The Development and Validation of the Religious/Spiritually Integrated Practice Assessment Scale

Background and Purpose: Research suggests integrating clients’ religion/spirituality (R/S) in practice, when desired by the client, has the potential of improving health and mental health outcomes (Koenig, et al., 2012). Most Americans (80%) consider themselves at least somewhat religious (Pew Charitable Trust, 2012) and prefer their helping professional assess and integrate their R/S into treatment. The majority of clinically trained helping professionals are social workers; however, only 35% have received training on integrating clients’ R/S into practice (Canda & Furman, 2010), suggesting social workers may not be adequately trained or open to integrating this important aspect of client culture into practice. Additionally, there has been limited evaluation of these training efforts, and existing scales designed to assess these views or practices have not established factorial validity or focused on the use of empirically supported approaches. As such, this study reports on the development and validity of the Religious/Spiritually Integrated Practice Assessment Scale (RSIPAS), designed to measure practitioners’ overall orientation to integrating R/S into practice. This multidimensional scale contains four subscales, assessing practitioners’ self-efficacy, attitudes, perceived feasibility and behaviors related to integrating clients’ R/S in practice.

Three research questions guided this study: 1)Does the RSIPAS have content and criterion validity?; 2)Can responses to the RSIPAS be explained by four factors (self-efficacy, attitudes, behaviors, and perceived feasibility)?; and 3)Can responses to the RSIPAS be explained by four first-order factors and one second-order factor (orientation toward integrating clients’ R/S into practice)?

Methods: The RSIPAS was developed and reviewed by experts in the field of R/S and health, then piloted with 13 social workers. The scale and related background questions were then administered to a national sample of 1,000 randomly selected social workers who advertised their services on HelpPRO. The online survey link was sent via email and mail, using Dillman’s sampling methods, to solicit participation. The final sampling frame was adjusted to 984 given bounce-back emails, with 482 respondents, yielding a  49% response rate. Confirmatory factor analyses (CFA), testing both first and second order factors, were run in Mplus to assess the model’s fit, using Kline’s recommended goodness of fit indices.

Results: The majority of practitioners were white females and an average of 57 years old, which is similar to NASW membership demographics. Data was missing completely at random. The CFA was run, and low loading items (<.45, Comrey & Lee, 1992) were removed. Modification indices were used to correlate theoretically similar error terms to improve the model. Both first and second order models had adequate fit: First-order: χ²(724,N=470)=2,653.06, p=.000, CFI=.93, TLI=.92, RMSEA=.075(low90=.072,high90=.078), Second-order: χ²(726,N=470)=2,753.89, p=.000, CFI=.92, TLI=.92, RMSEA=.077(low90=.074,high90=.080). Findings support the scale’s content, criterion, discriminant, and factorial validity. Cronbach’s α for the overall scale was .95, with each subscale α ranging .84-.91.

Conclusions and Implications: The findings from this study suggest the RSIPAS may be a useful instrument for identifying social work practitioners’ views and behaviors related to integrating clients’ R/S in practice in surveys, or to evaluate the outcomes of training or educational programs focused on R/S and practice.