Ethnicity-of-Interviewer Effects Among Mexican-Americans and Anglos

A general interpersonal deference explanation for ethnicity-of-interviewer effects is tested with a random survey of Anglos and Mexican-Americans using Anglo and Hispanic interviewers and questions related to Mexican-American culture. The theory was most clearly supported among Anglo respondents. The mixed support found among Mexican-Americans suggests that deference is strongest when questions deal with the culture represented by the interviewer, not the respondent. Stephen Reese and Pamela Shoemaker are Assistant Professors in the Department of Journalism at the University of Texas at Austin, 78712, where Wayne Danielson is Jesse H. Jones Professor and Tsan-Kuo Chang is a doctoral candidate. Huei-Ling Hsu is a doctoral student at the Institute for Communication Research at Stanford University. Public Opinion Quarterly Vol. 50:563-572 ? 1986 by the American Association for Public Opinion Research Published by The University of Chicago Press 0033-362X/86/0050-563/$2.50 This content downloaded from 157.55.39.255 on Mon, 23 May 2016 06:00:22 UTC All use subject to http://about.jstor.org/terms 564 STEPHEN D. REESE AND OTHERS viewer's race. Most studies supporting such a race-of-interviewer effect, however, have examined only blacks and whites. Hyman et al. (1954) found that blacks gave different responses to whites than to other blacks, presumably out of deference to the dominant "caste." Black responses to items regarding militant protest and hostility to whites are particularly sensitive to race of interviewer, while blacks' reports of discrimination, living conditions, and personal background are not (Schuman and Converse, 1971). Subsequent studies have supported a more general interpersonal deference theory, which accounts for interviewer effects on whites as well as blacks (Athey et al., 1960, Hatchett and Schuman, 1975-76, and Campbell, 1981). What's more, a face-to-face interview is not required; the race-of-interviewer effect in a telephone interview has been found which is similar in magnitude to that reported from a personal survey (Cotter et al., 1982). This finding is of particular interest, given the predominance of telephone surveys as a data-gathering method. Other ethnic groups have also been examined for interviewer effects, but the focus of these studies has been on nonracially sensitive items. Weeks and Moore (1981) found that a difference in ethnicity (Cuban, Chicano, Puerto Rican, and Chinese) between interviewers and respondents did not affect responses. Similarly, Welch et al. (1973) reported that Mexican-Americans did not differ in their responses to Anglos and other Mexican-Americans. Both studies were based on personal interviews about general items, supporting the basic theory that general deference does not occur for questions which aren't related to ethnicity. This study tests the general deference theory using a telephone survey of Mexican-Americans and Anglos, with both Hispanic and Anglo interviewers. Both general and ethnicity-related questionnaire items are examined, including questions not previously studied for interviewer effects-ethnic media behavior and ethnic environment. Possible bias in such items is particularly important, given their frequent use in measures of acculturation. On the basis of previous research it is hypothesized that no interviewer effects will be found for nonethnicity-related items (e.g., education), but that both Anglos and Mexican-Americans will defer to opposite-ethnicity interviewers on questions related to Mexican-American culture. In addition, it is hypothesized that among Mexican-Americans this deference will extend to their self-reports of ethnic environment and behavior. In other words, according to the general deference theory we would expect respondents to overreport ethnic behavior when interviewed by someone from the same ethnic background, and underreport it when interviewed by someone else. This content downloaded from 157.55.39.255 on Mon, 23 May 2016 06:00:22 UTC All use subject to http://about.jstor.org/terms ETHNICITY-OF-INTERVIEWER EFFECTS 565