Limitations and directions for future research The results in this study have been obtained by using self-report questionnaires. To obtain meaningful results we used questionnaires that were validated and structurally equivalent across cultural groups as much as possible. However, in studies with immigrant children it has been found that using teacher and parental reports may provide results that are different from those obtained in self-reports (Xxxxxxxxx et al., 1990; Xxxxxxx & Xxxxx, 1993; Xxxxxxx et al., 2003). As schools only had a limited amount of time to participate in research we could not use a longitudinal design. The hypothesized directions of effects that are presented in this book are all derived from theoretical frameworks and prior research, but without a longitudinal design we cannot be sure that the hypothesized direction of the effects is correct. A third important point of critique has to be made with regards to the samples. Immigrant students are known to drop out of junior vocational education more often than national students, 16.8 percent as opposed to 11.2 percent (Herweijer, 2009); this may have contributed to finding confirmation for the immigrant paradox. This study provides an overview of the acculturation and adaptation of the largest group of immigrant adolescents in the Netherlands, those in junior vocational education. When compared with other adolescents in junior vocational education these immigrant adolescents are doing rather well. However, when compared to a nationwide sample, the under representation in higher education suggests that the position of immigrant adolescents remains to be improved. As such, future studies should address the question why immigrant adolescents have a positive pattern of adaptation, but fail to perform as well as national students in terms of education.
Limitations and directions for future research A study of this nature, where links are sought between a range of variables that are essentially predictors, is best served by regression analysis. However due to extensive recruitment difficulties, large enough Ns were not possible to make regression feasible. The scarcity of fathers with OCD in major mental health facilities identifiable as a recruitment strategy and in general was curious and the answer could be twofold. It is well known that men with OCD are less likely to marry and have children (Castle, Xxxxx et al. 1995); therefore this scarcity could be a genuine reflection of reality. However given the nature of the participating father’s obsessional thoughts (the majority concerning harming their children or family) it is also possible they stay away from statutory services through fear of negative consequences. This may explain why the majority of OCD fathers were recruited via adverts rather than through mental health services. Additionally it is possible that the burden of taking part in research for the few that were undergoing treatment was prohibitive for these fathers. This is supported by the fact that the one father who was recruited via an IAPT service dropped out of the project. In being able to run regression analysis it would also be desirable to have measures of maternal factors to see what proportion of variance in COCD scores is accounted for by each parent as in the Kilic studies (Xxxxx, Xxxxxxx et al. 2003; Xxxxx, Xxxxx et al. 2011). Given the fathers group also have significant levels of comorbidity (although similar to those identified in the literature e.g. by Xxxxxxx, Xxxxx et al. (2001) this type of analysis would also allow more sophisticated examination of the role of comorbidities in child outcomes. Another route to teasing apart the effects of comorbidity would be to examine a group of fathers with moderate to severe depression with the same measures and compare results to the OCD fathers. A related issue is the recruitment of the participating fathers in terms of volunteering. Informal conversations at first contact revealed that the reasons fathers took part in the study fell into two groups: 1. Those who believed OCD was genetic and little to do with parenting 2. Those who felt their OCD had developed as a result of their own parenting and were concerned the pattern would repeat in the next generation. Whilst both groups were concerned about their children developing OCD, these ideas are likely to have influ...
