Social closure among the higher educated: trends in educational homogamy in 55 countries

Abstract

Loglinear analysis is used to assess for 55 countries—and for two age groups within these countries—the degree of educational homogamy with regard to a specific boundary in the educational structure, which separates persons with a relatively high educational level from the rest of the population. The aim of the research is to gain insight in the factors that affect the degree of openness or closure of the group of persons with a high educational level. In all 55 countries, the tendency to marry across the selected boundary is restricted. However, there are rather large differences among the countries in this respect. In more modern countries, Protestant countries, and countries in which the group of higher educated is larger (and hence less exclusive), the boundary is found to be more open than in other countries. In most countries, intermarriage levels are higher among younger couples than among older couples. This trend towards more openness is stronger in more rapidly developing countries and in countries where the group of higher educated is smaller.

Introduction

In this paper, the marriage behavior of people with a high educational level—and hence with good prospects on the labor market—is studied for a large number of countries, which range from the least developed to the highest developed countries in the world. A central objective of the research is to gain insight in the societal factors that influence the degree of openness or closure of the higher levels of the social structure of societies. According to the “closure thesis” (Goldthorpe, 1980; Parkin, 1971), the privileged groups in a society tend to use their resources to protect their positions and they will not let in more persons than are needed to fill up the open places. One of the ways in which they accomplish this is by marrying within their own group (compare Weber, [1921] 1972). This idea is supported by many empirical studies, which show that marriage partners tend to resemble each other with regard to status characteristics like the social class of their parents (Blau and Duncan, 1967; Kalmijn, 1991; Uunk, 1996), their educational level (Berent, 1954; Mare, 1991; Rockwell, 1976) and—if both spouses are employed—their occupational level (Hout, 1982; Lipset and Bendix, 1959; Smits et al., 1999).

This paper contributes to the literature on social openness by performing large-scale comparative research on educational assortative marriage (or educational homogamy) with respect to one specific boundary in the educational structure, which is the same over the whole range of countries under study. In earlier studies involving a large number of countries, global measures of educational homogamy were used, which gave insight into the importance of education in choosing a partner for the vast majority of the population of the countries, but not for specific educational groups or boundaries in the educational structure (e.g., Smits et al., 1998, Smits et al., 2000; Ultee and Luijkx, 1990). The educational structures of the 65 countries studied by Smits et al., 1998, Smits et al., 2000 even differed so much that it was not possible to use the same educational classification for all of them. Smits et al. therefore, had to use relative educational categories, which depended on the educational structures of the countries.

The 55 countries studied in the current paper are largely the same as those studied by Smits et al., 1998, Smits et al., 2000. However, this time the large difference among the educational structures of the countries is not considered to be a problem which has to be solved, but as an advantage which will be helpful for addressing the basic research problem of this paper: gaining insight in the factors that influence the openness or closure of the group of persons with a high educational level.

The boundary in the educational structure on which we focus is the one between persons with completed secondary or higher education and people with incomplete secondary or lower education. In highly developed countries, this boundary is somewhere in the middle range of the educational structure and does not seem to be very special. However, in the least developed countries included in this study, this boundary separates a very small and exclusive group of highly educated persons from the large majority of the population with less education. This means, that by comparing the degree of intermarriage across this boundary over the whole range of countries, we can see what happens to the boundary of such an elite group when, with increasing development of the country, the group becomes less special.

Although intermarriage across a boundary in the educational structure involves two groups of individuals—those with more and those with less education—, the strength of the tendency to marry across the selected boundary is expected to reflect largely the marriage preferences of the higher educated. In the competition for a highly educated partner, individuals who have a high educational level have more to offer and hence have a better chance to be successful. The stronger the preference for a highly educated partner among the highly educated, the more each member of this group has to offer at the marriage market and hence the higher the probability that they will marry within their own group. The preferences of the persons with a lower educational level have much less influence on this process. That means that, in the context of this paper, educational homogamy with regard to the selected boundary in the educational structure and social closure among the higher educated are to a large extent equivalent. Throughout the paper, therefore, both terms will be used interchangeable.

The analyses presented here are restricted to social closure in the strict sense, that is, the tendency of persons with a high educational level to marry within their own group. Whether the persons with a high educational level who marry outside their group tend to marry a partner with a more or less closely related educational level cannot be studied with the available data. Because of the large differences among the educational structures of the countries, it is not possible to further subdivide the group of lower educated in a satisfactory way.

In the following and next sections, first some background information on the importance of educational homogamy and social closure for social inequality is given. After that, the significance of comparative research in this area is discussed and the research questions are presented. The theoretical part ends with the formulation of several hypotheses about effects of economic development, dominant religion, and the size of the group of higher educated on (trends in) educational homogamy. In the empirical part of the paper, the degree of educational homogamy among older and younger couples is estimated for each of the 55 countries. For this purpose loglinear analysis is used, a technique which makes it possible to control for the differences among the educational structures of the countries. To test the hypotheses about the differences in homogamy and in trends among the countries, characteristics of the countries are added as explanatory variables to the loglinear models.