From Life Style to Value Systems to Simplicity

ABSTRACT - The present paper discusses the use of value dimensions for the explanation of consumer behaviour.

Citation:

Flemming Hansen (1998) ,"From Life Style to Value Systems to Simplicity", in NA - Advances in Consumer Research Volume 25, eds. Joseph W. Alba & J. Wesley Hutchinson, Provo, UT : Association for Consumer Research, Pages: 181-195.

Advances in Consumer Research Volume 25, 1998      Pages 181-195

FROM LIFE STYLE TO VALUE SYSTEMS TO SIMPLICITY

Flemming Hansen, The Copenhagen Business School and Gallup A/S

[Numerous people at Gallup, Denmark have been instrumental in the development of this project. I would like to thank all.]

ABSTRACT -

The present paper discusses the use of value dimensions for the explanation of consumer behaviour.

In the paper, dimensions used in different value systems are compared. Basically, it is found that when one forces a value system into a two-dimensional solution, almost regardless of the question asked, one ends up with two quite similar dimensions, the one ranging from modern to traditional, the other relating to individualism versus valuing social factors, or egoism versus hedonism.

This conclusion has led to the development of a very simple and operational system which basically relies upon identifying two major dimensions, with the added possibility of studying a larger number of values, interests and behavioural grouping, which can be read into the two-dimensional space.

In the paper, it is described how the two-dimensional solution defines nine segments and a more detailed account of the development of the system is given as well as its validity is documented based upon experiences in such different areas as media, personal demographics, consumption patterns, political affiliation, etc.

Following this, the reliability is demonstrated using the Danish data from 1994, 1995 and 1996 value dimensions. Based uponapprox. 10.000 observations from each year it is found how the respondents in one year in more than 95% of the cases are assigned to the same segments the following year regardless of whether a model based on 1994, 1995 or the 1996 data is used.

1. BACKGROUND

In the early days of consumer behaviour research, there was great interest in explaining consumer choices based upon psychological personality measurements. Instruments such as Edwards’ (1954) Personal Preference Schedule (EPPS), the Minnesota Personal Preference Schedule (Graham, 1990) or Thurnstone’s (1953) Temperament Schedule (TTS) were among the scales used. The success, however, was limited compared with the use of demographic variables alone. In a classical study, Evans (1959) was able to improve the discrimination between Ford and Chevrolet owners only slightly with the use of psychological variables. Later Kassarjian (1965) concluded that we are able to explain 5-10 per cent of the variation in consumer behaviour with the use of psychological measurements. One might suggest that it could be fruitful to start looking for variables that can explain the remaining 90-95 per cent.

Lifestyle research grew out of this and other observations. Rather than using standardised psychological tests, people like Wilson (1966), Wells (1966) and others thought that developing instruments based upon consumer and consumption-relevant items could bring forward consumption-relevant traits that would be more useful for studying consumer behaviour. Working with AIO (Attitudes, Interest and Opinions) variables, they identified segments such as the "the happy housewife," "the price-conscious shopper", "the affluent consumer", etc. However, this work, too, had limited success in explaining consumer choices, particularly at brand level. Moreover, the segments did not seem to be very stable over time, and depending on the particular questionnaire used, the segments came out quite differently.

In the late 1970s and the early 1980s, researchers began to work with more general values in an attempt to identify variables that could explain consumer behaviour and which at the same time were relatively stable over time. Tolman’s (1951), Maslow’s (1959) and Rockleach’s (1973) work on values was influential, as was Riesman’s (1961) "The Lonely Crowd," which introduced the concept of "inner" and "other directedness."

The Stanford Research Institute’s Values and Lifestyles (VALS) program was one of the, first applications of psychographical classification to consumer behaviour. Later, a parallel system called RISC emerged in Europe, and still later, a second, revised VALS system was introduced in the States. French researchers also developed their own systems, particularly the group around CCSA. Another early U.S. system is Yankelovich’s MONITOR with dimensions much like RISC.

These early systems were quite complicated and worked with a large number of more or less overlapping values. Other research agencies came out with their own value programmes, most of them less complicated than the early ones. Today, 6-8 standard systems can be identified in Scandinavia alone, and in addition to this, work has been published on value systems developed for special product groups or special segments. However, in most of the models, the complicated communication of findings and problems with the maintenance of stability in the dimensions over time and across countries has led to simplifications. The main tendency has been to develop a two-dimensional space in which consumption, respondents and values are placed. We shall review some of these systems here.

The best documented value system is VALS. It was developed at the Stanford Research Institute and has subsequently become the most used system in The States. The primary dimensions in the system are strugglers (with minimal resources) versus actualizers (with abundant resorces) and principle oriented versus action oriented (Peter & Olsen, 1993).

The original RISC version was inspired by the work with MONITOR. Later RISC has undergone a number of changes and until recently relied primarily on only two dimensions: "openness to change" versus "resistance to change" (modern versus traditional) and on "an ethical" versus "a hedonistic" dimension one which also could be labelled inner versus outer-directedness (Hassan and Ladet 1994). An example of the RISC representation in the two dimensions is shown in fig. 1.

CCA is a French system which works with dimensions characterised as "change" versus "stability," and "material" versus "immaterial" values (Askegard 1993). In contrast with the commercial value systems for which very little information is publicly available a study of interesting dimensions has been published by Grunert & Juhl (1995) based upon work by Schwartz and Bilsky (1990). Also here a two dimensional representation: "openness to change" versus "conservation" and "self-transcendence" versus "self-enhancement" can be found. The theoretical considerations underlying this system are highly relevant in connection with the Kompas system. The Kompas system is a commercial segmentation instrument introduced by Gallup in Norway, Denmark and Sweden. Also this system, to be described in more details subsequently, works with two dimensions: Modern versus traditional and individual versus socially values..

Danish Attitudes (danske holdninger) is a RISC-inspired system. Again we find an East/West dimension concerned with modern versus traditional attitudes and a South/North dimension concerned with individuality and inner-directedness versus collectivity and outer-directedness (Vilstrup 1959). A similar Danish RISC version is Minerva, recently introduced by A.C. Nielsen/AIM (Nielsen, 1997).

FIGURE 1

RISC/ACE DIMENSIONS (HASSAN AND LADET 1994)

Valuescope is a Swedish system developed by Hans Zetterberg (1995). Zetterberg was one of the founders of the RISC system. Again we find a modern versus a traditional dimension, but in contrast to the other systems, there are two more dimensions; one relating to fidelity versus pragmatism and the other reflecting humanism versus materialism (Fig. 2). The same two dimensions can be derived from the data on the basis of which RISC, VALS, Kompas and other systems are constructed, but when the data are forced into two dimensions only, they collapse into an inner/outer-directedness or egoism/altruism dimension. Actually RISC is presently introducing a third dimension into its system.

Among the more special systems are those developed by Bruns° et al. (1996) on food-related behaviour, by Williams (1991) on alcoholic drinks, and by Rentorff (1995) on fashion products. The dimensions that appear in these and other specialized applications are similar to those described above.

Table 1 summarises the dimensions of the different systems.

Drawing upon the Danish experience one might, like Zetterberg, add a third variable to the description of the segments. As can be seen from fig. 3, after the first three variables in the Danish value battery, each of the following variables accounts for not much more variance than a single question can explain. However, in this system it has been chosen also to concentrate on two dimensions, of which the second one more or less combines social versus individual orientation and altruism versus egoism.

2. KOMPAS: A BASIC TWO DIMENSIONAL VALUE SYSTEM

In Denmark a value system labelled Gallup Kompas has been developed. Basically, the system is based on a large media/marketing data base including opinions, values, interests, etc. App. 1000 items are included in the questionnaires.

The structure of the total questionnaire is shown in Table 2. In deriving the Kompas value system, particularly the attitude, interest and activity variabes are important. However, further analysis of the values and their application can draw upon information contained in the remaining questionnaire.

Data collection is related to the Danish Media Survey, a study conducted with the use of 50.000 annual CATI interviews. A subset of the respondents from this study of 17.603 (1996) respondents are completing a self-administered questionnaire including the measures in Table 2. All information of the upper 3 parts of Table 2 is gathered from all respondents, whereas the information on brands and durables is divided into four groups where each group of questions is administered with one quarter of the total sample..

The focus of Gallup Kompas is to give a picture of the basic values in the population. Any model is a simplification of the reality, intended to offer understanding of the basic tendencies and differences. This is the "raison d’Otre" and the limitation of any model. This is also the aim and the limitations of this Kompas model. Kompas gives a very valuable insight and overview, but we do not believe that one model fits all situations. There will be markets and situations where it is wise to produce more specific models as supplements to the basic model.

The system is based on a two-step factor-analytical procedure resulting in two basic dimensions constructed on the basis of 22 questions. The dimensions reflect "modernism" versus "traditionalism" and "social" versus "individual" orientation. Each respondent is given a score on each of the two dimensions.

FIGURE 2

THREE-DIMENSIONAL VALUE SPACE, DEFINING A TOTAL OF 8 (2x2x2) DIFFERENCE SEGMENTS (HERE EXAMPLIFIED WITH THE "MATTER OF FACT "SEGMENT" IN THE SOUTH-EAST LOWER CORNER) (ZETTERBERG, 1995)

TABLE 1

MAJOR DIMENSIONS IN DIFFERENT VALUE SYSTEMS

The derivation of the two dimensions is described in a little more detail here. In the questionnaire, values are measured on 6 point scales, interests (in various activities, product-issues etc.) on 5 point scales and activities are self-reported frequencies on a pre-coded scale ranging from 1 daily over weekly, 5-6 times a week, 3-4 days a week, once or twice a week, less frequent, never. The Kompas system which is the basis for the present paper was developed based on the 1994 questionnaire. Later some changes have occurred in the contents of the questionnaire. Particularly in early 1995 systematic work was done to secure that the value battery was broadly representative of values in society, a work which led to some additions in the battery included in the version presently in use.

In the process of developing the dimensions, analyses were carried out on attitudes, interest and behavioural measures separately and on the three groups of measures combined. It soon became evident that the solutions that emerged were quite similar regardless of the battery used, and when the three batteries were combined the value questions were by far the most discriminating and contributed by explaining most variance. Consequently it was decided to derive the solution based upon the value items alone. In this process the initially total number of 120 values was reduced to 90 values. This was done by eliminating a large group of values (31) particularly relating to telecommunication. Subsequently the battery was further reduced to 59 items. This was done by eliminating items that were obvious duplicates, items that is was judged might be misinterpreted and items with low discriminative power. With these 59 items it was found that two (or three) dimensions explained a large proportion of the total variance (Figure 3). At this point it was decided to focus on a two-dimensional solution and to add to that a more detailed value structure profile (to be described in a later section).

FIGURE 3

AMOUNT OF VARIANCE EXPLAINED IN DANISH 1995 VALUE DATA BY SUCCESSIVE FACTORS (59 VARIABLES, 6000 RESPONDENTS)

TABLE 2

DATA IN THE GALLUP MARKETING INDEX 1996 (N=APP. 17,600)

For the purpose of simplifying future analysis of the data it was desirable to reduce the number of variables from the initial 59 ones. This was done by concentrating on those items relating significantly to the two first unrelated factors in the initial solution and with the highest discriminatory power. The latter meant that items were excluded where respondents tended all to agree or to disagree. Actually, in this process, a number of otherwise "obvious" items were deleted. Seemingly they were too obvious to the respondents also.

This process led to the establishmentof a 22 item battery based upon which the two dimensions were identified. This battery is kept unchanged in the questionnaire from year to year. Thus by the time of the presentation, results with this battery from the first half of 1997 are available.

FIGURE 4

PLOT OF RESPONDENTS

EACH POINT REPRESENTS THE SCORE FOR ONE RESPONDENT ON THE DIMENSIONS: TRADITIONAL VS. MODERN AND SOCIAL VS. INDIVIDUAL. THERE ARE APP. 6000 POINTS.

The solution presented here, and used in future work and in commercial applications, is the two-dimensional varimax rotated solution based upon the 22 selected items. With this solution a factor score co-efficient for each respondent on each of the two dimensions is computed. When these scores are plotted they result in a diagram like the one illustrated in Figure 4.

At this point it should also be mentioned, that a further reduction of the number of questions necessary to identify the dimensions has been made. A total of eight has been selected. This is done with the purpose of making the instrument usable in various ad-hoc projects without overloading the questionnaire with too many value items. The items here are selected from the 22 ones with four relating to each of the two dimensions: Two strongly positively loaded, two negatively loaded for each of them. This reduced battery, it turns out, reproduces the grouping of the respondents based upon the larger questionnaire very well. Actually, in 1996, 98 per cent of the respondents were grouped similarly when the reduced battery was used as with the grouping based upon the 22 questions.

In making the model, one of the most important decisions is the number of groups the population is divided into. There is always a difficult balance between realism (as many groups as possible) and simplicity (as few groups as possible). As seen in Figure 5 we have chosen to divide the population into 9 groups. This is done by disregarding the more or less neutral centre and by dividing the remaining respondents into 8 pie slice segments. These segments have been labelled as shown in Table 3. Each respondent has a score on each of the two dimensions and a number indicating the segment to which he/she belongs. These very simple data can be used in different analyses of the respondents.

To get a feeling for the nature of the diagram, it is possible to look at respondents depending upon various criteria. In Fig. 6 the voters of major political parties are illustrated. "A" The Social Democratic Party, is socially oriented and traditional. "Z" is a very traditional right wing party. "C" is the major conservative party opposing the social democrats. "B" are the social liberals, while "F" are more left-wing oriented socialists.

In the same way, geographical regions are shown in Fig. 7. The capital Copenhagen is more modern and the most rural regions are more traditional.

In Fig. 8, occupation is shown. Independent business persons and students are individually/modern oriented. Workers are traditional/socially oriented and pensioners are traditional and individually oriented. Finally white collar workers are modern socially oriented.

Illustrated in a slightly different manner, income or the different segments is shown in Fig. 9. Here, like in VALS, the more affluent are found in the modern segments.

Fundamentally however, Kompas is an instrument to analyse the profile of different consumers, users of the different brands etc. When this is done, a diagram like the one for BMW owners in Denmark shown in Fig. 10 emerges. Here the colour and the size of the pie slices signify the relative importance of the particular segments. The ownership is more widespread among the modern individually oriented and low among the traditional socially oriented consumers.

FIGURE 5

DIMENSIONS AND SEGMENTS IN THE GALLUP VALUE SYSTEM

TABLE 3

THE SIZE OF THE 9 KOMPAS SEGMENTS IN DENMARK

FIGURE 6

POLITICAL PARTY AFFILIATION

FIGURE 7

THE LOCATION OF DIFFERENT DANISH REGIONS

FIGURE 8

THE LOCATION OF DIFFERENT OCCUPATIONAL GROUPS

FIGURE 9

INCOME IN THE DIFFERENT KOMPAS SEGMENTS

FIGURE 10

KOMPAS DIAGRAMME OF BMW OWNERS IN DENMARK

3. A THEORETICAL FOUNDATION

The work by Grunert & Juhl (1995) is mentioned earlier. Underlining their thinking is a model proposed by Schwartz & Bilsky (1990) and in more detail by Schwartz (1992). The model which originates in traditional psyhological and social-psychological value theory proposes 10 value domains ordered along two dimension, one being a conservation versus openness to change and the other a self-transcendence versus self-enhancement dimension. Dimensions that appear to be very similar to the individual/social and modern/traditional dimensions of the Kompas system.

The Schwartz value domains and the location in the two-dimensional value systems are illustrated in Figure 11. The division of this universe differs slightly from that of Kompas so that the triangular value domains do not have a one to one relationship with the similar Kompas value domains. The relationship, however, is obvious. "Self-direction" and "universalism" reflect the modern/socially oriented segment, and "achievement and power" reflect individually orientation as well as "security," "tradition," and "conformity" reflecting the traditional value domains are examples.

The similarity between the 57 values measured to identify the 10 motivation domains in the Schwarz system and the 59 values measured in the Kompas system to identify the Kompas segments is interesting also. Even though the Schwartz scales are based on more basic value questions items are quite comparable. For instance "to have a guiding principle in my life" (Schwartz) corresponds to "I have more direction than my friends" (Kompas), similarly "equality (equal opportunities for all)" (Schwartz) is quite similar to "everybody should have equal rights" in the Kompas system, and so is "ambitious (hard-working, aspiring)" (Schwartz) similar to "I spend a lot of energy on my job" in the Kompas system.

A study where the Schwartz value inventory is administered to the same respondents who have answered the Kompas value questionnaire is in progress. This will enable a more direct analysis of the relationship between the value dimension, the triangular domains and the individual items of the two systems.

An interesting study using the Schwartz value inventory has been reported recently (Beckmann & Kilbourne 1997). Here the value inventory is related to the power triad of the "Dominant Social Paradigm" based on a political, an economical and a technological dimension (Beck 1995, Kilbourne 1995). These three dimensions relate intuitively to the modern-traditional as well as the individual-social dimensions, and it is shown that they relate significantly to those in a manner, where the economic dimension of the "Dominant Social Paradigm" relates positively to the openness to change dimension and self-enhancement dimension. In the same manner the technical dimension of the social paradigm relates negatively to self-transcendence (and positively to self-enhancement). Finally the political dimension reflecting legislation, democracy and political equality, the democracy aspect relates positively with traditionalism.

FIGURE 11

VALUE DOMAINS AND THE STRUCTURAL RELATIONS AMONG THEM (BECKMANN & KILLBOURNE 1997)

4. THE NINE SEGMENTS

Using the information in the total questionnaire (Table 2) the segments can be described as follows:

The moderns

The moderns are much drawn to all new things. For instance, their homes are veritable showcases for the most recent advances in consumer electronics. To the moderns, the distinction between what is HOT and what is NOT is crucial. In the typical modern family, the male head of the household holds an advanced degree and works as a senior executive, while his wife is self-employed. They have two children, both in their teens. They are pro-NATO, career-oriented and vote Liberal just like mum and dad.

Unlike earlier generations of teenagers, the young moderns desperately seek security. Typical representatives of "The gapless generation," they love advertising and dream about the sixties and seventies. Rebellious behaviou, body piercing and The Artist Formerly Known As Prince are NOT. Morten Korch films (popular Danish comedies from the fifties) and invitations in thick envelopes are HOT.

The moderns belong to the affluent part of the population and are great spenders when it comes to durable goods. The most recent acquisition could well be a black Peugeot 406, which looks great in the driveway in the nice neighbourhood where they live. Once inside the house, you are likely to come across on impressive selection of PCs, fax machines, cellular phones and answering machines. All in a home which blends classical modernism with kitsch and a little bit of romance. On the table you will find a quality daily newspaper, professional journals and the Danish equivalent of the Financial Times. The CD player plays soft rock and the tickets for next weekend’s trip to Prague have just arrived. To the moderns, knowledge and culture is power. They play golf and tennis and participate in other socially acceptable leisure activities such as going to the opera.

As with other things, the moderns have an expensive taste in clothes. Their wardrobes feature designer clothes labelled Hugo Boss, Calvin Klein, etc. Manufacturers of exclusive personal care items are also well represented in their homes. The nondurable consumer goods include Italian gastronomic specialities, some of which have been bought in Denmark, while others-olive oil and spices-have been brought home on the frequent trips abroad. The possession of several credit cards makes it easier for the moderns to act on impulse when they come across something they would like to buy. But then again, expensive habits are not a problem for those who can afford them.

For all 9 segments descriptions like the one above exists. For the sake of brevity, they are only sketched here.

The traditionals

There are many elderly people in this segment, and many single-person households. After a long life together, the husband has died, leaving his wife to fend for herself. The home is cosy. In these families cooking and baking are popular activities. For many traditionals, there is no lack of free time to pursue these interests: In many cases, one or both household members are out of job. The children have long since moved out.

The individually oriented

For better and for worse, the slightly derogative term "bourgeois" is apt here. The individually-oriented are frequently mature men. There are two people in the house which they normally own themselves. Only few have a higher education, but many are trade educated. They are formally married with an annual household income in the range of DKK 200-400.000. They can afford a car or motorbike, which is used for driving around in Jutland, where half of the individually oriented live.

The individually oriented are the kind of people who will only call the plumber if a tornado has passed by. They handle power drills, grinding machines and a wide selection of saws and other tools for indoor and outdoor repairs and maintenance expertly. Some of the tools are not inexpensive. They play pools, lotto, etc. and have a savings account in the bank.

The socially oriented

This segment has more women than men. On the door there are 3-4 names. The children are mostly between 13-30 years. The socially-oriented are often white-collar workers employed within the public sector. They are active and engaged people. They prioritise closeness, honesty and healthy human relationships. They often live in cities. Politically they are affiliated with the social democrats or the left-of-centre parties. They are "green" idealistic families where everybody helps with everything. The man in the house knows how to cook, do house-cleaning and care for the children. The socially oriented are united by a great interet for aesthetic, artistic, cultural and intellectual activities.

The modern/socially oriented.

Materialism is not exactly "in" here. It applies to them as persons and it applies to their homes. They are, however, well educated and many university graduates are found in this segment. There are two or more adults in the household. The man is employed within the public sector in the higher salary range. The woman is independent or homegoing. Money are not spent in the same way as it is done by the moderns. The car is not quite as trendy, the expensive durables are fewer-less is a possibility. The modern social oriented are more idealistic. When they talk about recent developments, it is not so much e-mail or getting on "the net" that they are concerned with, as it is with intellectual cultural questions and not the least being together with other people.

The modern/individually oriented

This is a segment with quite large households. They are more self-confident than most, with a sensible balance between private consumption and a healthy financial situation. The segment comprises many young success-oriented men with good theoretical backgrounds and careers. The housewife is self-employed and there are three or more children in the home, a few them below 13 years of age. The other children are apprentices or students..

For better or worse the modern individually oriented represent the PC boom and are among the 900,000 households that do not accept less than 8 megabyte RAM on their 486 PC. The printer is the latest model, and many have CD-ROM drives and modems.

For shopping the family uses the car. Frequently a slightly used one from a classified advertisement in one of the major dailies. The car is also used when some of the family members go play golf, tennis, badminton or the like. There are no less than 93% of the modern individually oriented who are actively engaged in sports and exercise.

Traditional/oriented

The traditional individually oriented have little economic power and have the prime of life behind them. The children have left home. The attitudes and lifestyles of the traditional individually oriented are quite predictable and stable. They spend a lot of money on pools and lotto. They find that too little is done for the elderly and the weak, and they think that the gap between rich and poor has grown too wide. The government ought to help families get back to the good old roles. Also, the government ought to keep an eye on the doings of the business community.

Traditional/socially oriented.

A greying but also living group of the population. Almost half of the group are sixty years old or more and they have a grade school education behind them. Rarely more. They are pensioners or people not in the labour market. There are no or few children. Many are widows or widowers living in apartments in larger cities. They do not have an impressive income-but they have a remarkable ability to get the most out of what is available. One out of 4 has a cottage, a timeshare apartment or a swimming pool.

5. THE RELIABILITY OF THE SEGMENTATION.

The segments as described here can be analysed along more than 1000 different variables. The above, however, gives an idea of the nature of the 8 major segments. An important question is of course to what an extent the segments exist over time. Do they chnge from one point in time to the next?

To test this we have performed an analysis on the Danish data in which we estimated the two dimensions based upon critical items central to the definition of the two major variables. This was done for 1994. Subsequently the same factor-scores were used on the data from 1995 to estimate the location of the individuals. The analysis showed more than 95 per cent of the respondents were located in the same segment as that which they were assigned to, based on the factor scores derived directly from the data from 1995. Later the same findings emerge when comparing 1995 with 1996. The ability to reproduce the segments from one period to the next is in other words remarkable. Similar dimensions are derived also from minor selections from the larger questionnaire such as "fashion" or "telecommunication" related items and for segments such as "woopies". It is worth adding also that the Danish, Norwegian and Swedish solutions agree very well.

6. 18 FINER VALUE DIMENSIONS

The basic idea behind the system is of course that there are two relatively stable fundamental dimensions that can be used over time across surveys and across countries to describe individual populations. Much like the way in which sex, age, income, etc. are used. However, being derived from a large database including close to 1000 items, lots of information is not extracted, and could be added to the two basic dimensions of the system. One such possibility is to make a value battery based upon a larger number of the values measured in the questionnaire. When that is done with the 59 values the following 18 finer value dimensions emerge.

1. Socially responsible: Reflects concern with the environment, interest in how other people are doing in society and concern with local policy.

2. Multicultural: Positive attitude towards refugees and immigrants. Interested in foreign policy issues.

3. Technology buffs: Aware of the possibilities that modern technology offers, and completely "on line" with the latest developments in technology.

4. Self-assertive (extroverts): Progressive, outspoken persons with strong opinions about everything. Often innovators and trend-setters.

5. Advertising-conscious: Scan the print and electronic media for specials and bargains.

6. Ecological concern: Avoid goods with artificial flavourings, shop for environmentally correct and ecologically correct goods. Prefer to buy goods produced by companies known for their high moral and ethics. Conscious about their physical well-being and their eating habits: the green consumers.

7. Rational: Think rationally and rarely do anything unplanned. Make decisions based upon facts only and always consider price when purchases are made.

8. Liberal: Concerned about the role of the individual in relation to the government. The unions have too much power, the public sector is too loosely managed, and it is too easy to get social welfare and other public benefits. Favour privatising.

9. Quality consciousness: Like luxury. Spend more money on branded goods, and find it important to be well dressed. Think that there is a connection between price and quality.

10. Traditional role-holders: The man decides at home. The woman cooks, while the man reads the newspaper.

11. Family orientation: The family is more important than work. Family and friends are appreciated instead of status and career.

12. Local orentation: Read local news and prefer to get it from newspapers rather than from television.

13. Environmental concern: Do more to recycle waste. Too little is done to protect nature.

14. Time pressure: Being stressed vs. easily being able to have time for all.

15. Political Concern: Takes part in political debates, prefers news in dailies.

16. Materials: Money is a measure for success, wants more material goods.

17. Status: Success is important. I am concerned with status.

18. Price concern: Shops in discount stores, goes for offers, reads retail ads.

FIGURE 12

18 FINER VALUES PLACED IN THE BASIC TWO DIMENSIONAL VALUE DIAGRAMME

These values (some with positive as well as negative ends) are shown in Figure 12.

It is possible to ask to what an extent these more detailed values are stable over time. This has been tried by running parallel factor analyses on the entire 1994 database, the spring database of 1995 and the third quarter of 1995. Some of the most important factors are shown in Table 4. It is obvious that within a time span like the one we are working with here, these dimensions remain relatively stable and can be measured repeatedly.

7. INTEREST AND ACTIVITY DIMENSIONS

It is also possible to analyse further the relationship between the interests and the consumer behaviour variables of the questionnaire in more details. This can be done by identifying interests and consumption factors using a procedure similar to the one described above in connection with the derivation of the 18 value dimensions. When this is done a total of 9 interest dimensions emerge. These are

Political debate

Travel

Listening to classical music

Technology

Sports television viewing

Entertainment

Fashion and cosmetics

Health

Home and newspaper and television news

TABLE 4

COMPARISON OF 2 (OUT OF 18 FINER VALUES) ON FACTOR LOADINGS THROUGH DIFFERENT TIME PERIONDS (1994-1995: DANISH DATA)

Like with the values a factor-score coefficient for each respondent can be computed for each of the interest scales. When this is done the "average respondents" can be plotted in the modern-traditional, social-individually orientation diagram like done in Figure 13. Here each point represents the average score on the two basic Kompas dimensions for that third of the respondents scoring highest on the different interest scales.

It appears that the technology interested are found among the modern-individually oriented ones, the health interested among the socially-traditionally oriented ones, the political debate interested among the socially-modern oriented ones and the home and television, focused ones in the more traditional value segments.

Similarly one can make an activity battery based upon the large number of consumer activities.

When a similar analysis is carried out on the consumer behaviour items (activities) a total of 16 factors emerge:

Electoral orientation

Classic and culture

Individual sport

Fitness

Healing school education

Going out in the evening

Team sport

Sailing

Visual orientation

Prestige sport

Classified ads reading

Home-making

House and garden

Church

Gambling and lotto

Again for each respondent a score can be computed for each of the scales. Concentrating the one third scoring highest on each of the activity factors their average score on the modern/traditional and social/individual basis values can be computed. When this is done a diagram like the one in Figure 14 results.

8. CONCLUSION

In the present paper, we have presented efforts made to arrive at a simple stable solution to value measures which can be applied over time, across countries and across different surveys. After having been through most of the available published value studies from later years, it is our conviction that there are two or perhaps three basic dimensions which will come out almost regardless of what questions are asked. If this is so, it might be useful to put more effort into refining these two or three dimensions. Further research on this is in progress.

Presently the Kompas system is a highly flexible research tool for application on the Gallup Marketing Index Data Bases. Also good experiences have been made with the measurement of the two basic dimensions in other countries and surveys using a limited number of questions.

FIGURE 13

INTEREST DIMENSIONS LOCATED IN 2 DIMENSIONAL KOMPAS SPACE

FIGURE 14

ACTIVITY DIMENSIONS LOCATED IN THE 2-DIMENSIONAL KOMPAS SPACE

So far, the stability of the 2 basic dimensions over time and across countries have been remarkable. Obviously there are limits here, however. Future work is needed to develop methods to cope with emerging changes.

The System can be seen to tie in with basic psychological personality and value theory. Also here, however further work has to be done.

LIST OF LITERATURE

Askegard, S. (1993) Livstilsundersogelser, Odense Universitet.

AIM/Nielsen (1997) Minerva Snapshots, AIM/A.C.Nielsen A/S, Copenhagen.

Beck, U. (1995) Ecological enlightment. Essays on the politics of risk society. Atlantic Highlands, N J: Humanities Press.

Beckmann, S.C. & Kilbourne, W.E. (1997) "The Interplay between the Dominant Social Paradigm and Value Systems: Influences on Danish Business Students’ Environmental Concern," Working Paper, Department of Marketing, Copenhagen Business School, Denmark, July.

Brunso, K., Grunert, K.G. & Bredahl, L. (1996) "An Analysis of National and International Consumer Segments," Working Paper, MAPP, Aarhus Business School, Denmark, January.

Edwards, A.L. (1954) Personal Preference Schedule: Manual, New York: The Psychological Corporation.

Evans, F.B. (1959)"Psychological and Objective Factors in the Prediction of Brand Choice. Ford versus Chevrolet." ,Journal of Business, Vol. 32, pp. 340-69.

Graham, J.R. (1990) MMPI-2: Assessing personality and psychopathology, New York: Oxford University Press.

Grunert, S.C. & Juhl, H.J. (1995) "Values, Environmental Attitudes and Buying of Organic Food," Journal of Economic Psychology 16, 30-39.

Hassan, Larry & Ladet (1994) Esomar Conference Proceedings, Copenhagen..

Kassarjian, H.H. (1965) "Social Character and Differential Preference for Mass Communication," Journal of Marketing Research, Vol. 2, pp. 146-53.

Kilbourne, W.E. (1995) "Consumption ideology and the ecologsical crisis: Is demarketing consumption the solution?" Presented at the International Association of Business and Society, Vienna/Austria, Summer.

Maslow, A.H. (Ed.) (1959) New knowledge in human values, New York, Harper.

Peter, P.J. & Olsen, J. (1993) Consumer Behavior and Marketing Strategy, Third edition, Irwin Horne word, Illinois.

Rentorff, M. (1995) "Mode og Forbrugervcerdier (Fashion and Segmentation)," Thesis, Copenhagen Business School, Denmark.

Riesman, D. (1961) The Lonely Crowd, New Haven: Yale University Press.

Rokeach, M. (1960) The Open and the Closed Mind, New York: Basic Books.

Rokeach, M. (1973) The nature of human values, New York: Free Press.

Schwartz, S.H. (1992) "Universals in the content and structure of values: Theoretical advance and empirical tests in 20 countries," In: M.Zanna (Ed.), Advances in experimental social psychology, Vol. 25 (pp. 1-65), San Diego, CA: Academic Press.

Schwartz, S.H. & Bitsky, W. (1990) "Toward a Theory of Universal Content and Structure of Values," Journal of Personality and SociaI Psychology, 58, 878-891.

Thurstone, L.L. (1953) Examiner Manual for the Thurstone Temperament Schedule, Chicago: Science Research Associates.

Tolman, E.C. (1951) Purposive Behaviour in Animals and Men, New York, Appleton-Century-Crofts.

Vilstrup Research (1996), Newsletter, Copenhagen.

Wells, W.D. (1966) "General Personality Tests and Consumer Behavior," In J.W. Newman (ed.): On Knowing the Consumer, New York: John Wiley and Sons, Inc.

Williams, J. (1991) "Constant Questions on Constant Meaning" Marketing and Research Today, August.

Wilson, L.L. (1966) "Homemaker Living Patterns and Marketplace Behavior-a Psychometric Approach," In J.S. Wright and J.L. Goldstucker (eds.): New Ideas for Successful Marketing, Chicago: American Marketing Association.

Zetterberg, H. (1995) "Valuescope: A Three Dimentional Value System," in European Advances in Consumer Research, Hemming Hansen (Ed.) Vol 2, pp. 163-192.

----------------------------------------