Abstract
This article investigates the effect of the rural-urban divide on mean response styles (RSs) and their relationships with the sociodemographic characteristics of the respondents. It uses the Representative Indicator Response Style Means and Covariance Structure (RIRSMACS) method and data from Guyanaa developing country in the Caribbean. The rural-urban divide affects substantial mean RSs differentials, and it moderates both their relationships with and the explanatory power of the respondents' sociodemographic characteristics. Within-country research is therefore subject to substantial rural-urban RSs bias, and it is hence imperative that researchers control RSs in such studies. Previous research findings should also be reexamined with RSs controlled. In addition, joint modeling of culture, RSs, and their sociodemographic predictors may clarify some of the conflicting results about their effects in the cross-cultural research literature.
Original language | English |
---|---|
Pages (from-to) | 97-115 |
Number of pages | 19 |
Journal | Educational and Psychological Measurement |
Volume | 74 |
Issue number | 1 |
DOIs | |
Publication status | Published - Feb 2014 |
Externally published | Yes |
Keywords
- response styles
- rural and urban culture
- survey research
- Guyana
- STRUCTURAL EQUATION MODELS
- MEASUREMENT INVARIANCE
- MARKETING-RESEARCH
- RATING-SCALES
- FIT INDEXES
- EXTREME
- ACQUIESCENCE
- BIAS
- EQUIVALENCE
- COVARIANCE