2010 - Student and School Predictors of High-Stakes Assessment in Science

Attribution: Maerten-Rivera, Jaime, Myers, Nicholas, Lee, Okhee, & Penfield, Randall
Researchers: Jaime Maerten-RiveraNicholas MyersOkhee LeeRandall Penfield
University Affiliation: University of Miami
Email: jmaerten-rivera@miami.edu
Research Question:
This study examined both student and school predictors of science achievement as measured by a high-stakes state test.
Published: 1
Journal Name or Institutional Affiliation: Science Education
Journal Entry: Vol. 94, No. 6, pp. 937-962
Year: 2010
Findings:
  • Student background predictors combined to account for 25% of the student-level variation in science achievement, and each of the background predictor variables was a statistically significant predictor of student-level science achievement.
  • Results indicate that females scored lower than males, Blacks and Hispanics scored lower than Whites, students on free or reduced price lunch scored lower than students not on free or reduced price lunch, ESOL students scored lower than non-ESOL students, and ESE students scored lower than non-ESE students.
  • Reading and mathematics jointly accounted for an additional 58% of the student-level variation in science achievement.
  • School predictors combined to explain 85% of the variation in average school science achievement.
  • School racial and SES composition were not significant predictors of achievement
Keywords: Academic AchievementCompositionRacial CompositionReadingScienceSES CompositionRegions: SouthMethodologies: QuantitativeResearch Designs: Secondary Survey DataAnalysis Methods: Multilevel Models Sampling Frame:Large Urban School District
Sampling Types: PopulationAnalysis Units: SchoolStudentData Types: Quantitative-Cross Sectional
Data Description:
  • Research was conducted in a large urban school district in the southeast United States with a student population displaying a high level of linguistic and cultural diversity.
  • Data were collected for all fifth-grade students in the district from all public elementary schools, except charter and alternative schools.
  • Information on each school (i.e., percentage of beginning teachers, school size, and average class size) was available for all schools in the district.
  • 23,854 fifth-grade students from 198 elementary schools were included in the analyses.
  • DV: Fifth-grade science achievement (measured by student scores on a high-stakes state test in science)
  • IV: Ethnicity, eligibility for free or reduced-price lunch, reading score, mathematics score, exceptional student education, percentage of beginning teachers, average class size, percentage of female students, percentage of Black students, percentage of Hispanic students, percentage of student eligible for free or reduced-price lunch, percentage of students in exceptional student education, aggregated reading scores, aggregated mathematics scores
Theoretical Framework:
Relevance:
Archives: K-12 Integration, Desegregation, and Segregation Abstracts