The Relation of Ethnic Identification with Cigarette Smoking Among

US Urban African American and Latino Youth: a pilot study

Ethnicity & Health, Feb-May 98, Vol. 3 Issue 1/2, p. 135.

Authors: Vanessa Parker, Steve Sussman, David Crippens, Pam Elder and Donna Scholl

Abstract: Primarily African American and Latino seventh graders (N = 545) attending health and science classes at three urban Southern California junior high schools were administered a 13-item ethnic identification assessment during one class period, to investigate the relationship between ethnic identification and tobacco use. For African Americans, the perception of looking different from other ethnic groups was found to be associated with avoidance of peer tobacco use. Among Latinos, the perception of being liked by other ethnic groups, and enjoying interaction with people outside of their ethnic group, were found to be associated with negating peer influence for tobacco use. Watching television shows with main characters of same ethnic group was associated with being able to make friends without using tobacco, for both ethnic groups. For Latinos, ever trying a cigarette was associated with the perception of being liked by other ethnic groups. These results suggest there may be elements inherent in ethnic identification protective against peer influence for smoking among ethnic minority youth. Also, acculturation of Latinos to the majority population may increase their risk for experimentation with tobacco products.

Introduction

The establishment of one's personal identity is a major developmental task for adolescents. This developmental task may be particularly complicated for ethnic minority adolescents because they have the added demand of incorporating ethnic identity into their attainment of personal identity.[1] Ethnic identity refers to perceptions of shared unique cultural traditions and a heritage that persists across generations.[1,2] Ethnic group identification accords the individual a sense of belonging and group pride, and a set of prescribed norms, values, and social behaviors.[2-4]

The formation of one's ethnic identity can be viewed as a continuum, ranging from totally uncertain to absolutely certain. As youth explore the boundaries of their ethnic identity, they may engage in behaviors such as cigarette smoking which place them at an increased risk for negative health consequences. Where a youth is along this continuum, therefore, may be potentially important to consider in the development of prevention programming. Spencer and Markstrom-Adams argue that the formation of an ethnic identity during childhood influences psychological well-being among adults.[1] For example, some pressures which increase ethnic identification may lead to enhanced personal well-being. However, no publications exist which measure ethnic identification among adolescents and explore the relations of minority youths' ethnic identity to cigarette smoking-related variables. The present authors attempted to assess the components of ethnic identity among urban minority adolescents. In addition, they assessed the relations of ethnic identification with adolescent tobacco use-related variables.

Methods

Design

Seventh grade students attending health and science classes at three urban Southern California junior high schools in the cities of Los Angeles (South Central), Southgate (North), and Pomona (East) were selected for participation. The schools were selected so as to obtain a highly urbanized, minority sample which is reflective of the different regions within the Los Angeles Basin. With the support of each respective school's administration, teachers of health and science classes were provided personalized verbal and written invitations from the research team to participate in the study. All of the teachers (n = 4) of health and science classes agreed to participate, providing the research team with 20 classes of youth for inclusion in the study.

The study involved a two-group experimental design. Across the three schools, 20 classrooms were randomly assigned to either a discussion-only group (n = 263) or videotape condition (n = 282). Within the videotape condition, the 10 classrooms were randomly assigned to view a rap or soap opera-style videotape first, followed by a discussion.

Procedure

Five hundred and forty-five seventh grade students completed a pretest questionnaire, as part of a larger study on smoking prevention for urban minority adolescents through use of videotape messages. Several items on the test dealt specifically with variables postulated to be involved in the formation of ethnic identity. The pretest was administered during one class period (45 minutes) to all participants in the larger study. Within each condition, subjects were administered the pretest, provided the intervention (discussion only or video plus discussion), administered a posttest, and thanked for their participation. Informed consent was obtained through active parental and student assent for each of the subjects, either in writing or over the telephone. Overall, 75% of the enrolled students were consented; of the students consented, 85% participated in the assessment. Youth were instructed they could withdraw from the study at any time without suffering any penalty. More detail on the design and procedure is provided in Sussman et al.[5]

Instrument

Pretest Demographic and Smoking-Related Items

The pretest included items on gender, age, city of residence, with whom students live, ethnic group name, and lifetime use of cigarettes. The pretest also contained four smoking-related items which served as the smoking measures in the current study. Students were asked to indicate on a 5-point rating scale, ranging from (1) none to (5) all, how many of their friends smoke cigarettes. This scale was recoded to indicate whether or not any of their friends smoked (yes/no). On a 3-point scale, ranging from (1) no, never to (3) yes, more than once, students were asked whether they had ever smoked even a puff of a cigarette. This scale was recoded to indicate whether or not they had ever smoked (yes/no). Tobacco use avoidance self-efficacy was assessed by asking students to indicate on a dichotomous (yes/no) scale whether they could make new friends with students their age without using tobacco, and, on a separate dichotomous (yes/no) scale, whether they are able to avoid students their age when these peers use tobacco.

Ethnic Identification Items

Ethnic identity is typically measured with one item which asks the respondent to specify his or her race, from which ethnicity is determined by the investigator. A more comprehensive measurement of ethnic identity should involve the assessment of the ethnic group/groups with which an individual identifies, in conjunction with measures of domains of the culture/cultures with which the individual identifies. These measures might include, at minimum, self-identified ethnicity,[6-8] religiosity,[9] familiarity with the culture of the self-identified ethnic group affiliation and that of the host culture[10-12] social activities,[12] fatalism,[9] language preference,[7,10,13,14] language proficiency,[13] values and attitudes,[15-17] and health-seeking behavior.[18] The components of ethnic identity measured by the items in this study include self-identified ethnicity, language/vernacular preferences, the degree of homogeneity of social activities, and familiarity with self-ascribed ethnicity. The items used in this study to measure ethnic identification were derived from the combination of a review of the literature, focus group discussions with adolescents, interviews with African American researchers, and pilot testing for readability, comprehensibility, relevance, and linguistics with 40 ethnic minority youth who either attended a local urban academy or an urban dance instruction center.[19-24] A total of seven items were retained (Table 1).

The pretest was structured such that the 13-item scale directly followed a question about the students' own ethnic group name. Students then were asked to answer the ethnic identification items based upon the ethnic group name with which they identified in the previous question. The students' ethnic group names item consisted of 10 alternative categories. Three categories, 'African American/Black', 'African', and 'Caribbean' were recoded as 'African American'. One category, 'Latino/Hispanic', composed a Latino category. Finally, the remaining categories, 'Native American', 'Asian American', 'Pacific Islander', 'White/Anglo (Non-Hispanic)', 'Middle Eastern' and 'other' were all recoded as 'other'.

Analyses

A correlation matrix of the seven ethnic identification items was generated. An examination of the matrix revealed r/s values generally around 0.15. Only three of the items correlated sufficiently high with each other to provide a Cronbach's alpha of 0.66, if combined ('I prefer being in social groups where all of the people are from my ethnic group', 'People from my ethnic group are liked by people from other ethnic groups', and 'I enjoy being in social groups with people who are in different ethnic groups').

Because of the low overall alphas (0.50), we completed further analyses at an individual item level of analysis. Chi-square (χ2) models were calculated (df = 1) which examined the association of each of the seven ethnic identification items with each of the four smoking-related items separately for African Americans and Latinos (4 smoking-related items X 7 ethnic identification items x 2 ethnic groups = 56 analyses). Analyses were not calculated for the category 'others', because of their status as a multiethnic group and small sample size. Of 56 tests, 10 were significant at p </ = 0.05 and four additional tests reached a p </ = 0.01 level; three tests would be significant (at p </ = 0.05) by chance alone. The results are presented grouped by smoking-related item.

Results

The ethnic profile of the sample included 31% who identified as African American, 56% who identified as Latino, and 13% who identified as other (Native American, Asian American, White, mixed ethnicities). Females comprised 51% of the sample. Participants ranged in age from 11 to 14 years (mean = 12.6 years, SD = 1.35). Sixty percent of the sample lived with both parents, while 23% lived only with their mother, and 17% lived with another adult.

Socioeconomic status (SES) was assessed through ratings of parents' (or guardians') occupation. The most frequently endorsed of eight categories for male care-giver's occupation were skilled laborers (31%), minor professionals (14%), small business owners or managers (14%), and semi-skilled laborers (14%) smoking among ethnic minority youth. Most of the students' female care-givers were classified as minor professionals (23%), homemakers or unskilled laborers (21%), semi-skilled laborers (12%), or small business owners (11%). No ethnic differences were found for SES.

Cigarette smoking behavior was reported by 34% of the students, ranging from one puff to greater than five packs over their lifetime, and 4% of the students indicated that they are current smokers. Thirty-three percent reported that at least one family member smokes. Having friends who smoke was reported by 23% of the subjects. Ninety-one percent reported that they could make new friends without using tobacco; 86% said that they could avoid others who used tobacco. No differences were found between the African Americans and Latinos on these measures.

Avoidance of Peers When They Use Tobacco

As shown in Table 2, 65% of the African American sample perceived themselves as looking different than people who are from other ethnic groups and reported being able to avoid peers when they use tobacco, whereas only 20% did not perceive themselves as looking different but reported being able to avoid others' use (χ2= 4.23, p = 0.04). For the Latino youth, there were associations between being able to avoid peers when peers use tobacco and reports that (a) their ethnic group is liked (versus not liked) by other ethnic groups (70% vs 17%; χ2= 4.54, p = 0.03), (b) that they enjoy interacting (versus not interacting) with people outside of their ethnic group (81% vs 7%; χ2 = 4.04, p = 0.04), and (c) that they look different (versus not different) from other ethnic groups (64% vs 22%; χ2 = 3.21, p = 0.07). Fewer Latino youth reported being able to avoid peers when they use tobacco and also indicated preference for same-ethnicity social groups compared to those who indicated not preferring same-ethnicity social groups (20% vs 67%; χ2 = 4.97, p = 0.03).

Having Friends Who Smoke

Among the African American youth, there was a positive association between the presence of books in the house written by people of their ethnic group (versus no such books in the house) and being more likely to report having friends who smoke (16% vs 5%; χ2 = 4.05, p = 0.04). The converse trend was observed (though not significant at p < 0.05) for the Latino youth (10% vs 15%; χ2 = 2.75, p = 0.10). Also, for Latinos there was a significant negative association between the perception of being liked by those outside the ethnic group and having friends who smoke (22% vs 3%; χ2 = 7.56, p = 0.01).

Able to Make New Friends Without Using Tobacco

Watching television shows in which the main characters are members of their ethnic group (versus not watching such shows) was significantly associated with being able to make new friends without using tobacco among the African American youth (54% vs 33%; χ2 = 4.26, p = 0.04). For the Latino youth, there was a highly significant negative association between watching ethnic-specific television shows and being able to make new friends without using tobacco (16% vs 77%; χ2= 13.78, p = 0.0001). Also, for Latinos, there was a marginal positive association between the perception of being liked by those outside the ethnic group and being able to make new friends without using tobacco (74% vs 19%; χ2 = 3.04, p = 0.08). For Latinos, there also was a significant positive association between reporting enjoyment of being with those of other ethnic groups and being able to make new friends without using tobacco (86% vs 7%; χ2 = 5.56, p = 0.01).

Having Tried a Cigarette

The report that one's ethnic group is liked by other ethnic groups was positively associated with having tried a cigarette among the Latino youth (32% vs 6%; χ2 = 4.57, p = 0.03) and a similar trend was found among the African American youth (25% vs 6%; χ2 = 3.11, p = 0.08).

Discussion

Ethnic identity refers to a sense of group or collective identity based on one's perception that he or she shares a common racial heritage with a particular racial group,[10,25] Thus, as a construct, ethnic identity has the potential of providing a conceptual framework for interpreting the ongoing experience or fit between self and the environment.[1] The results of this study suggest that there may be some underlying elements of ethnic identification which serve a protective role against smoking among adolescents. Two results were in common for both African Americans and Latinos. First, the perception of looking different from other ethnic groups was associated with ease of avoiding peers' use of tobacco. Second, the perception of being liked by other ethnic groups was associated with having tried a cigarette. There are multiple explanations for these results--including restriction of social access to tobacco leading to a lower likelihood of using tobacco, or the perception of other ethnic groups as using relatively more tobacco. These findings also do support the contention that some components of ethnic identification are protective against tobacco use.

Two results suggest converse relations of some ethnic identification items with smoking-related items when comparing African Americans with Latinos. First, owning books written by same-ethnic group authors was associated with friends's cigarette smoking among African Americans, but with friends not smoking among Latinos. Perhaps such books among African Americans are rebellious in content and might be associated with rebellious behaviors whereas Latino-oriented books may reflect a lack of impact of the negative effects of acculturation on behavior. Second, while viewing ethnic-specific television programs among African Americans was associated with being able to make new friends who are not smokers, the converse relation between TV shows viewed with this same smoking-related item was found for Latinos. Perhaps, regarding TV viewing, seeing oneself in roles normally held by the majority ethnicity (e.g., entertainment) results in increased ethnic pride among African American youth, which manifests itself in perceived ability to make new friends while engaging in protective health behaviors (e.g., not smoking). The relatively lower number of Spanish language/ culture television programs available across television stations may account for the converse effect of same-ethnic group television characters on smoking-related variables for Latinos. The interpretations of this set of results, of course, are speculative.

Several results applied only to Latinos. If they perceived themselves as being liked by other groups, they were relatively less likely to report friends as being smokers, and were more likely to report being able to avoid peers who were smoking and make new friends who were not smokers. Enjoyment or preferring to interact with persons from other ethnic groups, or not preferring to interact with their own ethnic group, were associated either with ease of avoiding situations where peers smoke or with making new friends who were not smokers. On the other hand, as with African Americans, if they perceived themselves as being liked by other ethnic groups, they were more likely themselves to smoke. Perhaps, these results indicate that effects of acculturation are relatively strong for Latinos.