Strengths and challenges of intervention research in vocational rehabilitation: an illustration of agency-university collaboration

Journal of Rehabilitation, April-June, 2005 by Fabricio E. Balcazar, Cecile Lardon, Christopher B. Keys, Curtis Jones, Margaret Davis

Participants set an average of 3.5 goals (range 3 to 4) in areas like employment, education, health, and social life. The average weighted goal attainment scores showed no significant differences between the groups because all participants were able to attain some of their goals. The cumulative weighted goal attainment scores were used to compare goal attainment among the four groups. These data are summarized in Table 3. A one-way ANOVA performed to compare between group differences yielded a significant treatment effect [F(3,50) = 4.24, p < .01]. Tukey HSD Post Hoc Tests revealed that the group that received both training and follow-along counselor support performed significantly better than the training-only group (Mean difference = 61.56, p < .04) and the active treatment control group (Mean difference = 78.03, p < .008). These findings partially confirmed our second hypothesis, given that the mean difference between the training plus follow-along support and the follow-along support-only group was not significant. The third hypotheses could not be confirmed because the Tukey HSD Post Hoc Tests also showed no significant differences between the training-only group and the control group (Mean difference = 16.47, ns) and the follow-along support and the control group (Mean difference = 28.88, ns).

Our analysis of the effect sizes of the goal attainment results show that when compared to the active control group, there was a meaningful effect of follow-along services (.50) and a moderate effect of training (.29). Follow-along services appear to have a compounding effect when training services were also provided, as indicated by the large effect size (.96) when comparing the group that only received training services to the group that received training and follow-along services. Although not quite as pronounced, the differences in goal attainment by consumers who received follow-along services only and those who received both follow-along services and training was still strong (.79). Our analyses of effect sizes lead us to conclude that while both training and follow-along services are beneficial to consumers, follow-along services are the more critical to success in attaining personal goals.

Table 4 summarizes the employment data for the four groups at baseline and follow-up. The follow-up employment data were collected approximately 22 months after the beginning of the intervention. Given the small sample size and the categorical nature of these data, a non-parametric test (Wilcoxon Signed Ranks Test) was used to determine that overall, there were significant differences in employment before and after the intervention (Z = 3.162, p < .002). The training-only group showed significant employment gains from baseline to follow-up (Z = 2.000, p < .04), while the training plus counselor follow-along support group showed a trend toward improvements in the desired direction (Z = 1.732, p < .08). The groups that did not receive training showed no significant employment gains.