musical trends existed within Lansing radio. Table 13 shows radio station reach for the Lansing area, according to age groups used by Arbitron. Age groups from the Arbitron sample 71 .ucoaoun uoxums use» now .6: .666666 66600 «.6 .6: .0666606 6. .6: .6666606 . comma codumun c“ cum aoaon ooumflm i ... 66 666 ochoundq anon Hem can zoom usnmmHo oaoo odaom\uodoao Houdnumao wae\m>oz xoom ooucoduo sends mo: oeucoundq anew game\n>oz anacooo xuucooo xuom conceauo Renae oaoo ofiaom\noaoao 66 666 6066 0666\06666 xuom o>ummoumoum unsuom m bN m . . m . ma «H ma mm ma Nm ma ma . on Hm m . . vN . mm +om mclmm we Hm m ma mH mm vN h mm «Mimm cm '0‘ mm Hm ma «Nina throu>3 «eaoooa Zhiozzk thimxxz tarmaxt tarmGXB thioxnk «uzdlmbt Shiznht 2¢IXH63 ZhIQBHB tdlafiuk «ZhIUOHb 26-2666 26-6606 zaixxaz 26-:666 6620606 1660056 66. 6660663 60606 6666606 66666 66 06606 72 were analyzed because this is where the most variance took place within the data gathered by the survey instrument. Sissors and Surmanek (1982) define reach as the number of persons exposed to a particular media vehicle.) When each station's reach is cross tabulated against age category, some interesting patterns appear. Generally, as age in- creases, the appeal of rock music (all types) appears to decrease. This trend is especially evident with WVIC-FM, WJXQ-FM and WGOR-FM. WKKP, WMMQ, WIBM and WFMK all play sub-genres, if you will, of rock music that showcases older artists and packages the music within a format that appeals to certain demographic segments. WITL-FM, the Country music station of the Lansing market, has a listening pattern that fits the preference pattern for Country music seen in Table 3. As sample age increases, the listenership of WITL increases. This same pattern is seen with Country music preference in Table 3. A similar trend for Easy Listening may be seen with WJIM-FM. WJIM does not rank among the top nine stations in terms of audience reach until the Arbitron 35-49 category sample. The reach then increases 14 percentage points with the 50+ segment. In Table 2, as age increased, the liking for Easy Listening increased as well. WOOD-FM out of Grand Rapids did not register in the top nine stations in Lansing until the 50+ category. The reach for WKAR-FM, which plays Classical music, remains fairly steady across the 25-50+ age spectrum. The 73 50-65 age group in Table 3 had the highest mean for Classi- cal music, so it is difficult to make a comparison here between the station reach and music approval as the age groupings are different. Some caution must be exercised when comparing Tables 5 and 13. The station formats must be taken into consideration here. Radio is so firmly entrenched as a music medium that the influence of format may never really be factored out. The trends seen in Table 5 could be the result of the radio formats in the Lansing market, and not actual music prefer- ence. In order to determine if the sample's music preferences carried over into their television viewing, the respondents were asked to indicate how often they watched particular music-oriented programs. The programs in Table 14 were chosen because they represented a wide range of musical styles. Table 14 shows the self-reported frequency in which the respondents watched the programming in question. Again, the number of valid cases does not equal 217 because not all of the respondents were familiar with all of the programs. Also, some of the programs (as indicated in the table) are only available on CATV or on the fringes of the Lansing market, which also lowers the number of valid responses. At first glance, it is apparent that the majority of the respondents replied "Never" for all of the television programs listed. The "Never" column total ranges anywhere from 37.1% to 93.9%. This would appear to indicate that the 74 .ooa. how AOOHL mad .ooav.mom .ooH. mad “OOH. «ma AOOH. «ha Aces. wow AQOHV HmH .oon. QON Aocfiv mom .cou. cod AOOH. mod .ccav mma “OOH. had «av Ouam> Ao.mmv NMH .m.vm. mun Ad.mm. and an.ab. and Aa.mmv vNH Ao.om. ems .m.ovv no Ao.mb. mad Ao.omv mvd Ad.vmv MHH Am.mbv mvd Amp. HHH goo. mod Ad.bmv mp Aavuo>oz Ao.onv mm ..H.HH. NN 16.666 66 16.66. 66 16.66 6 16.6. 66 16.66. 66 16.66. 66 16.66. 66 16.666 66 16.66. 66 16.66. 66 16.6. 66 16.66. 66 166666066 .mmma .xumounom nouns .uouuou Heaven» ammo m a. 66:6 >660 :6 . “N.m. mH Am.HHV mm Am.mH. NM .m.m. h .m.H. N Am.mv v Am.oH. mm Am.h. HA Ah.ov ma Av.vav om A~.6V o Am.m. o “5.0. m .m.ma. Nm va.mmuoo .m.v. m .m.o. 5H .m.vv OH .H.mv m .ov o .m.o. H Ab.ov ma «n.o. .v.H. .v.av AH.~. Av.m. .>.o. .5.NH. mm Assauuoaom HIDQ'MMH Am.n. m Am.cv a Ao.ov 0H .H.NV v .o.ov H Ac. 9 am.vH. on .o.~v m AN.h. ma an... m “N.cv c .6.Nv v Ab.m. m Ah.cH. mm .6. 666566 augmuoaou> c06m~>onuh an: no: 606666 6060 660664 06: 6:6 6606 6>66 0:662 666666 .0666 scone aaaoonum Duns: one uoucuo enou:aq o>6q «sauna Hoom x603 mucouamq Occumccmm .uosc 6666666 6666606: .66: 60066 60:66 eunmuuh unmuz moucmfluomuom umouo anemone vH manna 75 audience for musical programming is indeed a small one. However, if one looks at the data in a slightly different manner, some interesting observations can be made. If the cumulative valid response rates for the choices "Often," "Sometimes" and "Occasionally" are summed, the picture begins to take on new meaning. Table 15 shows the cumulative percentages for the responses "Always" through "Occasionally." The program with the largest cumulative valid response rate is greer Eerrprneneee (PBS) with 43.7%. Liye frpm Linepln genrer (PBS) was next with 42.2%, followed by Liye rrpn rne Mer (PBS) with 29.3%. At the bottom of the list were Emcee! Night mime 0.6.11.0 where (8.8%) The Meme Sonool (2.9%) and Eppny Men; (2.3%). A few caveats need to be considered when examining this set of data. green Perrornances, Liye rrpn Lincoln genrer and Liye firpn rne Meg do not always feature music. Sometimes these programs will feature drama, dance or musical comedy. However, Classical music performances are often shown on these programs; therefore, they were included in the ques- tionnaire. Also, Austin giry Limirs was carried by WKAR-TV during the 1988-89 season but was dropped for the 1989-90 season. The program is still carried by other PBS affiliates in markets adjacent to Lansing for the 1989-90 season (Detroit, Grand Rapids and Mount Pleasant). nee Men is also carried outside of the Lansing market. Migntrlight was canceled by its parent network, USA, before the fall 1989 season began. Even with these cautions, it is interesting to 76 mama .aumounom mcauam nouuou puma mouse «a 66:6 >660 :6 6666666>6 . xuucsoo huucooo HmuuummHo NNMO 66666 0:6 560666 HmunmumHo amuummmao 66666 6:6 860666 60666 666m \6666 666 6066\666 noosum> 6066\666 6066\666 Hanauano 66606 0666: 6666 c.5H m.v~ m.m~ m.m m.N m.N N.~v m.m m.ha H.o~ m.oH m.HH H.HH b.mv a 660 an: 00: 606866 6060 666666 06: 6:0 6666 6>66 6066666..0:osz .66 .0666 6:666 ..666606 0666: 6:6 .000 6660666..6>66 .66606 6666 x603 mucouama ocmumocmm .uoe¢ 6666666 6666:06: .66: 60066 60:66 .6eosflmuzoaz moucmsuomuum umuuo summoum .060: no zhaamco«umuuo: mo oucomom o>HumHosoo ma manna 77 note that the three programs that came out on top of the survey list frequently featured classical music. The respondents were also asked to mention any other performing arts programs that they watched. Table 16 shows the top nine programs that were mentioned. The program Evening at Eggs (PBS) was cited 13 times, more often than any other program. The next program most often mentioned were Lansing Symphony and MSU Symphony replays that air on WKAR—TV. NBC's Eriday flight Videos and Video Sou; (BET) were cited by the sample three times each. Thirty-one different programs were mentioned by the 217 survey participants. Again, the program most often given by the respondents (with unaided recall) was a Classical music program. While there are a variety of different musical styles available for viewing, it seems as though the persons that watch Classical music on television are quite loyal to those programs that feature the music they enjoy. MUSIC ON TELEVISION Table 17 shows the ratings for those performing arts programs that aired in the Lansing market during the Nielsen sweeps periods from November 1988 through July 1989. (Cable programs and those programs not originating in the Lansing market were not included, as that information was not avail- able.) A rating is the actual percentage of televisions in the entire market tuned into a program. The program with the highest rating was a Michigan Skyline special featuring Iabie lg Television Viewing1 1) 2) 3) 4) 5) 5) 7) 8) 9) 1 Question was "Are there any other musically oriented programs 0 am Evening at Pops (PBS) WKAR Symphony Replays Friday Night Videos (NBC) Video Soul (BET) Grand Ole Opry Nashville Live (TNN) Diana Ross Special (HBO) Star Search (Ch.47 IND) Holiday Specials (various) 22 programs with that you do watch? 78 I: 13 I" 3501;; K 79 MHMMWM 2mm mat 1 13.). L~7___he‘=lre2 13.). mg Skyline: P. Richards 4 3 Jazz Evening at Pops 3 8 Classical Lawrence Welk 2 6 Easy Listening Live fr. Lincoln Center 2 5 Classical The Music School 2 4 Classical Friday Night Videos 1 21 Pop/Rock In Performance 1 3 Classical In Recital 1 3 Classical Newport Jazz 1 3 Jazz On Stage 1 3 Easy Listening Austin City Limits 1 2 Country 1 A rating is equal to the number of TV households tuned into a program divided by the total number of TV households in a market. 3 A share is equal to the number of TVs tuned into a program divided by the number of TVs in use in the market. 8O Patti Richards, a Lansing area jazz singer. This program was seen by approximately 8,940 households (nieleeh Vieweze ih Erefiile). Eyehihg EL.EQE§ had a four week average rating of 3 during July, 1989. Approximately 6,705 households in the Lansing market saw this show. Lawrence Weih, Liye frem Lihglh genhe; and The Musie School all had ratings of 2, and the rest of the programs had ratings of one. Out of the top five, three shows featured classical music. The share trend results are somewhat different. A share is defined as the percentage of televisions actually in use tuned to a partic- ular program. Epiday nigh; yigeee (NBC) had the highest audience share, followed by Eyehihg eh Pepe and Lewhehee Welk. While Erigey nigh; Videos had a much larger market share than Evening eh Pepei this was because nggay nigh; Vigeos is on at 1:30 AM, a time slot which traditionally has fewer households watching television (Sissors and Surmanek, 1982). When ratings are compared, Eyenihg eh Pepe performs better in terms of total households reached than figigey EiQDS yigee_. Hence, the programs in Table 17 are ranked according to ratings. What makes these ratings results interesting is that the two local productions that did not fair well with the sample (The flgeie Seheel and niehigeh Skyline) in terms of aided recall actually did well in the ratings. Both programs that performed well were specials that featured Patti Rich- ards (Shyiihe) and Ralph Votapek (ugeie Seheel), both well known local artists. How well the artist is known by the 81 market may indeed have a strong impact on the performance of the television program in the ratings. Also, national popu- larity must also be considered. The Boston Pops are a well known orchestra and have had their own television program for several years. An examination of the correlations between the musical style preferences and the individual program viewership yields some interesting results. Appendix B lists the corre- lation matrix between musical styles and television pro- grams. Since the scales used to measure these two variable types are opposite (see Appendix A), within the confines of this particular matrix a negative correlation indicates a potential positive relationship between variables. The correlations between classical music preference and each of the classically oriented television programs follow the general pattern seen so far. The correlation between QEQQL Eegfogmenees and Classical music was -.4415 (p<.001), indicating that a positive relationship appeared to exist between the two. The correlation between Liye IIQE Licoln Center and Classical music was -.4891, and the correlation between Liye e; hhe Me; and Classical music was -.4554. Conversely, Heavy Metal (.2611), Soft Rock (.2888) and Progressive Rock (.2958) all had positive correlations with Classical music, indicating a potential negative relation- ship among these variables (p<.001). At the p<.001 level, these correlations are quite strong indeed. It would appear to be safe to conclude that the stronger the liking for 82 Classical music, the more the respondent is likely to watch these programs. Folk music (-.1927, p<.01), Rhythm & Blues (-.1572 p<.05), Easy Listening (-.1216, p<.05) and Big Band (-.3428, p<.001) were all significantly correlated with QIQQL Reg; fiezmeheee as well, indicating a possible positive relation- ship. Folk and Big Band also appear to be positively related to Liye {rem Liheeln genie; and Liye £19m hhe Me; as well. Country music preference had significant correlations with the two programs that featured Country music. The correlation between Country and Austin Qihy Limihe was -.4295 (p<.001). The correlation between Country and gee flew was -.3535 (p<.001). There appears to be a strong positive relationship between Country music and Country television programming. Folk (—.2242, p<.01), Rhythm & Blues (-.1675, p<.01) and Big Band (-.1227, p<.05) also had-slightly lower correlations with Austih Qihy Limihe. Each of these styles appears to be positively linked as well. The Folk and Rhythm & Blues styles seem to fit in with the music offered on the program. The significant correlation between Ageeih gihy Limihe and Big Band is not as easily explained; the correla- tion between Leggehee Welk and Country music is also signif- icant (-.1816, p<.01), so the apparent relationships are most probably not a statistical aberration. Age could be the common factor among these variables, since age had an influence upon the music style preferences involved. While there is no significant correlation among 83 age and Austin Sihy Lining, there are significant correla- tions among age, Country music and Leynenee Weik. It may be that age is the common link between Anehin Qihy Linihe and Big Band music. Spiritual music was positively connected with flee hey (r= -.2777, p<.001): a connection between Spiritual and Country has been seen in occupation and age groupings (see Tables 3 and 5). Jazz Fusion (-.2470, p<.01) and New Age (-.2073, p<.01) both correlated significantly with David Sanborn's jazz program Seneey Nighh, indicating a potential positive rela- tionship. However, both styles had opposite correlations with Lawrence Wei_. Fusion also had an opposite significant correlation with flee Hey. Rap had a possible positive relationship with Sen; Tnein (r= -.494o, p<.001); E2221 Bea: (r= -.3431, p<.001); Senee BEES! SSA (r= -.2286, p<.01) and Aneriean Sandshand (r= -.2070, p<.01). All of these programs feature some form of dance music, so these statistics make sense. Dance/Disco also had significant correlations with these programs that indicated a potential positive relationship exists with those television programs as well. Progressive Rock appears to be positively related to See; Innin. However, the only other significant correlations between Progressive Rock and the television programs are those indicating possible nega- tive relationships with Shea; Perfognanees, Michigan Sky- line, Lewrence Welk, Live inen Lincoln Qenter, Live finen hhe nee, Aushin Qihy Lining and See Haw. None of these programs 84 feature any form of rock music per se, with the possible exception of Country/Rock "crossover" artists on Austin gihy Limits and flee Hey. The results given in Appendix B could lead one to conclude that those persons who do watch music programming on television will watch that form of music which appeals to them. Most of the musical styles either had positive rela- tionships with programs that featured that particular style (or a related style), or had a negative relationship with those programs that did not. It seems that music preference may be used as a predictor for what type of music program- ming a viewer may watch. One caution must be taken into consideration. Most of the programs in Table 15 were not heavily watched, and only four of the music styles in Table 1 had means above the expected mean of 5.5 on a 10 point scale. So it would appear that the strong correlations are anchored in the dislike of the style, as well as the televi- sion show that goes with it. Even so, those persons that do like Classical or Country music appear to be watching pro- gramming that will show those music styles. MEDIA PREFERENCES FOR ENJOYING MUSIC Assuming that music preference may be used, to some extent, to predict what type of music programming will be viewed by a given individual, it would be of use from a programming standpoint to know if a market segment exists that is predisposed to watching music on television, and, if 85 so, what type of music they enjoy. Table 18 shows media preferences indicated by the respondents when asked to choose between two means of enjoying music. The respondents were asked to indicate a preference for enjoying a musical performance of an artist. There were no "forced" responses: "neutral" and "don't know" were allowed. All other things being equal, live performance was the medium of choice for the sample. Concerts fared better than watching performances on television, purchasing records, listening to the performance on the radio, and renting a musically oriented video. Nearly three quarters (74.7%) of the sample indicated that they would rather attend a concert than watch it on television. When asked to chose between concerts and purchasing records, 68.4% said they would rather go to the concert. When offered a choice, 80.9% said they would rather attend a concert than watch a videotape of an artist. Additionally, 73.4% preferred concerts over listening to a performance on the radio. Television was preferred over radio by 56.7% of the sample. However, television did not fare as well against listening to records: 41.9% preferred television, 40% pre- ferred records, and 18.1% were neutral. It would seem that television ranks between concerts and radio as a means of enjoying music, and that it is about even with records. Interestingly enough, 65.6% of the sample either agreed or strongly agreed with the statement "I like watching live concerts on television," which may or may not be a point of 86 .906. OHN .oou. mom .996. com .oou. and .006. and .oon. whH .ooH. ¢o~ .oon. OHN .ood. 56H .ooa. OHN .9966 cam AOOH. OHN .006. mow .vv cunu>6lv .v.n. n Am.o. H ...H. m An.mH. vm .m.w. MA .m.om. Hm .m.H. N A~.oH. mm .m.~. v .m.ma. ma .6.m. ma .N.°o. won .a.m. ma .vn. Ab Am.o~. av Am.~m. HHH .m.o~v mm Am.Ho. med ~v.~. m .o.m~. on AH. N .6.mu. mm Am.d. v .o.. we .v.~. m Add. MN .uumnaa.m .d. 00606660 : cmzu Hozuflu 0:60) we GEO» unedunmno « ...6. 6 .6.66. 666 .6.66. 666 6666: 66666 .6.66. 66 .6.66. 666 .6.6. 66 66666660 6666>6666 666666 .6.66. 66 .6.66. 66 .6. 66 66666> 6666660 666666 .66. 66 .6.66. 66 .6.66. 66 .66> 6666 66>6 666666 6:6 ...66. 66 .6.66. 66 .6.6. 66 .66> 6666 66>6 666666 666 ...66. 66 .6.66. 66 .6.6. 6 666666 66>6 66666 666 .6.66. 66 .6.66. 66 .6.6. 66 666666 66>6 666666 666 .6. 66 .6.66. 66 .6.6. 6 6666660 666666 66>6 66666 .6.6. 66 .6. 66 .6.6. 6 6666660 66>6 6666> 6666 .6.66. 66 .6.66. 66 .6.66. 66 66666 66>6 >6 .6.66. 66 .6.66. 66 .6.66. 66 666666 66>6 6666660 .6.66. 66 .6.66. 66 .6.6. 66 >6 66>6 666666 .66. 66 .6.66. 66 ...66. 66 «>6 66>6 6666660 .6. 6666662 .6. 66666 .a. 66666 .6 noncouououd 6600: 66 o6nma 87 encouragement for the fine arts television producer. Videocassette preference among the respondents was apparently not high. Neither renting nor purchasing video- cassettes compared favorably against purchasing records or concert tickets. This more than likely ties in with the television medium's ranking against concerts and recordings. A videocassette is going to have the same medium related limitations as over the air television. Also, the questions referred to videocassette purchase and rental, as opposed to time shifting programs for later viewing. Taping a program for later viewing requires less effort than leaving the home and renting a tape. It is worth noting that purchasing recordings or concert tickets and renting/purchasing a videotape require (in most circumstances) about the same amount of effort in terms of leaving one's home and going to a distribution outlet. This would appear to make the com- parison between the two somewhat fair. However, the question of time shifting was not addressed, so the VCR use data must be viewed with caution. The respondents were also asked if they liked music video programming where the music's meaning was interpreted for them visually. The responses were close together, with 35.5% agreeing that they enjoyed such programming, 30.5% disagreeing and 22.5% indicating that they were neutral. This question refers to the abstract or non-performance music video, as opposed to music programming that shows an actual performance. Performance based-programming was the 88 basis for the remainder of the questions. FINDING AN AUDIENCE What may be of interest to the television programmer is whether any relationships exist among the media preferences and musical styles. Appendix C shows the SPSS generated correlations between the individual music style preferences and media preferences. Of particular interest to this study are the results from the comparisons among television, concerts, radio and recordings. When records and television were compared, only Easy Listening (.2159, p<.001) and Spiritual (.1734, p<.01) had significant positive correla- tions with the preference for television. Jazz (-.1448, p<.05), Jazz Fusion (-.2779, p<.01), Progressive Rock (-.2344, p<.01), Classical (.1484, p<.05), New Age (-.2149, p<.01) and Folk (-.2530, p<.001) all correlated negatively with the preference for television. This could be interpret- ed to mean that those persons who enjoyed Easy Listening would be more inclined to watch a musical performance on television than those that enjoyed Jazz, Jazz Fusion, Pro- gressive Rock, Classical and Folk. When the respondents were offered a choice between television and radio, New Age (.2009, p<.01) and Rap (.1583, p<.05) had significant negative correlations with televi- sion, while Easy Listening (-.1581, p<.05) had a significant positive correlation with television. (Note that the posi- tive/negative sign is influenced by the scale of the ques- 89 tion, and should be kept in mind when interpreting the results.) Again, this could be interpreted to indicate that those persons who enjoyed Easy Listening would be more inclined to watch music performance on television than radio. When television and concerts were compared, only two styles had significant correlations with television. Pro- gressive Rock (-.2006, p<.01) had a significant negative correlation with television, while Easy Listening (.1265, p<.05) had a significant positive correlation with televi- sion. Since the majority of the sample (74.7%) preferred concerts to television, it may be that this preference was evenly spread among the various music styles except for the two indicated. Soft Rock (.1745, p<.01), Jazz (.1298, p<.05) and Progressive Rock (.1759, p<.05) all had significant negative correlations with the liking of television as a means of enjoying a concert (see TV LIKE in Appendix B). Easy Listen- ing (-.1529, p<.05) and Big Band (-.2205, p<.01) had signif- icant positive relationships with television. What makes these correlations interesting is that 65.6% of the sample indicated liking the idea of televised concerts yet only two styles were significantly and positively linked to televised concerts. This is similar to what occurred when television and concerts were compared. Country Music had high positive correlations with videocassette use in two instances. Country had a 90 significant positive correlation (-.1397, p<.05) both when purchasing videocassettes was compared to purchasing records and when renting videocassettes was compared with purchasing records (.2122, p<.01). Progressive Rock had significant negative correlations with both renting (-.3266, p<.001) and purchasing (.1508, p<.05) videocassettes when compared with purchasing records. Classical music also had significantly negative correlations with buying (.1374, p<.05) and renting videocassettes (-.1367, p<.05) when compared to buying records. New Age had a negative correlation with buying videos as well (.1659, p<.05). Spiritual (-.1387, p<.05) and Easy Listening (-.2230, p<.01) had significant negative correlations with buying videos as opposed to records. They also had negative correlations with renting videos as well (.1579, p<.01 and .2043, p<.05 respectively). Dance/Disco was the only other style to correlate with video rentals (-.1308, p<.05). Given these results, it would appear that Easy Listen- ing is the only style to have a consistent positive rela- tionship with over the air television as a means of enjoying musical performance. Country and Dance/Disco seemed to have positive relationships with videocassette rentals when compared to record purchasing. These were the only styles that were positively related to the video-based media. Some caution must be exercised when interpreting this data. These questions by their nature assume that the choice is made in a vacuum of sorts, that is to say "all things 91 being equal." This vacuum does not exist outside of the laboratory. Several factors must be examined in addition to the media choice. For example, according to the data, most people given a choice prefer to go to concerts. However, the survey questions did not take cost or distance traveled into account when asking about concerts as a media preference. Concerts are an intimate experience for the attendee; tele- vision may be able to put the performer right in the living room, but it is difficult to capture the essence of a live performance within a 19 inch screen. The survey instrument also did not ask any of the respondents if setting aside money for attending a concert was a high or low priority item. Different results might have occurred if the respond- ents were asked to consider the associated costs when re- sponding to the questions. Another factor to consider is the technical limita- tions of the television medium. Until recently, television used to be a monaural medium, and consequently, suffers in comparison to radio, concerts and recordings, which are all binaural presentations. Since music is supposed to appeal to the ear, it seems reasonable that those persons wishing to listen to it would want as high a quality as possible. Furthermore, those persons in poor reception areas without cable or a home satellite receiving dish would have to tolerate a lesser quality picture when viewing at home. This results in a poor video and audio signal, and presumably, a lesser quality viewing experience. 92 Another consideration is that most of these questions refer to musical performance so as to provide a common frame of reference for the respondent. Some of the music featured on programs like MTV and Friday Night yigggg is presented in an abstract form that is supposed to provide the viewer with a point of view, or interpretation, of a particular song. This video art may attract a different type of viewer than a program that features straight performances. In fact, the original concept behind MTV was that of a radio program that could be watched on television (Desinoff, 1987). This study does not inquire as to how music videos are used (if at all) and how this impacts upon the perception of music perform- ance on television. This perspective needs to be addressed in a separate study. Even with the cautions given above, it would appear that Country and Easy Listening music appear to have a symbiotic relationship of sorts with the television medium. The results in Table 19 and Appendix C appear to confirm this relationship. Furthermore, Country, Spiritual, Easy Listening and Big Band all had significant positive correla- tions with the number of television hours watched by the sample. With p<.001, the correlation between Country and hours spent watching television per day was .2674; with Spiritual, .2207; with Easy Listening, .2686 and with Big Band, .2016. It would appear that those persons who rated these styles highly also watched more television than the rest of the sample. Conversely, Heavy Metal (-.1823, p< .01) 93 Jazz Fusion (-.2361, p<.001) and New Age (-.2573, p<.001) all had negative correlations with the number of television hours watched in a day, indicating that those persons who rated these styles highly watched less television than did the rest of the sample. Interestingly enough, the correla- tion between age and time spent watching television was .3920 (p<.001), indicating that as age increased the likelihood of watching more television per day increased. The relationship between age and Country, Spiritual, Easy Listening and Big Band was seen in Table 5, and the rela- tionship between the these styles and the various television programs in the survey was seen in Appendix B. The correla- tions between music preference and time spent watching television appear to fit in with all of the other informa- tion about television viewership and music preference given in Tables 15 through 19 and the Appendices, so it appears that these observations regarding Easy Listening, Country, Big Band and Spiritual are not flukes. Chapter V CONCLUSIONS REVIEW Music has become a viable programming tool for the television medium within the last ten years. It is hypothet- ically possible to develop a television program that would appeal to any kind of musical interest. This study set out to determine if a relationship exists among music prefer- ence, music consumption and fine arts television viewing in order to identify potential market niches relevant to such programming. One must first determine the target market's music preferences and then design a format that caters to that preference. A literature search yielded several legitimate methods of measuring music preference by style, along with methods for finding overall preference patterns within a population. Several instances where music style significantly interacted with media use were documented within the literature as well. These relationships, or market niches, were the basis for the commercial success of music programming on televi- sion. Results from a 1988 Harris poll indicated that televi- sion could have an impact in delivering the performing arts to the general public and that Public Television was widely regarded as a source for such programming. Two samples were drawn from the Lansing area, one 94 95 consisting of WKAR-TV members, and one of non-members. A telephone survey instrument was used to gather data for analysis. CONCLUSIONS Given the results as presented in Tables 1 through 18, it would appear that those persons who are inclined to watch musical performances on television will watch the same kind(s) of music that they listen to on the radio, see in concert and purchase in recordings. This relationship ap- peared to be the strongest with respect to Classical music preference. Classical music received the highest preference mean for the entire sample, indicating that the style was well received. More persons reported attending Classical concerts than any other style in the sample; the type of music that garnered the largest total television audience was also Classical. It seems that those persons that enjoy Classical music will seek it out no matter what the presen- tation medium is, be it radio, recordings, concerts or television. This is also true of the other musical styles examined in this study. Positive relationships were seen between the various forms of Jazz and television programs like Sunday Night with Qayig Sanborn, which feature Jazz performances. Positive relationships were also established between the preference for Country music and programs like Austin City Limits and gee flaw; this also held true for the 96 liking of Easy Listening and the Layrgngg ngk §g9_. The apparent relationship among the music style prefer- ences and the self-reported television viewing habits was also such that persons seemed to avoid those shows that featured music that they disliked. For example, certain music styles, such as Rap and Dance/Disco, were among the styles that received the lowest preference means from the sample, indicating that they were not well received by the sample. Those television programs that were designed to appeal to Rap or Dance/Disco fans were not claimed to have been watched as much by the sample as were the Classical- based programs. Finally, the correlations between music style preferences, media preferences, and television program viewing patterns yield information that should be of inter- est not only to the fine arts television producer, but also to concert promoters, radio station programmers, and video- cassette distributors. Market Segments It would appear safe to conclude that certain patterns exist within the music preference spectrum of the sample. The sample music preference means for each of the styles varied widely with age. Only Rhythm & BLues and Dance/Disco preference did not vary with age. This impact should be kept in mind when developing market goals for music programming in general, and the Lansing market in particular. 97 Furthermore, the factor analysis as shown in Table 11 did yield some groupings of musical styles that could be consid— ered to be valid market segments. These groupings should be useful in terms of developing programming strategies for the electronic media so as to appeal to as broad a base as possible. The first, and apparently strongest, of these groups is the Classical music segment. This style had the highest mean in the sample, had the largest concert attendance cluster and had the strongest self-reported television viewing with programs such as gggn; 2g;;n;nnngg_. The Classical segment appears to be loyal to the genre, well-educated, somewhat older than the rock based music segments and more likely to be members of WKAR-TV. It would appear that, according to the results given, the current WKAR programming practices, such as broadcasting Evening n; Enpg, are reaching their intended audience. Another market segment in terms of music preference would be Easy Listening. Those persons who enjoy Easy Lis- tening tend to be somewhat older than the rock music seg- ments and indicate a preference for using television as a means of enjoying music. The Lawnencg ngk gnny, carried by WKAR, is a program that may appeal to the person that enjoys Easy Listening. The Country Western segment is also quite distinct within the sample. The Country listener appears to be older 98 than the rock listener, has completed less schooling on the average than the Classical segment and may have a tendency to use their VCR to enjoy musical programming. This segment is also apparently loyal; while WKAR-TV does not currently air Angtin Qity Limits, the sample is apparently finding the program on other PBS stations in adjacent markets or on cable television. Also, when Angtin gity Limits was airing in the Lansing market (November, 1988), it had a rating of 1; while not spectacular, this was better than some of the music programming that was airing at the same time. Country may be a potential market niche for WKAR, given this appar- ent loyalty. Another possibly viable segment is that comprised of those persons who expressed an enjoyment of Jazz, Jazz Fusion, Rhythm & Blues and Electronic/New Age. This segment does not demonstrate a positive relationship with television medium usage; rather, it would appear that this segment prefers using recordings over television as a means of enjoying music. This segment appears to be somewhat younger than the Classical, Country and Easy Listening segments and has a potential positive relationship with education. There seems to be no apparent short term gain in offering televi- sion programming towards this segment. If the reasons why this segment does not use television to enjoy music could be determined, then perhaps programming could be developed that can overcome these objections, turning a potential market liability into a market opportunity. CO 81'. 99 The two remaining segments, Heavy Metal and the group consisting of Dance/Disco, Spiritual and Rap, are somewhat enigmatic in nature. The Heavy Metal group consists mainly of males, and the style received a higher rating from those who were not members of WKAR TV. The second group made greater use of television in general. Dance and Rap fans, in particular, watched those programs that featured dance oriented music such as pang; EQIL! Egg. The preference for Spiritual music had a positive correlation with the amount of television watched per day. Television viewing habits may or may not be the factor that drew this segment of dissimi- lar styles together in the factor analysis. In any case, this internally diverse segment may be better served through cable television, which is a much more segmented medium than broadcast. An additional point worth pondering is that programming featuring Spiritual music might work in a Sunday morning slot on cable or broadcast television. OUTSIDE INFLUENCES Influence of Radio Programming The segmentation of Rock and Roll music into Heavy Metal, Progressive Rock, Soft Rock, Rap and Dance/Disco in this and other studies reflects the influence of radio programming practices in the study of music preference. Radio offers potential advertisers distinct market segments, and has developed formats to attract those segments. As a result, the styles mentioned above grew out of this 100 formatting practice. This segmentation has introduced noise into the communication process between interviewer and respondent. The respondents may have been responding to radio format more than musical style; this could have had a strong impact on the results. Any central tendencies that may have existed with a Rock and Roll continuum may have been eliminated by this segmentation. The conclusion of Deihl et al. (1983) that music preference is broad rather than specific would appear to support this supposition. The factor analysis as seen in Table 11 may have been affected by this trend; this would explain why Progressive Rock and Soft Rock almost loaded properly onto two different factors. Consequently, segmenting the Rock and Roll spectrum may reduce the impact Rock music might otherwise have in the television marketplace, because broadcast television is a broad-based medium, unlike radio or cable television. It would appear that the influence of radio programming was not entirely eliminated from the results. It might have been more effective to follow the example set by Music Television and query people about artists as opposed to musical style (Desinoff, 1987); in this fashion, it could have reduced radio's impact somewhat. However, using artists' names in the survey instrument could have resulted in a survey too unwieldy to gather the needed data in an efficient manner. 101 Radio Listening Patterns Another factor to consider is how radio is actually used. Radio is regarded by many to be a background medium (Eastman, Head and Klein, 1989). Television, on the other hand, typically requires that one actually pay attention to the visuals; it is difficult to relegate television to the background. The criteria for station selection are somewhat different as well. Eastman et al. indicate that a radio station's total format package is what attracts a particular listener, while the music being played plays a lesser role. The survey instrument attempted to put television and radio on equal ground as means of enjoying a musical performance. This may not be the case. Most of the music heard on the radio is prepackaged. Consequently, the typical listener may not expect any type of live performances on the radio, and would not consider radio when looking to enjoy a live per- formance. Conversely, music programming on television often features performances, e.g. Anstin gity Limitg. Even so, there are exceptions in radio. National Public Radio still broadcasts regular live performances of Classical music and has featured other live programming in the Folk domain, such asammgmign. Radio as a Preference Predictor The implications for these observations are signifi- cant. Radio station ratings may not be a very good indicator of music style preference in and of themselves because of 102 the format influence discussed above; however, if a station that showcased a particular style of music had a consistent ratings performance with a particular audience demographic, then radio station ratings may be used as an analytical tool to focus on a larger music preference picture. Two cases in point are Country and Classical stations. They both have consistent and loyal audiences. The market share itself is not as significant as the consistency of those numbers. The consistency would appear to aid in identifying possible music style preference segments in a market. Conversely, the absence of a particular music style in the local radio market may indicate that a music style preference segment may not exist. OTHER IMPACTS Demographics No significant differences existed between the WKAR and non-WKAR segments of the sample in terms of time spent using radio, television or home stereo equipment. There were also no significant differences between the two segments in terms of the number of recordings (both records and cassettes) owned either. The only significant demographic difference occurred among age groups (see Table 1). It would appear that the sample is somewhat homogeneous; most of those surveyed were white, had management/professional or other salaried jobs and have attended at least some college (but not necessarily completed their degree). Mi 103 Michigan State University The geographic proximity of Michigan State University and Lansing Community College may have had an impact upon the attitudes and opinions of the sample. The majority of the sample had completed some college or more in terms of education, so such an impact is possible. Both institutions offer a variety of performing arts programs. The majority of the performances offered at Michigan State University fea- ture Classical music. A significant portion of these per- formances showcase the MSU School of Music; these programs are offered to the general public for free. Consequently, Classical music is more accessible to the Lansing general public via minimal opportunity costs in terms of ticket price and distance traveled to reach the performance site. Regular exposure of this style of music to a community could have a potential impact on that community's perceptions of that style. Another component of this exposure is the exist- ence of public radio and television stations affiliated with universities in general, and Michigan State University in particular. Many public television and radio stations are affiliated with a university or college. Traditionally, these serve as outlets for performing arts programming, usually Classical in nature. This makes Classical music available free of charge via the airwaves in addition to free concerts with easy access. All of these factors may have contributed to the high preference mean for Classical music. 104 PTV Preconceptions Another potential impact is people's preconceptions regarding Public Television. Fifty-seven percent of those persons participating in a recent Harris poll said that they could expect to find arts programming (drama and music) on Public TV (Lou Harris and Associates, 1988). While over 40% claimed to watch thnt_2gttntmnngg§ occasionally or more (see Table 15), it may be that they are reinforcing a possi- ble preconception that those persons that like Classical music should watch such programs on PTV. This preconception could even be stronger if the respondent is a member of WKAR Television. They may feel that they must "justify" their membership by reporting that they do watch such programming. Perhaps if the question regarding WKAR membership were put off until the demographics section the results regarding program viewing would have been different. Television Program Variables One last caveat to consider is that this study explores only one aspect of performing arts television, that of music preference. There are several other factors that must be considered when purchasing or producing such programming. These include production variables, such as lighting and set design, shot sequencing, audio design and costuming, to name a only a few. The artist being featured also must be considered. Lee: the ard mat su of be 105 Local artists Patti Richards and Ralph Votapek fared well in the Lansing market. It is logical to assume that both Rich- ards and Votapek would not do as well in the ratings if those programs were offered well outside of the Lansing market, where they are most popular. If a local act were of sufficient quality and popularity, such an act could be offered on television on a local (or perhaps regional) basis, regardless of the style of music being performed. Hence, both a jazz artist (Richards) and a Classical artist (Votapek) received good ratings for a performing arts pro- gram. On the other hand, nationally well-known performers such as Itzhak Perhlman and the Boston Pops Orchestra also gathered sizable audiences, and would most likely do so in any of the top 100 markets. Again, musical genre might not matter as much, as enough people would have heard of the performer to garner a sizable audience. One must also consider the actual pieces being per- formed. Certain compositions will appeal to an audience no matter who is performing them, so long as the playing is professional. Sometimes the popular music industry takes advantage of this phenomenon with lesser known artists performing remakes of established popular songs, called "covers." Bands that perform in local establishments supple- ment their own compositions with covers while performing; in this way, the patrons are kept happy, and the band has a test ground for its own works. Orchestras such as the Boston Pops play mostly compositions that are well-known to their audie Orche its p succe if we Tele‘ blin tion the ther nan: mor pro of prc rat ed ta: pr. be 106 audience. Also, the Michigan State University Symphony Orchestra plays at least one popular composition in each of its programs when it performs. In any case, the odds of success would appear to be in favor of a television program if well-known works of any style are being performed. Television Programming Factors Program promotions must also be factored in when assem- bling a market offering for local broadcast. Program promo- tion becomes more and more crucial at the local level. If the artist featured is not nationally, but locally, known, then the promotional package must compensate for the lack of name recognition with a nationally known performer. Further- more, a lot of performances are one time events, making promotions even more crucial to the marketing mix. Knowledge of the music groupings and their associated demographics within a market can assist in developing a solid promotional plan, as certain media, e.g. radio, have strong and unique demographic characteristics (Sissors and Surmanek, 1985). Essentially, all of the other mass media can assist in promoting a performing arts television program, especially radio, as it is demographically and musically well segment- ed. Free publicity (news releases, et cetera) should also be taken advantage of wherever possible, especially if the promotions budget is small or nonexistent, which could often be the case for small commercial or public stations. One last consideration is the actual time slot in which co: ei 1c ti rn 107 the program airs. A program could face very strong competition in a given slot, or could benefit (or not) from either the program leading to it, or the program that fol- lows. Also, a significant portion of arts programs are one time performances; this could make gathering an audience difficult. One exception is the program Eygning gt 29mg, a weekly performance show. Its weekly status, which makes it easier to find, may be one reason it performed well (for a fine arts program) in the ratings. These are all factors that should be carefully considered when interpreting the results seen in Chapter 4. SUMMARY Although there are several internal and external cau— tions associated with this study, there appears to be enough evidence to support the general conclusion that music pref- erence, music consumption and performing arts television programming/audiences are related. Furthermore, it has been shown in this study (and others) that music preference tends to group itself into segments containing related styles. The style segments for this study were Classical, Easy Listen- ing, Country, Jazz/Fusion/Blues/New Age, Heavy Metal, and the Rap/Dance/Disco/Spiritual group. Knowing these groupings can be very useful in terms of program development. Each of these groups has an associated audience demographic, as well as related media preferences for enjoying music. Frog gra' she ing of pr an SC 111‘. a: 108 Program Development If the intention is to develop performing arts pro- gramming for television, then the following guidelines should be considered: Artists that fit in the Easy Listen- ing, Classical and Country genres have a reasonable chance of success. Classical and Country listeners are very loyal, and will seek out those types of music on a variety of presentation media. The liking of the Easy Listening style and time spent watching television were positively related, so those persons who enjoy Easy Listening acts may watch them on television. As far as the rest of the style groups are concerned, an examination of the local electronic media t and performance lounges will have to be made. If a particu- lar style is not offered anywhere, either as a live act or on a local radio station, it is possible that this type of music would not fare well on television either, regardless of the caliber of the artist in question. For example, if there is no jazz programming available on radio or televi— sion or in lounges that regularly host bands, then jazz would probably not succeed well in that particular area on television. Once the decision of music style has been made, a performer must then be selected. A local station producing its own programming would probably be best off featuring local artists. The local artist offers two advantages: they usually have a loyal following, and they cost less to bring 109 to the taping site than nationally known performers. Perhaps corporate sponsorship would become available if the perform- ances could be broadcast or videotaped at a local performing establishment; the resulting publicity would be beneficial for the production and the establishment hosting it. Another possibility would be to involve a local radio station that plays the same type of music that is being performed and offer a stereo simulcast. Such a move would provide an instant avenue for publicity, as well as additional funding, if needed. If a local station were really intent on showcas- ing national level talent, then grants or syndicated pro- gramming would have to be sought out. Finally, a coordinated promotional package should be assembled. The riskier the production, e.g. live versus tape-delayed, the greater the need for good publicity. Promotions will be especially critical for a one time per- formance. The individual programming executive will also have to determine the best day and time for such a program. This will more than likely vary from market to market. Recommendations for WKAR-TV Keeping these guidelines in mind, a couple of recommen- dations can be made to PBS affiliate WKAR. Classical music programming such as gtgnt Eetfgrmanges should continue to be offered. Easy Listening programming such as antgngg Emit should be offered as well given the relationship between Easy Listening and television use. Anstin Qity Limits should be brought back to WKAR since the Country music segment of the mark sea: aPP‘ seg spe and 3P re qr 110 the population is seeking the program out in adjacent markets. WKAR may also wish to engage in further audience re- search to determine if a program mix can be found that will appeal to the Jazz, Jazz Fusion, Rhythm & Blues and New Age segment. Jazz is available in the Lansing market as radio 6 specialty programming on WKAR-FM and WMMQ-FM, along with audio services offered by United Cable in East Lansing (KKGO-FM (Los Angeles, CA.) and Galactic Radio) and the Green Door, a music lounge which features Jazz and Blues i performances. An audience apparently exists for jazz in the Lansing area; the ratings from the Patti Richards special appear to bear this out. WKAR could engage in focus group research to see if this is a feasible market segment. Focus groups are held in high regard in the business community as a good way to acquire an understanding of consumer percep- tions and attitudes (Kotler, 1985). Given the positive relationship discussed earlier among preference for Jazz/Fusion/Blues/New Age music and recording use, another research possibility would be to call local record shops to obtain estimates regarding Jazz title sales as compared to other music style sales. A strong sales performance may indicate that this market segment is worth pursuing. Another segment to explore would be those persons who like Folk music, as WKAR members had a significantly higher preference for Folk than did non-members. While Folk did not establish itself as a stand alone style in Table 11, it did try vial hav out Fee sur Fi( C01 :31 111 try to load with Classical and Country, which are both viable market segments for WKAR. Folk would also appear to have a loyal audience as well, as there are two consistent outlets for Folk music in the Lansing area: the Michigan Festival, a folk/cultural "fair" that takes place every summer at Michigan State University, and the Ten Pound E Fiddle, a Folk music advocacy group that sponsors Folk concerts in the Lansing area. - _ \' «FRIEND. _ The above recommendations, if implemented, could have a 2 positive impact on WKAR television's audience, or any other % local audience. The results as given have demonstrated that Public Television is not wasting its time by broadcasting musical programming. Also recall that over 50% of those persons surveyed by Lou Harris and Associates in 1988 claimed that if fine arts programming were offered, they would watch it. (Lou Harris and Associates, 1988). If the audience is offered something they would like to see, then they most likely would watch it. If existing audience seg- ments can be identified and properly served, then the ties between the local broadcaster and its surrounding community can be strengthened, further serving the public interest. Finally, Public Television can strengthen its current posi- tion as an arts advocate in the broadcast industry. APPENDIX A SURVEY INSTRUMENT ' i .‘O .1 .1) music preference + tv habits Hello, I'm with the MSU Department of Telecommunication, and we' re conducting a study on the kinds of music people like and their media habits. This will only take about ten minutes to complete. First.... Are you 18 years old or more? 1: Yes 2=No [IF THE ANSWER IS 2, THEN SKIP TO QUESTION 61] How many radios do you have in your home? (ENTER NUMBER) ? ____ (6-7) Are you a current member of WKAR-TV? 1=Yes 2=No (8) Yes .. 1 No 2 Do you listen to a radio at work? 1=Yes 2=No (9) Yes .. 1 No ... 2 How many hours per day do you listen to the radio? (ENTER NUMBER, OR 99 FOR DK/NR) ? (10-11) 112 11.3 0.6 How many days per week do you listen to the radio? (ENTER NUMBER, OR 9 FOR DK/NR) (12) One .... 1 Two .... 2 Three .. 3 Four ... 4 Five ... 5 Six .... 6 Seven .. 7 8 9 DK/NR .. 0.7 How many televisions do you have in your home? (ENTER NUMBER) ? (13-14) [IF THE ANSWER IS 0, THEN SKIP TO QUESTION 12] 0.8 How many hours per day do you watch television? (ENTER NUMBER, OR 99 FOR DK/NR) ? (ls-16) 0.9 How many days per week do you watch televsion? (ENTER NUMBER, OR 9 FOR DK/NR) (17) One .... 1 Two .... 2 Three .. 3 Four ... 4 Five ... 5 Six .... 6 Seven .. 7 .. ..... 8 DK/NR .. 9 0.10 Do you have a VCR? 1=Yes 2=No (19) Yes .. 1 No ... 2 1414 0.11 On which day do you most often watch televsion? 1=SU 2=MON 3=TU 4=WED 5=TH 6=FRI 7=SAT (20) Sunday ...... Monday ...... Tuesday ..... Wedenesday .. Thursday .... Friday ...... Saturday .... \lO‘UI-bUNH 0.12 Please rate each of the following music styles from 1 to 10, s using one if you don't like it at all, and 10 if you like it a 6 lot. If you have never heard of a particular style, please tell me. (USE 99 FOR NF) Heavy Metal? (21-22) Soft Rock? (23-24) Jazz? (25-26) Jazz-Fusion? (27-28) Progressive Rock? (29-30) Country Western? “' (31-32) L Classical? (33-34) Electronic/ New Age? (35-36) Folk? (37-38) Rhythm 8 Blues? (39-40) Dance (Disco)? (41-42) , Reggae? (43-44) Christian/Spiritual? (45-46) Easy Listening? (47-48) Big Band? (49-50) Rap? (51-52) 0.13 00 you have a favorite type of music we haven't covered? If so, what is it? 0.14 Who are your three (3) favorite artists or groups? .15 .16 .17 .18 .19 .20 .21 11.5 Do you presently own or have access to a stereo? 1=Yes 2=No (53) Yes .. 1 No ... 2 How many recordings-- records, tapes, & CD's-- have you purchased in the last year? (ENTER NUMBER UP TO 999) ? (54-56) How many records-- LP's and 45's-- do you own? (ENTER NUMBER UP TO 999) ? (57-59) How many pre-recorded tapes do you presently own? (ENTER NUMBER UP TO 999) ? (60-62) How many compact discs do you own? (ENTER NUMBER UP TO 999) ? (63-65) How many hours per day do you spend listening to your records, tapes or CD'S? (ENTER NUMBER OR 99 FOR DK/NR) ? (65-66) How many concerts have you attended in the last year? (ENTER NUMBER) ? (67-68) [IF THE ANSWER IS 0, THEN SKIP TO QUESTION 23] 2116 0.22 Please tell me which concerts you attended. (ENTER THE FIRST FIVE ONLY) [IF THE ANSWER TO QUESTION 10 IS 2, THEN SKIP TO QUESTION 24] 0.23 Please give me your best guess as to how many musically oriented videos you have rented in the last year. (ENTER NUMBER, OR 99 FOR DON'T KNOW OR NO RESPONSE) ? (71-72) 0.24 Please tell me whether or not you agree with each of the following statements. Tell me if you STRONGLY AGREE, AGREE, are NEUTRAL, DISAGREE, OR STRONGLY DISAGREE with the statement. I would rather see a concert in person than watch it on television. 1=SA 2=A 3=N 4=DA 5=SDA 9=NR (73) Strongly Agree ..... 1 Agree .............. 2 Neutral ............ 3 Disagree ........... 4 Strongly Disagree .. 5 ................... 6 ................... 7 ................... 8 No Response ........ 9 0.25 I would rather see a concert in person than listen to a live recording. 1=SA 2=A 3=N 4=DA 5=SDA 9=NR (74) Strongly Agree ..... 1 Agree .............. 2 Neutral ... ...... ... 3 Disagree ..... ...... 4 Strongly Disagree .. 5 ................... 6 ............ ....... 7 ...... ............. 8 No Response ..... ... 9 117 0.26 I would rather listen to a recording of an artist than watch them on television. 1=SA 2=A 3=N 4=DA 5=SDA 9=NR (75) Strongly Agree ..... Agree .............. Neutral ............ Disagree ........... Strongly Disagree .. ......OOOOOOIOIOOOO No Response ........ \OmdQUlbb-JNH 0.27 I would rather watch a concert on television than listen to it on the radio. 188A 2=A 3=N 4=DA 5=SDA 9=NR (76) Strongly Agree ..... Agree ...... ..... ... 2 Neutral ............ 3 Disagree ........... 4 Strongly Disagree .. 5 .............. ..... 6 ................... 7 8 9 No Response . ....... [IF THE ANSWER TO QUESTION 10 IS 2, THEN SKIP TO QUESTION 29] 0.28 I would rather rent a videotape of an artist than see them in concert. 1=SA 2=A 3=N 4=DA 5=SDA 9=NR (77) Strongly Agree ..... Agree ..... ........ . Neutral ... ......... Disagree ..... ...... Strongly Disagree 000...... 0000000000 No Response ........ \DmflmmbUNH 118 0.29 I would rather listen to a concert live on the radio than attend in person. 1=SA 2=A 3=N 4=DA 5=SDA 9=NR (78) Strongly Agree ..... 1 Agree ........ ...... 2 Neutral ............ 3 Disagree ........... 4 Strongly Disagree .. 5 ................... 6 ................... 7 ............ ....... 8 No Response ..... ... 9 0.30 I would rather purchase a recording of an artist than a concert ticket. 1=SA 2=A 3=N 4=DA 5=SDA 9=NR (79) Strongly Agree ..... 1 Agree .............. 2 Neutral ............ 3 Disagree ........... 4 Strongly Disagree .. 5 ................... 6 ................... 7 ................... 8 No Response ........ 9 [IF THE ANSWER TO QUESTION 10 IS 2, THEN SKIP TO QUESTION 34] 0.31 I would rather purchase a video of an artist than a sound recording. 1=SA 2=A 3=N 4=DA 5=SDA 9=NR (80) Strongly Agree ..... Agree ......... ..... Neutral ............ 1 2 3 Disagree ..... ...... 4 Strongly Disagree .. 5 .....OOOOIOOOOOCOO. 6 ........... ..... ... 7 8 9 No Response ........ 0.32 Q.33 Q.34 119 I would rather purchase a concert ticket than rent a video of an artist. l=SA 2=A 3=N 4=DA 5=SDA 9=NR (81) Strongly Agree ..... Agree .............. Neutral ............ Disagree ........... Strongly Disagree .. No Response .. ..... . \DmslmLfl-bUNH I would rather purchase a sound recording of an artist than rent a video. 1=SA 2=A 3=N 4=DA 5=SDA 9=NR (82) Strongly Agree ..... 1 Agree ......... ..... 2 Neutral ............ 3 Disagree ........... 4 Strongly Disagree .. 5 ......OOCOOOOOOOOOO 6 ......C..... ....... 7 ......O... ...... .0. 8 No Response ........ 9 I like music videos that provide me with an interpretation of a song. 1=SA 2=A 3=N 4=DA 5=SDA 9=NR (83) Strongly Agree ..... 1 Agree ......... ..... 2 Neutral ....... ..... 3 Disagree ........... 4 Strongly Disagree .. 5 .............. ..... 6 ................... 7 ................... 8 No Response ........ 9 a. 6.] Y. 6.6. _" 120 0.35 I like watching live concerts on television. 1=SA 2=A 3=N 4=DA 5350A 9=NR (84) Strongly Agree ..... Agree . ....... ...... Neutral ............ Disagree ........... Strongly Disagree .. No Response ........ ‘OQQGU-bUNH Q.36 I enjoy music a lot. 1=SA 2=A 3=N 4=DA 5=SDA 9=NR (85) Strongly Agree ..... Agree .............. Neutral ............ Disagree ........... Strongly Disagree .. ......OOOODOOOOOOOO No Response ........ mmqmmbuww Q.37 Do you have cable television in your home? 1=Yes 2=No (ENTER NUMBER) (86) Yes .. 1 No .0. 2 Q.38 Please tell me how often you watch each of the following television programs. We'll use the scale: OFTEN, SOMETIMES, OCCASIONALLY, RARELY OR NEVER. How often do you watch Great Performances? 1=OF 2=s 3=OC 4=R 5=N 9=DK/NR (87) Often ........ ........ 1 Sometimes ...... ..... . 2 Occasionally . ..... ... 3 Rarely ............... 4 Never ................ 5 ..................... 6 ..................... 7 ..................... 8 Don't Know/No Resp. .. 9 [IF THE ANSWER TO QUESTION 37 IS 2, THEN SKIP TO QUESTION 41] 12]. 0.39 How often do you watch Night Flight? 1=OF 2=S 3=OC 4=R 5=N 9=DK/NR (88) Often ................ Sometimes ............ Occasionally ......... Rarely ............... Never ................ Don't Know/No Resp. .. \DQQO‘U'O-UNH Q.40 How often do you watch Dance Party USA? l-OF 2=S 3=OC 4=R 5=N 9=DK/NR (89) Often ................ Sometimes ............ Occasionally ......... Rarely ............... Never ................ Don't Know/No Resp. \oooqosuusuww Q.41 How often do you watch Michigan Skyline? 1=OF 2=S 3=OC 4=R 5=N 9=DK/NR Often ......... ....... Sometimes ............ Occasionally ......... Rarely ...... ...... ... Never ...... ....... ... Don't Know/No Resp. .. \omumuabumt-a 1122 Q.42 How often do you watch American Bandstand? 1=OF 2=S 3=OC 4=R 5=N 9=DK/NR (91) Often ................ Sometimes ............ Occasionally ......... Rarely ............... Never ................ Don't Know/No Resp. .. moumwauww 0.43 How often do you watch Lawrence Welk? 1=OF 2=S 3=OC 4=R 5=N 9=DK/NR (92) Often ................ Sometimes .. ....... ... Occasionally ......... Rarely ............... Never ....... ......... Don't Know/No Resp. .. \OQQGU-fiuNt—J [IF THE ANSWER TO QUESTION 37 IS 2, THEN SKIP TO QUESTION 45] 0.44 How often do you watch Soul Train? 1=OF 2=S 3=OC 4=R 5=N 9=DK/NR Often . ............... 1 Sometimes .... ...... .. 2 Occasionally ......... 3 Rarely ............... 4 Never ........ ..... ... 5 6 7 ....OOOOOOIOOOOOOOOOO 8 Don't Know/No Resp. .. 9 1123 Q.45 How often do you watch Live from Lincoln Center? 1=OF 2=S 3=OC 4=R 5=N 9=DK/NR (94) Often ................ 1 Sometimes ............ 2 Occasionally ......... 3 Rarely ............... 4 Never ................ 5 OOOOIOOODOOOOOOOOOOOO 6 ......OOOOOOOOOOOOOOO 7 8 9 Don't Know/No Resp. .. Q.46 How often do you watch The Music School? 1-OF 2=S 3=OC 4=R 5=N 9=DK/NR (95) Often ................ Sometimes ... ...... ... Occasionally ......... Rarely ............... Never ................ ......OIIIOCIOOIOOOO. Don't Know/No Resp. .. tomqmmsuup [IF THE ANSWER TO QUESTION 37 IS 2, THEN SKIP TO QUESTION 48] Q.47 How often do you watch Ebony Beat? 1=OF 2=S 3=OC 4=R 5=N 9=DK/NR (95) Often ......... ...... . 1 Sometimes ............ 2 Occasionally ...... ... 3 Rarely ............... 4 Never ......... ....... 5 ............ ...... ... 6 ............... ...... 7 .............. ..... .. 8 Don't Know/No Resp. .. 9 124 0.48 How often do you watch Sunday Night w/ David Sanborn? 1=OF 2=S 3=OC 4=R 5=N 9=DK/NR (97) Often ................ Sometimes ............ Occasionally ......... Rarely ............... Never ................ 00000000000000.000000 00000000000000.000000 Don't Know/No Resp. .. mooslmuuauwp Q.49 How often do you watch Live from the Met? 1-OF 2=S 3=OC 4=R 5=N 9=DK/NR (93) Often ................ Sometimes ............ Occasionally ......... Rarely ..... ..... ..... Never ................ 00.000000000000000... 0000000000000... ..... 00000000000000 0000000 Don't Know/No Resp. .. \OmQO‘Ul-bUND-I’ Q.50 How often do you watch Austin City Limits? 1=OF 2=S 3=OC 4=R 5=N 9=DK/NR (99) Often ...... .......... Sometimes ..... ..... .. Occasionally ......... Rarely ............... Never ................ Don't Know/No Resp. .. \OCDQO‘U‘buNH 1125 0.51 How often do you watch Hee Haw? 1=OF 2=S 3=OC 4=R 5=N 9=DK/NR (100) Often ................ 1 Sometimes .. ........ .. 2 Occasionally ......... 3 Rarely ............... 4 Never ................ 5 ..................... 6 ..................... 7 .......... .......... . 8 F Don't Know/No Resp. .. 9 0.52 Are there any other music shows that you watch that we haven't mentioned? If so please tell me what they are. . MT. 0.53 Do you regularly watch Public Television? 1=Yes 2=No (102) Yes .. 1 No 000 2 [IF THE ANSWER TO QUESTION 7 IS 0, THEN SKIP TO QUESTION 54] 0.54 One more set of questions and we'll be finished. How many people in your household? (ENTER NUMBER) (103-104) [IF THE ANSWER IS 1, THEN SKIP TO QUESTION 57] 0.55 Do you have any children living at home? 1=Yes 2=No (104) Yes .. 1 No 000 2 [IF THE ANSWER IS 2, THEN SKIP TO QUESTION 57] Q.56 Q.57 Q.58 Q.59 Q.6O 1126 How many children do you have? (ENTER NUMBER) (106-107) I am going read to you some income categories. Please stop me when I get to the one that best describes your total household ncome. (108) $10,000 or less .. $10,000- 20,000 .. $20,000- 30,000 .. $30,000- 40,000 .. $40,000- 50,000 .. 550,000+ ......... No response ...... \OCDNIO‘Ul-DUNH I am now going to read to you some categories regarding education. Please stop me when I get to the one that best describes you. (109) High School ......... 1 Some College ........ 2 College Graduate .... 3 Some Graduate Work .. 4 Adavanced Degree .... 5 .................... 6 OOOOOOOOOOOOOOOOOOOO 7 8 9 00000000000000.0000. No response . ........ How Old are you? (ENTER NUMBER OR 999 FOR NO REPONSE) ? (110-112) What is your Race? (113) White/Caucasian ...... Black/Afro—American .. Native American ...... Asian-Pacific Amer. .. Hispanic ............. Oriental ............. Arabic/Middle East ... Far East ............. No response .......... \OQQO‘U'IbUNI-I' .-.U1 ‘- 127 0.61 I will now read to you some categories for occupation. Please stop me when I get to the one that best describes you. (114) Part time hourly ..... Full time hourly ..... Management/Profess. .. Other Salaried ....... Unemployed ........... Retired ........ ...... Technical ............ Student .............. No response .. ..... ... \DOQO‘UI-bUNO-J 0.62 Thank you for participating, we appreciate your time. (TERMINATE CALL) RECORD GENDER OF RESPONDENT 1=Male 2=Fema1e (115) Male .... 1 Female .. 2 APPENDIX B CORRELATIONS OF MUSIC STYLE PREFERENCE WITH TELEVISION PROGRAM TITLES ‘ln‘in .1" IL“ “in. CORRELATIONS /VARIABLES metal to rap with greatper to heehaw /OPTIONS 2 5. Correlations: GREATPER METAL .2611 ( 191) P- .000 SOFTROCK .2888 ( 195) P- .000 JAZZ -.0895 ( 196) P- .106 FUSION .0311 ( 107) PI .375 (Coefficient / (Cases) / 1-tailed " . " is printed if a coefficient Page 17 Correlations: GREATPER PROGROCK .2958 ( 154) P8 .000 COUNTRY .0607 ( 196) P8 .199 CLASSIC -.4415 ( 197) P= .000 NEWAGE .0106 ( 157) P- .448 NITEFLIT ( 128) P- .163 -.0677 ( 130) P3 .222 .0290 ( 131) P3 .371 .1089 ( 70) p= .185 NITEFLIT DANCEUSA -.0945 -.0658 ( 103) ( 114) P= .171 P3 .243 .0287 -.1113 ( 132) ( 146) P- .372 P= .090 .1615 .2141 ( 132) ( 146) p. .032 P3 .005 -.0392 -.0023 ( 107) ( 119) P3 .344 P= .490 DANCEUSA -.l398 ( 141) P- .049 -.0790 ( 144) P- .173 .0230 ( 145) P- .392 .0775 ( 75) P- .253 Significance) SKYLINE .1285 ( 184) p- .041 .0775 ( 188) p- .145 .0099 ( 189) Pa .446 .0439 ( 104) p- .329 cannot be computed SPSS/PC+ SKYLINE .1652 ( 148) P8 .022 -.0034 ( 189) P3 .481 -.2415 ( 190) P8 .000 -.0694 ( 153) P= .197 (Coefficient / (Cases) / 1-tailed significance) " . " is printed if a coefficient cannot be computed Page 18 Correlations: GREATPER FOLK -.1927 ( 193) P- .004 BLUES -.1572 ( 197) P8 .014 DANCE -.0476 ( 194) P- .255 SPSS/PC+ NITEFLIT DANCEUSA .1669 .2172 ( 130) ( 143) F: .029 p- .005 .0024 .0645 ( 132) ( 146) p- .489 p- .220 -.2194 -.2115 ( 129) ( 143) pa .006 p- .006 £128 SKYLINE -.1696 ( 186) P3 .010 -.0785 ( 190) P3 .141 -.0980 ( 187) P8 .091 AMBAND -.1313 ( 202) P” .031 -.1079 ( 207) P- .061 .0382 ( 208) pa .292 .1466 ( 113) P= .061 AMBAND .0033 ( 164) P= .483 -.1421 ( 208) P= .020 .0877 ( 208) P= .104 .0901 ( 167) P= .123 AMBAND .1815 ( 205) P- .005 ( 209) P= .179 -.2200 ( 204) p= .001 WELK .1567 ( 201) P8 .013 .2659 ( 206) p: .000 .1766 ( 207) P= .005 .2907 ( 112) P= .001 1/22/90 WELK .4085 ( 163) P= .000 -.1816 ( 207) P= .004 -.1572 ( 207) P= .012 .1538 ( 166) P= .024 1/22/90 WELK -.0180 ( 205) P= .399 -.0250 ( 208) P= .360 -.1184 ( 203) P: .046 .129 REGGAE .0225 -.1356 .0247 .0099 .0233 .0509 ( 160) ( 106) ( 119) ( 155) ( 167) ( 166) PI .389 PI .083 PI .395 PI .451 PI .382 PI .257 (Coefficient / (Cases) / 1-tailed Significance) ” . ” is printed if a coefficient cannot be computed page 19 SPSS/PC+ 1/22/90 Correlations: GREATPER NITEFLIT DANCEUSA SKYLINE AMBAND WELK SPIRIT -.0978 -.0225 .1698 -.O310 -.1179 -.2434 ( 195) ( 131) ( 145) ( 188) ( 207) ( 207) PI .087 PI .399 PI .021 PI .337 PI .045 PI .000 EASYLIS -.1216 -.0331 .1349 -.0314 -.1206 -.2213 ( 194) ( 129) ( 143) ( 187) ( 205) ( 204) PI .046 PI .355 PI .054 PI .335 PI .043 PI .001 BIGBAND -.3428 -.0550 .1105 -.1347 .0062 -.3254 ( 195) ( 131) ( 145) ( 188) ( 206) ( 205) PI .000 PI .267 PI .093 PI .033 PI .465 PI .000 RAP .0972 -.0781 -.2286 -.0914 -.2070 .0476 ( 177) ( 116) ( 128) ( 170) ( 183) ( 182) PI .099 PI .202 PI .005 PI .118 PI .002 PI .262 (Coefficient / (Cases) / 1-tailed Significance) " " is printed if a coefficient cannot be computed Page 20 SPSS/PC+ 1/22/90 Correlations: SOULTRAN LINCOLN MUSSCHOO EBONYB SANBORN MET METAL -.1683 .2709 .0447 .0616 -.1181 .1777 ( 146) ( 199) ( 168) ( 128) ( 186) ( 198) PI .021 PI .000 PI .282 PI .245 PI .054 PI .006 SOFTROCK -.1514 .2124 .1056 -.0129 .0293 .2283 ( 149) ( 204) ( 172) ( 130) ( 191) ( 203) PI .033 PI .001 PI .084 PI .442 PI .344 PI .001 JAZZ -.0170 -.0411 -.0781 .0771 -.0910 -.1267 ( 150) ( 205) ( 173) ( 131) ( 192) ( 204) PI .418 PI .279 PI .154 PI .191 PI .105 PI .035 FUSION .0256 .0516 -.0413 .1035 -.2470 -.0324 ( 82) ( 112) ( 95) ( 71) ( 107) ( 110) PI .410 PI .294 PI .345 PI .195 PI .005 PI .368 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 21 SPSS/PC+ 1/22/90 Correlations: SOULTRAN LINCOLN MUSSCHOO EBONYB SANBORN MET PROGROCK -.1549 .3503 .0674 .0803 -.0599 .2267 ( 119) ( 161) ( 137) ( 101) ( 154) ( 161) 1130 PI .002 .1598 ( 204) PI .011 -.4554 P: .000 .0082 ( 163) p- .459 1/22/90 MET -.1817 ( 201) PI .005 -.1157 ( 205) PI .049 .0266 ( 200) P= .354 -.0627 165) ( P= .212 PI .046 PI .000 PI .217 PI .212 PI .230 COUNTRY .0245 .1571 -.0597 -.1028 .0043 ( 151) ( 205) ( 173) ( 132) ( 192) PI .383 PI .012 PI .217 PI .120 PI .476 CLASSIC .0895 -.4891 -.0656 -.0659 -.0139 ( 150) ( 206) ( 174) ( 132) ( 192) PI .138 PI .000 PI .195 PI .226 PI .424 NEWAGE .0337 .0763 -.0561 .0176 -.2073 ( 124) ( 166) ( 138) ( 107) ( 154) PI .355 PI .164 PI .257 PI .429 PI .005 (Coefficient / (Cases) / 1-tailed significance) " . " is printed if a coefficient cannot be computed Page 22 SPSS/PC+ Correlations: SOULTRAN LINCOLN MUSSCMOO EBONYB SANBORN FOLK .0217 -.2096 -.0299 .0113 .0288 ( 148) ( 202) ( 170) ( 129) ( 189) PI .397 PI .001 PI .349 PI .450 PI .347 BLUES -.0989 -.0550 -.0888 -.1600 -.0427 ( 151) ( 206) ( 174) ( 132) ( 193) PI .113 PI .216 PI .122 PI .033 PI .278 DANCE -.1639 -.0392 -.2380 -.1732 -.0705 ( 147) ( 202) ( 171) ( 129) ( 189) PI .024 PI .290 PI .001 PI .025 PI .167 REGGAE -.2127 .1115 -.1131 -.1372 -.0325 ( 124) ( 165) ( 143) ( 113) ( 159) PI .009 PI .077 PI .089 PI .074 PI .342 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 23 SPSS/PC+ Correlations: SOULTRAN LINCOLN MUSSCHOO EBONYB SANBORN SPIRIT -.0593 -.1017 -.0886 -.2426 -.0475 ( 150) ( 204) ( 172) ( 131) ( 191) PI .236 PI .074 PI .124 P- .003 PI .257 EASYLIS .0091 -.0726 -.0844 -.0052 -.0418 ( 148) ( 202) ( 171) ( 129) ( 191) PI .456 PI .152 PI .136 PI .477 PI .283 BIGBAND .1617 -.3485 -.0799 .0700 -.1018 ( 149) ( 204) ( 172) ( 131) ( 190) PI .024 P- .000 PI .149 PI .214 PI .081 RAP -.4940 .0767 -.1920 -.3471 -.0962 ( 132) ( 182) ( 157) ( 116) ( 172) PI .000 PI .152 PI .008 PI .000 PI .105 (Coefficient / (Cases) / 1-tailed Significance) 1/22/90 MET -.0892 ( 203) P= .103 -.0138 ( 202) P- .423 -.2635 ( 202) PI .000 -.0065 ( 181) PI .465 13]. " . " is printed if a coefficient cannot be computed Page 24 SPSS/PC+ 1/22/90 Correlations: AUSTIN MEEHAW METAL .1604 .0092 ( 192) ( 200) PI .013 PI .449 SOFTROCK .0482 -.0051 ( 197) ( 205) PI .251 PI .471 JAZZ .0195 .1403 ( 198) ( 206) PI .393 PI .022 FUSION -.0185 .2148 i ( 109) ( 113) PI .424 PI .011 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed page 25 SPSS/PC+ 1/22/90 Correlations: AUSTIN HEEMAW PROGROCK .2244 .1731 ( 157) ( 164) PI .002 PI .013 COUNTRY -.3535 -.4295 ( 198) ( 206) PI .000 PI .000 CLASSIC -.0520 .1184 ( 198) ( 206) PI .233 PI .045 NEWAGE .0157 .0728 ( 161) ( 167) PI .422 P- .175 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 26 SPSS/PC+ 1/22/90 Correlations: AUSTIN HEEHAW FOLK -.2242 .0594 ( 195) ( 203) PI .001 PI .200 BLUES -.1675 -.0299 ( 199) ( 207) PI .009 PI .335 DANCE .0318 -.1667 1£32 ( 194) ( 202) p- .330 p- .009 REGGAE .0250 .1756 ( 162) ( 167) PI .376 PI .012 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 27 SPSS/PC+ 1/22/90 Correlations: AUSTIN HEEHAW SPIRIT -.1569 -.2777 ( 197) ( 205) PI .014 PI .000 EASYLIS -.0450 -.1625 ( 195) ( 203) PI .266 PI .010 BIGBAND -.1227 -.0517 ( 196) ( 204) PI .043 PI .231 RAP .0444 -.0004 ( 175) ( 183) PI .280 PI .498 (Coefficient / (Cases) / l-tailed Significance) " . " is printed if a coefficient cannot be computed page 28 SPSS/PC+ 1/22/90 This procedure was completed at 8:56:44 finish. End of Include file. APPENDIX C CORRELATIONS OF MUSIC STYLE PREFERNCE WITH MEDIA PREFERENCES correlations /variab1es metal to rap with convstv to musicenj /options 2 5. Correlations: CONVSTV COVSREC RECVSTV TVVSRAD VIDVSCON RADVSATT METAL -.0534 -.0262 -.0896 -.0366 .0579 .0660 ( 202) ( 202) ( 203) ( 203) ( 172) ( 203) PI .225 PI .356 PI .102 PI .302 PI .225 PI .175 SOFTROCK -.0227 .0374 -.0830 .0698 -.0373 .1274 ( 207) ( 207) ( 208) ( 208) ( 175) ( 208) PI .372 PI .296 PI .117 PI .158 PI .312 PI .033 JAZZ -.1077 -.1005 -.1448 .1067 -.0002 .0817 ( 208) ( 208) ( 209) ( 209) ( 176) ( 209) PI .061 PI .074 PI .018 PI .062 PI .499 PI .120 FUSION -.0667 -.0520 -.2779 .1027 .0057 .0632 ( 114) ( 114) ( 114) ( 114) ( 94) ( 114) PI .240 PI .291 PI .001 P- .138 PI .478 PI .252 (Coefficient / (Cases) / l-tailed Significance) " . " is printed if a coefficient cannot be computed Page 17 SPSS/PC+ 1/22/90 Correlations: CONVSTV COVSREC RECVSTV TVVSRAD VIDVSCON RADVSATT PROGROCK -.2006 -.0376 -.2344 .0921 .0921 .2099 ( 164) ( 164) ( 165) ( 165) ( 137) ( 165) PI .005 PI .316 PI .001 PI .120 PI .142 PI .003 COUNTRY .0603 .0690 .0700 -.0443 -.0678 -.0511 ( 208) ( 208) ( 209) ( 209) ( 176) ( 209) PI .193 PI .161 PI .157 PI .262 PI .186 PI .231 CLASSIC -.0832 -.0905 -.1484 .1079 .0264 .0460 ( 208) ( 208) ( 209) ( 209) ( 176) ( 209) PI .116 PI .097 PI .016 PI .060 PI .364 PI .254 NEWAGE -.0718 -.0471 -.2149 .2099 -.0029 .0612 ( 168) ( 168) ( 168) ( 168) ( 140) ( 168) PI .177 PI .272 PI .003 PI .003 PI .486 PI .215 (Coefficient / (Cases) / 1~tailed Significance) " . " is printed if a coefficient cannot be computed Page 18 SPSS/PC+ 1/22/90 Correlations: CONVSTV COVSREC RECVSTV TVVSRAD VIDVSCON RADVSATT FOLK -.0370 .0099 -.2530 .0888 -.0310 -.0125 ( 205) ( 205) ( 206) ( 206) ( 174) ( 206) PI .299 PI .444 PI .000 PI .102 PI .342 PI .429 BLUES -.0486 -.1092 -.0377 -.0162 -.0009 .0671 ( 209) ( 209) ( 210) ( 210) ( 177) ( 210) PI .242 PI .058 PI .293 PI .408 PI .495 PI .166 DANCE .0452 -.0043 .0908 -.0500 -.0913 -.0202 ( 204) ( 204) ( 205) ( 205) ( 174) ( 205) PI .260 PI .476 PI .098 PI .238 PI .115 PI .387 1133 134 REGGAE -.0426 .0129 -.0679 .1354 -.0825 .0136 ( 167) ( 167) ( 167) ( 167) ( 141) ( 167) PI .292 PI .434 PI .192 PI .041 PI .165 PI .431 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed page 19 SpSS/pc+ 1/22/90 Correlations: CONVSTV covsazc RECVSTV TVVSRAD VIDVSCON RADVSATT SPIRIT .0878 .0130 .1734 -.0946 -.0349 -.0952 ( 207) ( 207) ( 208) ( 208) ( 175) ( 208) p- .104 p- .426 p- .006 p- .087 pa .323 pa .086 EASYLIS .1265 .0094 .2159 -.1581 -.0728 -.0430 ( 205) ( 205) ( 206) ( 206) ( 173) ( 206) pa .035 p- .447 p- .001 p- .012 pa .171 ps .270 BIGBAND .0928 .0169 .0090 —.0453 -.0148 -.1085 ( 206) ( 206) ( 207) ( 207) ( 174) ( 207) p- .092 p- .405 ps .449 p- .258 p- .423 p= .060 RAP -.0064 -.0560 -.0503 .1583 -.0329 .0888 ( 183) ( 183) ( 184) ( 184) ( 154) 184) ( PI .466 PI .226 PI .249 PI .016 PI .343 PI .115 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed page 20 SPSS/PC+ 1/22/90 Correlations: RECVSTIK VIDVSREC TIKVSVID RECVSVID VIDSONG TVLIKE METAL -.0548 .0746 -.0067 .0002 -.1623 .0791 ( 202) ( 171) ( 154) ( 155) ( 194) ( 203) p- .219 p- .166 p- .467 p= .499 p= .012 p= .131 SOFTROCK -.0783 —.0221 -.0172 .0063 -.1030 .1745 ( 207) ( 174) ( 156) ( 157) ( 198) ( 207) p- .131 p= .386 p- .416 p- .469 p= .074 p= .006 JAZZ -.0556 .0171 -.0352 -.1022 -.0566 .1298 ( 208) ( 175) ( 157) ( 158) ( 199) ( 208) p- .212 p- .411 p- .331 p- .101 p- .213 Re .031 FUSION -.1679 -.0049 .0144 -.1601 .0037 .0880 ( 113) ( 95) ( 85) ( 85) ( 109) ( 113) PI .038 PI .481 PI .448 PI .072 PI .485 PI .177 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 21 SPSS/PC+ 1/22/90 Correlations: RECVSTIK VIDVSREC TIKVSVID RECVSVID VIDSONG TVLIKE PROGROCK -.1039 .1508 -.0846 -.3266 -.1371 .1759 ( 164) ( 137) ( 123) ( 124) ( 158) ( 164) 1135 PI .093 PI .039 PI .176 PI .000 PI .043 COUNTRY -.0467 -.1397 .1012 .2122 -.0708 ( 208) ( 175) ( 157) ( 158) ( 199) PI .252 PI .033 PI .104 PI .004 PI .160 CLASSIC .1044 .1374 -.0302 -.1367 .1542 ( 208) ( 175) ( 157) ( 158) ( 199) PI .067 PI .035 PI .354 PI .043 PI .015 NEWAGE -.0496 .1659 .0161 -.2370 -.0772 167) ( 141) ( 127) ( 128) ( 161) PI .262 PI .025 PI .429 PI .004 P- .165 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 22 SPSS/PC+ Correlations: RECVSTIK VIDVSREC TIKVSVID RECVSVID VIDSONG FOLK -.0522 .0549 -.0870 -.1141 .1893 205) ( 173) ( 156) ( 157) ( 196) PI .229 PI .236 PI .140 PI .077 PI .004 BLUES -.0103 -.0145 .0413 .0753 .0073 209) ( 176) ( 158) ( 159) ( 200) PI .441 PI .424 PI .303 PI .173 PI .459 DANCE .1091 -.1308 .0501 .0756 -.1257 204) ( 173) ( 156) ( 157) ( 196) PI .060 PI .043 PI .267 PI .173 PI .040 REGGAE .0162 .0350 .1063 -.1512 -.1523 166) ( 140) ( 128) ( 129) ( 160) PI .418 PI .341 PI .116 PI .044 PI .027 (Coefficient / (Cases) / 1-tailed Significance) " " is printed if a coefficient cannot be computed Page 23 SPSS/PC+ Correlations: RECVSTIK VIDVSREC TIKVSVID RECVSVID VIDSONG SPIRIT .0807 -.1387 -.0838 .1579 -.0837 207) ( 174) ( 157) ( 158) ( 198) PI .124 PI .034 PI .148 PI .024 PI .121 EASYLIS .0997 -.2230 -.0193 .2043 -.0413 ( 205) ( 172) ( 155) ( 156) ( 196) PI .078 PI .002 PI .406 PI .005 PI .283 BIGBAND .0418 -.0843 .0344 .0371 .1698 ( 206) ( 173) ( 155) ( 156) ( 197) PI .275 PI .135 PI .336 PI .323 PI .009 RAP .1003 .0045 .0164 .0623 -.1750 183) ( 154) ( 138) ( 139) ( 175) PI .088 PI .478 PI .424 PI .233 PI .010 (Coefficient / (Cases) / 1-tai1ed significance) P: ( p: .012 0937 208) .089 -.0740 pg 208) .144 .0292 p- 167) .354 1/22/90 TVLIKE .0751 ( P: 205) .142 .0310 ( P: 209) .328 .0612 205) ' .192 .0255 166) .372 1/22/90 TVLIKE ( p. ( P: ( p= ( P: .0995 207) .077 .1529 205) .014 .2205 206) .001 .0200 184) .394 1136 " . " is printed if a coefficient cannot be computed Page 24 Correlations: MUSICENJ METAL - e ( p- 0605 203) .195 SOFTROCK .0167 ( p- 208) .405 JAZZ -.1056 ( p. 209) .064 FUSION -.1031 ( p- 114) .137 SPSS/PC+ 1/22/90 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 25 Correlations: MUSICENJ PROGROCK -.1325 ( p. COUNTRY .1219 ( p. CLASSIC -.1621 ( p- NEWAGE -.2260 168) .002 ( p- 165) .045 209) .039 209) .010 SPSS/PC+ 1/22/90 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Correlations: MUSICENJ FOLK -.0338 206) .315 ( p- BLUES -.1159 210) .047 ( p- DANCE -.1402 SPSS/PC+ 1/22/90 2137 ( 205) PI .022 REGGAE -.1114 ( 167) PI .076 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 27 SPSS/PC+ 1/22/90 Correlations: MUSICENJ SPIRIT .0556 ( 208) PI .213 EASYLIS .0263 ( 206) p- .354 BIGBAND -.1419 ( 207) PI .021 RAP -.0911 ( 184) PI .109 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed page 28 SPSS/PC+ 1/22/90 This procedure was completed at 9:08:11 finish. End of Include file. APPENDIX D CORRELATIONS OF MUSIC STYLE PREFERENCE WITH MEDIA USE correlations /variables metal to rap with RADIOS RADHRS RADDAY TVNUM TVHRS TVDAY MUSICBUY RECOWN TAPEOWN MUSICLIS CONCERTS INCOME SCHOOL AGE /options 2 5. RADDAY RADIOS .0840 204) .116 Correlations: METAL ( p3 .0837 ( 209) PI .114 SOFTROCK -.0172 210) .402 JAZZ ( P- .0250 ( 115) PI .395 FUSION (Coefficient / (Cases) / 1-tailed Significance) RADHRS ( .1842 202) PI .004 -.0080 ( 207) PI .454 -.0876 ( 208) PI .104 -.1568 ( 114) PI .048 ( .0052 202) P- .471 .1062 207) ( p- .064 .1312 207) ( p- .030 ( .0736 114) PI .218 TVNUM -.0830 ( 204) PI .119 .1491 ( 209) PI .016 -.0766 ( 210) PI .135 -.2525 ( 115) PI .003 TVHRS -.1823 ( 203) ps .005 -.0964 ( 208) PI .083 -.1328 ( 208) PI .028 -.2361 ( 114) PI .006 TVDAY -.0655 ( P: 204) .176 .0009 ( Pg ( P: ( P: 209) .495 1492 210) .015 0765 115) 0208 " . " is printed if a coefficient cannot be computed Page 17 Correlations: RADIOS PROGROCK .1217 ( 166) PI .059 COUNTRY -.0063 ( 210) PI .464 CLASSIC -.1272 ( 209) PI .033 NEWAGE .0612 ( 169) PI .215 (Coefficient / (Cases) / 1-tailed Significance) RADHRS ( .0057 164) PI .471 ( .0785 208) PI .130 ( 207) PI .265 -.0134 ( 168) PI .432 SPSS/PC+ RADDAY ( .1000 164) PI .101 -.0525 ( 207) PI .226 ( .0492 206) PI .241 ( 166) PI .486 TVNUM -.0726 ( 166) PI .176 .0400 ( 210) p- .282 -.1208 209) .041 'f" .0639 ( 169) PI .205 " . " is printed if a coefficient cannot be computed SPSS/PC+ Page 18 Correlations: RADIOS FOLK -.1205 ( 207) PI .042 BLUES -.0757 ( 211) PI .137 DANCE .0690 ( 206) ( RADHRS -.0642 205) PI .180 ( -.0336 209) PI .314 ( -.0781 204) RADDAY ( -.0054 204) PI .469 ( -.0473 208) PI .249 ( .13 -.0845 203) 8 -.1610 ( 207) PI .010 -.0329 ( 211) PI .317 .1493 ( 206) TVHRS -.2601 ( 165) PI .000 .2674 ( 208) R: .000 -.0167 ( 207) PI .406 -.2573 ( 168) PI .000 TVHRS .0429 ( 205) PI .271 .0982 ( 209) PI .079 .0803 ( 204) 1/29/90 TVDAY 'UA 0923 166) .118 .0971 '0’" T!" II II "GA 210) .081 .0361 209) .302 .0825 169) .143 1/29/90 TVDAY "GA II ”A N ( .0040 207) .477 .0895 211) .098 .0719 206) 2139 PI .162 PI .133 PI .115 PI .016 PI .127 PI .152 REGGAE -.0617 .0134 -.1372 -.1351 -.1273 -.2625 ( 168) ( 167) ( 166) ( 168) ( 167) ( 168) PI .214 PI .432 PI .039 PI .040 PI .051 PI .000 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 19 SPSS/PC+ 1/29/90 Correlations: RADIOS RADHRS RADDAY TVNUM TVMRS TVDAY SPIRIT -.1594 -.0107 -.1636 -.1898 .2207 .0430 ( 209) ( 207) ( 206) ( 209) ( 207) ( 209) PI .011 PI .439 PI .009 PI .003 PI .001 PI .268 EASYLIS -.0471 .0694 -.0072 .0985 .2686 .1758 ( 207) ( 205) ( 205) ( 207) ( 206) ( 207) PI .250 PI .161 PI .459 PI .079 PI .000 PI .006 BIGBAND -.0713 -.0157 -.0110 .0188 .2016 .1269 ( 207) ( 205) ( 204) ( 207) ( 205) ( 207) PI .154 PI .412 PI .438 PI .394 PI .002 PI .034 RAP .0851 -.0119 -.0145 .0538 -.0562 -.1446 ( 184) ( 182) ( 183) ( 184) ( 183) ( 184) PI .125 PI .437 PI .423 PI .234 PI .225 PI .025. (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed page 20 SPSS/PC+ 1/29/90 Correlations: MUSICBUY RECOWN TAPEOWN MUSICLIS CONCERTS INCOME METAL .1157 -.0858 .0293 -.0071 -.0605 -.0413 ( 203) ( 201) ( 201) ( 198) ( 203) ( 192) PI .050 PI .113 PI .340 PI .460 PI .195 PI .285 SOFTROCK .0434 -.0155 .0814 -.0390 -.1716 .2572 ( 208) ( 206) ( 206) ( 203) ( 208) ( 195) PI .267 PI .412 PI .122 PI .290 PI .007 PI .000 JAZZ .0710 .2536 .0225 .0960 -.0203 .0492 ( 209) ( 207) ( 207) ( 204) ( 209) ( 196) PI .154 PI .000 PI .374 PI .086 PI .385 PI .247 FUSION -.0402 .1453 .0257 .0715 -.0090 .0902 ( 114) ( 114) ( 114) ( 112) ( 114) ( 109) PI .335 PI .061 PI .393 PI .227 PI .462 PI .176 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 21 SPSS/PC+ 1/29/90 Correlations: MUSICBUY RECOWN TAPEOWN MUSICLIS CONCERTS INCOME PROGROCK .0940 .0538 .0493 .0223 -.1292 .1124 COUNTRY CLASSIC NEWAGE (Coefficient / (Cases) / 1-tailed Significance) ( 165) p- .115 -.0201 ( 209) p: .386 -.0080 ( 209) p- .454 .1590 ( 168) p- .020 ( 164) p- .247 -.0495 ( 207) p- .239 .1229 ( 207) p- .039 .2042 ( 167) p- .004 14() ( 164) p- .265 .0455 ( 207) p- .258 -.0331 ( 207) p- .318 .0223 ( 167) p- .388 ( 161) p- .390 -.1653 ( 204) p- .009 .0685 ( 204) p- .165 .0625 ( 165) p- .213 " . " is printed if a coefficient cannot be computed Correlations: FOLK BLUES DANCE REGGAE MUSICBUY -.0635 ( 206) P3 .182 .0236 ( 210) P3 .367 ( 205) Pg .168 -.0039 ( 167) P‘ .480 RECOWN .1585 ( 204) pa .012 .1899 ( 208) pa .003 -.1004 ( 203) R: .077 .1374 ( 167) pa .038 SPSS/PC+ TAPEOWN -.0037 ( 204) PI .479 .0142 ( 208) pa .419 -.0087 ( 203) PI .451 ( 167) P3 .304 MUSICLIS -.0819 ( 201) p- .124 .0205 ( 205) p- .385 .0740 ( 200) pa .149 .0485 ( 164) pa .269 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Correlations: SPIRIT EASYLIS BIGBAND MUSICBUY -.0991 ( 208) p- .077 -.1641 ( 206) p- .009 -.1029 ( 207) p- .070 -.0087 ( 184) PI .453 RECOWN -.1039 ( 206) p- .069 -.1452 ( 204) PI .019 .0237 ( 205) PI .368 .0195 ( 183) p- .397 SPSS/PC+ TAPEOWN -.0887 ( 206) p- .102 ( 204) P- .440 ( 205) P3 .179 -.0572 ( 183) p- .221 MUSICLIS -.0488 ( 203) p- .245 .0226 ( 201) p- .375 .0467 ( 202) p- .255 .0443 ( 180) p- .278 ( 165) - .049 -.1575 ( 209) .011 .2870 209) ' .000 'UA I -.0233 ( 168) PI .382 CONCERTS .1016 ( 206) PI .073 -.0530 ( 210) p .222 .1136 205) .052 F". .0286 167) I .357 'UA I CONCERTS -.0551 ( 208) PI .215 -.0948 ( 206) PI .088 .0259 ( 207) PI .355 -.0156 ( 184) p= .417 157) .081 .0952 196) .092 "UA II .0086 196) .453 .U" I! .1898 PI .008 1/29/90 INCOME .0244 ( 193) .368 .0149 197) .417 .Uf‘ N -.0208 192) .387 -.0894 162) .129 1/29/90 INCOME -.2781 ( 195) p= .000 .0242 193) .369 'UA II .0455 195) .264 ’UA II .0235 174) .379 .U" II 141. (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed page 24 SPSS/PC+ 1/29/90 Correlations: SCHOOL AGE METAL .0120 -.3659 ( 201) ( 200) PI .433 PI .000 SOFTROCK .0578 -.4519 ( 205) ( 205) PI .205 PI .000 JAZZ .1868 -.1417 ( 206) ( 206) PI .004 PI .021 FUSION .2660 -.3323 ( 112) ( 113) PI .002 PI .000 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 25 SPSS/PC+ 1/29/90 Correlations: SCHOOL AGE PROGROCK .1586 -.5937 ( 162) ( 163) PI .022 PI .000 ( 206) ( 206) PI .000 PI .000 CLASSIC .1782 .3042 ( 206) ( 206) PI .005 PI .000 NEWAGE .1909 -.2998 ( 165) ( 166) PI .007 PI .000 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 26 SPSS/PC+ 1/29/90 Correlations: SCHOOL AGE FOLK .2387 .2044 ( 203) ( 203) PI .000 PI .002 BLUES .0416 .0238 ( 207) ( 207) PI .276 PI .367 142 DANCE -.1153 .0629 ( 203) ( 202) p- .051 p- .187 REGGAE .1876 -.2454 ( 165) ( 166) PI .008 PI .001 (Coefficient / (Cases) / 1-tailed Significance) " . " is printed if a coefficient cannot be computed Page 27 SPSS/PC+ 1/29/90 Correlations: SCHOOL AGE SPIRIT -.2624 .2591 ( 205) ( 205) PI .000 PI .000 EASYLIS -.2609 .2721 ( 203) ( 203) PI .000 PI .000 BIGBAND -.0631 .4367 ( 204) ( 204) PI .185 PI .000 RAP -.0159 -.3107 ( 183) ( 182) PI .416 PI .000 (Coefficient / (Cases) / 1-tai1ed Significance) Page 28 SPSS/PC+ 1/29/90 This procedure was completed at 8:02:16 finish. End of Include file. BIBLIOGRAPHY AND GENERAL REFERENCES 1979 Blalock, Herbert M. Jr. figgig1.§tati§§ig§. 2nd ed. New York: McGraw-Hill. 1979 Bruner, Gordon C. "The Association Between Record Purchase VOlume and Other Music Related Characteris- tics." 29.9.9191: 19.19.19 and 599191;:- 1979 V01 6:3- pp. 235- 241. 1983 Deihl, E. Roderick: Schneider, Michael J. and Petress, Kenneth. "Dimensions of Music Preference: A Factor Analytic study-” Repular.nusis.and.5991951. Vol. 3: pp. 41-49. 1987 Denisoff, R. Serge. "Psychographics and MTV: An interview with Marshall Cohen." 292913;;unsig and Sggigty. 1987 Vol. 11: pp. 27-34. 1980 Dixon, Richard D. "Suggested Scales for the Measurement of Musical Involvement and Genre Tastes." £99912: Essie and 5991952. 1980 V01 7:4- pp. 223-244. 1989 Eastman, Susan Tyler: Head, Sydney w. and Klein, Levis. Breadsastzsahle.Eregramminsi.§tratesiss and .zcagtiggg‘ Belmont: Wadsworth Publishing Company. 1988 Pass, Thomas Gerard. Explerins.89919.2r9ferense: An W 51:11:12 a 1.981.998 fer 21mm- Master's Thesis, Michigan State University. 1985 Pink, Edward L.: Robinson, John P. and Dowden, Sue. "The Structure of Music Preference." ignznalIgf .semmunisasien- July. 1985- pp- 318- 1981 Hair, Harriet I. "Verbal Identification Of°Music ConceptS-" learnal 91.89519 899999198. Vol. 29. nO-l- pp. 11-21. 1987 Hall, William E. and Blau, Judith R. "The Taste for Popular Music: An Analysis of Class and Cultural Demando" Eeeular:MRais.and assists. V917: pp- 31-49- Hargreaves, David J. "The Effects of Repetition of Liking for Music." ignznglyg1,uugig Educatign. Vol. 32 No. 1. pp.35-47. 1987 Hartman, John K. and Tucker, Raymond K. "Music- Video Viewing in the 18-35 Yr. Old Age Group From a Uses and Gratifications Perspective." 2gpglg;_un§1g_gng 1988 Jones, Steve. "Cohesive but not Coherent: Music Videos, Narrative and Culture." 222313; Magi; and figgiggy, Vol. 12:4, pp. 15-30. 143 144 ‘ 1986 Kerlinger. Fred N. Eeundatiens 9f.Behaxieral Beeeezeh. 3rd. ed. New York: Holt, Rinehart and Win- ston. 1984 Kotler. Philip. Marketins.uanassmsnti end eenezel. 5th ed. New Jersey: Prentice- planning Hall, Inc. 1983 LeBlanc, Albert and Cote, Richard. "Effects of Tempo and Performing Medium on Children's Music Prefer- ence." Journal of Music 299289190. Vol. 31. No. 1. ppe 57-66 e 1982 LeBlanc, Albert. "An Interactive Theory Of Music Preference." Jeanne; QI.MR§1Q.TD§I§R¥~ 1982 Vol. 19:1. pp. 28-45. 1978 Lindsay, Carl A.: LeRoy, David J. and Novak, Theresa A. "Public Television: Whose Alternative?" Enhlég 18 Regieu, March/April. pp. - - 1989 Mchy, Steven. Lectures 9 October and 13 November. Communication Arts & Sciences 892, genenmez,fieme Enter: tainment. 1987 Melton, Gary Warren and Galician, Mary-Lou. "A Sociological Approach to the Pop Music Phenomenon: Radio and Music Video Utilization for Expectation, Motivation and Satisfaction." Bennie; Mania and Seeie: 32. 1987: Vol. 11: pp. 35-46. '1988 National Research Center for the Arts. "Americans and the Arts V: Nationwide survey of public opinion." Louis Harris and Associates, Inc. 1988. 1988, 89 A. c. Neisen Research Bureau. Neileen,yieneze in.2refile 1987 Oglesbee, Frank W. "Eurythmics: An Alternative to sexism in music videos." £99918: Musis.and 6921952. V01. 11:2, ppe 53-64e 1989 Radio Research Consortium. "Analysis of Metro Diary Mentions." Lansing-East Lansing Market, Spring 1989. 1980 Scheussler, Karl E. MB:1£§1.IE§§£ end fieeie: Esenemis Basksreund- New York. Arno Press. 1986 Sissors, Jack 2. and Surmanek, Jim. Megie zlenning 2nd ed. Lincolnwood: NTC Business Books. 145 1987 Walker, James R. "The Context of MTV: Adolescent Entertainment Media use and Music Television." 222918: Music and assists. Vol. 11:3. pp- 1-10- 1985 Weinstein, Jordan. "V for Victory: America's hottest idea in television is right here in Central Massachusetts." ageineee flezeeegez. August, 1985. pp. 25-28. 1984 Wells, John D. "Music Television Video and the Capacity to Experience Life." Bennlez,nneie and Seeie; 3!. Vol. 9:4, pp. 1-6. 1983 WOlfe, Arnold S. "Rock on Cable: On Music Televi- sion: The First Music Video Channel." Beagle: nueie fieeiegx. 1983 V01. 9: (1): pp. 41-50. . HIGAN STRTE UNIV IILIIILWILUIII I all)" H IIIIIIJIIIQIL|7l|l7WIES