Overview

Objectives
This project examined whether where women live influences the likelihood of them starting a family. Spatial variations in fertility have been noted within several European countries, including Britain. In particular, low fertility rates tend to be clustered in central urban areas whereas the surrounding peri-urban fringes contain pockets of relatively high fertility. These findings led the research team to hypothesise that there are important local 'cultures' of fertility, which influence an individual's perception of normal or desirable fertility behaviour through mechanisms of social learning. The main objectives of the research were to define and measure local 'cultural' contexts that are meaningful in relation to fertility behaviour, and then link this geography to individual fertility histories from the British Household Panel Survey (BHPS). In addition, it investigated whether there is a significant association between local context and time to first birth after accounting for a range of other factors, including characteristics of housing and selective residential relocation.

Methods
The first stage of the research used geo-coded small area data from birth registration and the 2001 census to identify spatial clusters of higher, average and lower fertility throughout Britain. Spatial patterns of fertility were identified using data for 40 thousands small areas in Britain. The G* Statistic (Getis and Ord 1996) was used, the results were mapped using GIS software and a classification of local areas was produced.

In the second stage, local contexts were linked to individual fertility histories from the BHPS (1999-2008) to produce a longitudinal dataset for analysis of the timing of first births. The study focused on a sub-sample of women of reproductive age (16 to 44) who were childless and not pregnant in 1999. Their risk of conception leading to first births between 1999 and 2008 was observed and resulted in an analytical sample of ~3800 women and 665 events during the observation period.

Findings
Findings have confirmed differences between urban and rural areas, and by settlement size. This had already been observed in other European countries, with lower fertility in cities and higher fertility in less densely populated areas. Women living in local areas within each of these fertility categories have distinctive age-specific fertility profiles. In lower fertility areas, for example, women on average have births later and their fertility is lower at all ages compared to those in higher fertility areas. This local fertility geography has the potential to influence the fertility behaviour of individuals who live in (or move into) these areas through social learning.

The findings indicate that individual life paths respond to a variety of social processes operating at different scales, and that these influences vary by birth order. They support the hypothesis that local fertility context is significantly associated with time to first birth. This is the case even when demographic and socio-economic characteristics of women and their partners, along with housing conditions and patterns of spatial mobility, are taken into account. However, we find no evidence that local fertility contexts influence transitions to higher order births.

Therefore, local social interactions, including social learning, may be playing an important role in shaping individual fertility behaviour through their influence on the transition to first birth.