Effects of Individual and Product-specific Variables
Received November 1987 Revised March 1988
Sevgin A. Eroglu
Georgia State University, and
Karen A. Machleit
University of Cincinnati To date, considerable research effort has been made to determine whether country of origin affects consumers' product evaluations. Previous studies in industrial (for example White, 1977; White and Cundiff, 1978) and consumer buying (see for example, Andersen and Cunningham, 1972; Lillis and Narayana, 1974; Gaedeke, 1973; Han and Terpstra, 1988) have shown the country of origin to be a salient cue in buyers' evaluations of product quality. However, due to other conflicting findings, some questions have been raised about the theory and methodology of the research (Bilkey and Nes, 1982; Jaffe and Nebenzahl, 1984) as well as whether the country of origin effect even exists (Johansson et al.; Schellinck, 1986). The conflicting findings in the literature have been attributed, in part, to the methodological limitations of past studies (Bilkey and Nes, 1982). One of the major limitations Bilkey and Nes point out about this research stream concerns the number of product cues considered. Most studies used a single cue, country of origin, as the only information on which respondents based their evaluations. This not only created internal and external validity problems, but also prohibited the assessment of how much influence the country of origin cue has in the presence of other product cues. Second, the authors contend that respondents were typically asked to base their product evaluations on only verbal references to products. The use of an actual, as opposed to an imaginary product, might have produced completely different results. A third major limitation concerns the general lack of attention to reliability and validity issues. The inadequate or non-existent reliability assessment and the ability of subjects to guess the purpose of the research might have resulted in overstated, understated, or spurious results. Recently Han and Terpstra (1988) and Johansson et al. (1985) addressed some of the above limitations pointed out by Bilkey and Nes (1982). Both studies adopted a multicue approach to alleviate the problem of using country of origin as a single cue. Han and Terpstra, for example, included seller's brand name among
International Marketing Review 6,6 28
the evaluative cues and found the salience of country of origin cue in consumers' product evaluations. Johansson et al., included some other factors such as respondent demographics and product familiarity in their analysis. Their findings indicated less significant country of origin effects than have been found in previous studies. The authors, however, provided insights about appropriate research methodology before making premature conclusions against country of origin influences. The purpose of this article is to address some of the aforementioned theoretical and methodological issues about the country of origin research. In particular, it attempts to: (1) use the cue paradigm as a conceptual framework for determining the relative perceived predictive value of country of origin as a quality indicator when other salient product cues are present; (2) investigate the extent to which this influence depends on selected individual and product variables that prior research has suggested to be important in cue utilisation and, (3) eliminate some of the methodological limitations noted by Bilkey and Nes (1982) when testing the posited relationships. These relationships are examined in a causal framework to simultaneously consider a larger number of influences than have been previously addressed. Background This research integrates previous findings on country of origin impact in an information-theoretical perspective which focuses on cue characteristics and cue...