Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Purpose: The global popularity of solution focused therapy is now at a point that its acceptance as an evidence-based model must evolve beyond studies described as preliminary (Gingrich & Eisengart, 2000). One sign of progress would be the development of a fidelity instrument, measuring the extent to which delivery of an intervention adheres to the model originally developed (Mowbray, Holter, Teague, & Bybee, 2003). The aim of this study therefore, was to test the reliability and validity of a new fidelity instrument for therapists and clients of solution focused therapy (de Shazar, 1982, 1985, 1988; de Shazar et al., 1986). Testing was accomplished through the parallel development and utilization of the Solution Focused Therapy Fidelity Instrument Consumer (SFT-FIC) and the Solution Focused Therapy Fidelity Instrument Therapist (SFT-FIT) in a community mental health clinic at the University of Texas at Arlington. Earlier versions were sent to an international solution focused therapy listserv to obtain feedback and establish face and content validity. Method: The method employed pilot testing the two instruments to compare the therapist's and client's perception of adherence to the intervention model. The study of therapist and consumer fidelity ratings involved a multiple time series design with two or more instrument completions per case, over a period of eight months. Twelve therapists completed 69 SFT-FIT instruments and 91 clients completed 101 SFT-FIC instruments Results: A reliability analysis and a covariance matrix determined the internal consistency of questionnaire items. An overall alpha coefficient (alpha = .89) was calculated with a standardized item alpha determined for the eighteen items contained within the parallel forms of the scalesacross the time series. The analysis also used scores from the first observation in the time series to assure independence of responses with the reliability coefficient remaining stable (alpha = .88). The overall scale mean (96.78) and standard deviation (17.36) were calculated. The mean value for specific items was derived and the mean of item variances (301.52) also determined. A principal component factor analysis was performed to determine the least number of factors that can account for the common variance (correlation) of a set of variables. This was used to identify the number of factors included in the survey instrument. The components (factors) were extracted using Varimax rotation with Kaiser Normalization with eigenvalues set at a value of 1.00 with an estimate of total explained variance (52.43%). Communalities values suggested that items were interrelated in the matrix and represented a one-dimensional scale. Further analysis of pair wise correlations was completed to determine whether the item inter-correlations between variables were less than would be required to consider the elimination of an item from the survey instrument. Implications: A reliable, single factor scale for measuring intervention fidelity to solution focused therapy was found across multiple therapists and client families Consequently, agencies have opportunities to evaluate program delivery while social work researchers might design studies that meet strict efficacy criteria using fidelity measures that have been tested.