Project abstract:

Mental health challenges in childhood can have significant long-term consequences. It is therefore essential that mental health symptoms are accurately assessed. One method for increasing the accuracy of mental health assessment is to integrate multiple perspectives (van Dulmen and Egeland, 2011). However, multiple reports rarely align, leaving mental health professionals with the important task of reconciling disparate information. These discrepancies are even more common within child welfare settings (Makol et al., 2020; Parker et al., 2018), which is particularly problematic given that mental health symptom reports are used in case planning. Several methods have been used for integrating multiple reports, including interpreting the average score, highest score across reports (LaPalme et al., 2020), and selecting specific questions from respective reporters (Thakur & Cohen, 2019); however, there is no standard approach (Martel et al., 2017). While several studies have been conducted examining the most effective method for integrating multiple reporters in child welfare settings, none to date has been longitudinal (Thakur & Cohen, 2019). Additionally, the majority of studies have examined specific disorders rather than the broad range of mental health symptomology common among the child welfare population (Thakur & Cohen). Finally, no studies examine the breadth of ages among those receiving child welfare involvement, instead examining discrete periods of development, like early childhood or adolescence (McWey et al., 2018). The current seeks to address these gaps by examining how child and parent reports of internalizing symptoms, externalizing behaviors, and trauma responses compare to caseworker assessments. Using structural equation modeling, the current study will compare how the accuracy of reporting changes over the course of ten years of child welfare involvement. These findings will support child welfare caseworkers in integrating disparate reports in order to more effectively help children and youth access the services they need.

Description of summer undergraduate work including timeline:

• May 17 - 28: The undergraduate will familiarize him/herself with the literature about the topic. This will include:
• Reading 5 articles about structural equation modeling
• Reading 5 articles about mental health outcomes in child welfare
• Reading 5 meta-analyses about discrepancies in reporters
• Read the LONGSCAN manual
• May 31 - June 11: The undergraduate will complete two Data-camp classes and build the LONGSCAN database in SPSS.
• June 14 - July 2: The undergraduate will finalize the LONGSCAN database and support me in running preliminary analyses.
• I will instruct the student in how to interpret the data and have them practice writing up the results.
• July 5 - July 23: Together, the undergraduate student and I will create the code for structural equation modeling in R.
• I will instruct the student in how to interpret the data and have them practice writing up the results, for which I will provide feedback.
• The undergraduate will also create tables to conveying the results from preliminary analyses, for which I will provide feedback.
•  July 26 - August 6: The undergraduate will review my dissertation proposal, including the literature review and methods section, in order to develop the abilities to critically engage with research and provide feedback.
• The undergraduate will also create tables to convey the results from the structural equation models, for which I will provide feedback.
• August 9 - August 20: The undergraduate will review the results section in order to develop the abilities to critically engage with research and provide feedback.