Development and Initial Validation of the Self-Care Practices Scale (SCPS)

Background/Purpose

There has been increasing acknowledgment that self-care is essential for effective, ethical practice and promotes resiliency among practitioners, offsetting occupational stress. However, the extant literature reflects notable absence of a clear and well-validated definition of self-care. In order to begin to address this gap, Lee and Miller (2013) offered a conceptual definition of self-care for social work that distinguishes personal self-care practices from professional self-care practices, and also clearly highlights the contingent relationship between the two. This paper presents the results of a confirmatory factor analysis of the Self-Care Practices Scale (SCPS), created by the authors and informed by Lee and Miller’s conceptual definition of self-care.

Methods

A 71-item version of the SCPS was created via a comprehensive review of the self-care literature, including concepts of wellness, resilience, and coping. The 71-item version was piloted with a sample of social work students (n = 129), yielding internal validity for the scale (α = .932), the personal self-care subscale (α = .865), and the professional self-care subscale (α = .906). Results of the pilot test and a thorough review of the scale to eliminate redundancy informed elimination of 23 items. The resulting 38-item SCPS is a Likert-scaled (“0” to “4”) instrument designed to assess the frequency of both personal and professional self-care practices. Higher scores reflect greater frequency of self-care practices. Data was collected via mailed survey from a random sample of Masters level social workers who are members of NASW; 492 provided complete data on the SCPS. The sample was randomly divided into two subsamples of 246. Analyses were conducted in two stages with Mplus 6.2. Stage I used a model-generating approach whereby a confirmatory factor analysis (CFA) was run on Sample 1, followed by modifications to the model and a new CFA to assess fit of the revised model. Decisions on model modification were made based on quantitative results as well as qualitative assessment of item quality. The process was used with each subscale alone then with both subscales simultaneously in a two-factor model. This iterative process was repeated until adequate model fit was achieved for a two-factor model. Stage II used Sample 2 to cross-validate the final model. This was a necessary step due to the danger of capitalizing on idiosyncrasies of the data.

Results

Stage I resulted in an 18-item version of the SPCS with 9 items assigned to each subscale. Fit indices demonstrated good fit (CFI = .97, TLI = .96, RMSEA = .04, SRMR = .04). The Stage II CFA also resulted in a good fit, although the fit indices were less robust (CFI = .92, TLI = .91, RMSEA = .05, SRMR = .05).

Conclusions/Implications

The results support the factor validity of an 18-item SCPS. To the authors’ knowledge, this is the first validated instrument to measure the frequency of self-care practice in both personal and professional domains. The SCPS provides a mechanism to increase systematic investigation in this understudied area that can be useful in investigating the role of self-care in practice.