Cultural and sub-cultural differences in reliability : An empirical study in Japan and Malaysia

Purpose – The purpose of this paper is to focus on testing hypotheses derived from theories presented by leading scholars regarding possible sources of differences in reliability coefficients, using cross-sectional data from Japan and Malaysia. Design/methodology/approach – A survey was conducted in Malaysia and Japan, both of them characterized as having collectivist values. Convenience sampling and the same method of survey administration, personal interviews by trained interviewers, was used in both countries. Findings – The tests of the hypotheses regarding cultural, subcultural, and aging effects provide some results in line with the hypotheses, but they raise additional concerns about the possible sources of measurement differences and the applicability of theories and measures developed in one country to other countries and subcultures within countries. The major findings include the alpha reliability coefficient of the self-esteem scale (SSE) which is hypothesised to be lower than the alpha for the material value scale (MVS) and satisfaction with life scale (SWLS) in the Malaysian and Japanese samples, but this difference only occurs for the Japanese sample. Research limitations/implications – The study findings highlight the challenges that researchers confront when conducting cross-cultural studies. Originality/value – The present study systematically tests the reliability of measures developed in the West and employed in different countries (especially Asian countries). Although several assumptions and speculations have been made regarding the sources of potential bias in measures used when studying cultural and subcultural differences, little empirical work supports these views. Our study fills this gap.