2015 - Narrowing Pathways? Exploring the Spatial Dynamics of Postsecondary STEM Preparation in Philadelphia, Pennsylvania

Attribution: Edmunds, Kimberly A., Pearsall, Hamil, & Porterfield, Laura K.
Researchers: Hamil PearsallKimberly A. EdmundsLaura K. Porterfield
University Affiliation: Temple University
Email: kedmunds@researchforaction.org
Research Question:
What geographical factors are associated with the postsecondary STEM preparation of students from underrepresented groups in the School District of Philadelphia from middle to high school?
Published: Yes
Journal Name or Institutional Affiliation: The Urban Review
Journal Entry: Vol. 47, No. 1, Pp. 1-25
Year: 2015
Findings:

– These analyses find strong relationships among math performance, a key indicator of college readiness for courses of study in STEM, and neighborhood factors within school catchment areas. For example, high percentages of unemployed residents are negatively correlated to math performance, while high median household income is positively correlated with math performance. These relationships vary spatially across middle and high school catchment areas.
– Percent non White or Asian, percent IEP, and percent economically disadvantaged are strongly and significantly correlated to math performance. Each of these variables is negatively correlated to math performance.
– From the neighborhood variables, Median household income, percent unemployed, percent of neighborhood with high school diploma only, and percent of neighborhood with a Bachelors Degree are significantly correlated. Median household income and percent with a Bachelors Degree are positively associated, while percent unemployed and percent with high school diploma only are negatively associated.
– . SAT performance and AP enrollment were negatively associated with diploma earning while positively associated with bachelor’s degree attainment.

* Correlations between composition within neighborhood and school and math readiness.

Scholarship Types: Journal Article Reporting Empirical ResearchKeywords: ContextHigh SchoolMathNeighborhoodSTEMUrban SchoolsRegions: NortheastMethodologies: QuantitativeResearch Designs: Secondary DataAnalysis Methods: Cluster AnalysisCorrelation analysisGeographically-weighted regressionOLS Sampling Frame:Students in Philadelphia
Sampling Types: Non-Random - PurposiveAnalysis Units: School Catchment AreaData Types: Quantitative-Cross Sectional
Data Description:

The authors utilize the Pennsylvania System of School Assessment (PSSA) data in math from 2007 to 2008 (Grade 8) and 2010-2011 (Grade 11). Students in Pennsylvania take the PSSA for reading and math in grades 3-8 and 11. To correspond with the school performance data, the authors used American Community Survey (ACS) data 5-year estimates from roughly corresponding years, based on the data availability (2005-2009 and 2007-2011). This study’s sample includes 25 neighborhood high schools.

The authors examined 8th and 11th grade groups at neighborhood middle and high schools in Philadelphia and their corresponding catchment areas. As the unit of analysis in this study, the school catchment area constitutes a neighborhood boundary. Elementary schools are contained in middle school catchments and middle schools are contained in high school catchments. Catchment areas and school feeder patterns are subject to change based on shifts in population size, school reconfigurations, and closings. The authors acknowledge that by also excluding charter schools, which in recent years have been rapidly expanding in Philadelphia, as well as private and parochial schools, they limit the analyses to a small subset of educational institutions.

The authors included demographic information on the students taking the PSSA exam: race/ethnicity, English language learners (ELL), special education (students with an individualized education plan, or IEP), and economically disadvantaged (eligible for free or reduced price lunch).

The housing density variable included the percentage of each school catchment area that was covered by high-density housing. They also used park acreage as an additional neighborhood characteristic, frequently associated with quality of life.

Other variables include: Percent ELL of the school, percent individualized education plan (IEP) of the school, percent economically disadvantaged of the school, percent Black of the school, percent Latino of the school, percent White of the school, percent Asian of the school, percent non Asian or White at school, percent high school diploma only of the neighborhood, percent bachelors degree of the neighborhood, median household income, and percent unemployed.

The dependent variable is math performance which is measured by the results from PSSA

Theoretical Framework:
Relevance:Factors Related to STEM readiness
Archives: K-16 STEM Abstracts