TITLE/TITRE: The Best Interests of Children: An Evidence-Based Approach (Toronto: University of Toronto Press, 2009).
AUTHOR/AUTEUR: Paul Millar
REVIEW/CRITIQUE: Book Review by Gene C. Colman SOURCE/SOURCE: Canadian Journal of Family Law / Revue canadienne de droit familial
CITED/CITÉ: (2010) 26 Can. J. Fam. L. 269 - 277/ (2010) 26 Rev. Can. D. Fam. 269 - 277
1 What variables influence Canadian courts when they grant custody orders? What variables should influence Canadian courts? Is there any empirical data that would inform the policy maker and the decision maker of the need for relevant principles to apply to the decision making process? Paul Millar was most recently a post-doctorate fellow in the department of community health sciences at Brock University. In his new book, The Best Interests of Children, Dr. Millar creatively and expertly makes sense of the available empirical data. The author presents a persuasive and well thought out analysis that explains child custody outcomes in Canada.
2 Proponents of equal shared parenting and their detractors alike revel in citing statistics. Each side of the debate can marshal statistics to support their respective positions. What distinguishes Dr. Millar's work is his expert use of two mega-sample databases, his statistical analysis of those databases, and his effortless explanation of what the data means in common sense language.
3 Dr. Millar applied to the Canadian Department of Justice, pursuant to Canada's Freedom of Information Act. He was able to obtain data from the Central Divorce Registry of the Centre for Justice Statistics (He also obtained additional data from the National Longitudinal Survey of Children and Youth ("NLSCY")). The data that he obtained from Justice Canada included over 2.5 million divorces in Canada between June 1, 1996 and the end of September, 2002. After he excluded some of the data on a variety of grounds, he was left with a sample of 673,450 children and 378,390 cases. He examined the outcome of custody cases according to three categories: no custody, joint custody, and sole custody. The data was further broken down according to multiple variables including the following:
- grounds for divorce;
- number of children;
- child's age at divorce;
- family position;
- region;
- marriage length;
- age difference between husband and wife;
- wife's age at marriage;
- wife's age at divorce;
- gender;
- who was petitioner and who was respondent; and
- whether a contested hearing was held.
4 Dr. Millar also provides us with a basic understanding of the importance of child support to household income. This is the first attempt to gauge the effects of child support on household income in Canada using a representative database. In this case, the source was the Survey of Labour and Income Dynamics ("SLID"), another database maintained by Statistics Canada.
5 Dr. Millar's empirical findings, based on a multidimensional analysis of the data, are nothing less than startling. The statistical data clearly demonstrate that time and again, the custody result seems to be tied to gender no matter what configuration of factors is examined. The clearest predictor of result is gender. For example (see pages 24 - 25), Dr. Millar finds that mothers gain some form of custody (whether sole or joint) in 89% of cases, while men completely lose custody in 67% of cases, although losing legal custody does not necessarily imply they lose all contact with their children. Custody results from this study do not include data on access or contact with the children. They include court orders pursuant to a hearing as well as so called "consent" orders, since divorce negotiations are conducted "in the shadow of the law", that is, within the context of the expected outcome in a court of law. Dr. Millar compares this to a previous 1979 study based on custody data from 1975 (pages 25 - 26) where the results were virtually identical with respect to sole custody situations. It would therefore appear that the 1986 amendments to the Divorce Act had little effect on modifying the effects of gender on sole custody outcomes.
6 Dr. Millar finds that where there is a contested hearing, it improves the chances for sole custody for both parents. The probability for success of mothers, however, is significantly greater, such that fathers have five times less probability of succeeding in gaining custody than do mothers at a contested hearing. The focus of the book is on custody outcomes and the data being analyzed is with respect to custody claims only. The extent of access claimed and granted would surely make for an interesting follow up study.
7 Mothers are approximately 27 times (2,600%) more likely to obtain sole custody than no custody, and have five times greater odds of achieving joint custody than no custody. Most mothers were the petitioner, and indeed, being the petitioner increases the odds of obtaining both joint and sole custody over no custody by 71% and 281%, respectively. Dr. Millar finds that being the petitioner increases the odds of gaining sole and joint custody by 280% and 70%, respectively, independent of parental gender.
8 Dr. Millar's research determines that where the grounds for divorce are physical or mental cruelty, this appears to reduce the chances for joint custody significantly (see page 30). The longer the marriage, the greater the chance there is for joint custody over no custody. The more children there are in the marriage, the less chance there is for an order of joint custody. Millar examines the interplay between gender with: length of marriage, grounds for divorce, and whether a hearing was held. With respect to all three factors, gender has a significantly greater statistical effect on the result, with mothers still having about 142 times greater odds of achieving sole custody and a 12 times greater chance of achieving joint custody, compared to no custody, when there is no hearing, the husband has not filed for divorce on the basis of adultery, and when the marriage was very short.
9 A surprising finding in this respect is that the double standard is apparently alive and well in Canada's courts. If the husband alleges adultery on the part of the wife, the wife's chances of achieving joint or sole custody are reduced by 37% and 58%, respectively. Feminist legal analysis has claimed that where fathers claim custody they tend to achieve that result more often in the courts. Millar found some support for this. The data reveals that a contested hearing reduces the chances of mothers obtaining custody by 8% for joint custody and 16% for sole custody.
10 Millar finds that if the "mitigating factors" of adultery, marriage length, and contested hearing are not present, the mother's chances of obtaining some form of custody would be even higher than the 2600% chance previously noted.
AUTHOR/AUTEUR: Paul Millar
REVIEW/CRITIQUE: Book Review by Gene C. Colman SOURCE/SOURCE: Canadian Journal of Family Law / Revue canadienne de droit familial
CITED/CITÉ: (2010) 26 Can. J. Fam. L. 269 - 277/ (2010) 26 Rev. Can. D. Fam. 269 - 277
1 What variables influence Canadian courts when they grant custody orders? What variables should influence Canadian courts? Is there any empirical data that would inform the policy maker and the decision maker of the need for relevant principles to apply to the decision making process? Paul Millar was most recently a post-doctorate fellow in the department of community health sciences at Brock University. In his new book, The Best Interests of Children, Dr. Millar creatively and expertly makes sense of the available empirical data. The author presents a persuasive and well thought out analysis that explains child custody outcomes in Canada.
2 Proponents of equal shared parenting and their detractors alike revel in citing statistics. Each side of the debate can marshal statistics to support their respective positions. What distinguishes Dr. Millar's work is his expert use of two mega-sample databases, his statistical analysis of those databases, and his effortless explanation of what the data means in common sense language.
3 Dr. Millar applied to the Canadian Department of Justice, pursuant to Canada's Freedom of Information Act. He was able to obtain data from the Central Divorce Registry of the Centre for Justice Statistics (He also obtained additional data from the National Longitudinal Survey of Children and Youth ("NLSCY")). The data that he obtained from Justice Canada included over 2.5 million divorces in Canada between June 1, 1996 and the end of September, 2002. After he excluded some of the data on a variety of grounds, he was left with a sample of 673,450 children and 378,390 cases. He examined the outcome of custody cases according to three categories: no custody, joint custody, and sole custody. The data was further broken down according to multiple variables including the following:
- grounds for divorce;
- number of children;
- child's age at divorce;
- family position;
- region;
- marriage length;
- age difference between husband and wife;
- wife's age at marriage;
- wife's age at divorce;
- gender;
- who was petitioner and who was respondent; and
- whether a contested hearing was held.
4 Dr. Millar also provides us with a basic understanding of the importance of child support to household income. This is the first attempt to gauge the effects of child support on household income in Canada using a representative database. In this case, the source was the Survey of Labour and Income Dynamics ("SLID"), another database maintained by Statistics Canada.
5 Dr. Millar's empirical findings, based on a multidimensional analysis of the data, are nothing less than startling. The statistical data clearly demonstrate that time and again, the custody result seems to be tied to gender no matter what configuration of factors is examined. The clearest predictor of result is gender. For example (see pages 24 - 25), Dr. Millar finds that mothers gain some form of custody (whether sole or joint) in 89% of cases, while men completely lose custody in 67% of cases, although losing legal custody does not necessarily imply they lose all contact with their children. Custody results from this study do not include data on access or contact with the children. They include court orders pursuant to a hearing as well as so called "consent" orders, since divorce negotiations are conducted "in the shadow of the law", that is, within the context of the expected outcome in a court of law. Dr. Millar compares this to a previous 1979 study based on custody data from 1975 (pages 25 - 26) where the results were virtually identical with respect to sole custody situations. It would therefore appear that the 1986 amendments to the Divorce Act had little effect on modifying the effects of gender on sole custody outcomes.
6 Dr. Millar finds that where there is a contested hearing, it improves the chances for sole custody for both parents. The probability for success of mothers, however, is significantly greater, such that fathers have five times less probability of succeeding in gaining custody than do mothers at a contested hearing. The focus of the book is on custody outcomes and the data being analyzed is with respect to custody claims only. The extent of access claimed and granted would surely make for an interesting follow up study.
7 Mothers are approximately 27 times (2,600%) more likely to obtain sole custody than no custody, and have five times greater odds of achieving joint custody than no custody. Most mothers were the petitioner, and indeed, being the petitioner increases the odds of obtaining both joint and sole custody over no custody by 71% and 281%, respectively. Dr. Millar finds that being the petitioner increases the odds of gaining sole and joint custody by 280% and 70%, respectively, independent of parental gender.
8 Dr. Millar's research determines that where the grounds for divorce are physical or mental cruelty, this appears to reduce the chances for joint custody significantly (see page 30). The longer the marriage, the greater the chance there is for joint custody over no custody. The more children there are in the marriage, the less chance there is for an order of joint custody. Millar examines the interplay between gender with: length of marriage, grounds for divorce, and whether a hearing was held. With respect to all three factors, gender has a significantly greater statistical effect on the result, with mothers still having about 142 times greater odds of achieving sole custody and a 12 times greater chance of achieving joint custody, compared to no custody, when there is no hearing, the husband has not filed for divorce on the basis of adultery, and when the marriage was very short.
9 A surprising finding in this respect is that the double standard is apparently alive and well in Canada's courts. If the husband alleges adultery on the part of the wife, the wife's chances of achieving joint or sole custody are reduced by 37% and 58%, respectively. Feminist legal analysis has claimed that where fathers claim custody they tend to achieve that result more often in the courts. Millar found some support for this. The data reveals that a contested hearing reduces the chances of mothers obtaining custody by 8% for joint custody and 16% for sole custody.
10 Millar finds that if the "mitigating factors" of adultery, marriage length, and contested hearing are not present, the mother's chances of obtaining some form of custody would be even higher than the 2600% chance previously noted.
Comment