Interviewing in a Mexican-American Community: An Investigation of Some Potential Sources of Response Bias

In: Public opinion quarterly: journal of the American Association for Public Opinion Research, Band 37, Heft 1, S. 115-126

Abstract

The problem of response bias in survey res is addressed. Data from a survey of Mexican-Amer's in 2 Nebraska counties are analyzed to assess the influence of diff aspects of the interview situation on response consistency. 3 specific hyp's are tested: (1) Responses of Mexican-Amer's interviewed in English differ systematically from responses of Mexican-Amer's interviewed in Spanish. (2) Responses of Mexican-Amer's interviewed by Anglos differ systematically from responses of Mexican-Amer's interviewed by Mexican-Amer's, & by Anglos & Mexican-Amer's working together. (3) Responses of Mexican-Amer's born & raised in the US differ systematically from responses of Mexican-Amer's born & raised in Mexico. Initial diff's consistent with the hyp's, were found but disappeared when educ & age were controlled. Minor exceptions appeared to be accounted for by factors other than those examined here. Contrary to initial expectations, responses to both health & pol'al questions were unaffected by these independent variables. The findings re country of origin were more tentative; however, no signif diff's in response to any of the health questions were noted once educ & age were controlled, & only one pol'al question showed a diff when educ & age were controlled. 4 Tables. M. Maxfield.