The family information section within the General Lifestyle Survey (GLF) and the General Household Survey (GHS) provides the longest running time series of reliable statistics on partnerships and relationships in Great Britain. It is one of the only government surveys that collects data on family history, which is vital for understanding partnership transitions and partnership stability. Data have been collected since 1979, allowing us to look at changes in relationship patterns over 32 years.
In 1979 questions on marital history were introduced on the GHS for both men and women. Questions were also introduced for women aged 18 to 49 relating to pre-marital cohabitation before the current or most recent marriage. In 1986 these questions were extended to both men and women aged 16 to 59 and every marriage past and present. In 1998 a further question was added to find out the number of past cohabitations not ending in marriage and in 2000 new questions were included on the length of past cohabitations not ending in marriage. The 2011 GLF continued to collect information about marital history and periods of cohabitation from adults aged 16 to 59.
How the data are used and their importance
GLF and GHS data on family formation have been extremely valuable for understanding the changes that have occurred in society over the last four decades. For example, changes in the proportion of cohabiting partners and in the stability of relationships have highlighted policy issues, such as the rights of cohabiting couples and the welfare of children. For these reasons, the GLF and GHS are widely used by researchers working within and outside Government. An example is the Population Trends research article, Cohabitation and marriage in Britain since the 1970s, written by the Centre for Population Change (CPC). The article uses GHS and GLF data to provide an overview of trends in marriages and relationships over several decades and emphasises the importance of good information on family trends.
The GLF estimates of the England and Wales population by partnership status are used to inform and quality assure cohabitation estimates, which themselves feed the assumptions made for cohabitation projections. Estimates and projections of the cohabiting population are made alongside the publication of marital status projections and are used by government departments for statistical modelling relating to housing policy and benefits policy. The cohabitation and marital status estimates for England and Wales are published on the ONS website.
Information about the marital status of all adults aged 16 or over in the household is collected in two stages. First, the marital status of all adults aged 16 and over is collected from the person answering the household questionnaire (usually the Household Reference Person (HRP) or their partner). At the second stage, each household member aged 16 to 59 is asked detailed questions about their marriage and cohabitation history. For this stage, respondents are given the option of self-completion, particularly if the interviewer judges that a lack of privacy might affect reporting. In 2011, around 17% of respondents chose to self-complete the questions.
For the 2009 and 2010 survey years, if a respondent was in the longitudinal sample of the GLF, but had not answered the second stage section of the questionnaire in the previous year, their full marriage and cohabitation history was not collected. As a consequence, the proportion of full responses to the marriage and cohabitation sections of the questionnaire is lower for these years.
De facto marital status (that is, including cohabitation) is the legal marital status of the respondent unless the respondent was currently cohabiting with someone else; in which case cohabiting is the de facto status. Cohabiting couples are people who live together as a couple in a household without being married to each other. Respondents who were single, widowed, divorced or separated but who were cohabiting are here classified as cohabiting, rather than by their legal marital status. Those who were not cohabiting have been classified by their legal marital status.
Table 5.1 shows the de facto marital status of men aged 16 and over in 2011: 52% were married, 1% were in a civil partnership, 11% were cohabiting, 27% were single, 3% were widowed and 6% were either divorced or separated. Among women aged 16 and over the estimates are: 49% were married, less than 1% were in a civil partnership, 11% were cohabiting, 21% were single, 9% were widowed and 10% were either divorced or separated. The proportions of men and women who were married and cohabiting were very similar but due to differences in life expectancy a larger proportion of women were widowed or divorced/separated.
In 2011, 15% of both men and women aged 16 to 59 were currently cohabiting. Among men, those aged 25 to 29 were more likely to be currently cohabitating than any other age group (33%), this difference is not statistically significant when compared with men aged 30 to 34 (25%). Within all other age groups fewer than 20% of men were cohabiting. Similarly, among women aged 16 to 59, those aged 25 to 29 were also more likely to be cohabitating than any other age group (37% of women aged 25 to 29 were cohabiting compared with 2% to 22% in the other age groups). Among adults aged 16 to 59 who were not married, around three in ten men and women were currently cohabiting (28% of men and 29% of women). A higher proportion of younger women (aged under 30) were currently cohabiting than men in each age group. For example, 55% of non-married women aged 25 to 29 were cohabiting compared with 43% of men in this age group. This is likely to be because men form relationships with women younger than themselves on average. For example, the average age at first marriage for women was 30.0 in 2010 but 32.1 for men. The percentage of men and women cohabiting falls after the ages 30-34 as more people get married. Previous research shows that younger people cohabiting tend to have not been married, whereas older people cohabiting tend to be either divorced or widowed.
Data from 2010 and 2011 have been combined to provide a large enough sample to analyse current cohabitation by age and legal marital status. Among non-married men aged 16 to 59, those who were divorced and those who were single were the most likely to be currently cohabiting; 31% of divorced men and 27% of single men compared with 13% of separated men and 3% of widowed men. Among non-married women aged 16 to 59, those who were single were most likely to be currently cohabitating (31%), followed by those who were divorced (24%), separated (9%) and widowed (3%).
As noted earlier, women aged 18 to 49 were the first to be asked questions on cohabitation in the GHS. This section looks at the trends over time for this age group.
Over the 32 years that this data have been collected, the proportion of women aged 18 to 49 who were married has fallen from around three-quarters in 1979 (74%) to slightly fewer than half in 2011 (47%). There was a steady decline between 1979 and 2003 and since then the proportion has changed little. Over the same period, the proportion of women who were single (that is, who had never been married) has increased steadily from 18% in 1979 to 43% in 2011.
The decline in proportion of women who were married contrasts with the increase in the proportion of non-married women aged 18 to 49 who were cohabiting between 1979 and 2011. The proportion of non-married women aged 18 to 49 who were cohabiting increased from one in ten (11%) in 1979 to one third (34%) in 2011. This increase occurred between 1979 and 2001, with the proportion rising steadily from 11% to 32% over that time period. Since 2001 the proportion of non-married women aged 18 to 49 who were cohabiting has ranged between 28% and 35%. A contributing factor to the increase in the proportion of cohabiting non-married women since 1979 has been the increase in the proportion of single women cohabiting. Among single women aged 18 to 49, the proportion cohabitating more than quadrupled from 8% in 1979 to 36% in 2011. Again this increase was steepest between 1979 and 1998, rising from 8% to 31%, and since then the proportion has remained between 29% and 37%.
In households with children, children may be dependent or non-dependent. Dependent children are defined as those aged less than 16 living with at least one parent, or aged 16 to 18 in full-time education, excluding all children who have a spouse, partner or child living in the household.
In 2011, among women aged 16 to 59, 57% of married women had at least one dependent child in their household compared with 45% of cohabitating women, 20% of single women, 30% of divorced women and 60% of separated women (the difference between the proportions found for married and separated not statistically significant).
In 2011, as in previous years, non-married women aged 16 to 59 who had dependent children in their household were more likely than those without dependent children to be cohabiting. 44% of non-married women who had at least one dependent child living with them were cohabiting, compared with 22% of non-married women without dependent children. This difference is largely due to the proportions of single women in each group; over half (54%) of single women who had dependent children living with them were cohabiting, compared with slightly fewer than a quarter (23%) of single women without dependent children.
Since 1998, the GHS/GLF has asked a question to find out the number of past cohabitations not ending in marriage. Since 2000, questions have also been included to establish the length of past cohabitations not ending in marriage. These periods of completed cohabitation do not include the current relationship of a respondent living as a couple at the time of interview.
With the exception of those who chose the self-completion option, married and cohabiting respondents might have been interviewed in the presence of their partner. Therefore, it is possible that previous cohabitations may be under-reported for these groups.
In 2011 among adults aged 16-59, 16% had had at least one completed cohabitation that did not end in marriage; the majority had only one completed cohabitation not ending in marriage; (12 %of adults aged 16-59); 3% had two; and 1% had three or more.
As in previous years, the proportions reporting past cohabitations not ending in marriage varied by current marital status for both men and women. Married people were less likely than other respondents to report these kinds of relationships (12% of men and 11% of women) compared with those who were cohabiting (26% of men and 23% of women), single (16% of men and 25% of women) or divorced (23% of men and 22% of women).
First cohabitations that did not end in marriage tended to be longer than second cohabitations. Among adults aged 16 to 59, 36% of first cohabitations lasted for less than two years compared with 46% of second cohabitations. The difference was more marked among women (33% of first cohabitations lasted for less than two years compared with 46% of second cohabitations) than among men (40% compared with 45%).
Among adults aged 16 to 59 who have cohabited, the average length of time for the first cohabitation not ending in marriage was 46 months compared with 34 months for the second cohabitation.
Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: firstname.lastname@example.org