The negative impact of COVID-19 on students persists three years after the onset of the pandemic. A substantial body of research shows that students have yet to recover to their learning trends before the pandemic (Callen et al. 2023, Lewis & Kuhfeld 2023, Cohodes et al., 2022). This suggests that students face a lengthy road to recovery, and there is a need to identify additional means to help students accelerate their learning. It is worth noting that the detrimental effects of the pandemic are reflected through multiple channels including academic growth, school attendance, as well as social-emotional wellbeing. Other than examining the change of trajectory of students' academic achievement after the pandemic, there is a lack of studies focused on how the pandemic affected attendance or students' social-emotional learning. There is a growing awareness that there is an epidemic of chronic absenteeism at the aftermath of pandemic. The rate of chronic absenteeism has reached 25% on average in 2022, 10 percentage points higher relative to pre-COVID trend. In the meantime, the pandemic has also taken a toll on students' social and emotional wellbeing. According to a recent survey conducted in 2022, 65% of parents taking the survey consider their child's happiness and emotional well-being as the top concern. To our knowledge there has not been evidence on how the pandemic affected students' social-emotional wellbeing. Based on the evidence provided by prior studies that students' social-emotional competence (SEC) predicts academic success in school (Oberle et al. 2014, Deming, 2017, Caprara et al. 2000, Schonert-Reichl 2019, Kanopka 2024), it is to be expected that the impact of the pandemic on students' social-emotional learning will also reflect on their academic achievement. There are three research questions we would like to address in this study. 1. How has the pandemic changed the school attendance? 2. Is students' social-emotional learning a driving force of the prevalence of chronic absenteeism? 3. How the relationship between students social-emotional learning and academic achievement has changed before and after COVID-19 pandemic. The heterogeneity of the relationship between social-emotional learning and academic achievement across subgroups of students categorized by race/ethnicity, gender, socioeconomic status. To answer these questions, we are going to use longitudinal dataset from California's CORE Districts including both standardized state test scores, attendance data, as well as a comprehensive set of measurement of social-emotional competence. The advantage of this dataset is that it covers multiple school years spanning from before pandemic period 2017-2018 until most recent year 2022-2023 which allows us to compare the trends of the academic achievement as well as the development of social-emotional learning before and after the pandemic to evaluate how the causal relationship between social-emotional learning and academic achievement has changed before and after pandemic. The data is also most suitable to examine the heterogeneity of the relationship between social-emotional learning and academic achievement across subgroups of students categorized by race/ethnicity, gender, socioeconomic status. The CORE districts dataset is particularly suited for this type of heterogeneity analysis since students in the CORE Districts come from diverse demographic backgrounds. Many of them are from historically underrepresented minority groups who were disproportionally impacted by the pandemic. In comparison to the national public-school population, our sample consists of a much higher percentage of Latinx 73% vs. 28% at the national level, and higher proportion of students eligible for free or reduced-priced lunch 72.2% vs. 55% nationwide. Prior studies have shown that the development of social-emotional learning from Grade 4 to Grade 12 vary by these dimensions such as race/ethnicity, gender, socioeconomic status, as well as grade level. For instance, among the four constructs of social-emotional learning, three of which, i.e., self-efficacy, self-management and social awareness do not increase monotonically as students advance through grade levels. Female and male students report different patterns of development in self-management, social awareness, and self-efficacy in middle and high school. It is worth noting that these findings are based on the data prior the pandemic. We develop our primary models using the tools of multilevel models, allowing for the contributions of students nested within schools and school nested within districts. These models are closely related to value-added growth models in that they control for prior determinants of student achievement, including lagged achievement, student characteristics, and school, district, and year effects. Chetty, Friedman, and Rockoff (2014) and Kane and Staiger (2008) show that models of this type provide valid measures of teacher quality (and indirectly, school quality). We consider how growth models, which we use here as a more general term than value-added, are affected by factors essentially new because of the COVID pandemic. The models embody features of evaluation models and standard growth and value-added models. The evaluation lens prompts us to view the COVID pandemic as a type of treatment (or intervention) and the pre-COVID period as a control. This work has broad policy implications for policy makers and educators. Our hypothesis is that low school attendance might be the driving force behind the disappointing academic recovery from COVID. If the data and empirical analysis would confirm this is the case, then the estimate of the impact of social-emotional learning on attendance and on academic achievement will shed light on the potential cause of the sluggish academic recovery. Hence, our study will provide evidence on how to best mobilize resources to accelerate academic recovery process to bring students back on track.