Morgan, Paul L., Farkas, George, Hillemeier, Marianne M, & Maczuga, Steve
Science Achievement Gaps Begin Very Early, Persist, and Are Largely Explained by Modifiable Factors
Pennsylvania State University; University of California, Irvine
- How large are general knowledge gaps occurring in kindergarten, and to what extent do these continue to occur by the end of first grade?
- As children move from third to eighth grade, what is their typical initial level (i.e., intercept) and rate of achievement growth (i.e., slope) in science?
- Are these gaps consistent with stable, cumulative (i.e., gap increasing), or compensatory (i.e., gap decreasing) achievement growth trajectories? How do these initial third-grade science achievement levels and third- to eighth-grade growth trajectories vary by children’s race, ethnicity, language, and family SES status? How are a more general set of child- and family-level characteristics, including parenting quality, related to typical levels of third-grade science achievement in the United States as well as to achievement growth from third to eighth grade?
- To what extent are the third-grade science achievement gaps, as well as third- to eighth-grade science achievement growth, explained by such modifiable factors as general knowledge, reading and mathematics achievement, and behavioral self-regulation? How much of children’s later science achievement can be predicted by their first-grade achievement-related knowledge, skills, and behaviors?
- With the aforementioned first-grade predictive factors accounted for, how important are the modifiable factors of children’s subsequent reading and mathematics achievement, and behavioral self-regulation at each of third, fifth, and eighth grades to their science achievement during these grades?
- To what extent does a school’s academic climate and racial, ethnic, and economic composition explain children’s science achievement, over and above the afore- mentioned child- and family-level factors?
Journal Name or Institutional Affiliation:
Vol. 45 No. 1 Pp. 18-35
Children who are Black scored .62 of a standard deviation lower than children who are White on standardized science tests. The gap was .29 of a standard deviation for Hispanics. Racial and ethnic gaps were evident despite statistical control for SES and for non-English language use in the home, which them- selves were associated with, respectively, higher and lower science achievement.
The general knowledge gaps initially observed on the basis of children’s race, ethnicity, language use, and family SES in the spring of first grade were mostly attributable to the large size of these gaps already occurring at kindergarten entry.
Kindergarten general knowledge was the strongest predictor of first-grade general knowledge, which in turn was the strongest predictor of children’s science achievement from third to eighth grade.
Large science achievement gaps were evident when science achievement measures first became available in third grade. These gaps persisted until at least the end of eighth grade. Most or all of the observed science achievement gaps were explained by the study’s many predictors
Early-appearing gaps may be exacerbated by other modifiable factors. These include whether children also experience lower reading and mathematics achievement as they age and the racial-ethnic composition of the schools they attend.
This suggests that policies and practices designed to address the nation’s large and long-standing science achievement gaps may also need to be multifaceted and address lagging reading and mathematics achievement, lower behavioral self- regulation, and school racial segregation which is currently increasing.
Journal Article Empirical Research
Achievement Gap, Minorities, Poverty, SES, Science
Method of Analysis:
Unit of Analysis:
Parent, School, Student
The database analyzed is the public-use file of the Early Childhood Longitudinal Study, Kindergarten Class of 1998-1999 (ECLS-K), a nationally representative cohort of children who entered kindergarten in 1998. The analytic sample consists of 7,757 students who had at least one science test score.
DV: Science test score
IV: Race, family SES, child age in months, sex (child is male), child's mother is not married, test scores, reading test score, mathematics test score, approaches to learning, percent minority in school, percentage free lunch in school.