A CROSS NATIONAL COMPARISON OF THE RELIABILITY OF SELECTED MEASUREMENTS FROM CONSUMER SURVEYS* Harry

Cross national surveys are frequently undertaken for purposes of making comparisons among markets with respect to factors relevant to marketing policies. A potentially troublesome threat to the validity of conclusions drawn from such comparisons is that due to differences in the reliability of measurements arising from linguistic and conceptual non-equivalencies of questionnaire instruments. This paper investigates the question of whether the reliability of measures commonly used in marketing surveys differs cross nationally. Results are presented from a five country study. Significant between-sample reliability differentials were uncovered but no systematic tendency for particular national samples or language groups to exhibit consistently high or low reliability across different types of measures and variables was observed. Between-sample reliability differentials appeared less likely to occur for measures of "hard" variables (e.g. demographics) than for measures of "soft" variables (e.g., life style/ attitudinal variables). Implications of these findings for the design and analysis of cross national surveys are discussed.