In lieu of an abstract, here is a brief excerpt of the content:

228 Appendix D Methods for Judicial Opinion Study Judicial opinions that apply the Hague Convention to legal disputes involving both child abduction and domestic violence served as the data for this study. We used judicial opinions because they include detailed information about the parties involved in Hague litigation, and because they are the official records of who prevailed in the legal proceedings. The opinions in our sample were identified through online LexisNexis and Westlaw searches. Typical search terms included physical, violence, harm, assault, fight, punch, harass, and abuse. In all, 57 different combinations of terms were used to identify cases where domestic violence might have occurred. All searches included reference to the Hague Convention. State and federal court opinions—at all court levels—were obtained through our online searches, although no us Supreme Court opinions were located (Abbott v. Abbott [2010] occurred after our search). The results of the LexisNexis and Westlaw searches were supplemented by a search of the International Child Abduction Database (incadat) for disputes involving domestic violence that were litigated in the United States. incadat was established by the Permanent Bureau of the Hague Conference and includes judicial decisions from courts in 34 countries that interpret the 1980 Hague Convention. Search terms mirroring those used in the LexisNexis and Westlaw searches were used for the incadat search. The online search initially yielded 306 opinions. We then took several steps to construct the final sample. First, we reviewed the opinions to determine whether they actually discussed domestic violence and, in fact, litigated the Hague Convention. Opinions that did not meet these criteria were removed. For example, opinions discussing child abuse without allegations of domestic violence were removed, as were those that simply cited or made reference to the Convention without actually issuing a legal judgment about it. It should also be noted that none of the legal opinions in our sample was primarily focused on determining whether domestic violence existed. Domestic violence was regarded by the courts as part of each case’s factual background. Second, we narrowed our sample to include only published opinions, as is typical of empirical legal research (Chew & Kelley, 2005; Perry, Kulik, & Bourhis, 2004; Choi & Gulati, 2008). Methods for Judicial Opinion Study | 229 Data Analysis To facilitate the descriptive analyses, we developed a coding form to help extract needed information from the opinions in our sample. This form was used to gather demographic information about the parties; information about the disputes, including residence and citizenship of parents and children; children’s genders and ages; the severity and frequency of domestic violence; and the presence of child maltreatment. Data were also gathered about key legal issues and decisions, such as the judge’s decision on the children’s habitual residence, the Hague defenses that were litigated during the dispute, and whether the petition was granted or denied. All gathered data were placed into Microsoft Excel and frequencies were determined. Our content analysis focused on evaluating the portions of the opinions that addressed the grave risk defense. Content analysis was selected because it is well suited to the empirical study of legal texts (Hsieh & Shannon, 2005). Indeed, it resembles the process of legal reasoning, including the systematic reading of materials, identifying and coding their consistent features, and drawing inferences about their use and meaning (Hall & Wright, 2008; Lens, 2008). Content analysis has previously been used to study a variety of legal phenomena addressed in judicial opinions, such as use of the primary caretaker standard in custody disputes (Mercer, 1998), the admissibility of expert evidence (Merlino, Murray, & Richardson, 2008), and criminal sentencing of youth sex offenders (Bouhours & Daly, 2007). The content analysis proceeded in two phases. First, we searched for specific terms and phrases that were associated with concepts in our coding sheet, including child maltreatment, the severity and frequency of domestic violence, and whether children testified during the court proceedings. In addition, the research team’s understanding of the lethality of domestic violence sensitized it to the importance of an abuser’s threats to kill his partner or children. As a result, we searched the opinions for any threats made by abusers. Second, it became clear that the opinions included several key concepts that were not included in the coding form, but which were critical to the courts’ decision making. These emergent concepts were recorded as they appeared, and the opinions were subsequently searched for them. An iterative process of identifying and coding key concepts related to the grave risk defense and then returning to the...

Share