Two Peoples Divided By a Common Language: Consumer Alienation Among British and Americans

ABSTRACT - In this study, the authors used the Allison Alienation Scale to measure the similarities/differences in consumer attitudes toward the marketplace in the U. S. and the U. K. In an effort to show that the construct was similarly understood by people in both countries, factor analysis solutions of the 34 items of the questionnaire were compared and found to be quite similar. Subsequent MANOVA tests indicate that the American sample was much more alienated than the British sample. Marketing implications of this difference were discussed.

Citation:

Martha R. McEnally and William L. Tullar (1995) ,"Two Peoples Divided By a Common Language: Consumer Alienation Among British and Americans", in E - European Advances in Consumer Research Volume 2, eds. Flemming Hansen, Provo, UT : Association for Consumer Research, Pages: 157-162.

European Advances in Consumer Research Volume 2, 1995      Pages 157-162

TWO PEOPLES DIVIDED BY A COMMON LANGUAGE: CONSUMER ALIENATION AMONG BRITISH AND AMERICANS

Martha R. McEnally, University of North Carolina at Greensboro

William L. Tullar, University of North Carolina at Greensboro

ABSTRACT -

In this study, the authors used the Allison Alienation Scale to measure the similarities/differences in consumer attitudes toward the marketplace in the U. S. and the U. K. In an effort to show that the construct was similarly understood by people in both countries, factor analysis solutions of the 34 items of the questionnaire were compared and found to be quite similar. Subsequent MANOVA tests indicate that the American sample was much more alienated than the British sample. Marketing implications of this difference were discussed.

An understanding of cultural diversity in international markets is critical to the development of international marketing strategies. Comparison of markets and cultures is difficult, however, because of language and cultural differences and, more subtly, because of the connotations of words and phrases. Frequently, researchers have relied on back-translation of questionnaires and scales in order to ensure that the "meaning" of the questions had not been lost. This process only really ensures that the questionnaire's meaning in the original language has not been lost. It does not ensure that the "foreign" population understands the questions in the same manner as the original population, nor that they use the same frame of reference to answer the questions as the original population does (Price-Williams, 1986).

Recently, Riordan and Vandenberg (1993) found that back-translation does not ensure comparability in the meaning and interpretation of standard scales when translated from English to Korean. They concluded that "researchers should no longer treat the stability of measurement continua as a 'given' in comparative cross-cultural research." (Riordan and Vandenberg, 1993; p.4) This means that some earlier cross-cultural comparisons which assumed equivalence of measurement across cultures could be wrong. Earlier research may have identified differences between cultures as if they were on a continuum when in fact the two groups understood the construct being measured in a wholly different way (Riordan & Vandenberg, 1993).

But what about cultures with the same language and similar values? Would Americans and British use the same frame of reference when interpreting questionnaire items? Using values as the basis for describing a culture (Hofstede, 1980), Americans and the British might be construed as similar since they share many of the same values and for the most part, they use the same language. Do these two cultural groups have the same attitudes toward the marketplace and can the attitude(s) be measured with a standardized instrument?

To answer these last two questions, the researchers administered a standardized instrument, the Alienation Scale developed by Neil Allison, to two samplesCone composed of Americans and the other of British citizens. Comparison of the results enables us to determine the similarity of the alienation construct between the groups and whether the two sets of respondents interpreted the scale in a similar fashion. If there is commonality of interpretation, then meaningful consumer alienation comparisons can be made between the two samples.

LITERATURE REVIEW

Numerous scholars primarily in the fields of sociology and psychology have written about alienation (Dean, 1961; Seeman, 1959 and 1971; Nettler, 1955). These scholars reached several conclusions: (1) alienation should be measured in regard to some reference; (2) that feelings of alienation toward the political system are not equivalent to feelings of alienation toward the marketplace, for example and (3) alienation has several dimensions: powerlessness, normlessness, social isolation, and self estrangement. In the 1970s, several marketing researchers attempted to measure alienation in regard to the marketplace.

There were three branches of alienation research in marketing. First, Pruden, Shuptrine and Longman (1974), Pruden and Longman (1972) and Shuptrine, Pruden and Longman (1977) measured alienation using a sixteen point scale. Because their primary purpose was to compare levels of alienation between groups such as Anglo-Americans, Blacks, Mexicans and businessmen, they did not rigorously test their scale for reliability.

Lambert (1980, 1981) and Durand (1985) measured consumer dissatisfaction and alienation. The 1980 study focused on dissatisfaction and specific consumer issues. The 1985 study used a 27 item scale that included all the dimensions listed previously plus political efficacy and trust. Although the scales were tested for reliability, they were mainly used to relate alienation to advertising, general dissatisfaction, consumerism and groups of consumers defined using demographic characteristics.

Only Allison (1978) attempted rigorous development of a valid and reliable scale for measuring alienation. To measure the four dimensions of alienation, he generated 115 statements which were then evaluated by a panel of 35 judges. This procedure resulted in a fifty item scale which was administered to a stratified-by-area sample of Austin, Texas residents. Allison used principal components and varimax rotation to develop a four factor solution which he deemed the most interpretable analysis. This solution, however, only explained 38% of the variance and only 21 of the 50 items loaded to a high degree on only one of the four factors. He then tested whether alienation consisted of only two factors which were an antibusiness and a probusiness attitude. In testing the internal consistency of the entire scale, he obtained an alpha of .8576 which led him to conclude that alienation should be redefined as a unidimensional construct. Finally, he tested the reliability of the scale and obtained the highest alpha coefficient, +.8802, with a scale of 35 items. Since then, researchers have retained the 35 items but rescaled them into several factors.

Then, Bearden, Lichtenstein and Teel (1983) used Allison's thirty five item scale in a survey of a consumer panel data. Factor analysis of their data resulted in a three factor solution with the factors being business ethics (a combination of normlessness and powerlessness), informed choice , and self-estrangement. Reliabilities for each of these subscales were .84, .67, and .60. They also examined the dimensionality of the scale using confirmatory factor analysis and concluded that the three factor solution appeared to be a better representation of the data. Singh (1991) also used the Allison Alienation Scale and analysis of his data produced the same three factor solution as the Bearden et. al. study. Singh also compared the alienation scale to the Lundstrom and Lamont consumer discontent scale and found that the two were highly similar. Therefore using the alienation scale should tap the same underlying sentiments of the population as consumer satisfaction/discontent measures would.

TABLE 1

COMPARISON OF RESPONDENT SAMPLES

A search of the literature did not reveal any previous use of the alienation scale with British consumers, nor did we find any other articles comparing the levels of consumer alienation or discontent between Americans or British. Therefore, we anticipate that the same three factor solution found by previous researchers would be obtained for both groups; thereby indicating that the structure of alienation (or possibly consumer discontent) is the same or similar in the U. S. and Great Britain.

METHODOLOGY

For this study, a mail survey was conducted in a southeastern American city and a British city located north of London beyond commuting distance. The two cities were highly similar in size and in resident's socioeconomic characteristics. Samples were randomly selected from the City Directory in the U. K. and the Electoral Register in the U. K. Twelve hundred questionnaires were mailed in the U. S. and 259 usable responses were received for a response rate of 21.6%. One thousand questionnaires were mailed in the U. K. and 239 usable responses were returned for a response rate of 23.9%.

The questionnaire consisted of several sections. The first two dealt with purchase of grocery items. The third contained Allison's Alienation Scale and the fourth asked for the following demographic information: gender, age, marital status, educational level and income.

Because the questionnaire was originally designed for use in the American market, several minor changes were deemed necessary before the questionnaire could be employed in the U. K. Examples of these changes are the substitution of "tick" instead of "check" in the instructions and confectioneries and crisps for snack foods.

The two samples are compared in Table 1. There were more males in the American sample; the American sample tended to be slightly older; and the Americans were more likely to have gone to college and graduate school. Incomes are difficult to compare, but the Americans appear to have a somewhat higher income.

Results and Discussion

Table 2 shows the means and standard deviations for the two samples. Although it is not often customary to include summary statistics on the actual items themselves, we believe, for reasons that will become apparent later that comparison across items makes the most sense. Moreover, the individual item means reveal the "real" differences between the two samples. We have included the full text of all the items in Appendix 1.

TABLE 2

MEANS AND STANDARD DEVIATIONS OF BRITISH AND AMERICAN SAMPLES

As we may see from Table 2, many of the items are quite close in the two samples, but several of them are remarkably different. For instance, "same product" shows approximately 1/3 of a standard deviation difference while "shop pleasant" shows more than 3/4 of a standard deviation difference. However, before we can speculate effectively on the means, we must first settle the question of whether this questionnaire has a similar meaning for both samples.

Perhaps the best first step to asking this question is to examine the factor structure in both samples. As noted above, several authors (Bearden, et al, 1983; Singh, 1991) have concluded that a three factor solution is the best representation of the alienation construct. Accordingly, we extracted the principal components and used a varimax rotation as prescribed by Bearden, et al (1983). The three factor solution is rather disappointing in both the British and the American sample accounting for 32% and 31% of the variance, respectively. Clearly, it is difficult to claim that a factor solution is a good representation of the items when only 32% of the variance is accounted for. Table 3 shows the two three factor solutions side by side for the American and British samples.

The two solutions are remarkably similar. In fact if we consider .40 as a significant loading and only count whether a loading is significant or not, the two solutions match on more than 90% of the loadings. There appears to be a great similarity between the two three factor solutions.

However, as we noted above, these solutions account for less than 1/3 of the variance in their respective samples. What if a more representative factor structure were derived? If Kaiser's Criterion of eigenvalues above 1.0 is used, the British sample produces 10 factors and the American sample produces 11 factors. Neither of these is particularly parsimonious and moreover neither of these solutions exceeds 62% variance explanation.

TABLE 3

COMPARISON OF PRINCIPAL COMPONENTS BETWEEN BRITISH AND AMERICAN SAMPLES

In accordance with the procedure pioneered by Riordan and Vandenberg (1993), we attempted to examine the fit of the two three factor solutions with the three factor solution of Bearden, et al (1983). Table 4 shows the results of the LISREL confirmatory factor analysis.

As can be seen from Table 4, neither of the samples fits the Bearden, et al. solution. That is, when the Chi-square is significant it means that there are significant differences between the factor structures. In addition, we tried to fit the American data to the three factor American solutionCand even that produces a chi-square of 814.62, significant at the .001 level also. The problem is that the three factor solution is just not representative enough of the data to use in confirmatory factor analysis. However, the 10 or 11 factor solution requires the estimation of too many parameters from the 34 variables and LISREL cannot fit a model. Thus we are left with judgment as to whether the British and American samples shared a common understanding of the items.

We hold that in fact comparison of the three factor solutions shows a common understanding, and there are some real differences between the samples on certain items. That is, we hold that the constructs are perceived in the same way in the two populations, but there are some real inter-country differences in alienation. We turn next to MANOVA to explore the differences between the two groups.

TABLE 4

LISREL CONFIMATORY FACTOR ANALYSIS RESULTS

As may be seen from the last column of Table 5, 22 of the 34 contrasts are significant. If one compares these results with Table 1 where the means are shown, another result emerges. The American means are lower than the British in all but five cases. Of these five cases, only one is significant at the 5% level. If one examines the scale of the questions in Appendix 1, it is apparent that the larger numbers indicate more disagreement. Thus it may be concluded that the British are significantly less alienated than American consumers. On the biggest differences across the two samples, "Same product", "Shopping Pleasant", "Mass Production", "Embarrassing", "Sale Bargain", "Brands Same", and "Beyond Means" represent the largest effects. All of these show the British having on larger average scores: That is they disagree with the alienation statements substantially more than Americans do.

The overall MANOVA taking all of the items together is, of course, significant at the .01 level with all the multivariate statistics (Pillai, Hotelling, and Greatest Latent Root). Thus, there is no question of an overall significant effect. The British consumers are significantly and practically less alienated than American consumers are. Moreover, this difference extends across most of the various facets of alienation.

CONCLUSIONS

In summary we conclude that the alienation scale does not yield a parsimonious principal components solution. This fact makes it impossible to test the instrument equivalence using Riordan and Vandenberg's (1993) method. However, examination of the three factor solution seems to indicate that the British and American solutions resemble each other rather strongly. In fact, if we consider only whether a loading is significant or not, the two solutions agree on over 90% of all the loadings.

If we accept equivalence in understanding between the British and American samples, then it appears that there are some very substantial differences between the two groups. The British sample is, on 29 of 34 items, more positive about the marketplace in general, and marketing and advertising practices in particular. The differences are quite substantial as well as being statistically significant. Several of the differences reach almost a full standard deviation. In two groups of this size that is a very substantial difference.

TABLE 5

MULTIVARIATE AND UNIVARIATE TESTS OF DIFFERENCES BETWEEN COUNTRIES

REFERENCES

Allison, Neil K., "A Psychometric Development of a Test for Consumer Alienation from the Marketplace," Journal of Marketing Research, November 1978, 565-75.

Bearden, William O., Lichtenstein, Donald R., and Jesse E. Teel, "Reassessment of the Dimensionality, Internal Consistency and Validity of the Consumer Alienation Scale," American Educators Conference Proceedings, 1983, 35-40.

Dean, Dwight G. "Alienation: Its Meaning and Measurement," American Sociological Review, October 1961, p. 753-758.

Durand, Richard M. and Zarrel V. Lambert. "Alienation and Criticisms of Advertising," Journal of Advertising, vol. 14, No. 3, p. 9-17.

Hofstede, G. 1980. Culture's Consequences: International differences in work related values. Beverly Hills, CA: Sage.

Lambert, Zarrel V. "Consumer Alienation, General Dissatisfaction, and Consumerism Issues: Conceptual and Managerial Perspectives," Journal of Retailing, 56. (Summer 1980), p. 3-24.

Lambert, Zarrel V. "Profiling Demographic Characteristics of Alienated Consumers," Journal of Business Research, 1981, vol. 9, p. 65-86.

Nettler, Gwynn. "A Measure of Alienation," American Sociological Review, December 1957, p. 670-77.

Price-Williams, D. R. Cultural Psychology. In Lindzey & Aronson, E. (eds.) Handbook of Social Psychology (1986) p. 993-1042. New York, NY: Random House.

Pruden, Henry O. and Douglas S. Longman, "Race, Alienation and Consumerism," Journal of Marketing, July 1972, p. 58-63.

Pruden, Henry O., Shuptrine, F. Kelly and Douglas S. Longman. "A Measure of Alienation from the Marketplace," Journal of the Academy of Marketing Science, Vol. 2 (Fall, 1974) p. 610-619.

Riordan and Vandenberg, "A Central Question in Cross-Cultural Research: Do Employees of Different Cultures Interpret Work-Related Measure in an Equivalent Manner?", Academy of Management Proceedings, 1993.

Seeman, Melvin. "On the Meaning of Alienation," American Sociological Review, Vol. 24 (December 1959), p. 783-791.

Seeman, Melvin. "The Urban Alienations: Some Dubious Theses from Marx to Marcuse," Journal of Personality and Social Psychology, Vol. 19 (1971), p. 135-143.

Shuptrine, F. Kelly, Pruden, Henry O. and Douglas S. Longman, "Alienation from the Marketplace," Journal of the Academy of Marketing Science, Vol 5. (Summer 1977), p. 233-248.

Singh, Jagdip, "Redundancy in Constructs: Problem, Assessment, and an Illustrative Example," Journal of Business Research, 22, 1991, 255-290.

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