Dykeman (2003) addresses the prevalence of verbal and physical aggression in the classroom among minors coping with parental separation or divorce. The author concedes that maladjustment following the destruction of the family unit varies by student age, gender, how harmonious the relationship between the parents was, the degree of parental control exercised, and whether the family environment was conflict-ridden in the first place. Nonetheless, non-resilient children act out their aggression and resentment at school, too.
In addition to the primary and secondary interventions commonly practiced by school counselors, the author pondered the possibility of family systems intervention minimizing, if not eliminating, the need for referral to tertiary-type special education programs in the community agency.
The intervention consisted of a family-system, conflict resolution model administered by community agency counselors to 21 students (with their custodial parents) already referred by counselors for disruptive behavior at five junior high schools in a mid-sized midwestern city. Fifteen student-parent dyads eventually completed treatment that typically lasted three months. The study instruments consisted of the Reasoning, Verbal, and Physical Aggression subscales of the Fischer and Corcoran Conflicts Tactics Scale, as well as checklist-aided teacher observation.
Dykeman reported that the two-tailed paired samples t-test yielded values of 4.29 and 3.62 for verbal reasoning and verbal aggression, respectively, at 14 degrees of freedom. Both values were significant at p < 0.01, superficially suggesting that the improvement from baseline to post-treatment follow-up might have occurred by chance variation alone less than once in a hundred replications of the field experiment. At the same time, the computed chi-square value for disruptive behavior observed by teachers was significant at p < 0.05. This suggests that the improvement in classroom verbal aggression or disruptive behavior could have occurred just once in 20 experimental runs had random variation been the only factor at work. Since the subjects stayed the same and the only difference from onset to the conclusion was the treatment itself, the inference is that family systems intervention mitigates the situation and may reduce the need for special education referral.
The most serious flaw of this field experiment is the reliance on the ordinary t-test to infer a conclusive result. Given the small net base of 15 dyads and large standard deviations relative to the means, the researcher blithely ignored the fact that t-tests are fundamentally based on the assumption of normality of distribution and homogeneity of variances. The more rigorous alternative was to employ the nonparametric equivalent for paired-sample analysis, the Wilcoxon T-Test.
Secondly, the researcher assumes that self-reported frequencies for verbal reasoning, verbal aggression, and other disruptive behavior are valid measures of maladjustment after a child’s parents have divorced. For the given time frame of six months, the reported frequencies are so low as to defy belief.
A third, even more, a dismal inference may be drawn from the findings that the gains in self-reported use of verbal reasoning (“improved” from a frequency of three in six months to four in a similar period at follow-up), reduced frequency of verbal aggression in the course of disputes with parents (dropping from 4.9 to 3.9, on average, over the same reference period), and the frequency of physical aggression are all too marginal or inconsequential. Perhaps, the intervention was not right or the observation period was too short. Such results are not enough in themselves to avoid having to refer troubled children to tertiary interventions. One wonders then whether Dykeman wasted time on an extra intervention stage that does not prevent special referral anyway. Perhaps, academic theorizing and licensed professional counselor resources are better devoted to preventing the root cause, the all-too-casual propensity for divorce.
References
Dykeman, B.F. (2003). The effects of family conflict resolution on children’s classroom behavior. Journal of Instructional Psychology, 30 (1) 41-46.