Background/Context: The early school years are an important time for building family-school relationships (Sibley & Dearing, 2014). As such, family engagement is a top priority of organizations that serve young children, such as Head Start. Children whose parents report being involved in their school demonstrate greater social competence and academic achievement than children of parents with lower levels of school contact (Epstein et al., 2009). Yet, these engagement behaviors vary depending on parents' cultural beliefs about best ways to support their children's learning. Low-income Latino families in particular tend to have lower school engagement compared with other racial/ethnic groups (Sibley & Dearing, 2014). Other research demonstrates that Latino families incur educational, linguistic, and cultural barriers to their participation in school (Turney & Kao, 2009), and instead utilize home-based forms of engagement (Auerbach, 2007). Head Start is an organization serving many Latino families (nearly 40% of participants, U.S. DHHS, 2018a) and has explicit strategic aims of family engagement (U.S. DHHS, 2018b). It may therefore be particularly good at overcoming some of the aforementioned barriers for Latino families. The purpose of this study is to examine the effectiveness of Head Start in engaging Latino families both during the early care and education (ECE) period and in elementary school. Purpose/Objective/Research Question: The purpose of the current study is to identify patterns of Latino family engagement in school-based and home-based academic activities, as well as school-home connections, on a national level. It further examines the effectiveness of Head Start -- an agency that serves many Latinos and focuses on family engagement -- in engaging Latino families during the ECE period and families' subsequent engagement in elementary school. Lastly, it explores the relationship between family engagement in ECE and the elementary grades and the connection between engagement and children's cognitive, academic, and socioemotional skills. We have 2 primary research questions: (1) What are the patterns of Latino family engagement during the ECE and elementary years, and do they differ by Head Start participation?; and (2) Does Latino family engagement influence later cognitive, academic, and socioemotional skills? Setting: Data from two large, nationally representative samples -- the experimental Head Start Impact Study (HSIS) and the Early Childhood Longitudinal Study, Kindergarten Class of 2010-11 (ECLS-K:2011) -- were utilized. Table 1 lists baseline descriptive statistics of both samples. Population/Participants/Subjects: The main sample of interest in both datasets in Latino children, and thus the respective analytic samples have been restricted to only include these children. "HSIS." In line with national Head Start estimates (U.S. DHHS, 2018a), approximately 38% of the total HSIS child sample (N = 1,670) was classified at baseline through parent interview as Latino -- 1,005 in Head Start and 665 in the control condition. Table 1 provides descriptive characteristics of these participants and indicates there were no significant differences in baseline characteristics between Latino children in the Head Start and control groups. "ECLS-K:2011." The current analytic sample is restricted to Latino children who are not missing data on Head Start attendance, resulting in 2,970 children, 580 of whom attended Head Start (all N's are rounded to the nearest 10 in accordance with the requirements for the restricted use version of the ECLS-K:2011). Research Design: This study makes use of two key strategies to eliminate sources of bias. First, it uses the experimental HSIS to determine the causal role of Head Start in these family engagement processes. In addition, by also using the ECLSK:2011, we focus on replication in two large, nationally-representative samples. Data Collection and Analysis: As these are both secondary datasets, data collection has been completed. Analyses include complementary sets of descriptive analyses and residualized growth models on both datasets. In addition, given the richness and breadth of engagement variables that both HSIS and ECLS-K have in terms of the three of the key engagement measures -- home-based activities, school-based activities, and the home-school connection activities -- data reduction techniques such as factor analysis were conducted at the outset of data analysis for more parsimonious model estimation. Findings/Results: Our results are currently preliminary due to challenges arising from COVID-19. We fully expect these to be finalized by the SREE fall conference. For RQ1, we found that many of the family engagement practices were home-based in both samples. We further found higher levels of engagement for families of children who attended Head Start. We anticipate this influence to extend into the elementary grades. For RQ2, we found that Latino children whose families were more engaged in their schooling exhibited higher levels of academic and cognitive achievement, better social competence, and fewer behavior problems than those whose parents are less or not engaged. Conclusions: This study addresses important practical and policy questions. RQ1 is of central importance for Head Start as it helps to elucidate key patterns of engagement of Latino families. Additionally, as the current study tracks family engagement patterns from ECE to elementary school, it examines the coordination and collaboration between early childhood programs and K-12 education. Further, by examining associations between family engagement and later skills and recognizing the import of family engagement in Latino learners' early educational experiences, this study helps identify effective approaches to engage, support, and meet the needs of families.