Strong and weak family ties revisited: reconsidering European family structures from a network perspective

Abstract

Family systems appear to be an important factor framing people's individual behavior. Thus far, family systems have been primarily addressed on a macro regional level with indirect measures. Revisiting Reher (1998) and the family ties criterion, the main question of this paper is to examine to what extent we perceive family structures differently in Europe by taking direct measures of the structures of people's broader social networks into consideration. Based on the Survey of Health, Ageing and Retirement in Europe (SHARE), we derived two indicators of family regimes based on individual-level data regarding the density of ego social networks: contact frequency and geographic proximity among network members. We aggregated these data and mapped them on the NUTS 2 level regions for various locations in Europe. The results of our analyses exhibit that, based on these two network indicators, significant differences in family structures between European regions exist. These results confirm the classification of strong family Southern and comparatively weaker family Northern European regions to a large extent, though substantial regional differences in and between countries are also revealed. Our findings demonstrate that the classification of European regions largely depends on which indicator of network density we consider. This is particularly obvious in the Eastern European regions where the classification markedly differs according to the type of network indicator. Intriguingly, social networks in Central European regions can be characterized as rather loose, often even looser than the ‘traditional’ weak ties in Scandinavia. Family regimes can, therefore, be regarded as a construct of multiple dimensions of which one dimension may be classified as weak while the other can be strong at the same time.

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 — 43 citations exist upstream.

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