EFFECT OF LANGUAGE OF INTERVIEW ON THE VALIDITY AND RELIABILITY OF PSYCHOLOGICAL WELL-BEING SCALES

This study tested the hypothesis that the use of different languages i n a telephone survey could adversely affect the cross-cultural compara bility of standardized research measures. The data used in the analysi s were taken from the 1988 Notional Survey of Hispanic Elderly People (NSHEP) to test this hypothesis. The NSHEP survey contains data on 2,2 99 Hispanic persons aged 65 and older, and it was designed to study th e living arrangements, health status, use of services, and attitudes t oward aging problems held by the respondents. A telephone interview wa s used to collect the data. A majority of the respondents (86.6 percen t) were interviewed in Spanish, and 13.4 percent were interviewed in E nglish. LISREL-7 was the method of statistical analysis used to examin e the factor equality of the positive affect and negative effect stale s between the English and Spanish language respondents. Our findings i ndicate that there was no significant difference between the factor st ructure of the negative affect scale between the two groups; however, the factor structure associated with the positive affect scale demonst rated marked factorial differences between the two groups.