Establishing Factor Structure for a Measure of Moral Distress for Child Welfare Professionals

Background: Child welfare caseworkers are a vulnerable subset of social work professionals affected by moral distress. Predominantly studied in healthcare, moral distress occurs when a professional knows the ethically appropriate action but cannot take it due to constraints in their everyday work. Moral distress include feeling complicity in wrongdoing and violation of professional values; outcomes include physical illness, low job performance, and turnover. As there is no known validated moral distress measure for caseworkers and other social work professionals, the lead presenter developed the Measure of Moral Distress for Child Welfare (MMD-CW), which was adapted from the Measure of Moral Distress for Healthcare Professionals Epstein et al., 2019). This current study is part of the MMD-CW validation process and uses exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) to establish MMD-CW’s factor structure.

Method: Data comes from two state public child welfare agencies where the MMD-CW was piloted in 2022 as part of a workforce initiative’s comprehensive organizational health assessment (COHA) effort (N=907 caseworkers). The 13-item MMD-CW assesses child welfare unit- and systems-level root causes of moral distress; the frequency of experiencing morally distressing situations and associated levels of distress. Composite item scores were calculated by multiplying the frequency by the level. An exploratory factor analysis (EFA) was performed to identify a factor structure using a maximum likelihood method. Factors were extracted if eigenvalues were greater than 1. An oblique rotation method was used because factors were expected to correlate with one another. Items were removed if they did not measure intended factors or showed low factor loadings (≤.32) and/or high cross-loading.

Results: One item designed to measure the system-level root cause of moral distress was dropped from a final EFA because the initial results indicated that it loaded more strongly with other items that measured team-level root causes. The final EFA produced a two-factor model as hypothesized. The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy (.927) and Bartlett’s test of sphericity (χ² = 4062.63, df = 66, p < .001) suggested that the factor analysis was appropriate for this dataset. The first factor consisted of 7 related to system-level root causes (factor loadings: .61 – .94). The second factor involved 5 items related to team-level root causes (factor loadings: .43 – .82). The two factors extracted were correlated at a moderate level (r = .52), suggesting some overlapping but discriminant constructs. Also, the overall scale showed excellent internal consistency (α = .90), along with Cronbach’s alpha for each subscale (first factor = .89; second factor = .84).

Conclusion: Findings supported the hypothesis for a two-factor model for the MMD-CW. With demonstrated strong internal consistency, study results support moving to next steps in measurement validation, which include examining the convergent, discriminant, and concurrent validity of the MMD-CW for use in child welfare and other social work contexts. Validating the MMD-CW helps capture the presence of different root sources of moral distress for caseworkers and inform future prevention and intervention research to address moral distress for this essential workforce.