EFFECTS OF FAMILY AND SCHOOL INSTITUTIONS ON CHILD OUTCOMES FROM EARLY CHILDHOOD THROUGH ELEMENTARY SCHOOL

Abstract

Children from different family backgrounds receive unequal levels of investment in their development, prompting concern about social inequality trends. Complete explanations of inequality in child outcomes require serious treatment of both family and school institutions. This dissertation follows a three-paper format unified by a common theme concerning the joint effects of the home learning environment and school institutions on outcomes in early childhood and elementary school. Data come from the Early Childhood Longitudinal Study Program. The first paper focuses on the role of parents in tandem with preschool programs for disadvantaged children using a birth-cohort dataset. The results indicate the home environment and preschool programs are associated with kindergarten readiness when both are specified in the estimation model. The moderation analysis suggests the effect of parents’ emotional support is stronger for Head Start children than parent-care children. This paper assesses the psychometric properties of home environment measures based on rich, modern instruments from child psychology over traditional self-reports. The interaction findings between home environment and preschool are thus more nuanced and convincing. The second paper examines the elementary school stage, calling attention to the need for continuous intervention in school and family institutions for disadvantaged children, using the kindergarten-cohort dataset. The home environment and school context were associated with outcomes and evidence of an interaction was found. This paper systematically examined available information regarding school environment, constructing scales with substantive clarity and high measurement reliability. The growth curve framework allows a thorough answer to the central question regarding family and school institution interactions with school-specific random effects in shaping the growth curves. Applying a data-integration method, the third paper combines the birth-cohort dataset collected at an earlier time with a recent kindergarten-cohort dataset to answer the same questions from the first paper. School-based cohort surveys often cover one educational stage. Yet prior exposure to favorable or harmful environments may have long-term impacts. The exploration of one data combination method in this paper moves the research forward both methodologically and substantively. The results reiterate and update the importance of including both institutions to estimate kindergarten readiness for a more recent cohort.