What you should know about this study

This study is based on the General Social Survey (Cycle 15) on family history, conducted by Statistics Canada during 2001. Almost 25,000 Canadians aged 15 and over living in private households in the 10 provinces were asked to provide information about all their marital and common-law unions, on separation, divorce and death of their partners, as well as a wide array of background characteristics.

This article focuses on adults aged 25 and over who have been legally married a minimum of one time, and the likelihood that their marriage will end in divorce or separation. The analysis is based on about 14,550 respondents who have married only once, 1,750 who married twice and 140 who married more than twice. These respondents represent almost 14.8 million, 1.7 million and 137,000 Canadians aged 25 and over respectively.

Ever-married: Adults aged 25 and over who have been legally married at least once, regardless of their marital status (still married, divorced, widowed) at the time of the survey.

Once- , twice- and serially-married: Persons who, as of the time they were surveyed, had legally married once, twice or more than twice, respectively.

Dissolution: The end of marriage due to separation, divorce or annulment. (Widowhood is excluded.) Because this study examines the breakdown of the relationship rather than its legal termination, dissolution is defined to occur at the time of final separation from the spouse; in the small number of cases where marriage ended with an immediate divorce without a period of legal separation, it is the time at divorce. This category therefore includes respondents who were separated but whose divorce was not yet final; these individuals account for about 30% of all persons in this category.

Risk ratio: The predicted likelihood that an individual's marriage will end in separation or divorce, compared with a reference individual. The ratios were calculated using a proportional hazard model, a statistical technique that estimates the likelihood that an individual will experience an event (in this case, marital dissolution), given a certain set of explanatory variables.

In this study, the explanatory variables are: sex; age at start of marriage; age difference between spouses; whether the couple had lived together before marriage; the decade in which the marriage started; educational level at the time of marriage; whether there were children in the household during the marriage; religious affiliation; religious attendance; mother tongue and region of residence. The model also included variables that measured the respondent's attitudes to marriage, being part of a couple and having children, as well as whether they would stay in an irreparable marriage for the sake of the children (if their children were less than 15 years old).