Work Experience Profiles and Turnover Intention Among Korean Social Workers

Background
Retaining social workers is an important issue for workforce management because persistent turnover can make staffing and service delivery more difficult. In practice, social workers do not experience pay, work-life balance, career support, job satisfaction, and stress as separate conditions; these aspects of work tend to cluster together. Examining such combinations may therefore provide a more realistic understanding of turnover intention than looking at single factors one at a time. This study identifies work experience profiles among Korean social workers and examines whether turnover intention differs across those profiles.

Methods
This study used individual-level survey data from the 2024 Social Worker Statistical Yearbook in Korea (N = 3,290). Turnover intention was measured as the mean of four items, with one retirement-related item reverse-coded (Cronbach’s alpha = .885). Latent class analysis was conducted using six indicators: physical work environment, pay perception, work-life balance satisfaction, career development support, job satisfaction, and job stress. Models with two to five classes were compared to identify the best-fitting solution. Differences in turnover intention across classes were tested using ANOVA. The association between latent class membership and turnover intention was then examined using OLS regression with HC3 robust standard errors, controlling for gender, education, employment type, position, license level, age, tenure, and total career length.

Results
Among the tested models, the five-class solution showed the best fit (BIC = 36121.65, entropy = .638). Turnover intention differed significantly across latent classes (F = 225.17, p < .001, eta2 = .215), and the class effect remained significant after adjustment for covariates (robust Wald F = 211.54, p < .001). The highest turnover intention was observed in a class characterized by low pay perception, poor work-life balance, limited career development support, low job satisfaction, and high job stress (M = 3.70). The lowest turnover intention was found in a class with more favorable work conditions, stronger support, and lower stress (M = 2.04).

Implications
The findings suggest that turnover intention is linked to distinct combinations of work conditions rather than to single factors alone. For policy and practice, this implies that efforts to improve retention should move beyond uniform interventions and pay closer attention to groups facing multiple disadvantages at once, particularly low compensation, weak support, and high stress. A profile-based approach may help organizations and policymakers target workforce stabilization efforts more effectively.