How Family Resources, Structure, and Support Relate to Students' Math Performance in Remote Learning

• In Event: COVID-19 Experiences, Contexts, and Outcomes From The Nation’s Report Card

Abstract

The COVID-19 pandemic hit every student in one way or another. When schools closed and students were sent home, students increasingly had to rely on their parents for educational support. Studies on educational outcomes during the pandemic focused mainly on socioeconomic status (SES). They showed that children from low-SES families tended to perform worse during remote learning (Moscoviz & Evans, 2022). A recent study from the Netherlands investigated the role of family composition in students' learning growth and found that children in single-parent families suffered larger learning lags during the pandemic than children in two-parent families (de Leeuw et al., 2023). To understand how U.S. family characteristics relate to students' performance, this paper examines how family resources, structure, support, and communication related to students' math performance in remote learning.

Data are from the National Assessment of Educational Progress (NAEP) 2022 grade 8 mathematics assessment and student contextual questionnaire, which collected information on family characteristics. The sample is eighth graders who indicated that they had learned remotely in the 2020-21 school year and who had no missing values in all the included variables. The study includes family structure, resources, support, and communication as predictors, gender, race/ethnicity, SES, math interests, and math confidence as control variables. The study used descriptive and a series of regression analyses.

Preliminary results show that 73% of eighth graders recalled learning remotely in the 2020-21 school year. Among them, 60% lived in a two-parent family and scored 20 points higher in math performance than their peers in a single-parent home (17%). Similarly, this group's students scored 16 points and 23 points higher than students who lived in a stepfamily and a foster home, respectively. About 21% of eighth graders indicated that they never or hardly ever talked about things studied in school with a family member, compared with 40% talking 2-3 times a week or daily. Students who never discussed their study with a family member scored 14 points lower than those who often did that.

Regression results show that family structure, resources, support, and communication were all significant predictors of math performance, controlling for gender, race, and SES. The regression coefficients for family structure did not change much even when math interest and confidence were added to the models, suggesting that the effect of family structure cannot be explained away by math interest and confidence. Family resources are positively related to math performance, but the regression coefficient is small, and the value was reduced to half when math interest and confidence were added to the model. Students who got more support/help from their family members scored lower than those who got less. Given the minor effects of family resources and the negative relationship between support and math performance, our final models will remove those variables. We will compare the relationship between family structure and math performance between students who had remote learning and who did not. The comparison may prove that the family structure has a larger effect on students with remote learning experiences.

Authors

• Grace Ji Cheng, American Institutes for Research

• Xiaying Zheng, American Institutes for Research