Increasing homework compliance by using the guiding model for practice : an analogue study : a thesis presented in partial fulfilment of the requirements for the Master of Arts at Massey University, Albany
Homework assignments are considered a fundamental component of Cognitive Behavioural Therapy and are believed to be significant in assisting to produce and maintain treatment gains. However, gaining clients compliance to homework tasks remains a significant challenge. An analogue study of a single session relaxation intervention was conducted to test the guiding model for practice (Kazantzis, MacEwan & Dattilio, 2005); designed to provide therapists with a step-by-step guide of how to systematically administer homework in therapy. Forty four participants were randomly assigned to one of two conditions. The systematic condition (n = 21) was designed to administer homework following the guiding model, while the non-systematic condition (n = 23) followed standard therapy practice. Hypotheses posited that participants in the systematic group would display greater levels of engagement in homework; would have more positive beliefs in completing the homework; that greater adherence to the homework would correlate positively with reductions in anxiety; and that the systematic group would show a greater reduction in anxiety. In relation to engagement in homework the results found a statistically significant difference in the mean ranks of homework compliance between the two groups. The Mann-Whitney U result was 182 (z = -1.48) with an associated probability of .14, showing that participants in the systematic group did have higher levels of homework compliance. A MANOVA calculation was used to assess the systematic group for more positive beliefs in completing homework. The results found significant differences in two of the four Homework Rating Scale II (HRS) subscales; behaviour: F(1, 42) = 1.83, p = .184, partial eta squared - .042; and consequences/synthesis: F(1, 42) = 2.93, p = .094, partial eta squared = .065. The other two subscales of the HRS; beliefs and situation, were not administered differently between groups, providing further support for the difference of homework administration. Partial support was found for correlations between homework practice and anxiety. While three of the four correlations were significant, it was found that state anxiety actually increased as practice increased, however, trait anxiety was found to reduce as homework levels increased. No significant group differences were found in anxiety reduction. Implications of these findings are discussed.