Background/Context: For decades, practitioners, policymakers, and researchers have sought to improve the quality of early childhood education (ECE) to foster equitable experiences for young children. States are increasingly investing in evidence-based professional development programs for teachers, including practice-based coaching and early childhood mental health consultation (ECMHC). ECMHC is an intervention strategy that pairs a mental health professional (i.e., "consultant") with the adults (i.e., caregivers, teachers, and families) who work with young children. Previous research has found that ECMHC is effective at improving classroom quality, increasing young children's social and emotional skills, and decreasing teacher stress (Brennan et al., 2008; Downer et al., 2018). In Virginia, ECMHC services have been provided to educators for the past three years through a partnership with the Virginia Department of Education (VDOE). Our ECMHC model is focused on reducing exclusionary discipline, particularly for young children of color who are more likely to experience discipline practices such as expulsion at disproportionate rates (Albritton et al., 2019; Davis et al., 2020; U.S. Department of Education Office for Civil Rights, 2014). As part of ECMHC services in Virginia, consultants collect a variety of teacher and child data that serve dual roles: 1) to inform consultation and 2) to evaluate the effectiveness of consultation pre to post. These data collection efforts must balance the need for policy-relevant, rigorous evaluation research with the need for practical tools that are manageable for consultants and teachers. Data collection is a critical element of improvement efforts in early childhood education, including in coaching and consultation. However, commonly used early childhood classroom observational measures, such as the Classroom Assessment Scoring System (CLASS; Pianta et al., 2008) and Teaching Pyramid Observation Tool (TPOT; Hemmeter et al., 2018), are resource intensive, requiring intensive training for reliability and multiple hours of in-person observation. These measures are not always feasible for consultants and other providers, especially in early childhood settings that face challenges that have only worsened in the aftermath of COVID-19, such as high stress, staffing shortages, and lack protected planning and professional development time (Bassok et al., 2021; Markowitz & Bassok, 2022). In this transforming landscape, the field needs measures that are reliable, valid, and feasible for consultants and other providers to do in real-world applied settings. In our ECMHC model, we use a measure adapted from the TPOT, known as the TPOT-Short form (TPOT-S). This study is the first to examine the psychometric properties of this tool. Purpose: This paper examines the psychometric properties of an observational measure of social and emotional teaching practices in early childhood settings, the Teaching Pyramid Observational Tool-Short Form (TPOT-S). Results will focus on examining internal reliability and findings from a series of confirmatory factor analyses. Setting, Participants, and Intervention: The current study used data collected in the context of Virginia's ECMHC pilot during 2023 to 2024. Only teachers with TPOT-S data were included (n = 75 at pre-consultation). This sample included preschool teachers who were receiving consultation services for a child referred through an open referral system. Data Collection and Analysis: Before conducting observations, consultants participated in a virtual two-hour training that described each TPOTS items and scoring procedures. Consultants then practiced scoring TPOT-S items using classroom footage. At the beginning of consultation, consultants conducted TPOT-S observations in four 10-minute cycles. Following the observation, consultants entered their TPOT-S data into an online form, and the research team processed data using STATA. The TPOT-S includes 35 items across eight subscales (see Tables 1 and 2). Eight items were indicator variables (consultants rated yes/no), and 27 items were rated on a scale from never (1) to almost always (4). Our analyses sought to understand how the TPOT-S measure works in the current sample. First, to understand internal reliability, we calculated Cronbach's alphas for each subscale. Next, we used confirmatory factor analysis (CFA) to understand how the subscales were operating together as latent factors. For CFA analyses, we first explored model fit a one factor model. Next, we ran an eight-factor model, with each subscale representing separate latent factors, and compared model fit. Analyses were performed in Mplus. Model fit was assessed via commonly used fit indices (Kline, 2005), such as root mean square error of approximation, standardized root mean square residual, and comparative fit index. Findings/Results: Descriptive statistics and Cronbach's alphas of each subscale are reported in Table 1. Overall, reliability was acceptable (Alpha > 0.70) for the following five subscales: Classroom Arrangement, Organization, and Materials; Classroom Interactions Promoting and Maintaining Engagement; Teacher Warmth; Commands; and Supporting Peer Interactions and Problem-Solving. Reliability was less strong (Alpha > 0.60) for the three other subscales. CFA models are presented in Table 2. Our first model was a one factor CFA with each item loading on to one overall factor. Item 2 was dropped because the variance of this item was 0 (i.e., all teachers had the same score). Model fit was not acceptable. Next, we tested the fit of an eight factor CFA, with individual items loading onto each of the six subscales and allowing subscales to correlate. This eight factor CFA was compared to a one factor CFA. Item 2 was again dropped. Model fit was better than the one factor, but still not acceptable based on fit indices. We are continuing to refine the model, based on model modification indices. Conclusions: The early childhood education field needs observational tools that are valid, reliable, feasible at-scale, and useful to practitioners. To begin to address this, this study examines the psychometric properties of an observational measure of social and emotional teaching practices, the TPOT-S. We use the TPOT-S in our model of ECMHC for both data collection and to inform consultation. No study to date has explored the reliability and validity of this tool. Analysis is ongoing, and findings will inform changes to the measure, from both a researcher and practitioner perspective, to better capture teaching practices in early childhood settings.