# Psychometric Properties of a Survey of Knowledge and Attitude Change in Residential Staff Receiving Training in Trauma-Informed Care: The Modularized Think Trauma Evaluation Questionnaires

## TL;DR

This paper evaluates a survey tool to measure changes in knowledge and attitudes of residential staff after trauma-informed care training.

## Contribution

The study provides a reliable and valid survey tool for assessing trauma-informed care training outcomes.

## Key findings

- The M-TTEQs showed strong internal consistency with Cronbach’s alpha values between 0.88 and 0.95.
- Confirmatory factor analyses confirmed acceptable construct validity for the survey modules.
- Some items, especially reverse-worded questions, showed weaker performance and need refinement.

## Abstract

Trauma-informed care has received increased attention in the scientific literature and clinical practice (Becker-Blease, 2017; Purtle, 2020; Stokes et al., 2024); however, evaluation of the implementation and effectiveness of these efforts is limited (Purtle, 2020). This study addresses this gap by exploring the psychometric properties of the Modularized Think Trauma Evaluation Questionnaires (M-TTEQs). The M-TTEQs were developed to assess frontline residential staffs’ trauma-informed knowledge and attitudes before and after receiving Think Trauma, a National Child Traumatic Stress Network (NCTSN) training curriculum that consists of 4 discrete modules. This paper utilizes data from 1807 staff members at 20 Illinois child welfare residential care facilities who received Think Trauma training between 2020 and 2024. The internal consistency of M-TTEQs was assessed using Cronbach’s alpha, with results indicating strong internal consistency across all pre- and post-surveys (α values between 0.88 and 0.95), supporting the reliability of the measures. A subset of 155 participants who completed all 4 pre-training, and 153 participants who completed all 4 post-training M-TTEQs were included in confirmatory factor analyses (CFA) assessing the measures’ construct validity. CFA models demonstrated acceptable fit indices, indicating that the surveys measured the intended constructs for each module. Despite these acceptable psychometric properties, some items showed weaker factor loadings, particularly reverse-worded questions, suggesting the need for further refinement. This study contributes to the trauma-informed care literature by providing a tool with acceptable reliability and construct validity for assessing knowledge and attitude change related to the Think Trauma training curriculum.