Life-Style Research and Prime Time Television

LIFE-STYLE RESEARCH AND PRIME TIME TELEVISION

Alan Wurtzel, American Broadcasting Company

Over the past ten or fifteen years, lifestyle research has developed to the point where it has become a commonly accepted technique for the study of consumer behavior. While some have attempted to make fine distinctions between lifestyle research and psychographics, perceptual mapping, segmentation analysis, and so on, for my purposes today I would like to combine all of these terms into a single category since their approach and objectives are more or less the same. By lifestyle research I refer to the segmentation of the mass audience into smaller sub-groups through the use of various attitude, interest, or opinion scales. In other words, the discrimination of an audience by variables other than traditional demographics. The basic assumption being that the more we can know about the person to whom we are communicating, the more effective our communication. Demographics play an obviously important role but they provide us only with a single dimension while lifestyle research offers a richer and more multi-dimensional view of the consumer or, in the case of television, the viewing audience.

Certainly those of you who are familiar with market research in the areas of product development, marketing, and advertising do not have to be convinced of the usefulness and advantages which audience segmentation offers the technique has become a standard part of the market researcher's repetoire and the research literature is replete with examples of its use in these fields. However, when we look at the application of these techniques to television programming, a much different picture emerges. Only a handful of articles are currently in the literature and with the exception of its application for a few tangential areas such as media planning, lifestyle research is--to my knowledge--little used in television program development, scheduling or planning.

Perhaps one of the main reasons for this is due to the characteristics of the television medium itself. Television--and at this point I should indicate that I am referring to commercial broadcast television--is the last truly "ma ss" medium. It's no secret that for a program to be successful on any one of the three commercial national networks, it must reach a large, diverse, and heterogeneous audience. Under these circumstances, we lose the very advantages which segmentation analysis offers: that is, the ability to identify a small slice of the entire consuming public and to use this subgroup as a marketing or advertising target. Since it's probably more true than not that television tries to reach the largest audience possible (within some limitations), the idea of identifying smaller subgroups within the television audience never seemed to be a particularly useful approach.

In the past few years, however, the television business has undergone a number of significant changes. The concept of the television season has been drastically altered. New program forms have been developed. Program scheduling decisions must be made on a year-round basis and not simply during "pilot season," As the business has changed rapidly, so too we are finding that our audience is changing. The evolution of the society's norms and values is a continuing process. It's crucial for us to measure our society's pulse on a regular basis since people's attitudes, values, and opinions concerning a broad spectrum of societal issues invariably impacts on the kinds of shows they watch on television and the kinds of programs they want to see.

To provide management and producers with this kind of information, the ABC Developmental and Social Research Unit has explored some of the segmentation techniques which have been used successfully by market researchers in other fields. It's our feeling that standard demographics only provide one part of the total picture and the use of psychographics can offer a more complete understanding of our audience and how they perceive our programming. One of the major problems which we face is the development of those measures which will provide management with usable data and which do not simply reiterate .hat we can already discern from demographics.

Since this is a new area for us and because our work is still in the formative stages, rather than present a single study I would like to give you a few excerpted examples of what we are doing. As I understand the format, this conference is more a working seminar than a formal presentation of research data. I am hopeful that after hearing about some of the things we are doing, those of you who have experience with these techniques might be able to offer some comments and suggestions which would be helpful to us as we continue to explore ways in which we can use these research techniques. Finally, since much of the data is--at the present time, proprietary in nature, rather than provide specific figures and tables I would like to present our information in a more general way.

Let me begin with a very brief statement of the theoretical rationale which underlies our work in this area. Our approach is based upon what has commonly come to be called the uses and gratifications' paradigm. As we interpret it broadly, the particular attitudes, interests, and expectations which the viewer brings to the television set are crucial determinants to the success or failure of our television programs, Furthermore, it is quite possible (and has been demonstrated repeatedly in the academic research literature) that one program can gratify widely disparate audience expectations. In fact, it is this multidimensionality of programming which is the hallmark of a truly successful and enduring television program that "has something for everyone."

Our next assumption is that the viewing audience is comprised of a variety of sub-groups and chat these groups come together for various reasons to create the audience for any particular show. The attitudes and interests of this audience are not static, however, but are continually evolving and reflect the many social and moral value changes which our overall society is subject to. A better understanding of these changes is important to our developing successful programming and equally important to our best serving the needs and expectations of our viewers.

Finally, we must assume that for any program to succeed in today's highly competitive television environment, the show's situations, themes, and characters must continually change and evolve, otherwise the viewer may grow tired of the program. Of course, deciding exactly which directions to take in making these changes are important advantages.

Using the various segmentation techniques with which you are all familiar, it is possible for us to do the following:

1. Identify the psychographic characteristics of the audience at large;

2. Identify the specific uses and gratifica tions for various audience subgroups,

3. Develop a profile of the audience tor any particular program.

With this information, it should be possible not only to identify who is or is not watching a program but more importantly, why. Once we know this, we can use the information to make more informed and intelligent decisions about ways to improve a program, attract a new audience while retaining the old, or to provide directions which the snow's producers and writers should take the evolution of a program to avoid its becoming tired and stale.

One of the major methods we use to collect this information is through a "social research" questionnaire which is administered as a leave-behind after an in-person interview. Throughout the year, we conduct a variety of in-home national surveys and at the completion of the in-person phase, the respondent is left with a questionnaire containing a variety of activity, interest, and opinion scales, Some of these were developed in-house to tap specific areas of interest and have been tested and refined over repeated administrations. These scales segment the audience on such relevant areas as the uses and gratification of television, tolerance for various controversial themes and depictions, and the viewer's overall evaluation of television. We also include a number of standardized socio-psychological scales which have been validated. Through. repeated administrations we have found the "leave-behinds" to be quite reliable and since they are picked up a few days after the in-person interview by the interviewer, we obtain very high completion rates of over 90%.

One of the ways we have found to use this data is in the development of a program's audience profile. To do this, we asked respondents to select their favorite dramatic show and their favorite comedy show. Naturally, people watch a great many programs on a regular basis but our assumption was that the respondent's favorite program would provide us with insight into the make-up of the show's most stable and loyal audience.

For example, take three popular programs which at face value would appear to appeal to rather similar audience groups. The shows are EIGHT IS ENOUGH, LITTLE HOUSE ON THE PRAIRIE and FAMILY. Segmenting viewers of these shows by a variety of ABC-developed and other standardized socio-psychological scales enables us to develop a psychographic profile of the audience of each program to accompany the traditional demographics.

According to demographics, the three shows appeal much more to women then to men by a two to one margin. Viewing age breaks, we see that LITTLE HOUSE appeals primarily to an older audience, EIGHT IS ENOUGH appeals to a younger audience, and FAMILY is quite strong across all age groups with particular appeal to 18-24 year olds.

While this demographic information can be used to make some generalizations based upon experience with this data, psychographic profiles offer a number of richer and more detailed insights. For instance, the viewer of LITTLE HOUSE is likely to be at the high end of the authoritarian scale suggesting that they are rigid and cogruatic. They have a generally positive view of themselves and they prefer situations which are, in the main, stable and unchanging.

The EIGHT IS ENOUGH viewer is somewhat more moderate in terms of social values than the LITTLE HOUSE audience. EIGHT IS ENOUGH viewers also tend to be much more open-minded and flexible as they skew toward the low end of the authoritarianism scale. This suggests a higher level of tolerance for controversial themes or topics.

Finally, the EIGHT IS ENOUGH viewer has a negative self-image and tends to look toward television as a socializing agent.

Finally, the audience profile for FAMILY suggests a viewer who is also low in self-esteem but tends to be much more progressive on the social values scale than either of the two audience groups. FAMILY viewers, like EIGHT IS ENOUGH viewers, also look towards television as a means of learning about ways to behave in their own lives.

Using these thumbnail profiles, what are some of the programming insights we can obtain from the data? First, although the surface appeal and demographics of the three shows appear quite similar, in fact their audiences have somewhat different expectations. The older, very traditional and inflexible LITTLE HOUSE viewer is probably not very interested in seeing the show deal with many sensitive or controversial topics. On the other hand, the EIGHT IS ENOUGH and FAMILY audiences are much more progressive and flexible in their attitudes and values. We have found that these individuals are far more likely to accept program themes which do deal with relevant social issues.

Although these viewers may not always agree with the content of every television program, they are far more likely to accept the fact that programming should display a range of lifestyles and situations and they are willing to see life depicted in a more realistic way. Finally, both of these groups tend to look toward television to provide them with cues or information which they can use in their own lives. In other words, these viewers enjoy watching how the Lawrences or the Bradfords handle the television version of a situation which they may face in their own lives.

Developing an audience profile for a program is only one way we can use segmentation data. To take another approach, it's sometimes possible to examine how a program's audience distributes across a particular segmentation construct. This technique can indicate some of the elements which are important to a program's appeal and identify those audience groups who are not watching the show.

As I mentioned earlier, uses and gratification theory suggests that different people approach television for different reasons. The more a program can simultaneously satisfy a number of different groups, each with their own set of expectations, the more popular the show. Probably the most notable example of this phenomenon is ALL IN THE FAMILY. Research which was conducted when ALL IN THE FAMILY was not just a successful television show but also a nationwide phenomenon clearly illustrates this. The research indicated that the show appealed across the entire authoritarian range. High authoritarian viewers--characterized as dogmatic and intolerant--loved the show because they identified with Archie and felt that he successfully articulated their view of the world. At the same time, however, low authoritarians--who tend to be open-minded and socially liberal--equally enjoyed the program and felt that the depiction of a bigoted Archie Bunker held his ideas and values up to obvious satire and ridicule. These viewers perceived Gloria and Michael, the two liberal counterparts to Archie's conservativism, as winning each week's ideological debate.

When we conducted our survey last year, the program had undergone some radical changes, Although it had not yet been transformed into the present format of ARCHIE'S PLACE, both Michael and Gloria had left the show. When we examined the distribution of authoritarianism we found that the show retained its popularity among high authoritarians but that low authoritarians were no longer interested in the program. This lower level of interest was corroborated by a softening of the program=s previously solid ratings. If our theory is correct, one way to interpret the psychographic data is that without the liberal characterizations of Gloria and Michael to serve as a counterbalance to Archie's conservativism, low authoritarian viewers lost the program elements with which they had identified. Consequently, the program no longer satisfied their expectations and lost its appeal. Although an alternative explanation might suggest that after almost 10 years on television the show had grown stale, this still would not explain the disproportionate number of high authoritarians who continued to name the show as their favorite comedy.

Psychographic data is also helpful in providing information to the Broadcast Standards and Practices Department which is responsible for establishing guidelines as to the suitability of the content for all entertainment programming. On the one hand, BS&P must permit writers and producers the necessary latitude to deal with a variety of program themes, topics, and depictions. Some of these are naturally serious and realistic depictions of sensitive moral and social issues. On the other hand, the BS&P Department must also be aware that our programming is broadcast to a heterogeneous audience with widely varying notions as to what is or is not acceptable program fare. Using a number of ABC developed scales, we have been able to identify various attitudinal characteristics of the audience which relate to these questions and then have been able to relate them to the more familiar demographic variables which are still the norm in broadcast research.

One of the scales which we used was developed to tap the degree of tolerance a viewer has to the depiction of contemporary program material. A multiple discriminant analysis was used to create the various demographic groupings. The results showed that men were significantly more tolerant about television content than were women. This attitude was consistent both in terms of the depiction of sexual themes as well as an overall notion of tolerance to controversial subject matter. Looking at age, we were not surprised to see that in general, as viewer age groups increased, tolerance to sexual material decreased. However, there was one notable exception. Viewers in the 35-49 year old age group were far more conservative than any other age group including those 50 and above.

When we analyzed these findings in the light of other research we have conducted, a logical pattern emerges, however. First, viewers in the 35-49 year age group are the ones most likely to have children who are old enough to be exposed to such program content. Other studies we have conducted have indicated that the presence of children is a very important variable in determining the viewer's level of tolerance for controversial program material. Therefore, it's likely that this group is responding to the fact that they have children who may inadvertently or otherwise be exposed to program content of which their parents do not approve. The discriminant analysis revealed a number of other interesting facts about this group. First, they are not conservative, per se. On the one hand, they are supportive of the idea that television has an obligation to depict alternative lifestyles and values which may not be the societal norm. At the same time, however, they reject any relaxation of restrictions on the use of explicit language in television programs. We also found that this group is highly supportive of the use of advisory warnings which give the viewer information before the show is broadcast so he or she can make an individual decision about its suitability for themselves and their family. Therefore, one conclusion this study suggests is that controversial material may be acceptable to this group, provided sufficient safeguards are taken by appropriate scheduling times as well as the prudent use of advisories where their use is indicated.

Another of ABC scales was designed to tap the viewer's Television Uses and Gratifications. One of the constructs which this scale measured is the extent to which viewers identified with television characters. The discriminant analysis revealed that younger viewers (15-17) were very likely to identify with television characters as were the 18-34 year olds, though to a much lesser extent. Viewers 35 and above do not identify with characters at all.

Looking at sex differences, women are much more likely to identify with characters than men. While SEX and education levels were inversely correlated with identification was not related to the frequency of television viewing. The extent to which a viewer identified with characters they watched was not related to the amount of television they watched each day.

In terms of the audience's feeling that television is their favorite kind of entertainment and their overall evaluation of the medium, neither SEX nor AGE reliably predict attitudes but SEX and EDUCATION are both inversely correlated to an overall positive evaluation of television.

One of the most interesting results was tapping the audience 's social values and this was also obtained from ~he discriminant analysis. We found that in terms of sexual morality, 15-17 year olds and 18-34 year olds are almost identical in their progressive social attitudes. Thirty-five to forty-nine year olds appear rather ambivalent to the new sexual mores but the 50+ age group was decidedly opposed to the new standards of moral behavior.

Sex differences were apparent when we asked respondents whether or not they would support a return to a more traditional set of moral and social values. Men were likely to disagree with the statement suggesting their satisfaction with the present state of society's values. Women, on the other hand, were far more likely to believe that a return to more traditional standards was preferable. Interestingly enough, there was one area of changing societal standards which men did not approve of. One of the most significant changes in sex roles over the past decade has been the emergence of the working woman. Men are not especially happy with this development, however, and felt that a child was far better off if he or she had a full-time mother at home, Women felt just the opposite and believed that a working mother was not harmful to children as they grew up.

The role of the working woman as a discriminatory variable is most evident when we look at groupings along SES and EDUCATION lines. Both low SES and low EDUCATION groups felt that it was embarrassing to admit that a woman had to work to supplement the family income. Further, these groups also felt that a working mother had a detrimental effect on children, High SES and high Educated groups felt just the opposite and believed chat it was neither embarrassing nor harmful if their wives held full-time jobs.

As I mentioned earlier, we are only beginning to see ways to apply lifestyle techniques and the resulting data to our own set of research questions. The examples I'-a--mentioned are only a few of the approaches which are currently being explored. As an example of additional uses of this research approach, imagine the use of segmentation analysis to provide a viewer profile for various day parts. The question is whether people have different expectations for their television programming during prime-time access than they do during the dinner hour which precedes it or during actual prime-time which follows. If we are able to obtain these profiles and then subject possible programs to a similar profile analysis, it would be possible to match-up the best program which--based upon its perceived attributes--would best meet the audience's most salient expectations at any point in the broadcast day. This is made a lot easier if we are working with programming which already exists and with which the audience is already familiar. The development of entirely new programming is a much more complex matter although psychographics can obviously provide us with additional insight which could be helpful in the development of new shows,

Although our experience with lifestyle research is at this point limited, I can say that we have already been able to see the enormous potential which audience segmentation offers. When we combine these with standard demographics, the result is a richness and dimensionality in our research studies which enable us to examine a question with added insight and with increased perception.