Undergraduate Honors Thesis

Abstract

Residential segregation in Worcester, Massachusetts and several dimensions of its impacts on marginalized communities was evaluated using a geospatial analytical approach. Geographic Information Systems (GIS) and dynamic mapping methods were used to ask if racial segregation can be seen at the local level of neighborhoods. The causality between residential segregation, school segregation, and academic achievement of minority students was investigated. Worcester Public School's student enrollment data was used to determine school segregation patterns and 10th grade MCAS scores from 2014-2019 to rank order academic proficiency in the minority student population. The Worcester Public School 5-year average for Legacy MCAS subjects was compared to the state 5-year average. Segregation at the neighborhood level was measured using dot density to model the distribution of racial groups.

1.0 Introduction (abbreviated version)

The present study aims to present Geographic Information Systems (GIS) as an analytical approach to understanding the relationship between educational policy developments, academic achievement, and the geography of opportunity. The patterns and effects of residential segregation have been correlated with other social issues relating to poverty and race (Tate & Jones, 2017). The education system in America is based upon geographical constraints, community needs, and relevant enterprises. The boundaries of such a system have traditionally been defined by political, legislative, and cultural forces but the geography of education varies according to demographics. Differences in school programs, quality of teachers, quality of materials, student backgrounds, family-based support, and overall academic achievement translate into disparities of educational opportunities. There is a current need in the realm of education research to include spatial methodology (Lubienski & Lee, 2017; Hogrebe & Tate, 2012). Using GIS, the contextual factors concerning education and opportunity in America can be more comprehensively examined. Tobler's first law of geography brings to the forefront the significance of spatial relationships and academic achievement by acknowledging how proximity to something creates greater causality than distance does (Miller, 2004). Meaning that, the issues impacting academic achievement at the student level are more likely to be caused by factors according to the proximity of such factors to the school system. In Massachusetts, Worcester Public Schools uses street listings to determine what schools a student is eligible to attend, and it can be a challenge for parents to transfer their children into a different public school without moving residence or having a disciplinary issue. Henceforth, it is appropriate to postulate that the racial composition of neighborhoods is a proximal factor that impacts Worcester Public Schools’ academic achievement. Education research will find relevance in geospatial analysis because the causality of the contextual factors concerning academic achievement across public schools in urban regions calls for nuanced analysis of geographic space. By evaluating which Worcester Public Schools are minority segregated and which schools have historically fallen behind in the Massachusetts Comprehensive Assessment System (MCAS), the optimization of test preparation, support for educators, and efficiency of taxpayer investment will be further improved upon.

Worcester, Massachusetts needs a study that visualizes the dimensions of segregation and public-school academic achievement. Standardized tests are an increasingly popular method to determine school promotion, retention, and graduation by tracking student data across demographics. High stakes standardized tests are used to determine what future classes students are eligible for, what curriculum teachers should teach, and if a student can advance to the next grade or to graduate high school. The MCAS, ACT, PSAT, and SAT are examples of such tests. The manifested consequences of standardized tests include student academic outcomes, school district accountability, tailored curriculum reform, and defining the criteria for state funded resources (Schoenberg, 2019). Educational improvement initiatives are a shared responsibility between students, parents, educators, public officials, states, and school districts. Students are put into specialized tracks as a direct result of their test scores and many receive scholarships for college depending on their recorded proficiency levels. Standardized tests are used to establish a standard for student achievement and learning outcomes, but they are also used to highlight what subject areas school officials need to reform in order to promote a higher quality of education for all students. School policies in Massachusetts are developed in response to a school district’s MCAS scores but those policies also need to encompass the broader scope of knowledge and skills the test is intended to measure. Being able to show academic proficiency visually showcases how accomplished Worcester Public Schools has been in preparing their students for graduation and to succeed in higher education. The best method to do this is through the utilization of geospatial analysis with relevance to specific high school’s MCAS test score data due to the current scope of education research lacking interdisciplinary analysis of complex variables impacting the geography of academic opportunity. The city of Worcester is currently being sued by the National Association for the Advancement of Colored People over allegations of a discriminatory school board election system (Moulton, 2021). The elected School Committee is responsible for educational policies and reforms which directly impact the diverse student population, yet the composition of committee via the at-large election system denies communities of color from electing a candidate of their choice. GIS can be used to convey a more than ever politically relevant spatial relationship between neighborhood segregation and MCAS test score proficiency levels which the public may not even be aware of but would certainly be interested to learn.

2.0 Methods (abbreviated version)

Previous research has used several different geospatial approaches and techniques to investigate the dimensions of segregation (Reardon & O’Sullivan, 2004; Oka & Wong, 2014; Tate & Hogrebe, 2019). Residential segregation at the neighborhood level demographic data from census tracts provided by the 2010 U.S. Census was mapped using the ArcGIS Pro. It is important to note, political and demographic factors have influence over both census tract and school district boundaries. Ultimately, this means that the areal units of census tracts and school district boundaries can be modified based on differing interpretations of political decisions and data processing techniques. Some examples of the modifiable areal units problem in relation to GIS is the reshaping of voting districts via gerrymandering or the statistical biasing effect in landscape ecology scale observed when larger spatial units are aggregated. Therefore, the right spatial scale needs to be used for proper evaluation of segregation to be conducted by GIS practitioners.

A study done by Tate and Hogrebe (2019) showcased how the modified areal unit problem affected segregation due to arbitrary changes in school and district boundaries. Their study showed how composite population counts mitigates these effects while also proposing a novel approach to measure two dimensions of segregation called evenness-clustering and isolation-exposure. Future research concerning residential segregation and the academic achievement of Worcester Public Schools should include the investigation of these two dimensions of segregation mapped as single variables. The present study used a dot density analysis, where one dot equaled 50 persons, in order to determine the natural distribution of racial groups in Worcester. Additionally, the racial composition of all seven Worcester Public Schools containing grade 10 was investigated and graphed (see Figure 1). Grade 10 was selected because it is the final MCAS testing year and is often used to determine a student’s college preparedness and their ability to get a high SAT score. Minority groups evaluated at the school level were African American, Asian, and Hispanic students. Claremont Academy, North High and University Park Campus School were determined to be composed of over 75% of minority students. The present study defined these schools as minority segregated although it is important to note that while University Park Campus School has a special application process the school still follows a street listing system. The level of academic proficiency in English Language Arts, Mathematics, and Science/Technology during 2019 under the New Generation MCAS standards was evaluated for all seven public schools and made into an excel sheet. Then, the data was uploaded into ArcGIS Pro through the Excel to Table geoprocessing tool and the table was joined with the MASSGIS school shapefile. The Next Generation MCAS data was overlaid, one academic subject at a time, onto a racial dot density map showing Worcester’s residential segregation according to 2010 U.S Census data. Schools determined to be minority segregated were compared to the state average and other nonminority segregated schools in Worcester. The 2019 Next Generation MCAS proficiency levels were defined by the percentage of students per school that received a score of meeting expectations or exceeding expectations. Next Generation MCAS proficiency levels of each school was separated into three categories and then those categories were visualized using three colors: black, medium gray and light grey. Black represented a school with over 60% proficiency level in a test subject. Medium gray represented a school with less than or equal to 60% proficiency level in a test subject. Light gray represented a school with less than or equal to 30% proficiency level in a test subject. Next Generation MCAS scores were not compared to Legacy MCAS scores other than to show how the difference in measurement and scaling impacted the each of the three minority school’s scores (see Appendix G).

3.0 Results (abbreviated version)

The rate of success in the Legacy MCAS comparisons was determined by the percentage of students that received a score of advanced or proficient versus the percentage of students that received a score of needing improvement or failure. Figures 2 and 3 show the state and Worcester Public School Legacy MCAS five-year averages. Figure 2 shows Asian and White students have made up 60% of Worcester Public Schools Legacy MCAS proficiency or higher scores while Hispanic and Black students have made up 66% of Worcester Public Schools Legacy MCAS needing improvement or lower scores which is congruent with the state five-year trend seen in Figure 3. Figure 2 also shows that Black students in Worcester Public Schools have had a slightly better five-year average proficiency level compared to the state.

This study concluded minority segregated schools Claremont Academy and North High had an overall lower MCAS proficiency level in all three subject matters, but University Park Campus School did not follow this trend. Figure 4 shows the only school to have a greater than 60% proficiency level in the 2019 Next Generation MCAS Mathematics test was Worcester Technical High School. Figure 4 also shows that Burncoat Senior High School, Claremont Academy and North High School had a less than or equal to 30% proficiency level in the 2019 Next Generation MCAS Mathematics test. The remaining schools shown in Figure 4, University Campus Park School, Doherty Memorial School, and South High School had between 31-60% proficiency level in the 2019 Next Generation MCAS Mathematics test. The 2019 Next Generation MCAS Mathematics exam had the lowest proficiency level accomplished across all Worcester Public Schools compared to the Science/Technology and English Language Arts 2019 Next Generation MCAS exams. Figure 5 shows the only school to have a less than or equal to 30% proficiency level in the 2019 Next Generation MCAS Science/Technology test was Claremont Academy. Figure 5 also shows that University Campus Park School, Doherty Memorial, and Worcester Technical High School had over 60% proficiency level in the 2019 Next Generation MCAS Science/Technology test. The remaining schools shown in Figure 5, South High School, Burncoat Senior High School and North High School had between 31-60% proficiency level in the 2019 Next Generation MCAS Science/Technology test. The 2019 Next Generation MCAS Science exam had the highest proficiency level accomplished across all Worcester Public Schools compared to the Mathematics and English Language Arts 2019 Next Generation MCAS exams. Figure 6 shows the only school to have a greater than 60% proficiency level in the 2019 Next Generation MCAS English Language Arts test was Worcester Technical High School. Figure 6 also shows Claremont Academy and North High School had a less than or equal to 30% proficiency level in the 2019 Next Generation MCAS English Language Arts test. The remaining schools shown in Figure 6, South High School, Burncoat Senior High School, University Campus Park School, and Doherty Memorial had between 31- 60% proficiency level in the 2019 Next Generation MCAS English Language Arts test. The 2019 Next Generation MCAS English Language Arts exam had the most Worcester Public Schools in the middle range of proficiency level between 31-60%.

4.0 Discussion (abbreviated version)

The implications of the present study are that all Worcester Public Schools need moderate improvement in order to be more closely aligned with the state annual MCAS proficiency average especially concerning Latino and Black students. The Next Generation MCAS scores from Worcester Public Schools show an overall drop in high score rates even among Asian and White students who have historically been successful in achieving proficient or advanced scores on the Legacy MCAS. Claremont Academy has the highest overall need for improving test scores and Worcester Technical High school has the least. The racial composition school’s and the correlation between academic achievement has been studied in other regions with the conclusion that the more integrated a school population is the better overall test scores for the lowest achieving racial group meaning that a possible solution to test score discrepancies is to have school districts be as integrated as possible (Tate, 2008; Gulosino & D'Entremont, 2011; Ellen et al, 2012; Walsh et al, 2014; Tate & Hogrebe, 2019). The present study was foremost limited by the information available on the Massachusetts Department of Elementary and Secondary Education database whose most recent testing information was significantly impacted by the COVID-19 pandemic. The measurement and scaling differences between the Legacy and Next Generation MCAS also poses a substantial restriction on data analysis. Future guidelines on how to fairly compare between the two formats of testing must be established by the Massachusetts Department of Elementary and Secondary Education. Additional limitations came from the constrained racial group definitions. The best practice for future analysis would be to expand the racial categorization of Asian and Hispanic students to include specific nationalities in order to better comprehend these very broadly described groups in reference to all standardized testing (Lee, 2011). A previous study of Massachusetts Elementary schools used remote sensing to generate Spatial Generalized Linear Mixed Models which estimated the impacts of surrounding greenness on school-based performance. That study concluded that greenness of the school area had a measurable causality to school-wide academic performance (Wu et al, 2014). Remote sensing-based research into the causality between school academic performance and school green space can also be applied to Worcester schools. Additionally, a future study should conduct research into the specific MCAS scores for each racial group at the school level preferably at a time when there is much more Next Generation MCAS data available for comparison.

5.0 Conclusion (abbreviated version)

These findings show that Latino students underperform all other racial groups and Asian students outperform all other racial groups in both Legacy MCAS and Next Generation MCAS proficiency levels across all three test subjects. Claremont Academy has the highest minority student population and consistently has the lowest proficiency levels in all three test subjects. Worcester Technical High School has the most evenly distributed student population and consistently has the highest proficiency levels in all three test subjects. Worcester Public Schools have an average lower proficiency level in all three test subjects than the state per year but the five-year Legacy MCAS average revealed that Black students did better than the state average.

References

Ellen, I. G., O’Regan, K., Schwartz, A. E., & Stiefel, L. (2012). Racial segregation in multiethnic schools: Adding immigrants to the analysis. In W. F. Tate (Ed.), Research on schools, neighborhoods, and communities: Toward civic responsibility (pp. 67–82). Lanham, MD: Rowman & Littlefield Publishers

Fábos, A., Pilgrim, M., Said-Ali, M., Krahe, J., & Ostiller, Z. (2015, February). Understanding refugees in Worcester, MA. Retrieved February 09, 2021, from https://commons.clarku.edu/mosakowskiinstitute/32/

Gulosino, C., & D'Entremont, C. (2011). Circles of influence: An analysis of charter school location and racial patterns at varying geographic scales. Education Policy Analysis Archives, 19, 8. https://doi.org/10.14507/epaa.v19n8.2011

Hogrebe, M. C., Tate, W. F. (2012). Geospatial perspective: Toward a visual political literacy project in education, health, and human services. Review of Research in Education, 36, 88–115.

Hogrebe, M. C., & Tate, W. F. (2019). Residential Segregation Across Metro St. Louis School Districts: Examining the Intersection of Two Spatial Dimensions. AERA Open. https://doi.org/10.1177/2332858419837241 retrieved from, https://gold.worcester.edu:3171/doi/full/10.1177/2332858419837241#articleCitationDownloadC ontainer

Lee, P. (2011). The "Asian" category in MCAS achievement gap tracking: time for a change. Asian American Policy Review, 21, 19+. https://link.gale.com/apps/doc/A306598114/EAIM?u=mlin_c_worstate&sid=EAIM&xid=91163 c8a Massachusetts Department of Elementary and Secondary Education. Massachusetts School and District Profile: Worcester (03480000) https://profiles.doe.mass.edu/general/general.aspx?topNavID=1&leftNavId=100&orgcode=0348 0000&orgtypecode=5

Miller, H. J. (2004). Tobler’s first law and spatial analysis. Annals of the Association of American Geographers, 94, 284–289.

Moulton, C. (2021, February 8). NAACP chapter sues Worcester over school board election system that allegedly discriminates. Telegram & Gazette.

Nichols-Barrer, I., Place, K., Dillon, E., & Gill, B. P. (2016). Testing college readiness. Education Next, 16(3) https://gold.worcester.edu/login?url=https://gold.worcester.edu:2082/scholarly-journals/testingcollege-readiness/docview/1792130263/se-2?accountid=29121

Oka, M., & Wong, D. W. S. (2015). Spatializing segregation measures: An approach to better depict social relationships. Cityscape: A Journal of Policy Development and Research, 17, 97– 113. (downloaded as pdf)

Schoenberg, Shira. Massachusetts' New School Accountability Rankings: What Do They Mean for Your School?, MassLive, 29 Jan. 2019, www.masslive.com/politics/2018/09/massachusetts_new_school_accou.html.

Sohoni, D., & Saporito, S. (2009). Mapping School Segregation: Using GIS to Explore Racial Segregation between Schools and Their Corresponding Attendance Areas. American Journal of Education, 115(4), 569-600. doi:10.1086/599782 Retrieved from, https://gold.worcester.edu:2490/stable/10.1086/599782?pqorigsite=summon&seq=1#metadata_info_tab_contents

Tate, W. F., Jones, B. (2017). Anonymity no more: Seeing our neighbors in Ferguson and the implications for social policy. Educational Researcher, 46, 211–222.

Tate, W. F. (2008). Geography of opportunity: Poverty, place, and educational outcomes. Educational Researcher, 37, 397–411. Retrieved from, https://gold.worcester.edu:3475/docview/216911261?pq-origsite=summon

Tate, W. F., & Hogrebe, M. C. (2018). Show me: Diversity and isolation indicators of spatial segregation within and across Missouri's school districts. Peabody Journal of Education, 93, 5– 22

Walsh, M. E., Madaus, G. F., Raczek, A. E., Dearing, E., Foley, C., An, C., Lee-St. John, T. J., & Beaton, A. (2014). A New Model for Student Support in High-Poverty Urban Elementary Schools: Effects on Elementary and Middle School Academic Outcomes. American Educational Research Journal, 51(4), 704–737. https://gold.worcester.edu:2191/10.3102/0002831214541669

Wong, D.W.S. Implementing spatial segregation measures in GIS, Computers, Environment and Urban Systems, Volume 27, Issue 1, 2003, Pages 53-70, ISSN 0198-9715, https://doi.org/10.1016/S0198-9715(01)00018-7. Retrieved from, https://gold.worcester.edu:2088/science/article/pii/S0198971501000187?via%3Dihub

Wu, C.-D., McNeely, E., Cedeno-Laurent, J. G., Pan, W.-C., Adamkiewicz, G., Dominici, F., Lung, S.-C. C., Su, H.-J., & Spengler, J. D. (2014). Linking Student Performance in Massachusetts Elementary Schools with the "Greenness" of School Surroundings Using Remote Sensing. PLoS ONE, 9(10). https://link.gale.com/apps/doc/A418553291/OVIC?u=mlin_c_worstate&sid=OVIC&xid=eb4b1b 5c