In the Limitations and Future Research section of the paper, the authors acknowledge this issue. Their sample consisted of predominantly European and African-American boys from an urban community, identified at risk of developing behavioural problems. Furthermore, the fact that all the participants were male, means the findings of this study cannot be generalised to the population at large (Supplee et al., 2011). But as the authors state, this study provides a platform for future research, setting a precedent, not only for further longitudinal research in this area, but also for observing SRS within a natural …show more content…
Using both Eyberg Intensity and Problem Behaviour Scale, the boys were assessed for family risk and behavioural issues. ANOVAs were used to compare the control and treatment groups; no significant differences were found, allowing the researchers to collapse the groups into one sample (Supplee et al., 2011). Two approaches were used to examine the patterns of change in the boys’ SRS at ages 2-, 3-, and 4-. Variable-centred repeated measures ANOVAs measured the boys’ ratio of use of each SRS over time. Person-oriented, semi-parametric group based method (Nagin, 1999) was used to model the trajectory of the groups of boys’ distinct SRS use over time (Supplee et al., 2011). Because of attrition, the Expectation-Maximization (EM) imputation method was used to handle the missing data for the repeated measures ANOVAs. The PROC TRAJ program was used for the person-oriented analysis (Supplee et al.,