Probation officers'' compliance with the Youth Level of Service Inventory/Case Management Inventory (YLS/CMI): A multi-level study of post-implementation practice across Pennsylvania counties

Structured risk and risk/needs assessment tools represent the cornerstone of decision-making within the Risk-Needs-Responsivity (RNR) model, yet little research examines how RNAs are used in practice. The proposed research examines the use of the Youth Level of Service/Case Management Inventory (YLS/CMI) among juvenile probation officers in Pennsylvania counties, several years after its implementation. In particular, it examines how policies and practices vary across counties, how these policies and practices conform to the RNR model, and how state, county, and agency factors account for this variation. It will be conducted by Dr. Joel Miller at Rutgers University in direct collaboration with Dr. Carrie Maloney at East Stroudsburg University. It will work in partnership with the Council of Chief Juvenile Probation Officers (CCJPO) Juvenile Court Judges’ Commission (JCJC) in Pennsylvania and local county probation offices.
The two-year project will examine county operational policies, organizational dynamics, community/jurisdictional context, officer characteristics, and officers’ actual use of the YLS/CMI. It will begin with in-depth fieldwork across five diverse counties, selected for variation in size, urbanization, and levels of compliance with the YLS/CMI measured by an earlier statewide survey. It will include interviews with about 85 juvenile probation staff across rank using a semi-structured interview tool, reviews of key policy documents, reviews of about 100 YLS/CMI-related events with probation officers also using a semi-structured tool, about 200 hours of direct observation, and interviews with about 6 state juvenile justice leaders. In a later stage of the project, the study will distribute a follow-up electronic survey to the approximately 1,300 juvenile probation staff in the 66 Pennsylvania counties using the YLS/CMI.
We will analyze qualitative data using NVIVO software, triangulating sources to develop coding categories through a constant comparative method. Key YLS/CMI-related events from the qualitative/fieldwork data will also be incorporated into a spreadsheet and analyzed descriptively. Survey data will be analyzed in Stata to produce descriptive statistics and multivariate multilevel random effects models that will estimate latent variables from survey responses and deploy path models to assess patterns of mediation.
Final products will include a summary report to NIJ, qualitative data transcripts and survey data, a survey instrument for assessing the implementation of risk/need assessment tools, a policy brief, and a webinar for practitioners in Pennsylvania and nation-wide. Findings may also be directly used to inform state policies and
training in relation to the YLS/CMI through our state level partners.

This project contains a research and/or development component, as defined in applicable law.

ca/ncf