Problem Definition

• A third of the approximately 80,000 NYC 8th graders applying to high school sit for the SHSAT.
• Admitted students represent ~20% of test takers (5,000 students).
• Admitted students attend 45 “feeder” middle schools ~20% of the all middle schools.
• Specialized high schools are unbalanced demographically.
• The biggest imbalances are with admitted Black/Latino and Economically disadvantaged students.

Research Purpose

• Predict feeder middle schools using factors derived from the Literature Review.
• Predict the number of student acceptances from SPHS middle schools.
• Propose how feeder middle schools can meritocracies and demographically representative.

How can PASSNYC address the Diversification of NYC Specialized High Schools using Data Science?

• Sponsors a Kaggle Competition for Recommender System Solutions.
• Goal is to close the diversity gap at 8 NYC Specialized High Schools.
• Recommendations focus is on reaching the underserved students and preparing them for taking the SHSAT.

• Motivational theory:" argues that test-based accountability can catalyze improvement
• Alignment theory: argues that test-based accountability enables structural consistency among major components of the educational system
  
• Information theory: tells us that analytics can be used as a feedback mechanism to drive performance improvements
• Symbolism: emphasizes that systems of accountability signal important values to stakeholders
• Adverse Selection: concentrating SPHS admissions in a few feeder middle schools is a principal-agent problem where the agent (NYCDOE) has more information about school quality, student performance and academic options than the principals (the students). This can lead to a system failure resulting in a few students that benefit from best schools while the rest of the are left with lower-quality options.

• Predict feeder middle schools using factors derived from the Literature Review.

SPHS data for 570 middle schools was used for predictive models.

Two binary logistic regression model were developed with an accuracy of 94.49% and 93.98% respectively.

• Predict the number of student acceptances from SPHS middle schools.

• ZIBN Model 1: (SHSAT test) predicted 41 (24%) SPHS feeder schools of 171 schools chosen at random from the data set.

• ZIBN Model 2: (NY State test) predicted 53 (31%) SPHS feeder schools of 171 schools chosen at random from the data set.

• Significant Finding:
  Model 1 & 2 are negatively correlated indicating the SHSAT tests fewer feeder middle schools with high concentrations of offers. This shows an underrepresented and more dispersed set of students that would otherwise be admitted on the basis of their NY State test scores if that was the standard.

• Propose how feeder middle schools can meritocracies and demographically representative.

• Models 1 (SHSAT) & 2 (NY StateTest) provide accurate predictions of SPHS feeder middle schools using factors derived from the Literature Review.

• Acceptances from SPSH feeder middle schools were accurately predicted using a zero-inflated binomial model which discounts for overinflation of non-feeder schools.

• Replacing the SHSAT with the NY State test indicates more dispersed acceptance rates at underrepresented schools. The distinction between the SHSAT and NY State Test that matters is accessiblity and preparation for the test.

• Note that both tests measure academic merit but the SHSAT only caters to a select portion of the student population.

[22] Achim, Zeileis & Kleiber, Christian & Jackman, Simon. (2008). Regression Models for Count Data in R. Journal of Statistical Software. 27. 1-25. 10.18637/jss.v027.i08. Retrieved from https://cran.r-project.org/ web/packages/pscl/vignettes/countreg.pdf [7] Akerlof, George A., 1970. “The Market for”Lemons“: Quality Uncertainty and the Market Mechanism,” The Quarterly Journal of Economics, Oxford University Press, vol. 84(3), pages 488-500. [11] Biggs, John B.; Tang, Catherine Kim Chow (2011). Teaching for quality learning at university: what the student does. Maidenhead: McGraw-Hill. ISBN 9780335242757. [10,17,18,19] Corcoran, Sean Patrick & E. Baker-Smith,Christine 2018. “Pathways to an Elite Education: Application, Admission, and Matriculation to New York City’s Specialized High Schools,” Education Finance and Policy, MIT Press, vol. 13(2), pages 256-279, Spring. [15] Deci, E.L., Koestner, R. & Ryan, R.M. (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125, 627-668. (10) [3] Fleeter, Howard B. “The Impact of Local Tax-Based Sharing on School Finance Equity in Ohio; Implementation Issues and Comparative Analysis.” Journal of Education Finance 20, no. 3 (1995): 270-301. Accessed January 16, 2021. http://www.jstor.org/stable/40703928. [9] Hastings, Justine S. & Neilson, Christopher A. & Zimmerman, Seth D., 2012. “The Effect of School Choice on Intrinsic Motivation and Academic Outcomes,” NBER Working Papers 18324, National Bureau of Economic Research, Inc. [9] E. Pfaffelhuber (1972) Learning and Information Theory, International Journal of Neuroscience, 3:2, 83-88, DOI: 10.3109/0020745720914701 [8,20] Przemyslaw Nowaczyk and Joydeep Roy, “Preferences and Outcomes: A Look at New York City’s Public High School Choice Process,” New York City Independent Budget Office (October 2016). [2] “Specialized High Schools”. NYC Department of Education. 2021. Retrieved January 04, 2021. Retrieved from https://www.schools.nyc.gov/enrollment/enroll-grade-by-grade/specialized-high-schools [5] “Test Results”. NYC Department of Education. 2021. Retrieved January 04, 2021. Retrieved from https://infohub.nyced.org/reports/academics/test-results [13] Resnick, L. B., Rothman, R., Slattery, J. B., & Vranek, J. L. (2003-2004). Benchmarking and Alignment of Standards and Testing. Educational Assessment, 9(1-2), 1–27. https://doi.org/10.1207/s15326977ea0901&2_1 [12] Smith, Calvin (Novemeber 2008). “Design-focused evaluation”. Assessment & Evaluation in Higher Education. 33 (6): 631–645. doi:10.1080/02602930701772762. S2CID 144731064. [16] Spicer, John. Making Sense of Multivariate Data Analysis: An Intuitive Approach. India: SAGE Publications, 2005, p 135. Retrieved from http://www.sagepub.com/upm-data/5081_Spicer_Chapter_5.pdf [1,14] Supovitz, J. (2010). Is high-stakes testing working? @Penn GSE A Review of Research, 7(2), 3-8. Retrieved from http://www.gse.upenn.edu/review/feature/supovitz [4] Supovitz, J.A. & Klein, V. (2003). Mapping a course for improved student learning: How innovative schools systematically use student performance data to guide improvement. Philadelphia, PA: Consortium for Policy Research in Wiliam, D., & Leahy, S. (2006, April). A theoretical foundation for formative assessment. Paper presented at the National Council on Measurement in Education, San Francisco. [6] Yiping, Lai (2018). Target Schools & Action Recommended for PASSNYC. Retrieved January 04, 2021. "https://www.kaggle.com/laiyipeng/target-schools-action-recommended-for-passnyc