Cross-Cultural Studies of Families: Hidden Differences

Article excerpt

The conduct of research across cultures is made more complex for many reasons, such as definition of constructs, relevance of constructs, issues on which people are willing to report honestly, differences in diversity of population within countries/cultures, difficulty comparing samples due to incomparability of demographic indicators, etc. Of the problems faced in conducting cross-cultural studies of families, we wish to address two.

One is related to conceptualization and measurement of constructs. The problem may exist because researchers from one country have collected data in their own country and then found a colleague who could collect the same data in another country, and they have done this without consideration of the construct and/or its measurement in the second country. They have done this partly because of the assumption that characteristics of individual development and/or difference and properties of relationships are universal, much like the elements of the biological and physical world that are assumed to be universal.

The second problem is the assumption of homogeneity within cultures leading to the comparison across countries and thereby missing some of the subtlety of differences. There probably are some universals in the process of human development, and the institution of the family is surprisingly similar across countries, especially developed countries; thus, in cross-cultural research on families, significant differences often represent small real differences; that is, effect sizes are small. Still, we can see differences in social and political systems, living conditions, work conditions, and educational systems that are expected to play a role in shaping families. Indeed, simple, real life observations of families reveal what appear to be differences that are larger than those found by most researchers.


Drawing on the literature on sociological and psychological cross-cultural research, we planned a study of families in four countries. In this paper, we describe the complex, long-term process we used in co-constructing this study. We report some of the revelations we experienced from the beginning of the study to the interpretation of data analyses.

Why study this issue?

In our case, this study was initiated because Michael Matskovsky had been granted money to study families throughout the Soviet Union. The first phase of the study was to be "conducted in the Republic of Georgia. Matskovsky wanted to know how Soviet families compared to families in other countries, an issue on which there were little data at that time. When he and Walters first discussed the study, Walters was intrigued with the idea and agreed to go to Moscow to plan the study. It was apparent that Matskovsky and his colleague Tatyana Gurko were interested in studying young families. Not knowing the emphasis on young families in the Soviet Union, Walters considered this interest a fascinating evidence of altruism toward families. It was not until several years later when more information was available and after many conversations with Russian colleagues that Walters began to realize her mistake. It was not a problem, it is simply an example of lack of communication that derives from unexplained assumptions o f knowledge or understanding. It is probably not as dramatic an example as might be observed in studies where a researcher from one country goes to another country, finds a colleague to collect data, and assumes that constructs and purposes are mutually understood.

Role of theory and decisions about what to study

When studies are developed collaboratively, decisions about what to study can be difficult. In our case, we knew we wanted a broadly based study, but we had great difficulty with the effect of theory and the role of theory. The problem with the effect of theory was that we were working from different theoretical perspectives, using many of the same words, and meaning slightly different things. …


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.