Who intermarries in Britain? Explaining ethnic diversity in intermarriage patterns

Abstract

This paper investigates trends, patterns and determinants of intermarriage (and partnership) comparing patterns among men and women and among different ethnic groups in Britain. We distinguish between endogamous (co-ethnic), majority/minority and minority/minority marriages. Hypotheses are derived from the theoretical literatures on assimilation, segmented assimilation and opportunity structures. The empirical analysis is based on the 1988-2006 General Household Surveys (N = 115,494). Consistent with assimilation theory we find that, for all ethnic minority groups, the propensity to intermarry is higher in the second generation than in the first. Consistent with ideas drawn from segmented assimilation theory, we also find that substantial differences in propensity to form majority/minority marriages persist after controls for individual characteristics such as age, educational level, generation and length of residence in Britain, with men and women of Indian, Pakistani or Bangladeshi background having higher propensities to form endogamous partnerships. However, we also find that opportunity structures affect intermarriage propensities for all groups alike, with individuals in more diverse residential areas (as measured by the ratio of majority to minority residents in the area) having higher likelihood to form majority/minority partnerships. We conclude then that, beginning from very different starting points, all groups, both minority and the majority groups exhibit common patterns of generational change and response to opportunity structures. Even the groups that are believed to have the strongest community structures and the strongest norms supporting endogamy appear to be experiencing increasing exogamy in the second generation and in more diverse residential settings. This suggests that a weak rather than a strong version of segmented assimilation provides the best account of British patterns.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 140 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.