Developing an Index to Measure &Quot;Hispanicness&Quot;

Humberto Valencia, Texas Tech University
ABSTRACT - Items for an index to measure the rate of acculturation of Hispanics in the United States ("Hispanicness") are proposed and partially tested. The results tentatively support the index, but more testing is needed. Five specific suggestions for additional research are proposed in the discussion.
[ to cite ]:
Humberto Valencia (1985) ,"Developing an Index to Measure &Quot;Hispanicness&Quot;", in NA - Advances in Consumer Research Volume 12, eds. Elizabeth C. Hirschman and Moris B. Holbrook, Provo, UT : Association for Consumer Research, Pages: 118-121.

Advances in Consumer Research Volume 12, 1985      Pages 118-121

DEVELOPING AN INDEX TO MEASURE "HISPANICNESS"

Humberto Valencia, Texas Tech University

ABSTRACT -

Items for an index to measure the rate of acculturation of Hispanics in the United States ("Hispanicness") are proposed and partially tested. The results tentatively support the index, but more testing is needed. Five specific suggestions for additional research are proposed in the discussion.

INTRODUCTION

Interest in studying Hispanic consumers in the continental United States has increased in the 1980s among business and academia as a result of the increasing recognition given to the size and growth of this ethnic population. In 1983, the size of the Hispanic population was estimated at 15.9 million persons and this population grew 61 percent between 1970 and 1930 (Bureau of the Census (1984a, pp. 3, 4). Furthermore, Hispanics' aggregate income after taxes amounted to $62.5 billion for 1982 (Bureau of the Census 1984b). Thus, the significance of the Hispanic market to consumer marketers is obvious.

Because Hispanics are a subculture living within the context of a broader American culture, they are acculturating to one extent or another. Acculturation is defined in this discussion as the process of learning a culture different from the one in which a person was originally raised (Berelson and Steiner 1967). Similarly, Hispanics are also being assimilated by mainstream American society. Assimilation is defined here as the process of being accepted as genuine member of a new social group (Berelson and Steine 1907). Therefore, Hispanics in the United States are in a state of cultural flux.

Simple observation of Hispanic consumers would indicate that the rate of acculturation varies from one- individual to another. Thus, some members of this ethnic subculture appear to be "more Hispanic" than others. Therefore, the term "Hispanicness" is coined to refer the rate (or degree) of acculturation of Hispanic consumers living in this country.

Given that culture affects the consumer behavior of individuals (cf. Howard and Sheth 1969, Sturdivant 1973, Wallendorf and Reilly 1983, Henry 1976), and that Hispanic consumers are in a state of cultural flux, an important moderator of Hispanic consumer behavior would be their "Hispanicness". Therefore the purpose of this paper is to develop an index of "Hispanicness" applicable for analyzing Hispanic consumer behavior variations. The effectiveness of this index will then be tested in the context of a comparative study on shopping orientations between Hispanics and White non-Hispanics (Valencia 1982a, 1983).

PREVIOUS RESEARCH

Despite the apparent importance of an index to measure the rate acculturation, there are only a few studies that have discussed it in the marketing literature. Hair (1971, 1973a, 1973b, 1975) devised a measure of consumer acculturation for foreign students and later refined it for black consumers. These studies, however, are only tangentially useful to present one, because they deal with consumer acculturation and were developed for other subcultures. Wallendorf and Reilly (1983) and Reilly and Wallendorf (1984) investigated cultural assimilation of Mexican-Americans in the Southwest by comparing their food consumption patterns with those of income matched Anglos living in the same region and those of Mexicans living in Mexico city, using garbage content analysis. They found that Mexican-American consumption patterns are more of a unique cultural style than a blending of Mexican and Anglo patterns; and that a lag factor exists stemming from the dynamic aspects of culture and the perceptions of the same by the groups. However, the methodology of these two studies did not include the identification of the characteristics of assimilation of the respondents.

The study by Hirschman (1981) on American Jewish ethnicity investigated one important component of acculturation (i.e., self-perceived strength of ethnic identification) and found that the degree of ethnicity among Jews affected certain aspects of their consumer behavior. Hirschman also cogently points out as one of three generalizations about the literature on ethnicity and consumption:

Finally, in many instances subjects have been assigned to the ethnic group of interest or general population category on the basis of researchers perceptions of their membership in one or the other group. This may, of course, lead to incorrect assignment. Further exacerbating the problem is the fact that subjects are typically not asked the degree of identification they feel with a particular ethnic group. One black consumer may feel an specially strong commitment to black ethnicity, whereas a second may feel much less identification with this group. The degree of identification the individual feels with a given ethnic group may largely determine the level of commitment he/she experiences regarding the norms of the group and, thus, the degree of influence the group has on his/her actions and attitudes. (Hirschman 1981).

Thus one of the Items of an acculturation index should be the strength of ethnic self-identification. Hirschman's ethnicity index also included religion, because the correlation between Jewish culture and Jewish religion has been found to be high. With other ethnic groups, however, the congruence between religion and culture may not be as significant (Hirschman 1981). The latter appears to be the case with Hispanic culture. While the majority of Hispanics are baptized Roman Catholic, the strength of their religious beliefs and practices varies. Grebler, et al., (1970) found that a large number of Mexican-American Catholics' do not conform to the norms of the Church. Their mass attendance, for example, was substantially below the national average. Therefore, the religious dimension does not seem appropriate for the "Hispanicness" index.

The major indicator of the rate of acculturation among Hispanics appears to be the person's understanding of the English language (Olmedo and Padilla 1978), because it facilitates interaction with the predominantly English dominant culture. In other words, English language ability will accelerate the rate of acculturation.

Another indicator of "Hispanicness" should be the extent of Spanish language maintenance. Hispanics that speak Spanish are more likely to retain and interactively reinforce their culture. Greenberg, et al., (1983), for example, found that predominantly Spanish speakers are more likely to use Spanish language media (i.e., newspapers, other print, television, radio). The key indicator of Spanish language maintenance would be the extent of Spanish language spoken at home. This indicant is preferred over other language-spoken (Spanish vs. English) measures because the extent of English language usage is more situationally determined (e.g., English generally would be the predominant and necessary language at work).

Since some Hispanics are bilingual (47 percent according to Yankelovich, Skelly & White 1981), another language-related indicator of "Hispanicness" should be their language preference when conversing with someone who could communicate just as well in English or Spanish. This measure is independent of the recipient of the communication and, therefore, a more accurate reflection of language choice.

The length of time the individual has lived in the host American culture, the more likely he/she is to have learned the American way of life. However, this dimension does not take into account the length of enculturation in the original Hispanic culture. The combination of these two criteria into a ratio of length of residence in the U.S. divided by the respondents age would appear to be a stronger indicator of these dimensions than each indicant alone.

Finally, Hispanics married to non-Hispanic Americans are more likely to acculturate and adopt some of the cultural ways of their spouse. Hispanics married to Hispanics, on the other hand, are-more likely to reinforce each others "Hispanicness". Therefore, miscegenation should be included as an indicator of "Hispanicness".

Other sociocultural indices (e.g., occupational status, family size, income, education) could have been proposed for the "Hispanicness" index; however, they are more likely to measure assimilation than acculturation as Olmedo (1979) has pointed out.

METHODOLOGY

The six "Hispanicness" indicators were measured using the questions and coding scheme presented in the appendix. Reverse scoring was used on English language ability to restore directional consistency of the overall index. Lower scores in the "Hispanicness" index indicate lower acculturation. The index ranges between 6 and 23. Both Hispanics and white non-Hispanics were scored on the index; however, non-Hispanics were reverse scored on the strength of ethnic identification item.

The sample to test the index consisted 564 responses obtained from a mail survey sent to New York, Los Angeles, San Antonio and Miami. These cities were selected because they are representative of the three major Hispanic subgroups in this country and they account for about 42 percent of the U.S. Hispanic population. After excluding responses from other ethnic groups and deleting incomplete questionnaires, 178 Hispanic and 288 White usable responses were analyzed.

Since many Hispanics do not understand the English language or they do not feel comfortable with it, the questionnaire was translated to Spanish. To ensure proper translation of the instrument a systematic procedure was used. First, the questionnaire was translated to Spanish by a trained bilingual expert. Four independent bilingual translators (a Cuban, a Mexican-American, a Puerto Rican and an American fluent in Spanish) then back-translated the Spanish version into English for verification of the conceptual equivalence of the two language forms. Only minor differences were found in the back-translation and these were resolved by modifying the wording. Finally, a committee of bilinguals reviewed the parallel translations to insure idiomatic precision of the final forms. The final instruments were then pretested and found to be acceptable equivalents of each other.

Ethnic identification of the respondents was measured by self-report, rather than ascribed. This approach is recommended by Cohen (1978) and Hirschman (1981) because it represents the internal beliefs of the individual and therefore reflects their cultural reality. Self-report ethnic measures also eliminate the misclassification bias inherent in using the respondent's surname or country of birth approaches (Valencia, 1982b).

Because the focus of this study is on how the Hispanic consumer has become more "American like" acculturation is conceived as a one-way process. Thus, American culture is perceived as static, while Hispanic culture is in a process of change. Naturally, the host American culture is also changing and Hispanic culture affects the host culture over a period of time; but at a given point in time and from the viewpoint of Hispanics, he/she is adjusting to the host culture.

VALIDATION TESTS

Validity is concerned with whether an instrument measures what it is suppose to measure (Nunnally 1907). Many tests must be performed before confidence in the instrument develops. At this time, however, only limited insights into the construct validity of the "Hispanicness" index can be reported.

Hair (1971, 1973, 1975) has devised a consumer acculturation test (CAT). This test measures consumer acculturation on several marketing-related dimensions. Attitudes toward types of products (e.g., ready-mix, convenience), multiple brand offerings, packaging, standardization, price bargaining, and promotional techniques are measured with a Likert scale (5 point, Strongly Agreed-Strongly Disagree). A subset of 6 selected CAT items were identified in the pretest as a more parsimonious and manageable set out of the 36-item original instrument. A composite consumer acculturation measure was then created by summating the responses to the items and dividing this sum by the number of item responses. Since some respondents occasionally omit some items, this procedure permits maximum data utilization, yet provides a reliable measure (alpha = .73). A positive measure of association between the CAT and the "Hispanicness" measure would support the construct validity of the index. The Pearson product moment correlation between these two measures is + 0.166 (p < .037) among Hispanics.

Hispanics born abroad (in a Spanish-speaking country) who immigrated to the U.S. are less likely to be as acculturated as Hispanics born in the United States, because the enculturation process would be stronger in a Hispanic country than in the U.S. Therefore, a significant correlation between the index and country of birth (U.S. vs. Spanish-speaking country) would provide support for the validity of the index. The Spearman correlation coefficient between these two measures is 3.535 (p < .000) in the hypothesized direction. Furthermore, if the U.S. lived ratio indicator is not included in the index because it measures, in part, country of birth, the correlation is 0.477 (p < .001) in the hypothesized direction.

Differences in consumer behavior between Hispanics and Whites should be moderated by fie "Hispanicness" of the respondents. Therefore, monotonic transformations in the group means should be expected going from high "Hispanicness" to low "Hispanicness" to Whites, in either direction. This proposition was tested using results from a study on shopping orientations (Valencia 1982), after dividing Hispanics into two groups according to their index scores. Hispanic scores ranged between 6 and 19. Scores between 6 and 12 were classified as "high Hispanicness" an scores between 13-19 were classified as "low Hispanicness", which provided about an even split of the Hispanic sample. Twelve shopping Orientations were selected for this study, because they had alpha reliability coefficients of 0.6 - 0.9 for both the Spanish and English language responses. Additional cross-cultural reliability and validity checks (e.g., factor analysis, congruency coefficients) provided extra safeguards for the comparability of the data. As can be clearly seen in Table 1, all the differences are monotonic transformations across the three ethnic groups in the hypothesized way. This data also shows that the index, even though crudely used here, can be useful in further identifying differences between Hispanic acculturation groups.

TABLE 1

SHOPPING COMPARISONS USING THE "HISPANICNESS" INDEX

DISCUSSION

The proposed "Hispanicness" index is still in a primitive developmental stage and only through its future use and application will its validity and usefulness be thoroughly explored. The primary purpose of this paper has been the proposal of major indicators of "Hispanicness" rather than an exhaustive discussion of possible indicators of Hispanic ethnicity. Furthermore, the validity tests were convenience test, given the secondary nature of this research to the original purpose of the study.

The three tests of the index presented are not sufficient to establish its appropriateness. A brief discussion of some additional test follows. First, the implicit assumptions of the indicators must be examined. For example, a linear relationship between the indicators and "Hispanicness" has been assumed. However, Weinstock (1964) argues that there is an initial period of a few months when almost every week counts, but, after that, other factors predominate." Second, the index should be tested against other assumed characteristics of acculturation. For example, ethnic food habits are said to be the most resistant to change and acculturation is indirectly related to the size of ethnic population. Third, the dimensionality and redundancy of the index should be analyzed. The strength of ethnic-identification is an attitudinal dimension, whereas the rest of the items measure behaviors. Fourth, variations across the major subgroups should be examined. For example, Cubans as a group are the most recent arrivals and, therefore, would be expected to score lower on the index. Fifth, the weights attached to the six indicators have been arbitrarily assigned by the scoring system on the basis of the researcher's subjective judgment of importance. Since the "true" weights of these indicators are "a prior" unknown, more research is needed here.

APPENDIX

Strength of ethnic identification: "How strongly do you identify yourself with the ethnic or racial group you mentioned above? (?receding question was on-ethnic self-identification) 1-Very Strongly, 2-Strongly, 3-More or less, 4-Weak, 5-Very Weak."

English language ability: "As you may know, some people in the U.S. are bilingual. If you speak Spanish, please answer the next three questions. Would you say you speak English: 1-Very well, 2-Well, 3-Not well;little, 4-Not at all."

Spanish language spoken at home: "Would you say your family speaks Spanish at home: 1-All of the time, 2-Most of the time, 3-Sometimes, 4-Not at all."

Language Preference "If you had the chance to communicate with someone just as well in English or Spanish, which would you prefer to converse with them? 3-English, 1-Spanish, 2-Either Spanish or English."

Ratio of length of residence in the U.S.: "How long have you lived in the U.S.? All of your life, # Years" This answer was divided by the respondent's age and weighted by 4.

Miscegenation: "If married, with which ethnic or racial group does your spouse identify with?" Coded: (1) Hispanic/Hispanic spouse or Hispanic single (2) Hispanic/Anglo spouse (3) Anglo/Hispanic spouse (4) Anglo/Anglo or Anglo single

REFERENCES

Berelson, Bernard and Gary A. Steiner (1967), Human Behavior: An Inventory of Scientific Findings, New York: Harcourt. Brace. Jovanovich.

Bureau of the Census (1984a), Condition of Hispanics in America Today, Washington, D.C.: U.S. Department of Commerce.

Bureau of the Census (1984b), After-Tax Money Income Estimates of Households: 1982, Washington, D.C.: U.S. Department of Commerce, Current Population Reports (P-23, No. 137).

Cohen, Ronald (1978), "Ethnicity: Problem and Focus in Anthropology," Annual Review of Anthropology (7), 379-403.

Grebler, Leo; Joan W. Moore; and Ralph C. Guzman (1970), The Mexican-American People: The Nation's Second Largest Minority, New York: The Free Press, Chapter 19.

Greenberg, Bradley S.; Michael Burgoon; Judee K. Burgoon; and Felipe Korzenny (1983), Mexican Americans and the Mass Media, Norwood, New Jersey: ABLEX Publishing Corporation.

Greenberg, Bradley S. and Rolph Anderson (1973a), "Culture, Acculturation and Consumer Behavior: An Empirical Study," Combined Proceedings, 3.W. Becker and H. Becker, editors, Chicago: AMA, Series No. 341 423-28.

Hair, Joseph F., Jr., The Impact of the Acculturation Process on Consumer Purchasing Patterns, unpublished doctoral dissertation, University of Florida.

Hair, Joseph F., Jr. and Ratchford, III (1973b), "Socio-cultural Parameters of Black Buyer Behavior", in 1973 Southern Marketing Association Proceedings, Robert L. King, editor, 275-79.

Hair, Joseph F., Jr. and Ronald, F. Bush, and Paul S. Busch (1975), "Acculturation and Black Buyer Behavior", Combined Proceedings, Edward M. Mazzi, editor, Chicago: AMA, Series No. 37, 253-56.

Henry, Walter A. (1976). "Cultural Values Do Correlate in the Consumer Behavior", Journal of Marketing Research, 13 (May), 121-7.

Hirschman, Elizabeth C. (1981), "American Jewish ethnicity: Its Relationship to Some Selected Aspects of Consumer Behavior", Journal of Marketing, 45 (Summer), 102-10.

Howard, John A. and Jagdish N. Sheth (1969), The Theory of BuYer Behavior. New York: John Wiley & Sons, Inc.

Nunnally, Jum C. (1957), Psychometric Theory. New York: McGraw Hill

Olmedo, Esteban L., Joe L. Martinez, Jr.; and Sergio R. Martinez (1978), "Measure of Acculturation for Chicano Adolescents", Psychological Reports, 42, 159-70.

Olmedo, Esteban L. and Amado M. Padilla (1978), "Empirical and Construct Validation of a Measure of Acculturation for Mexican Americans", Journal of Social Psychology, 105, 179-187.

Olmedo, Esteban L. (1979), "Acculturation: A Psychometric Perspective," American Psychologist, 34 (November), 1061-1070.

Ramirez, Manuel; Alfredo Castaneda; and P. Leslie Harold (1974), "The Relationship of Acculturation to Cognitive Style among Mexican Americans", Journal of Cross-Cultural Psychology, 5 (December), 424-33.

Reilly, Michael D. and Melanie Wallendorf (1984), "A Longitudinal Study of Mexican-American Assimilation," In Advance in Consumer Research, Thomas C. Kinnear, editor, Provo: Association for Consumer Research, 735-40.

Sturdivant, Frederick D. (1973), "Subculture Theory: Poverty, Minorities, and Marketing", in Consumer Behavior: Theoretical Sources, Scott Ward and Thomas S. Robertson, editors, Englewood Cliffs: Prentice Hall, Inc., 469-520.

Teske, Raymond H.C. Jr., Bardin H. Nelson; and Maria B. Villareal (1977), "Mexican-American Identity: A Further Evaluation," International Review of Modern Sociology, 7 (July-December), 225-40.

Valencia, Humberto (1982a), Shopping Orientations Among Hispanics and Anglos in the United States, unpublished doctoral dissertation, Georgia State University.

Valencia, Humberto (1982b), "Self Report Measures Find Elusive Hispanic Consumers," Marketing News. 23 (May 14), 19.

Valencia, Humberto (1983), "Shopping Orientations Among Hispanics and Anglos in the United States: Abstract," Journal of Retailing, 59 (Spring), 100-1.

Wallendorf, Melanie and Michael D. Reilly (1983), "Ethnic Migration, Assimilation and Consumption," Journal of Consumer Research, 10 (December), 292-302.

Weinstock, S. Alexander (1964), "Some Factors that Retard or Accelerate the Rate of Acculturation: With Specific Reference to Hungarian Immigrants", Human Relations, 17 (November), 321-40.

Yankelovich, Skelly & White (1981), Spanish USA New York: SIN National Television Network.

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