2 45 PREDICTIVE PROPERTIES DYNAMIC INSTRUMENTS samples were selected as described previously, resulting in the final inclusion of 27 unique samples in this component of the meta-analysis (see Appendix B1, available in the online supplemental materials). The studies on which these 27 samples were based were conducted in Australia (1), Austria (2), Canada (12), New Zealand (1), the United Kingdom (8), and the United States (3), and were published between 2000 and 2014 (mode 2013, median 2010/2011). The vast majority of the papers (44) was published in English. The two remaining papers were published in German. The effect sizes of these studies were taken directly from the publications for 18 samples. For the remaining nine samples, effect sizes were obtained from the authors, who either conducted additional analyses or provided us with the raw data, allowing us to compute effect sizes. Most samples (21) included sex offenders who had offended against children, as well as sex offenders who had offended against adults. Four samples included only sex offenders who had offended against children, and for two samples their composition in terms of offender type was unknown. In 11 of the samples, all offenders received some form of treatment; 10 samples included only untreated offenders; five samples included both treated and untreated offenders, and for one sample, the treatment status was unknown. 2.3.3.3 Incremental validity For the evaluation of incremental validity of dynamic risk assessment instruments (Research Question 2), 22 papers were eligible for inclusion. Ten of these papers consisted of studies with (partly) overlapping samples. These overlapping samples were selected as described previously, resulting in the final inclusion of 16 unique samples in this component of the meta-analysis (see Appendix B2, available in the online supplemental materials). The studies on which these 16 samples were based were conducted in Australia (1), Austria (1), Canada (7), New Zealand (1), the United Kingdom (4), and the United States (2), and were published between 2004 and 2014 (mode 2012/2013, median 2011/2012). The majority of these papers (18) were published in English. Four were published in German. The effect sizes of these studies were taken directly from the publications for seven samples. For the remaining nine samples, effect sizes were obtained from the authors, who either conducted additional analyses or provided us with the raw data, allowing us to compute effect sizes. Most samples (13) included sex offenders who had offended against children, as well as sex offenders who had offended against adults. Two samples included only sex offenders who had offended against children, and for one sample the composition in terms of offender type was unknown. In six of the samples, all offenders received some form of treatment; six samples included only untreated offenders; three samples included both treated and untreated offenders, and for one sample the treatment status was unknown.
RkJQdWJsaXNoZXIy MjY0ODMw