JOSE Anita; O'LEARY K. Daniel; MOYER Anne;
Does premarital cohabitation predict subsequent marital stability and marital quality? A meta-analysis.
Journal citation/publication details
Journal of Marriage and Family, 72(1), February 2010, pp.105-116.
This meta-analysis was carried out with data from 16 studies reporting on marital stability outcomes and 12 studies on marital quality, the majority of which were from the USA or Canada. Overall, cohabitation had a significant negative effect on subsequent marital stability and quality, although there was no negative effect on marriage stability when cohabitees who subsequently married their partner were analysed separately. Conflicting results from individual non-US studies suggest that sociocultural factors may play an important role.
Research on cohabitation with a romantic partner and its associated marital outcomes has produced mixed results with some studies suggesting that the relationship between cohabitees are less stable, less committed and of a lower quality compared to marriages. This meta-analysis focuses on the view of cohabitation as a step in the mate selection process and compares the subgroup of cohabitees 'who ultimately marry to those who do not cohabit before marrying and examines marital outcomes by group in terms of stability and quality'.
What sources were searched?
Two electronic databases PsycINFO and SocINDEX were searched.
What search terms/strategies were used?
The following keyword search was conducted: cohabitation effect, cohabi* AND satisfy*, cohabi* AND quality, cohabi* AND stab*, cohabi* AND instability, cohabi* AND dissolution, cohabi* AND divorce, cohabi* AND separation, cohabi* AND premarital. The search was carried out in January 2009.
What criteria were used to decide on which studies to include?
Studies were eligible for inclusion if they were primary empirical studies comparing premarital cohabitees who subsequently married to couples who married without prior cohabitation. Only studies providing longitudinal or cross sectional data, including primary outcomes measuring postmarital stability or postmarital quality, and which presented data separately for cobabitees and noncohabitees were included. Studies were restricted to those published in English in peer review journals between 1970 and 2008.
Who decided on their relevance and quality?
Abstracts were assessed against the inclusion criteria. Authors were contacted when there was insufficient data for meta-analysis.
How many studies were included and where were they from?
A total of 973 articles was identified 270 appeared in both databases, 304 were found only in PsycINFO, and 399 were only found in SocINDEX. Twenty-six studies were considered for inclusion; the reasons for exclusion are specified in the text.
How were the study findings combined?
Meta-analyses were conducted using SPSS version 16 using formulas and macros according to Lipsey and Milson (2001), the Effect Size Determination Program (Wilson, 2001), and META software. Where studies analysed sub-samples of the same large data sets, the study with the largest sample size and most directly presented effect size estimate was used. Individual study effect sizes are presented for marital stability and marital quality in Table 1 and overall effect sizes and sensitivity analyses in Table 2.
Findings of the review
Sixteen studies, mainly from the USA (9 studies), assessed marital stability. The total sample size was 53,230 (median 2,571). The samples were largely Caucasian females aged 22 to 35 years. An overall effect size (OR) of 0.81 (p<.01, 95% CI .69 to .95) indicated that premarital cohabitees were significantly less likely to stay married compared with those who did not cohabit prior to marriage. Premarital cohabitation was not associated with subsequent marital stability for those who cohabited with the eventual marriage partner.
Twelve studies (ten from the USA and two from Canada) assessed marital quality. The total sample size was 5,768 (median 127.5). The samples were again mostly Caucasian females with a mean age of 28 years. Cohabitation before marriage was found to have a modestly strong negative effect on subsequent marital quality (standardised mean difference -.19, p<.01, 95% CI -.32 to -.07), similar results were obtained when analyses were limited to first marriages, and to cohabitees who eventually married their cohabitation partner.
'When cohabitation is conceptualized as a step in the mate selection process, it is associated with negative marital outcomes'. But 'cohabiting with the eventual marital partner may not be negatively associated with marital stability'.
Implications for policy or practice
None are discussed.