Organizational Inclusion, Commitment, and Turnover Among Child Welfare Workers: A Multilevel Mediation Analysis

Background and Purpose

Many organizational theories have conceptualized psychological processes that mediate between individual variables and group variables. However, little research has been conducted using MLM (multilevel modeling) analysis even though it may be more appropriate to deal with administrative data that are naturally nested (i.e., employees in an organization). In addition, MLM provides enhanced statistical power.

The purpose of the study is to assess whether or not organizational commitment, a key factor in retaining competent employees, mediated the effects of organizational inclusion (e.g., participation in decision-making, accessibility to organizational resources and communicability) on turnover intention in different organizations using MLM analysis. That is, the study examines the role of workers' attitudinal variables (e.g., organizational commitment) in the determination of workers' actual behavior (e.g., intention to leave, job withdrawal, etc.). In addition, the study examines the cross-level effects of organization-level inclusion and individual-level commitment and turnover intention among child welfare workers.

Methods

The study employed a cross-sectional design and data were collected from public child welfare workers via a self-reported online survey across a diversified Mid-Atlantic state (N=621, 56.5% response rate). MLM analyses were used to examine the multilevel mediation effects in the study. Valid and reliable measures captured organizational inclusion (alpha = .78), commitment (alpha = .91), and intention to leave (alpha = .84) which constructed the conceptual model. The analyses were conducted using restricted maximum likelihood estimation for MLMs (Bickel, 2007). To test the multilevel mediating effects, three-level MLMs were constructed for the two dependent variables, turnover intention and organizational commitment (Baron & Kenny, 1986; MacKinnon, 2008). The mediated effects for significance (95% confident limits) were computed based on MacKinnon's (2008) formula.

Results

The intraclass correlation (ICC) was 8.6%, which is explained by county membership with regard to the variance in turnover intention across counties. The individual-level demographic characteristics explained a modest proportion of the county-level variance in the individual-level turnover intention (36%). The county-level construct (organizational inclusion) accounted for all of the remaining county-level variance in turnover intention (64%). The conceptual model that worker's individual perception of commitment to the organization may play a role in mediating the relationship between organizational-level inclusion and individual-level turnover intention was supported with 95% confidence limits [-1.13 to -0.53].

Conclusion and Implications

Building upon prior research (Mor Barak et al., 2006), the study suggests that individual worker's organizational commitment plays a mediating role between organizational inclusion and individual turnover intention; that is, child welfare workers' turnover intention was reduced by their perception of organizational commitment. In order to tackle the high turnover rate, administrators or leaders in child welfare organizations need to focus on enhancing workers' commitment as well as innovating organizational structure, thereby fostering a greater sense of inclusion in the daily life of the organization. The findings suggest that interventions aimed at reducing turnover and retaining quality workers may be more effective with a combined approach of targeting individual attitudinal perceptions (e.g., organizational commitment, job satisfaction) with changing organizational structures (e.g., practices that embody organizational inclusion).