Ethnic differences in women’s social support networks: a latent class analysis

Presenter: Dharmi Kapadia, University of Manchester

Author: Dharmi Kapadia

Social support has been shown to be beneficial for good mental health, and social isolation to be detrimental. Often, aspects of social support networks are used as explanatory variables in epidemiologic studies, as explanations for mental health status. It is much less common for social support network characteristics to be modelled as outcome variables. Further, very few studies have examined the relationship between social support networks and ethnicity in women, and how mental illness may influence this relationship. Previous research in the UK has shown that Pakistani women have networks that are lacking in social support, leading to feelings of social isolation. Women from this ethnic group are also likely to experience greater negative aspects of networks compared with the majority White group.

In order to examine the relationship between ethnicity, mental health and social support networks in women, this paper uses data from Wave 2 of Understanding Society. Social network types were produced using latent class analysis, and their relationships with ethnicity and mental health were explored using multinomial regression.

Preliminary results show that there were four types of support network: extremely supportive, moderately supportive, no partner but supportive relatives and friends, and socially isolated. The results of the multinomial regression indicate that Pakistani women were more likely to have socially isolated networks compared with White British, White Irish and Indian women. High levels of mental illness were associated with having a more socially isolated network.