An Analysis of Moviegoers By Life Style Segments

Glen Homan, Leo Burnett U.S.A.
Robert Cecil, Leo Burnett U.S.A.
William Wells, Needham, Harper, and Steers
[ to cite ]:
Glen Homan, Robert Cecil, and William Wells (1975) ,"An Analysis of Moviegoers By Life Style Segments", in NA - Advances in Consumer Research Volume 02, eds. Mary Jane Schlinger, Ann Abor, MI : Association for Consumer Research, Pages: 217-230.

Advances in Consumer Research Volume 2, 1975      Pages 217-230


Glen Homan, Leo Burnett U.S.A.

Robert Cecil, Leo Burnett U.S.A.

William Wells, Needham, Harper, and Steers

[Data, staff, and computer time for this research were supplied by Leo Burnett U.S.A.]

[Glen Homan is Account Executive, Leo Burnett U.S.A.]

[Robert Cecil is Assistant Research Analyst, Leo Burnett U.S.A.]

[William Wells is Director of Corporate Research, Needham, Harper and Steers.]

The data in this study were used to demonstrate two points: the advantages of using life style data in addition to traditional demographics; and the advantages of sub-segmentation within a heavy user group. In the first phase of this analysis, heavy moviegoers were compared with nonmoviegoers on the basis of demographics and life style. In the second phase, the heavy moviegoers were separated into three groups via cluster analysis. These three groups were also compared on the basis of demographics and life style.


The findings reported in this paper are presented as empirical illustrations of two methodological points: (l) the advantages of using life style or "psychographic" information in addition to traditional demographic measures and (2) the advantages of sub-segmentation within a heavy user group. The paper shows that life style studies give a much fuller descriptive picture of the consumer than do demographics alone, that life style data may yield important dimensions that demographics alone may miss, and that life style data can give some idea of how products fit in the context of consumers' lives. It further shows that sub-segmentation both avoids the assumption that all heavy users are alike, and tells which variables really "go together" to characterize distinct consumer groups.

The data to illustrate these points come from a national study of 3700 married males, conducted by the Leo Burnett Co. in 1973. Respondents were members of the Market Facts Consumer Mail Panel. The sample is not intended to represent the entire movie-going population; it excludes married females and all single people.

The analysis was performed in two stages. In the first stage "heavy moviegoers" -- defined as those who had gone to the movies at least nine times in the past year -- were compared with nonmoviegoers - or people who had not gone to the movies at all in the last year. The groups were compared both in terms of demographics and in terms of life style, so as to give an idea of the differences between the two kinds of information. The life style analysis was based on two types of items: a series of 150 attitude, interest, and opinion statements (AIO's) on which the respondent indicated his degree of agreement or disagreement on a six-point scale; and a set of 56 activities on which the respondent indicated the number of times he participated in that activity in the past year.

The second stage of the analysis was a sub-segmentations by means of cluster analysis, among heavy moviegoers. This analYsis produced three segments of heavy moviegoers, described below. The purpose of this analysis was to show the advantages of sub-segmentation within a heavy user group.

Comparison of Heavy Moviegoers (N=758) versus Nonmoviegoers (N=1,122)

Demographics. Heavy moviegoers differ from nonmoviegoers on a number of demographic dimensions

Heavy moviegoers are younger: [Nonmoviegoers results are listed first, followed by Heavy Moviegoers results]

Under 25    1%   15%

25-34         7        32

Heavy moviegoers are better educated:

College graduate      8% 19%

Post-grad. degree    5      8

Heavy moviegoers have higher incomes:

$15,000-$24,999        14% 27%

$25,000 or more             3       7

Heavy moviegoers are more likely to live in an urban or suburban setting; nonmoviegoers are more likely to live in rural areas:

Central City*      14%      22%

Suburban*          17         30

Rural                    31         14

*In SMSA of 500,ooo or greater

Heavy moviegoers are more likely to be in professional or managerial occupations; nonmoviegoers are more likely to be craftsmen, farmers, or retired:

Professional/Manager    26%    36%

Craftsman, etc.               25        20

Farmer                           10          3

Retired                           25          3

Heavy moviegoers tend to belong to larger households.

2 member household                 53%     27%

3 member household                 18        24

4 member household                 14        25

5 or more member household   15        23

Life Style. As with demographics, a number of life style differences arise between heavy- and non-moviegoers. Numbers in charts are percent agreeing with each statement (i.e., answering 4, 5, or 6 on 6-point scale).

Heavy moviegoers are more ambitious and optimistic. They have a richer fantasy life and are more self-confident:

My greatest achievements are still ahead of me.                                                      49%    80%

Five years from now the family income will probably be a lot higher than it is now.  63      88

I expect to be a top executive within the next 10 years.                                           11       37

I would like to take a trip around the world.                                                            44       76

I'd like to spend a year in London or Paris.                                                             19       42

I would do better than average in a fist fight.                                                            33       55

I like to be considered a leader.                                                                               60       82

We will probably move at least once in the next five years.                                       24       49

Heavy moviegoers are more active and more socially-oriented:

I like parties where there is lots of music and talk.              37% 65%

I like to play poker.                                                           30     56

I would like to take a lesson in my favorite outdoor sport.  39     66

I always have the car radio on when I drive.                        49     82

Heavy moviegoers have more "swinging" interests:

"Playboy" is one of my favorite magazines.                                                   13% 48%

I think I'm a bit of a swinger.                                                                        13     33

There are situations where sex outside the marriage can be a healthy thing.   15    35

Heavy moviegoers are more inclined toward new and sporty possessions:

The new styles turn me on.     22%  46%

I like sports cars.                     29      58

Nonmoviegoers have more traditional values:

The U.S. would be better off it there were no hippies.                      73% 46%

Liquor is a curse on American life.                                                   59     36

I often wish for the good old days.                                                   63     43

A woman should not smoke in public.                                              62     36

There is too much violence on TV today.                                         77     57

If Americans were more religious, this would be a better country.     83    63

Nonmoviegoers are less secure and more pessimistic:

I don't like to take a chance.                                                                           71% 51%

I don't like to fly.                                                                                             40    25

I am really looking forward to retirement.                                                         75    58

A person should stay home whenever there is the slightest symptom of illness   55   28

Activities. Heavy moviegoers participate more often in virtually all activities monitored - except attending church:

                                         Avg. No. of times participated in past 12 months

Went bowling                                     3.4        9.8

Had a cocktail before dinner             11.4      18.5

Cooked outdoors                             10.6      19.3

Listened to a tape or record              22.5      42.5

Attended a sporting event                   4.3       10.3

Served wine with dinner                      3.1         8.0

Went swimming                                  3.7       11.4

Brought home work                            7.4       14.5

Attended church                                33.2       25.3

Conclusion. Life style information enables us to obtain a richer and more complete portrait of the heavy moviegoer than the information provided by demographic data alone. While some of the life style differences could have been inferred from the demographic profile, it seems highly unlikely that all of them could have been.


The second stage of the analysis was to employ a cluster analysis program that groups people together who respond in similar ways to a set of items. [The specific technique used in the Leo Burnett cluster analysis program is a sift-and-shift technique developed at Market Facts, Inc. by Rich Johnson and described in a speech given February 15, 1972 for the New York Chapters of the American Statistical Association and the American Marketing Association, and entitled "How Can You Tell If Things Are Really Clustered?"] In this case the items used were the 31 life style items on which the heavy moviegoers showed the most variance. -It seemed likely that if heavy moviegoers do indeed consist of several subtypes, these types would be revealed by analyzing the items upon which heavy moviegoers agreed least.

Membership in a given cluster is determined by the respondent's own answers, and is not predetermined by the researcher. The number of groups the program finds is predetermined by the researcher, however; so the researcher must compare several groupings to determine which is best. The criteria for selecting a solution were that each group should be coherent, internally consistent, reasonably large, and stable in a split-half test. A further criterion was that the cluster solution chosen should explain variance in the data better than differences produced by other cluster solutions. The three group solution fit these criteria. The groups were named the Mid American, the Mid Scale Swinger, and the Upscale Striver. They were examined for demographic differences and for differences on life style variables. Our sample included 272 Mid Americans, 216 Mid Scale Swingers, and 270 Upscale Strivers

The demographic comparisons elicited some clear differences among the subtypes of heavy moviegoers. Level of education completed by the Upscale Strivers was much higher than that for any other group. The pattern of family income was similar: the Upscale Strivers were highest.

The nonmoviegoers are clearly older than any of the three heavy moviegoer groups; sixty one percent of them are over 54 years of age. The Upscale Strivers are somewhat younger (their mean age is 39 ); and the Mid Scale Swingers are the youngest group (70% are under 35; 25% under age 25 ) . The Mid Americans, however, show no clear trends by age, being evenly spread across all age categories. The Mid Scale Swinger is most likely to live in an apartment (although only 16% of them do). Nonmoviegoers are least likely to live in an apartment (only 3% do). Nonmoviegoers and Mid Americans are much more likely to live on a farm than the other groups; Upscale Strivers are more likely to own more than one car.

Life Style. When looking at both life style and demographic differences, the differences between groups are most simply explained by the following model:



The vertical axis in this model is social class, which was reflected in the demographics, and is also reflected in the answers to a number of the life style questions. The horizontal axis represents a number of attitudinal dimensions not covered in the demographic description; the labels "liberal or conservative" summarize a variety of dimensions including attitudes toward sex, attitudes toward churchgoing, and political orientation. The model shown above is ar oversimplification because it doesn't show the changes in relationships between groups on some of the attitudinal dimensions. These changes in relationships will become clearer as we look at each of the groups in more detail.

First we will look at areas that differentiate nonmoviegoers from all types of heavy moviegoers. The earlier analysis assumed that if there was an overall difference between heavy moviegoers and nonmoviegoers, the difference would hold true for all types of moviegoers. This analysis is designed to show that there may be more than one type of heavy user of a product; and that different types of heavy users may behave differently (even like nonusers in some cases).

One area where nonmoviegoers were different from all three types of heavy moviegoers was in their view of the role of women. All three types of heavy moviegoers were more willing to accept changes in women's roles than were nonmoviegoers; as the following items show: [The percentages shown for all the AIO's in this paper represent agreement with the items, i.e., 4, 5, or 6 on a six-point scale between "Definitely Disagree" and "Definitely Agree."]


We also found that nonmoviegoers are more traditional in a number of other ways, on items ranging from family finance to the role of the church. The following list illustrates a few of these areas:


We found that all three types of moviegoers engage in many more activities than nonmoviegoers, as is illustrated by the following items: [The scores on activities are derived from a seven point scale measuring the number of times each respondent engaged in an activity in the past year. The figures shown here are the mean number of times members of each group engaged in a given activity in that year. The categories had different intervals, however; these "means" are approximations.]


This tendency extends across a wide range of items, the list could be extended if space permitted presentation of more data.

The final area of differences between nonmoviegoers and all three types of heavy moviegoers was in degree of self-confidence. All three types of heavy moviegoers are more self confident than nonmoviegoers, as these items show:


The heavy moviegoers have more generalized faith in themselves.

In summary, nonmoviegoers are different from all three types of heavy moviegoers in several respects. They are older than any of the groups of heavy moviegoers. They are also more traditional, less willing to accept changes in the role of women, less active, and less self confident.

The Mid American. The Mid Americans lie between the nonmoviegoers and the other two heavy moviegoer groups on the life style items. This is not to say that the Mid Americans do not have a distinct identity; the range from nonmoviegoers to the Mid Scale Swingers and/or Upscale Strivers is so great on many items that there is considerable room to fall in between. The Mid Americans use this latitude: In some areas their scores are similar to the scores of the other types of moviegoers; In other areas they resemble the nonmoviegoers.

Both the nonmoviegoers and the Mid Americans are more conservative about sex than the Mid Scale Swinger or the Upscale Striver:


This finding is particularly interesting when viewed with one of our earlier findings. The Mid American is more like the other heavy moviegoers in accepting changes in the role of women, but at the same time disagrees with them sharply on being liberal about sex. The Mid American also bounces back and forth between these two poles on a number of other items relating to traditional values and personal conservatism.

We have shown some of the items where the Mid American is similar to the other heavy moviegoer types; here are some items where the Mid American is more similar to the nonmoviegoers:


We also find that the Mid Americans have more limited aspirations than the other heavy moviegoers:


Further, the Mid American doesn't engage in much fantasy, and is generally rather cautious:


Finally, the Mid American does in fact lead a very stable, settled life, with a definite routine:


In summary, when we compare the Mid American to the Upscale Striver and the Mid Scale Swinger, we find that the Mid American is more conservative about sex, has more traditional values in a number of other areas as well, and has more limited aspirations. The Mid American is also more physically and mentally settled into a routine. He actively rejects both fantasy and major changes in his life (except retirement), and doesn't like taking risks.

The Mid Scale Swinger. The items that define the Mid American and nonmoviegoers at one end of a scale also define the Mid Scale Swinger and the Upscale Striver - at the other end. Stated briefly, the Mid Scale Swinger and Upscale Striver share the following characteristics: they are more physically mobile, lead active fantasy lives, don't follow a set routine, have high aspirations, and have more liberal attitudes toward sex, the role of women, and most other traditional American values.

The question that now arises is: what items differentiate the Mid Scale Swinger from the Upscale Striver, and which areas characterize each particular group? We already know a few demographic differences: The Upscale Striver has quite a bit more education and income, while the Mid Scale Swinger is younger. A number of life style variables differentiate between the two groups as well. The first of these is that the Mid Scale Swinger is very low on social and community activities, as shown in the following chart:


As a matter of fact, the Mid Scale Swinger is even lower on these activities than is the Mid American.

The Mid Scale Swinger is also materialistic, and somewhat dissatisfied and mistrustful.


Finally, the Mid Scale Swinger is more likely to do manual repairs on his house and car than the Upscale Striver:


He is, however, less likely to help with clean up projects around the house.

The Upscale Striver. Now let us turn to the areas that differentiate the Upscale Striver. First, the Upscale Striver is very hard working. He views his job as fulfilling and as a challenge, and he is very involved in his work.


The Upscale Striver is also the most active group in community affairs.


The Upscale Striver is also consistently highest on activities for personal enrichment and entertainment (except TV):


This group, in short, is tremendously active in many different areas. This finding is consistent with the finding that the Upscale Striver has very high self esteem and considers himself a leader in his community:


In addition, the Upscale Striver is innovative, and enjoys challenge and excitement in his life.


In summary, the Upscale Striver is hardworking, very involved in his job, and yet also very involved in his community and in his own personal enrichment and entertainment. The Upscale Striver is high in self esteem, and views himself as a leader, he is also innovative and likes challenges and excitement.


One of the purposes of this analysis was to show some of the types of information that life style items can add to the traditional demographics. The portraits of the groups were intended to illustrate the depth of information found in using life style items as opposed to using a demographic analysis alone.

The difference between life style analysis and demographic analysis should be particularly clear in the distinction between the Mid American and Mid Scale Swinger, where the only demographic difference is a slight difference in age. When life style items are used to compare the two groups, however, very clear differences emerge. Some of these differences, like the more traditional attitudes and orientation toward sex of the Mid American, might be posited from the slight difference in age. But we also find that the Hid American has more limited aspirations, engages in very little fantasy, and prefers (and has established) a life with much more routine. These differences cannot be expected on the basis of a slight difference in age. The Mid Scale Swinger, on the other hand is much lower on community and social activities, and is more materialistic, mistrustful, and dissatisfied with life. These differences cannot be projected from the demographic profile alone.

These data also illustrate that treating heavy users of a product or a service as a single group may greatly oversimplify the data. It is possible to conclude that there is a difference between heavy users and nonusers when that difference does not exist for one or more of the subgroups within the heavy user group. The following items illustrate the point:


On the other hand, it is also possible to conclude that there is no difference between heavy users and nonusers - when in fact the heavy user group resembles nonusers only because it consists of several subgroups pulling in opposite directions. That phenomenon is illustrated by the following examples:


The only way to get around these problems is to try one or more subsegmentations. It is our belief that these sub-segmentations are more likely to be successful if one uses a statistical technique which groups respondents on the basis of their own patterns of response, rather than on the researcher's a Priori ideas. That, however, is the topic of another paper.