Association for Public Policy Analysis & Management

Research examining parental household size and student academic achievement is prolific.  Parental household size is typically dichotomized between single parent households and two parent households.  Findings are generally consistent that children coming from two married parent households have increased academic achievement when compared to children from single parent households or children from two unmarried parent households (Artis (2007); Brown (2004); and Sun & Li (2011)).  Much of the research performed on this topic uses data drawn from the ECLS-K 1998-1999 data set.  Furthermore, several analyses use fixed effects models to attempt and achieve causal inference.  Findings from these studies suggest that divorce can negatively affect academic achievement and married parent households are generally better for promoting academic achievement (Amato & Anthony (2014) & Anthony, DiPerna, & Amato (2014)).  Together these analyses provide important evidence for both education and family support policy makers.  Specifically, policy makers can improve academic achievement for students by making investments in programs which not only promote marriage for cohabitating couples and single parents but also, support and encourage married parents to stay together and avoid the negative consequences of divorce.  Although these studies provide useful evidence, very little research has been conducted with the recently released ECLS-K 2010-2011 data set.  With the exception of certain annual shifts, the CDC/NCHS National Vital Statistics System (2015) reported an overall decrease of 14% in divorces and annulments nationally in the United States from 2000-2014. However, using the same time period (i.e., 2000-2014) this same system showed a decrease in marriage rates of 7.5%.  Together, this data suggests that while fewer couples are divorcing, fewer couples are also getting married.  Given this recent phenomena, it is important for researchers to leverage the newly released data set of the ECLS-K 2010-2011 to determine if changes in parental household size and student academic achievement also exist.  These types of analyses will provide important evidence based information for policy makers considering designing programs which promote academic achievement in young children.  In this paper, we use longitudinal data from the ECLS-K 2010-2011 data set obtained at four points in time to examine the effects of parental household size on students’ academic achievement.  Our analysis utilizes a cross-lag model and leverages the advantage bestowed with longitudinal data sets to help provide causal claims for the role of parental household size in determining student achievement gains for both mathematics and reading.