Translation and cultural adaptation of rating scales are two critical components in testing culturally and/or linguistically heterogeneous populations. Despite the proper use of these scales, challenges typically arise from respondents' language, culture, ratiocination, and characteristics of measurement processes. This study investigated factors affecting the rates of extreme response using direct measures and Spanish-speaking respondents from rural and urban settings. Issues of respondents' familiarity with rating scales, respondents' subjective categories, and scale characteristics were investigated and their relation to extreme response documented. The tendency to choose extreme portions of a rating scale seems to be rooted in the mismatch between scale characteristics and respondents' subjective categories, respondents' familiarity with rating scales, and communication norms. This article discusses implications of findings for validity of translated and adapted rating scales. (Contains 2 tables and 4 figures.)