Exploring the Effects of Individual and Organizational Characteristics and Reinforcement Strategies on the Use of Supervisory Coaching in Child Welfare Services

BACKGROUND/PURPOSE: This University/Agency Collaborative study evaluated transfer of learning (ToL) as reported by supervisors/managers in public and private child welfare settings following their participation in a coaching learning program. Coaching has been used as a strategy to support the uptake of skills within the workplace (Akin, 2016; Allen et al., 2020; Hatton-Bowers et al., 2015). This study sought to understand the extent that individual and organizational characteristics and reinforcement strategies influenced the uptake of coaching by supervisors. METHODS: A survey designed in Qualtrics was distributed during two time points: (10/18-1/19 & 12/19-3/20) to supervisors/managers who had completed a coaching learning program. Factors that had influenced ToL in past research were used as independent variables. The dependent variable asked how often supervisors used the full coaching process on a five-point scale from always to never. The survey was distributed to 2,050 eligible participants during the first round, and 1,114 eligible participants in the 2^nd round. Data collection was disrupted by the global pandemic in March 2020. Data from 1,346 participants in public child welfare (n=861, 64%), provider agencies (n=360, 26.7%), and unclassified (n=125, 9.3%) were used in analyses. RESULTS: Multiple linear regression analyses were conducted to explore relationships of individual and organizational characteristics and participation in reinforcement strategies on use of coaching. Analyses were conducted using IBM SPSS software (version 29.0.1.0). Each domain was independently tested to identify the strongest predictors to input in a final model. Perception of benefits of supervisory coaching on staff self-sufficiency; motivation to apply coaching; confidence and commitment to use coaching; the importance of coaching skills (i.e., presence, questioning, reflecting/clarifying), and a manageable workload were significant independent predictors of coaching. They were then input into a final model, which demonstrated good model fit F(8, 417) = 3.685, p <.001. The adjusted R² indicated that the model accounted for 4.8% of the variance in use of coaching. Confidence, b = 0.12, SE = 0.04, p <.01, perception of questioning as an important coaching skill, b = 0.16, SE = 0.08, p <.05, and motivation to apply learning in practice, b = -0.29, SE = 0.14, p <.05, were all significant predictors of using coaching in supervision. While all three individual characteristics were significantly associated with coaching use, standardized estimates revealed that confidence (b = 0.20) and importance of questioning (b = 0.12) were stronger predictors than motivation to apply learning (b = -0.10), which had an inverse relationship with coaching. Participants who attended reinforcement sessions reported increased confidence and commitment to coaching yet without significant impact on ToL. Qualitative thematic analyses using TextIQ in Qualtrics identified the top three barriers of time, workload, and emergencies to ToL. CONCLUSIONS/ IMPLICATIONS: Results indicated high rates of using coaching in supervision. Creating opportunities to enhance organizational characteristics may increase ToL. A limitation is that participants may not have accurately reported their actual use of coaching.