references.bib

@article{sorensen_big_2019,
	title = {“Big Data” in Educational Administration: An Application for Predicting School Dropout Risk},
	volume = {55},
	issn = {0013-161X, 1552-3519},
	url = {http://journals.sagepub.com/doi/10.1177/0013161X18799439},
	doi = {10.1177/0013161X18799439},
	shorttitle = {“Big Data” in Educational Administration},
	abstract = {Purpose: In an era of unprecedented student measurement and emphasis on data-driven educational decision making, the full potential for using data to target resources to students has yet to be realized. This study explores the utility of machine-learning techniques with large-scale administrative data to identify student dropout risk. Research Methods: Using longitudinal student records data from the North Carolina Department of Public Instruction, this article assesses modern prediction techniques, with a focus on tree-based classification methods and support vector machines. These methods incorporate 74 predictors measures from Grades 3 through 8, including academic achievement, behavioral indicators, and socioeconomic and demographic characteristics. Findings: Two of the assessed classification algorithms predict high school graduation and dropping out correctly for more than 90\% of an out-of-sample student cohort. Findings reveal a shift toward lower dropout incidence in regions hit hardest by the economic recession of 2008, especially for male students. Implications for Research and Practice: Machine-learning procedures, as demonstrated in this study, offer promise for allowing administrators to reliably identify students at risk of dropping out of school so as to provide targeted, intensive programs at the lowest possible cost.},
	pages = {404--446},
	number = {3},
	journaltitle = {Educational Administration Quarterly},
	shortjournal = {Educational Administration Quarterly},
	author = {Sorensen, Lucy C.},
	urldate = {2024-10-30},
	date = {2019-08},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\QMXG7N6T\\Sorensen - 2019 - “Big Data” in Educational Administration An Application for Predicting School Dropout Risk.pdf:application/pdf},
}

@article{wandera_predicting_nodate,
	title = {Predicting school performance using a combination of traditional and non-traditional education data from South Africa},
	abstract = {The application of big data analytics in education is transforming learning, teaching and administration in schools. Current Education Data Mining ({EDM}) research focuses on teaching and personalized learning in higher institutions mostly in western countries with limited research conducted in African countries. Most research has been conducted using small datasets, simple learning analytics techniques and machine learning black box models to predict students’ performance. Black box modelling approaches use complex structures which are difficult to be easily interpreted by stakeholders. We synthesize {EDM} approaches and tree based machine learning techniques to identify important features that can predict school performance across African countries such as South Africa. We apply {LightGBM} a gradient boosting framework and interpretable tree based algorithms on combined data sources from community surveys, school master lists and examination results to perform feature importance. The challenge faced in {EDM} research is limited education data sources, we merged different existing datasets from government reports and archives. We used community survey data to determine the standards of living in secondary schools within those communities. Cell phone internet, toilets, security, usable water sources, number of teachers and students, school location, and family head were identified as control variables impacting the attainment of schools. {LightGBM}, underlies the developed prediction model. It empowered the model with high accuracy, stability and easy interpretation hence outperforming {XGBoost}, decision tree and random forest algorithms.},
	author = {Wandera, Henry and Marivate, Vukosi and Sengeh, Moinina David},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\F2JLSK2L\\Wandera et al. - Predicting school performance using a combination of traditional and non-traditional education data.pdf:application/pdf},
}

@article{rebai_graphically_2020,
	title = {A graphically based machine learning approach to predict secondary schools performance in Tunisia},
	volume = {70},
	issn = {00380121},
	url = {https://linkinghub.elsevier.com/retrieve/pii/S0038012118302908},
	doi = {10.1016/j.seps.2019.06.009},
	abstract = {The main purpose of this paper is to identify the key factors that impact schools' academic performance and to explore their relationships through a two-stage analysis based on a sample of Tunisian secondary schools. In the first stage, we use the Directional Distance Function approach ({DDF}) to deal with undesirable outputs. The {DDF} is estimated using Data Envelopment Analysis method ({DEA}). In the second stage we apply machine-learning approaches (regression trees and random forests) to identify and visualize variables that are associated with a high school performance. The data is extracted from the Program for International Student Assessment ({PISA}) 2012 survey. The first stage analysis shows that almost 22\% of Tunisian schools are efficient and that they could improve their students’ educational performance by 15.6\% while using the same level of resources. Regression trees findings indicate that the most important factors associated with higher performance are school size, competition, class size, parental pressure and proportion of girls. Only, school location appears with no impact on school efficiency. Random forests algorithm outcomes display that proportion of girls at school and school size have the most powerful impact on the predictive accuracy of our model and hence could more influence school efficiency. The findings disclose also the high non-linearity of the relationships between these key factors and school performance and reveal the importance of modeling their interactions in influencing efficiency scores.},
	pages = {100724},
	journaltitle = {Socio-Economic Planning Sciences},
	shortjournal = {Socio-Economic Planning Sciences},
	author = {Rebai, Sonia and Ben Yahia, Fatma and Essid, Hédi},
	urldate = {2024-10-30},
	date = {2020-06},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\EEBUD9TM\\Rebai et al. - 2020 - A graphically based machine learning approach to predict secondary schools performance in Tunisia.pdf:application/pdf},
}

@article{rahal_rise_2024,
	title = {The rise of machine learning in the academic social sciences},
	volume = {39},
	issn = {0951-5666, 1435-5655},
	url = {https://link.springer.com/10.1007/s00146-022-01540-w},
	doi = {10.1007/s00146-022-01540-w},
	pages = {799--801},
	number = {2},
	journaltitle = {{AI} \& {SOCIETY}},
	shortjournal = {{AI} \& Soc},
	author = {Rahal, Charles and Verhagen, Mark and Kirk, David},
	urldate = {2024-10-30},
	date = {2024-04},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\882SZEN8\\Rahal et al. - 2024 - The rise of machine learning in the academic social sciences.pdf:application/pdf},
}

@incollection{katalinic_machine_2018,
	edition = {1},
	title = {Machine Learning in Education - a Survey of Current Research Trends},
	volume = {1},
	isbn = {978-3-902734-20-4},
	url = {http://www.daaam.info/Downloads/Pdfs/proceedings/proceedings_2018/059.pdf},
	abstract = {Nowadays, Machine Learning ({ML}) is one of the most promising application areas in a field of Information Technology where its application scope is almost unlimited. The application of machine learning in an education area is currently very interesting to researchers and scientists, and it is the main focus of our study. The aim of this paper is to evaluate the possibilities of applying and using machine learning in the education area. This paper identifies and analyses suitable literature, research papers and articles in order to determine their categorization in the field of education, to determine the current trends of using machine learning in education, and to determine its current and future applications.},
	pages = {0406--0410},
	booktitle = {{DAAAM} Proceedings},
	publisher = {{DAAAM} International Vienna},
	author = {Kucak, Danijel and Juricic, Vedran and Dambic, Goran},
	editor = {Katalinic, Branko},
	urldate = {2024-10-30},
	date = {2018},
	langid = {english},
	doi = {10.2507/29th.daaam.proceedings.059},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\3JNXFIGG\\Kucak et al. - 2018 - Machine Learning in Education - a Survey of Current Research Trends.pdf:application/pdf},
}

@article{sunderman_does_2009,
	title = {Does Closing Schools Cause Educational Harm? A Review of the Research},
	author = {Sunderman, Gail L and Payne, Alexander},
	date = {2009},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\IP396YL5\\Sunderman and Payne - Does Closing Schools Cause Educational Harm A Review of the Research.pdf:application/pdf},
}

@article{syeed_it_2019,
	title = {“It just doesn’t add up”: Disrupting official arguments for urban school closures with counterframes},
	volume = {27},
	rights = {https://creativecommons.org/licenses/by-sa/4.0},
	issn = {1068-2341},
	url = {https://epaa.asu.edu/index.php/epaa/article/view/4240},
	doi = {10.14507/epaa.27.4240},
	shorttitle = {“It just doesn’t add up”},
	abstract = {Mass school closures have become commonplace in urban school districts. To explain their actions, school system leaders often rely on a dominant frame that presents closures as an inevitable, data-driven, and politically neutral phenomenon in an educational landscape defined by shrinking budgets, demographic changes, and increased school choice. In response, research has typically focused on how communities tell counternarratives that seek to interrupt official accounts of school closures. Using a critical frame analysis of qualitative data from the 2013 school closure process in Washington, {DC}, I discuss another grassroots approach to disrupting school closures: counterframes. Drawing on Critical Race Theory and social movement theory, I discuss counterframes as discursive arguments that allow communities to directly challenge official rhetoric and offer alternatives. Findings show that communities in {DC} crafted counterframes that pushed back on the notion that the closures were inevitable, questioned the data guiding the process, and attempted to expose hidden agendas and interests behind shuttering schools. The article concludes with the relevance of counterframes to broader educational mobilizations as well as their limitations.},
	pages = {110},
	journaltitle = {Education Policy Analysis Archives},
	shortjournal = {{EPAA}},
	author = {Syeed, Esa},
	urldate = {2024-10-30},
	date = {2019-09-16},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\H2FNJJYE\\Syeed - 2019 - “It just doesn’t add up” Disrupting official arguments for urban school closures with counterframes.pdf:application/pdf},
}

@article{ewing_beyond_2022,
	title = {Beyond the Headlines: Trends and Future Directions in the School Closure Literature},
	volume = {51},
	issn = {0013-189X, 1935-102X},
	url = {https://journals.sagepub.com/doi/10.3102/0013189X211050944},
	doi = {10.3102/0013189X211050944},
	shorttitle = {Beyond the Headlines},
	abstract = {With the looming impacts of {COVID}-19 on district budgets, the growth of school choice options, and population shifts across urban, suburban, and rural contexts, an increasing number of districts have closed schools and more districts are expected to follow this trend. Rich scholarship has examined school closures; however, this field of research is limited in scope and methodological approach, and overwhelmingly focuses on the mass urban school closures of the mid-2010s. This offers a timely opportunity to consider new directions in the field. In this article, we identify trends in the scholarship on school closures by examining the empirical research in this area over nearly two decades. We conclude by offering a research agenda for future scholarship on school closures.},
	pages = {58--65},
	number = {1},
	journaltitle = {Educational Researcher},
	shortjournal = {Educational Researcher},
	author = {Ewing, Eve L. and Green, Terrance L.},
	urldate = {2024-10-30},
	date = {2022-01},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\5R3WEWA4\\Ewing and Green - 2022 - Beyond the Headlines Trends and Future Directions in the School Closure Literature.pdf:application/pdf},
}

@article{pearman_examining_nodate,
	title = {Examining Racial (In)equity in School-Closure Patterns in California},
	author = {Pearman, Francis},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\76PABCTP\\Pearman - Examining Racial (In)equity in School-Closure Patterns in California.pdf:application/pdf},
}

@article{hahnel_centering_2023,
	title = {Centering Equity in the School-Closure Process in California},
	author = {Hahnel, Carrie and Marchitello, Max},
	date = {2023},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\VLK2GFTB\\Hahnel - Centering Equity in the School-Closure Process in California.pdf:application/pdf},
}

@article{brazil_neighborhood_2022,
	title = {The neighborhood ethnoracial and socioeconomic context of public elementary school closures in U.S. metropolitan areas},
	volume = {103},
	issn = {0049089X},
	url = {https://linkinghub.elsevier.com/retrieve/pii/S0049089X21001320},
	doi = {10.1016/j.ssresearch.2021.102655},
	abstract = {Public school closures are increasing in number and size in U.S. cities. When public schools close, heated debates typically ensue. A central argument within this debate asserts that schools being closed are more likely to be located in minority, socioeconomically disadvantaged neighborhoods, and thus their abandonment has the potential for widening racial and socioeconomic gaps and exacerbating spatial inequality. Using school attendance boundary data in over 260 U.S. metropolitan areas, we examine the relationship between the locations of traditional elementary public school closures between 2010 and 2016 and neighborhood ethnoracial and socioeconomic composition in 2010 and their change over time. Our overall results indicate that closures are associated with lower neighborhood percent White and percent Hispanic and higher percent Black and socioeconomic disadvantage. While increasing percent White is positively associated with closure, we found little other evidence of a relationship between closure and other changes in ethnoracial and socioeconomic composition. However, the relationship between neighborhood context and closure varies across region and urbanicity, with closures associated with patterns of gentrification in urban areas, and exhibiting differential relationships with neighborhood {SES}, race and ethnicity across region.},
	pages = {102655},
	journaltitle = {Social Science Research},
	shortjournal = {Social Science Research},
	author = {Brazil, Noli and Candipan, Jennifer},
	urldate = {2024-10-30},
	date = {2022-03},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\MAEWK3XR\\Brazil and Candipan - 2022 - The neighborhood ethnoracial and socioeconomic context of public elementary school closures in U.S..pdf:application/pdf},
}

@article{weber_predicting_2020,
	title = {Predicting School Closures in an Era of Austerity: The Case of Chicago},
	volume = {56},
	issn = {1078-0874, 1552-8332},
	url = {https://journals.sagepub.com/doi/10.1177/1078087418802359},
	doi = {10.1177/1078087418802359},
	shorttitle = {Predicting School Closures in an Era of Austerity},
	abstract = {What factors do administrators consider when (dis)investing in public facilities? We model school closure decisions in Chicago from 2003 to 2013 with multinomial logit models that estimate the decision to close or “turnaround” schools as a function of building, student, geographic, political, and neighborhood factors during two mayoral administrations. The results from our specifications validate the “official” rationale for closures and turnarounds: Low test scores are associated with closures and turnarounds under Mayor Daley, and underutilization is associated with closures under Mayor Emanuel. However, our findings also reveal some distance between technical-rational decision making and the realities of capital budgeting under austerity. The race of students and proximity to both the Central Business District and charter schools also predicted closures. This suggests multiple, potentially conflicting, interests that school districts balance to serve the needs of school-age populations and taxpayers and also the potential for burdening already vulnerable populations with the negative effects of disinvestment.},
	pages = {415--450},
	number = {2},
	journaltitle = {Urban Affairs Review},
	shortjournal = {Urban Affairs Review},
	author = {Weber, Rachel and Farmer, Stephanie and Donoghue, Mary},
	urldate = {2024-10-30},
	date = {2020-03},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\MNFC4KEK\\Weber et al. - 2020 - Predicting School Closures in an Era of Austerity The Case of Chicago.pdf:application/pdf},
}

@article{howley_consolidation_nodate,
	title = {{CONSOLIDATION} {OF} {SCHOOLS} {AND} {DISTRICTS}},
	author = {Howley, Craig and Johnson, Jerry and Petrie, Jennifer},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\JC8FPIGJ\\Howley et al. - CONSOLIDATION OF SCHOOLS AND DISTRICTS.pdf:application/pdf},
}

@article{engberg_closing_2012,
	title = {Closing schools in a shrinking district: Do student outcomes depend on which schools are closed?},
	volume = {71},
	rights = {https://www.elsevier.com/tdm/userlicense/1.0/},
	issn = {00941190},
	url = {https://linkinghub.elsevier.com/retrieve/pii/S009411901100060X},
	doi = {10.1016/j.jue.2011.10.001},
	shorttitle = {Closing schools in a shrinking district},
	abstract = {In the last decade, many cities around the country have needed to close schools due to declining enrollments and low achievement. School closings raise concerns about the possible negative impacts on student achievement, neighborhoods, families, and teaching staff. This study examines an anonymous urban district that, faced with declining enrollment, chose to make student achievement a major criterion in determining which schools would be closed. The district targeted low-performing schools in its closure plan, and sought to move their students to higher-performing schools. We estimate the impact of school closures on student test scores and attendance rates by comparing the growth of these measures among students differentially affected by the closures. We use residential assignment to school as an instrument to address non-random sorting of students into new schools. We also statistically control for the contemporaneous effects of other reforms within the district. Results show that students displaced by school closures can experience adverse effects on test scores and attendance, but these effects can be minimized when students move to schools that are higher-performing (in value-added terms). Moreover, the negative effect on attendance disappears after the first year in the new school. Meanwhile, we find no adverse effects on students in the schools that are receiving the transferring students.},
	pages = {189--203},
	number = {2},
	journaltitle = {Journal of Urban Economics},
	shortjournal = {Journal of Urban Economics},
	author = {Engberg, John and Gill, Brian and Zamarro, Gema and Zimmer, Ron},
	urldate = {2024-10-30},
	date = {2012-03},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\GQLXANC9\\Engberg et al. - 2012 - Closing schools in a shrinking district Do student outcomes depend on which schools are closed.pdf:application/pdf},
}

@article{steinberg_effects_2019,
	title = {The effects of closing urban schools on students’ academic and behavioral outcomes: Evidence from Philadelphia},
	volume = {69},
	issn = {02727757},
	url = {https://linkinghub.elsevier.com/retrieve/pii/S0272775718302693},
	doi = {10.1016/j.econedurev.2018.12.005},
	shorttitle = {The effects of closing urban schools on students’ academic and behavioral outcomes},
	abstract = {Urban districts throughout the country are increasingly closing schools in response to declining enrollment and academic underperformance. We estimate the impact of public school closings in Philadelphia on student achievement and behavioral outcomes. While school closures had no effect on the average achievement of displaced students, achievement increased among displaced students attending higher-performing schools following closure. The achievement of students attending receiving-schools, however, was negatively affected by the receipt of displaced students. School absences increased significantly for displaced students following closure. We also find that the achievement of displaced and receiving-school students declined as the fraction of displaced students attending a receiving-school increased, and displaced students missed more days of school and received more suspension days the farther they traveled to their new school following closure. These findings suggest that the academic and behavioral consequences of closing urban schools depend on the school settings displaced and receiving-school students experience in the wake of closures.},
	pages = {25--60},
	journaltitle = {Economics of Education Review},
	shortjournal = {Economics of Education Review},
	author = {Steinberg, Matthew P. and {MacDonald}, John M.},
	urldate = {2024-10-30},
	date = {2019-04},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\8BSY39L8\\Steinberg and MacDonald - 2019 - The effects of closing urban schools on students’ academic and behavioral outcomes Evidence from Ph.pdf:application/pdf},
}

@article{pearman_gentrification_2023,
	title = {Gentrification, displacement, and academic achievement: A formal mediation analysis},
	volume = {115},
	issn = {0049089X},
	url = {https://linkinghub.elsevier.com/retrieve/pii/S0049089X23000601},
	doi = {10.1016/j.ssresearch.2023.102905},
	shorttitle = {Gentrification, displacement, and academic achievement},
	abstract = {Living in a disadvantaged neighborhood has long been known to adversely affect children’s academic achievement. Comparatively less is known about what happens to children’s academic achievement when disadvantaged neighborhoods gentrify. This study uses data from a nationally representative sample of children from the Panel Study of Income Dynamics (n = 1,163) along with counterfactual methods and a value-added design to examine how gentrification and residential displacement figure into children’s academic achievement patterns. This study provides a formal mediation analysis that decomposes the total effect of gentrification on children’s academic achievement into that which operates through residential displacement versus alternative pathways. This study finds that the effects of gentrification on children’s achievement patterns were concentrated amongst low-income children and were observed most strongly when gentrifiers were White. Low-income children exposed to gentrification saw declines in their academic performance trajectories, especially in math. These adverse effects were not found to be mediated by residential displacement. A comprehensive set of sensitivity analyses indicates that results were robust to unobserved confounding, alternative model specifications, different weighting strategies, and multiple measures of gentrification and displacement.},
	pages = {102905},
	journaltitle = {Social Science Research},
	shortjournal = {Social Science Research},
	author = {Pearman, Francis A. and Steyer, Lily},
	urldate = {2024-10-30},
	date = {2023-09},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\ABD2HFRM\\Pearman and Steyer - 2023 - Gentrification, displacement, and academic achievement A formal mediation analysis.pdf:application/pdf},
}

@article{green_we_2017,
	title = {“We felt they took the heart out of the community”: Examining a community-based response to urban school closure},
	volume = {25},
	issn = {1068-2341},
	url = {https://epaa.asu.edu/index.php/epaa/article/view/2549},
	doi = {10.14507/epaa.25.2549},
	shorttitle = {“We felt they took the heart out of the community”},
	abstract = {Massive school closures are occurring in urban school districts across the United States. Research suggests that school closures are the outcome of racialized neoliberal policies and decades of disinvestment that have left many urban districts with fiscal deficits and declining student enrollments. However, some urban communities have successfully organized against school closures and reopened neighborhood schools. As such, this study examines how leaders in a communityuniversity coalition in the Midwestern United States reopened a high school that was closed by its district. This case study draws on interviews and document data, and describes the forces that promoted school closure and its impacts on the community. Concepts from social capital and social network theories are used to guide the analysis. Findings indicate these leaders leveraged networks to negotiate a community-university social contract, took strategic and socially connected actions, and formed a community-driven education task force. This study offers implications for policy, future research, and communities in similar contexts.},
	pages = {21},
	journaltitle = {Education Policy Analysis Archives},
	shortjournal = {{EPAA}},
	author = {Green, Terrance},
	urldate = {2024-10-30},
	date = {2017-03-13},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\ZEJ5TP7X\\Green - 2017 - “We felt they took the heart out of the community” Examining a community-based response to urban sc.pdf:application/pdf},
}

@article{lee_school_2020,
	title = {School Closures in Chicago: What Happened to the Teachers?},
	volume = {42},
	issn = {0162-3737, 1935-1062},
	url = {https://journals.sagepub.com/doi/10.3102/0162373720922218},
	doi = {10.3102/0162373720922218},
	shorttitle = {School Closures in Chicago},
	abstract = {In 2013, the Chicago Board of Education closed 47 elementary schools, directly affecting 13,000 students and 900 teachers. The closures created employment uncertainty for closed-school teachers, and this article investigates the labor market consequences for teachers. We employ a difference-indifferences approach that compares the exit rates of closed-school teachers with teachers in schools that only experienced threat of closure. We estimate that the closures resulted in a near doubling of teacher exit among teachers in closed schools, particularly among low-performing teachers. We also find that, among closed-school teachers, Black teachers were more likely to return than White teachers. Given the nationwide trend of school closures for budgetary or performance reasons, this article has implications for strategic retention of effective teachers.},
	pages = {331--353},
	number = {3},
	journaltitle = {Educational Evaluation and Policy Analysis},
	shortjournal = {Educational Evaluation and Policy Analysis},
	author = {Lee, Helen and Sartain, Lauren},
	urldate = {2024-10-30},
	date = {2020-09},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\GF4QEKQJ\\Lee and Sartain - 2020 - School Closures in Chicago What Happened to the Teachers.pdf:application/pdf},
}

@article{pearman_school_2022,
	title = {School Closures and the Gentrification of the Black Metropolis},
	volume = {95},
	issn = {0038-0407, 1939-8573},
	url = {https://journals.sagepub.com/doi/10.1177/00380407221095205},
	doi = {10.1177/00380407221095205},
	abstract = {Largely overlooked in the empirical literature on gentrification are the potential effects school closures have in the process. This study begins to fill this gap by integrating longitudinal data on all U.S. metropolitan neighborhoods from the Neighborhood Change Database with data on the universe of school closures from the National Center for Educational Statistics. We found that the effects of school closures on patterns of gentrification were concentrated among black neighborhoods. School closures increased the probability that the most segregated black neighborhoods experienced gentrification by 8 percentage points and increased the extent to which these neighborhoods experienced gentrification by .21 standard deviations. We found no evidence that school closures increased the likelihood or extent that white or Latinx neighborhoods experienced gentrification. Substantive conclusions were consistent across multiple measures of gentrification, alternative model specifications, and a variety of sample restrictions and were robust to a series of falsification tests. Results suggest school closures do not simply alter the educational landscape. School closures are also emblematic of a larger spatial and racial reimagining of U.S. cities that dispossesses and displaces black neighborhoods.},
	pages = {233--253},
	number = {3},
	journaltitle = {Sociology of Education},
	shortjournal = {Sociol Educ},
	author = {Pearman, Francis A. and Greene, Danielle Marie},
	urldate = {2024-10-30},
	date = {2022-07},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\VHGALM2A\\Pearman and Marie Greene - 2022 - School Closures and the Gentrification of the Black Metropolis.pdf:application/pdf},
}

@article{kearns_status_2009,
	title = {‘The status quo is not an option’: Community impacts of school closure in South Taranaki, New Zealand},
	volume = {25},
	rights = {https://www.elsevier.com/tdm/userlicense/1.0/},
	issn = {07430167},
	url = {https://linkinghub.elsevier.com/retrieve/pii/S0743016708000594},
	doi = {10.1016/j.jrurstud.2008.08.002},
	shorttitle = {‘The status quo is not an option’},
	abstract = {This paper explores the impacts of proposed school closures on families in rural communities in the South Taranaki region of New Zealand. We situate this instance of educational restructuring in a critical policy context and present an account of its regional unfolding through drawing on local media coverage. We then interpret narratives gathered during an interview-based study of the proposed changes undertaken in 2003–2004. Our analysis highlights the impact of school closure for rural settlements in terms of affect as well as effects. More generally we reflect on the place of schools in the experience of place itself, as well as their contribution to social cohesion and the broadly defined health of a community.},
	pages = {131--140},
	number = {1},
	journaltitle = {Journal of Rural Studies},
	shortjournal = {Journal of Rural Studies},
	author = {Kearns, Robin A. and Lewis, Nicolas and {McCreanor}, Tim and Witten, Karen},
	urldate = {2024-10-30},
	date = {2009-01},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\7HR3TEBG\\Kearns et al. - 2009 - ‘The status quo is not an option’ Community impacts of school closure in South Taranaki, New Zealan.pdf:application/pdf},
}

@article{tieken_rethinking_2019,
	title = {Rethinking the School Closure Research: School Closure as Spatial Injustice},
	volume = {89},
	issn = {0034-6543, 1935-1046},
	url = {https://journals.sagepub.com/doi/10.3102/0034654319877151},
	doi = {10.3102/0034654319877151},
	shorttitle = {Rethinking the School Closure Research},
	abstract = {Recent mass closings of schools have rocked cities across the United States. Though these urban closures—and widespread community protests—have made headlines, rural schools have also long experienced and opposed the closure of their schools. A large body of research examines these urban and rural closures from a variety of perspectives, including their economic motivations and policy implications. This review reexamines this literature, looking across context to show how school closure can produce spatial injustice. Advocates argue that closures further academic opportunity, efficiency, and equality. But our analysis shows that closures are unevenly distributed, disproportionately affecting places where poor communities and communities of color live, and they can bring negative effects, harming students and adults and reducing their access to an important educational and community institution. We conclude with recommendations for research and practice.},
	pages = {917--953},
	number = {6},
	journaltitle = {Review of Educational Research},
	shortjournal = {Review of Educational Research},
	author = {Tieken, Mara Casey and Auldridge-Reveles, Trevor Ray},
	urldate = {2024-10-30},
	date = {2019-12},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\P8JAK5SY\\Tieken and Auldridge-Reveles - 2019 - Rethinking the School Closure Research School Closure as Spatial Injustice.pdf:application/pdf},
}

@article{stobbe_us_2023,
	title = {{US} births fell in 2023 to the lowest count in more than 40 years {\textbar} {AP} News},
	author = {Stobbe, Mike},
	date = {2023},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\TIUGRNIQ\\2023 - US births fell in 2023 to the lowest count in more than 40 years  AP News.pdf:application/pdf},
}

@article{peetz_as_2024,
	title = {As Enrollment Declines, Districts Consider Closing Schools},
	issn = {0277-4232},
	url = {https://www.edweek.org/leadership/as-enrollment-declines-districts-consider-closing-schools/2024/01},
	abstract = {Districts across the country are confronting a long-term decline in enrollments that has accelerated in recent years.},
	journaltitle = {Education Week},
	author = {Peetz, Caitlynn},
	urldate = {2024-10-30},
	date = {2024-01-08},
	langid = {english},
	keywords = {District Strategies, Enrollment, School Closures},
	file = {Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\IUPGY2VN\\01.html:text/html},
}

@online{bingamon_austin_2024,
	title = {Austin {ISD} Budget Shortfall Could Put School Closures on the Table},
	url = {https://www.austinchronicle.com/news/2024-06-07/austin-isd-budget-shortfall-could-put-school-closures-on-the-table/},
	abstract = {Desperate times may lead to desperate measures},
	titleaddon = {Austin Chronicle},
	author = {Bingamon, Brant},
	urldate = {2024-10-30},
	date = {2024-06-07},
	langid = {american},
	file = {Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\UQHFITH7\\austin-isd-budget-shortfall-could-put-school-closures-on-the-table.html:text/html},
}

@online{alejo_denver_2024,
	title = {Denver Public Schools seeks community input ahead of possible school closures - {CBS} Colorado},
	url = {https://www.cbsnews.com/colorado/news/denver-public-schools-seeks-community-input-ahead-of-possible-school-closures/},
	abstract = {{DPS} is expected to see six thousand fewer students by 2028, a drop of more than eight percent.},
	author = {Alejo, Anna},
	urldate = {2024-10-30},
	date = {2024-06-07},
	langid = {american},
	file = {Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\XRWLLMRD\\denver-public-schools-seeks-community-input-ahead-of-possible-school-closures.html:text/html},
}

@report{de_la_torre_when_2009,
	title = {When Schools Close  Effects on Displaced Students in Chicago Public Schools},
	institution = {Consortium on Chicago School Research at the University of Chicago Urban Education Institute},
	author = {de la Torre, Marisa and Gwynne, Julia},
	date = {2009-10},
	langid = {american},
}

@thesis{gordon_pittsburgh_2014,
	title = {Pittsburgh School Closures:  The Impact on Physical and Social  Neighborhood Dynamics},
	abstract = {School closures can have enormous effects on students and families living in a neighborhood. School districts in many urban areas in the United States have recently been faced with the challenge of how to deal with a shrinking school-aged population, budget crises, aging facilities and poor academic performance. This has led some districts to close large numbers of school buildings. A prime example occurred in Pittsburgh, Pennsylvania in 2006. This research will examine changes in population count, racial composition, and housing characteristics in three affected neighborhoods of Pittsburgh following the decision by the Pittsburgh Board of Education to close 22 schools that year. The research will examine how large vacant facilities affect the neighborhood’s physical and social dynamics. Data analysis, interviews and site analysis help to answer the research questions. The findings suggest that school vacancies have a negative effect on neighborhoods to varying degrees, and the neighborhood’s ability to cope with this loss is determined by a number of factors such as desirability of the area and community resources. Recommendations can help other cities to better plan for closures in the future to ensure the most equitable outcomes.},
	pagetotal = {48},
	institution = {Graduate School of Architecture, Planning and Preservation, Columbia University},
	type = {Master of Science in Urban Planning},
	author = {Gordon, Emily},
	date = {2014-05},
	langid = {american},
}

@online{tucker_oakland_2024,
	title = {Oakland Unified faces school closures again — three years after they sparked bitter fights},
	url = {https://www.sfchronicle.com/bayarea/article/oakland-unified-school-closures-again-19868666.php},
	abstract = {Oakland Unified faces school closures again — three years after they sparked bitter fights},
	titleaddon = {San Francisco Chronicle},
	author = {Tucker, Jill},
	urldate = {2024-10-30},
	date = {2024-10-30},
	langid = {english},
}

@online{noauthor_common_2024,
	title = {Common Core of Data ({CCD})},
	url = {https://nces.ed.gov/ccd/aboutccd.asp},
	abstract = {The Common Core of Data ({CCD}) is the Department of Education},
	urldate = {2024-10-31},
	date = {2024},
	note = {Publisher: National Center for Education Statistics},
}

@book{ewing_ghosts_2018,
	location = {Chicago},
	title = {Ghosts in the schoolyard: racism and school closings on Chicago's South side},
	isbn = {978-0-226-52602-7},
	shorttitle = {Ghosts in the schoolyard},
	pagetotal = {222},
	publisher = {The University of Chicago Press},
	author = {Ewing, Eve L.},
	date = {2018},
	keywords = {21st century, African Americans, Bronzeville (Chicago, Ill.), Education, History, Illinois Chicago, Low-performing schools, Public school closings, Racism in education, Walter H. Dyett High School (Chicago, Ill.)},
}

@book{oneil_weapons_2016,
	location = {New York},
	edition = {First edition},
	title = {Weapons of math destruction: how big data increases inequality and threatens democracy},
	isbn = {978-0-553-41881-1 978-0-553-41883-5},
	shorttitle = {Weapons of math destruction},
	pagetotal = {259},
	publisher = {Crown},
	author = {O'Neil, Cathy},
	date = {2016},
	keywords = {21st century, Big data, Democracy, Mathematical models Moral and ethical aspects, Political aspects, Social aspects, Social conditions, Social indicators, United States},
}

@book{buolamwini_unmasking_2023,
	location = {New York},
	edition = {First edition},
	title = {Unmasking {AI}: a story of hope and justice in a world of machines},
	isbn = {978-0-593-24185-1},
	shorttitle = {Unmasking {AI}},
	abstract = {"Dr. Joy Buolamwini is the self-described "Poet of Code" who has had a lifelong passion for computer science, engineering, and art-disciplines that, she felt, pushed the boundaries of reality. After tinkering with robotics as a high school student in Tennessee, to developing mobile apps in Zambia as a Fulbright fellow, Buolamwini eventually found herself at {MIT}. As a graduate student at the "Future Factory," Buolamwini's groundbreaking research revealed that {AI} systems-from leading tech companies-were consistently failing on non-male, non-white bodies. In Unmasking {AI}, Buolamwini goes beyond the news headlines about racism, colorism, and sexism in Big Tech to tell the remarkable story of how she uncovered what she calls "the coded gaze"-evidence of racial and gender bias in tech-and galvanized the movement to prevent {AI} harms by founding the Algorithmic Justice League. Applying an intersectional lens to both tech industry and research sector, Buolamwini shows how race, gender, and ability bias can overlap and render broad swaths of humanity vulnerable in our {AI}-dependent world. Computers, she reminds us, are reflections of both the aspirations and the limitations of the people who create them"--},
	pagetotal = {1},
	publisher = {Random House},
	author = {Buolamwini, Joy},
	date = {2023},
	keywords = {Social aspects, Artificial intelligence, Discrimination in science, Moral and ethical aspects, Philosophy, Sex discrimination in science},
}

@article{domingue_intermodel_2024,
	title = {The {InterModel} Vigorish as a Lens for Understanding (and Quantifying) the Value of Item Response Models for Dichotomously Coded Items},
	volume = {89},
	issn = {0033-3123, 1860-0980},
	url = {https://link.springer.com/10.1007/s11336-024-09977-2},
	doi = {10.1007/s11336-024-09977-2},
	pages = {1034--1054},
	number = {3},
	journaltitle = {Psychometrika},
	shortjournal = {Psychometrika},
	author = {Domingue, Benjamin W. and Kanopka, Klint and Kapoor, Radhika and Pohl, Steffi and Chalmers, R. Philip and Rahal, Charles and Rhemtulla, Mijke},
	urldate = {2024-10-31},
	date = {2024-09},
	langid = {english},
}

@article{van_der_laan_super_2007,
	title = {Super Learner},
	volume = {6},
	issn = {1544-6115, 2194-6302},
	url = {https://www.degruyter.com/document/doi/10.2202/1544-6115.1309/html},
	doi = {10.2202/1544-6115.1309},
	abstract = {When trying to learn a model for the prediction of an outcome given a set of covariates, a statistician has many estimation procedures in their toolbox. A few examples of these candidate learners are: least squares, least angle regression, random forests, and spline regression. Previous articles (van der Laan and Dudoit (2003); van der Laan et al. (2006); Sinisi et al. (2007)) theoretically validated the use of cross validation to select an optimal learner among many candidate learners. Motivated by this use of cross validation, we propose a new prediction method for creating a weighted combination of many candidate learners to build the super learner. This article proposes a fast algorithm for constructing a super learner in prediction which uses V-fold cross-validation to select weights to combine an initial set of candidate learners. In addition, this paper contains a practical demonstration of the adaptivity of this so called super learner to various true data generating distributions. This approach for construction of a super learner generalizes to any parameter which can be defined as a minimizer of a loss function.},
	number = {1},
	journaltitle = {Statistical Applications in Genetics and Molecular Biology},
	author = {Van Der Laan, Mark J. and Polley, Eric C and Hubbard, Alan E.},
	urldate = {2024-10-31},
	date = {2007-01-16},
}

@article{polley_super_2010,
	title = {Super Learner In Prediction},
	url = {https://biostats.bepress.com/ucbbiostat/paper266},
	journaltitle = {U.C. Berkeley Division of Biostatistics Working Paper Series},
	author = {Polley, Eric and van deer Laan, Mark},
	date = {2010-05-03},
	file = {"Super Learner In Prediction" by Eric C. Polley and Mark J. van der Laan:C\:\\Users\\mlchr\\Zotero\\storage\\DMUJJ8XQ\\paper266.html:text/html},
}

@online{noauthor_about_2024,
	title = {About {EDGE}},
	url = {https://nces.ed.gov/programs/edge/About#a},
	urldate = {2024-11-02},
	date = {2024},
}

@misc{baker_using_2022,
	title = {Using Demographic Data as Predictor Variables: a Questionable Choice},
	rights = {http://opensource.org/licenses/{MIT}},
	url = {https://osf.io/y4wvj},
	doi = {10.35542/osf.io/y4wvj},
	shorttitle = {Using Demographic Data as Predictor Variables},
	abstract = {Predictive analytics methods in education are seeing widespread use and are producing increasingly accurate predictions of students’ outcomes. With the increased use of predictive analytics comes increasing concern about fairness for specific subgroups of the population. One approach that has been proposed to increase fairness is using demographic variables directly in models, as predictors. In this paper we explore issues of fairness in the use of demographic variables as predictors of long-term student outcomes, studying the arguments for and against this practice in the contexts where this literature has been published. We analyze arguments for the inclusion of demographic variables, specifically claims that this approach improves model performance and charges that excluding such variables amounts to a form of ‘color-blind’ racism. We also consider arguments against including demographic variables as predictors, including reduced actionability of predictions, risk of reinforcing bias, and limits of categorization. We then discuss how contextual factors of predictive models should influence case-specific decisions for the inclusion or exclusion of demographic variables and discuss the role of proxy variables. We conclude that, on balance, there are greater benefits to fairness if demographic variables are used to validate fairness rather than as predictors within models.},
	publisher = {{EdArXiv}},
	author = {Baker, Ryan Shaun and Esbenshade, Lief and Vitale, Jon and Karumbaiah, Shamya},
	urldate = {2024-11-02},
	date = {2022-12-19},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\4CD5CU8Q\\Baker et al. - 2022 - Using Demographic Data as Predictor Variables a Questionable Choice.pdf:application/pdf},
}

@article{pearman_enrollment_2024,
	title = {Enrollment Down. Achievement Lackluster. Should This School Close?},
	issn = {0277-4232},
	url = {https://www.edweek.org/leadership/opinion-enrollment-down-achievement-lackluster-should-this-school-close/2024/06},
	abstract = {An equity researcher describes how coming district-reorganization decisions can help preserve Black communities in central cities.},
	journaltitle = {Education Week},
	author = {Pearman, Francis A.},
	urldate = {2024-11-02},
	date = {2024-06-26},
	langid = {english},
	keywords = {District Strategies, School Closures, Equity, Low Performing Schools, Race, Redistricting \& Consolidation},
	file = {Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\B77LI66D\\06.html:text/html},
}

@inproceedings{faisal_covid-19_2021,
	title = {Covid-19 and its impact on school closures: a predictive analysis using machine learning algorithms},
	url = {https://ieeexplore.ieee.org/abstract/document/9642617},
	doi = {10.1109/ICSCT53883.2021.9642617},
	shorttitle = {Covid-19 and its impact on school closures},
	abstract = {This research presents an extensive point of reference for investigating the operation of several machine learning ({ML}) algorithms in postulating the multiclass classification problem regarding the forthcoming effects of Covid-19 on school closures. With the prompt closure of schools across the world in response to this pandemic, school-going children and teenagers are ruptured both mentally and physically. Hence, {ML} has come across to be a reliable component to forecast the scenario effectively. A dataset from {UNESCO} is trained and tested by ten supervised {ML} algorithms. A comprehensive analysis among the predictive {ML} models was executed which bought satisfactory results with regard to accuracy, precision, sensitivity, F1 score, {ROC}-{AUC} by hyper parameter optimization. In this regard, grid search cross validation ({GridSearchCV}) was utilized in order to obtain the optimal parameters. However, the performance of Artificial Neural Network ({ANN}) was also investigated and compared with the supervised {ML} models where {ANN} displayed maximum accuracy of 80.37\%. After rigorous comparative analysis, Decision Tree ({DT}) portrayed the highest accuracy of 90.75\%. Hence, it is evident that machine learning algorithm holds strong promise in forecasting the upcoming scenario of school closures due to Covid-19 and can contribute significantly in decision making for the welfare of the education system.},
	eventtitle = {2021 International Conference on Science \& Contemporary Technologies ({ICSCT})},
	pages = {1--6},
	booktitle = {2021 International Conference on Science \& Contemporary Technologies ({ICSCT})},
	author = {Faisal, Fahim and Nishat, Mirza Muntasir and Mahbub, Md. Ashif and Shawon, Md. Minhajul Islam and Alvi, Md. Mahbub-Ul-Huq},
	urldate = {2024-11-02},
	date = {2021-08},
	keywords = {Education, Accuracy, Artificial neural networks, Classification algorithms, Covid-19, {COVID}-19, Machine learning algorithms, {ML} Algorithms, Multiclass Classification, Predictive models, School Closure, Sensitivity},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\BWCFBIG6\\Faisal et al. - 2021 - Covid-19 and its impact on school closures a predictive analysis using machine learning algorithms.pdf:application/pdf;IEEE Xplore Abstract Record:C\:\\Users\\mlchr\\Zotero\\storage\\TP8WYR83\\9642617.html:text/html},
}

@article{deangelis_education_2019,
	title = {The Education Marketplace: The Predictors of School Growth and Closures in Milwaukee},
	volume = {13},
	issn = {1558-2159},
	url = {https://doi.org/10.1080/15582159.2019.1595949},
	doi = {10.1080/15582159.2019.1595949},
	shorttitle = {The Education Marketplace},
	abstract = {Few evaluations have focused on the supply and demand within the education marketplace in a school choice environment. Because traditional public schools are not subject to the same level of competitive pressures as private schools, we expect that measures of school quality—enrollment, academic achievement, and safety—will be more likely to predict closures for private schools and public charter schools than Milwaukee Public Schools ({MPS}). We employ survival analysis using data from private, traditional public, and public charter schools in Milwaukee from 2005 to 2016. Data on enrollment trends, demographics, and academic performance from the Wisconsin Department of Public Instruction were combined with data from other sources on school safety and closure. Results from our models suggest: (a) enrollment losses drive school closure in all three sectors, (b) low academic achievement only predicts closure for private schools, (c) families choose schools based on academics in all three sectors, and (d) academics and school safety are positively correlated.},
	pages = {355--379},
	number = {3},
	journaltitle = {Journal of School Choice},
	author = {{DeAngelis}, Corey A. and Flanders, Will},
	urldate = {2024-11-04},
	date = {2019-07-03},
	note = {Publisher: Routledge
\_eprint: https://doi.org/10.1080/15582159.2019.1595949},
	keywords = {Private school, school choice, schooling supply, survival analysis},
}

@article{gilblom_charter_2021,
	title = {Charter School Closure in Ohio’s Largest Urban Districts: The Effects of Management Organizations, Enrollment Characteristics and Community Demographics on Closure Risk},
	volume = {10},
	rights = {Copyright (c) 2021 Elizabeth A. Gilblom,Hilla I. Sang},
	issn = {1927-5250},
	url = {https://ccsenet.org/journal/index.php/jel/article/view/0/45050},
	doi = {10.5539/jel.v10n3p1},
	shorttitle = {Charter School Closure in Ohio’s Largest Urban Districts},
	abstract = {This study builds on previous research investigating management organizations ({MOs}), charter school locations, and closure by examining the effects of {MO} type ({EMO}, {CMO} and freestanding schools), racial enrollment, student achievement, and the community characteristics surrounding each charter school in Ohio’s eight largest counties with the largest urban school districts on the likelihood of closure between 2009 and 2018. We conducted a discrete-time survival analysis using life tables and binary logistic regression. Findings indicated that freestanding charter schools experience higher risks of closure than {EMO} and {CMO} managed charter schools in those counties. Although they are more likely to close, freestanding schools have higher student achievement in math and reading. Higher math proficiency reduces the likelihood of closure by 2.8\%. However, community and enrollment characteristics are not statistically significant predictors of closure.},
	pages = {p1},
	number = {3},
	journaltitle = {Journal of Education and Learning},
	author = {Gilblom, Elizabeth and Sang, Hilla},
	urldate = {2024-11-04},
	date = {2021-04-06},
	langid = {english},
	note = {Number: 3},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\M8M9APJR\\Gilblom and Sang - 2021 - Charter School Closure in Ohio’s Largest Urban Districts The Effects of Management Organizations, E.pdf:application/pdf},
}

@article{hahnel_declining_2023,
	title = {Declining Enrollment, School Closures, and Equity Considerations},
	journaltitle = {Policy Analysis for California Education},
	author = {Hahnel, Carrie and Pearman, Francis A.},
	date = {2023-09},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\L8JTG738\\Hahnel - Declining Enrollment, School Closures, and Equity Considerations.pdf:application/pdf},
}

@article{oppong_predicting_2023,
	title = {Predicting Students’ Performance Using Machine Learning Algorithms: A Review},
	volume = {16},
	issn = {2581-8260},
	url = {https://journalajrcos.com/index.php/AJRCOS/article/view/351},
	doi = {10.9734/ajrcos/2023/v16i3351},
	shorttitle = {Predicting Students’ Performance Using Machine Learning Algorithms},
	pages = {128--148},
	number = {3},
	journaltitle = {Asian Journal of Research in Computer Science},
	author = {Oppong, Stephen Opoku},
	urldate = {2024-11-04},
	date = {2023-07-20},
	langid = {english},
	keywords = {data generated, educational data mining, machine learning, Predicting},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\Q2VPZMGW\\Oppong - 2023 - Predicting Students’ Performance Using Machine Learning Algorithms A Review.pdf:application/pdf},
}

@article{brummet_effect_2014,
	title = {The effect of school closings on student achievement},
	volume = {119},
	issn = {0047-2727},
	url = {https://www.sciencedirect.com/science/article/pii/S0047272714001509},
	doi = {10.1016/j.jpubeco.2014.06.010},
	abstract = {Many school districts across the country are shutting schools, but school closing policies remain a very controversial issue. The current study investigates the effects of school closing policies on student achievement by examining over 200 school closings in Michigan. Relative to the previous literature, the analysis uses a broader set of school closings to thoroughly investigate heterogeneity in treatment effects based on the performance level of the closed school. The results indicate that, on average, school closings in Michigan did no persistent harm to the achievement of displaced students. Moreover, students displaced from relatively low-performing schools experience achievement gains. The displacement of students and teachers creates modest negative spillover effects on the receiving schools, however. Hence, the closing of low-performing schools may generate some achievement gains for displaced students, but not without imposing spillover effects on a large number of students in receiving schools.},
	pages = {108--124},
	journaltitle = {Journal of Public Economics},
	shortjournal = {Journal of Public Economics},
	author = {Brummet, Quentin},
	urldate = {2024-11-05},
	date = {2014-11-01},
	keywords = {School closings, School quality, Student mobility},
	file = {ScienceDirect Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\RVPMHPPU\\Brummet - 2014 - The effect of school closings on student achievement.pdf:application/pdf},
}

@online{billger_demographics_2010,
	title = {Demographics, Fiscal Health, and School Quality: Shedding Light on School Closure Decisions},
	url = {https://www.iza.org/publications/dp/4739/demographics-fiscal-health-and-school-quality-shedding-light-on-school-closure-decisions},
	shorttitle = {Demographics, Fiscal Health, and School Quality},
	abstract = {In our current challenging budgetary environment, school closures remain a potentially attractive choice. With a large panel of Illinois schools from...},
	author = {Billger, Sherrilyn M.},
	urldate = {2024-11-05},
	date = {2010},
	langid = {english},
	file = {Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\LM8M3BAF\\demographics-fiscal-health-and-school-quality-shedding-light-on-school-closure-decisions.html:text/html},
}

@article{pearman_school_2017,
	title = {School Choice, Gentrification, and the Variable Significance of Racial Stratification in Urban Neighborhoods},
	volume = {90},
	issn = {0038-0407},
	url = {https://doi.org/10.1177/0038040717710494},
	doi = {10.1177/0038040717710494},
	abstract = {Racial and socioeconomic stratification have long governed patterns of residential sorting in the American metropolis. However, recent expansions of school choice policies that allow parents to select schools outside their neighborhood raise questions as to whether this weakening of the neighborhood–school connection might influence the residential decisions of higher-socioeconomic-status white households looking to relocate to central city neighborhoods. This study examines whether and the extent to which expanded school choice facilitates the gentrification of disinvested, racially segregated urban communities. Drawing data from the Decennial Census, the American Community Survey, the National Center for Educational Statistics, and the Schools and Staffing Survey, this study finds evidence that college-educated white households are far more likely to gentrify communities of color when school choice options expand. In particular, the expansion of school choice increases the likelihood of gentrification by up to 22 percentage points in the most racially isolated neighborhoods of color—more than twice the baseline likelihood for such communities.},
	pages = {213--235},
	number = {3},
	journaltitle = {Sociology of Education},
	shortjournal = {Sociol Educ},
	author = {Pearman, Francis A. and Swain, Walker A.},
	urldate = {2024-11-05},
	date = {2017-07-01},
	langid = {english},
	note = {Publisher: {SAGE} Publications Inc},
	file = {SAGE PDF Full Text:C\:\\Users\\mlchr\\Zotero\\storage\\43URBRUN\\Pearman and Swain - 2017 - School Choice, Gentrification, and the Variable Significance of Racial Stratification in Urban Neigh.pdf:application/pdf},
}

@article{hwang_what_2016,
	title = {What Have We Learned About the Causes of Recent Gentrification?},
	volume = {18},
	issn = {1936-007X},
	url = {https://www.jstor.org/stable/26328271},
	abstract = {Since 2000, increased gentrification in an expanding section of cities and neighborhoods has renewed interest from policymakers, researchers, and the public in the causes of gentrification. The identification of causal factors can help inform analyses of welfare, policy responses, and forecasts of future neighborhood change. We highlight some features of recent gentrification that popular understandings often do not emphasize, and we review progress on identifying some causal factors. A complete account of the relative contribution of many factors, however, is still elusive. We suggest questions and opportunities for future research.},
	pages = {9--26},
	number = {3},
	journaltitle = {Cityscape},
	author = {Hwang, Jackelyn and Lin, Jeffrey},
	urldate = {2024-11-05},
	date = {2016},
	note = {Publisher: {US} Department of Housing and Urban Development},
	file = {JSTOR Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\F2NJLYWB\\Hwang and Lin - 2016 - What Have We Learned About the Causes of Recent Gentrification.pdf:application/pdf},
}

@article{kirshner_tracing_2010,
	title = {Tracing Transitions: The Effect of High School Closure on Displaced Students},
	volume = {32},
	issn = {0162-3737},
	url = {https://doi.org/10.3102/0162373710376823},
	doi = {10.3102/0162373710376823},
	shorttitle = {Tracing Transitions},
	abstract = {Although closure is an increasingly common response to the problems of chronically underperforming urban schools, few studies have examined the effect of closure on displaced students. The authors used multiple methods to study the academic performance and experiences of Latino and African American high school students in the year following the closure of their school. Quantitative analyses show declines in the transition cohort’s academic performance after transferring to new schools. Qualitative findings help explain this pattern by describing students’ interpretations of the closure and their experiences transitioning to new schools. Overall, the case study suggests that closure added stressors to students who were already contending with challenges associated with urban poverty.},
	pages = {407--429},
	number = {3},
	journaltitle = {Educational Evaluation and Policy Analysis},
	author = {Kirshner, Ben and Gaertner, Matthew and Pozzoboni, Kristen},
	urldate = {2024-11-08},
	date = {2010-09-01},
	langid = {english},
	note = {Publisher: American Educational Research Association},
	file = {SAGE PDF Full Text:C\:\\Users\\mlchr\\Zotero\\storage\\9NWDBYJ4\\Kirshner et al. - 2010 - Tracing Transitions The Effect of High School Closure on Displaced Students.pdf:application/pdf},
}

@article{epple_superintendents_2018,
	title = {The superintendent's dilemma: Managing school district capacity as parents vote with their feet: Superintendent's dilemma},
	volume = {9},
	rights = {http://doi.wiley.com/10.1002/tdm\_license\_1.1},
	issn = {17597323},
	url = {http://doi.wiley.com/10.3982/QE592},
	doi = {10.3982/QE592},
	shorttitle = {The superintendent's dilemma},
	pages = {483--520},
	number = {1},
	journaltitle = {Quantitative Economics},
	shortjournal = {Quantitative Economics},
	author = {Epple, Dennis and Jha, Akshaya and Sieg, Holger},
	urldate = {2024-11-14},
	date = {2018-03},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\DBREB7SL\\Epple et al. - 2018 - The superintendent's dilemma Managing school district capacity as parents vote with their feet Sup.pdf:application/pdf},
}

@article{matz_using_2023,
	title = {Using machine learning to predict student retention from socio-demographic characteristics and app-based engagement metrics},
	volume = {13},
	url = {https://pmc.ncbi.nlm.nih.gov/articles/PMC10082180/},
	doi = {10.1038/s41598-023-32484-w},
	abstract = {Student attrition poses a major challenge to academic institutions, funding bodies and students. With the rise of Big Data and predictive analytics, a growing body of work in higher education research has demonstrated the feasibility of predicting ...},
	pages = {5705},
	journaltitle = {Scientific Reports},
	author = {Matz, Sandra C. and Bukow, Christina S. and Peters, Heinrich and Deacons, Christine and Dinu, Alice and Stachl, Clemens},
	urldate = {2024-11-14},
	date = {2023-04-07},
	langid = {english},
	pmid = {37029155},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\DNPVBAX8\\Matz et al. - 2023 - Using machine learning to predict student retention from socio-demographic characteristics and app-b.pdf:application/pdf},
}

@article{kemper_predicting_2020,
	title = {Predicting student dropout: A machine learning approach},
	volume = {10},
	issn = {2156-8235},
	url = {https://doi.org/10.1080/21568235.2020.1718520},
	doi = {10.1080/21568235.2020.1718520},
	shorttitle = {Predicting student dropout},
	abstract = {We perform two approaches of machine learning, logistic regressions and decision trees, to predict student dropout at the Karlsruhe Institute of Technology ({KIT}). The models are computed on the basis of examination data, i.e. data available at all universities without the need of specific collection. Therefore, we propose a methodical approach that may be put in practice with relative ease at other institutions. We find decision trees to produce slightly better results than logistic regressions. However, both methods yield high prediction accuracies of up to 95\% after three semesters. A classification with more than 83\% accuracy is already possible after the first semester.},
	pages = {28--47},
	number = {1},
	journaltitle = {European Journal of Higher Education},
	author = {Kemper, Lorenz and Vorhoff, Gerrit and Wigger, Berthold U.},
	urldate = {2024-11-14},
	date = {2020-01-02},
	note = {Publisher: {SRHE} Website
\_eprint: https://doi.org/10.1080/21568235.2020.1718520},
	keywords = {machine learning, Educational data mining, retention management, student attrition, student dropout prediction},
}

@article{oppong_predicting_2023-1,
	title = {Predicting Students’ Performance Using Machine Learning Algorithms: A Review},
	volume = {16},
	issn = {2581-8260},
	url = {https://journalajrcos.com/index.php/AJRCOS/article/view/351},
	doi = {10.9734/ajrcos/2023/v16i3351},
	shorttitle = {Predicting Students’ Performance Using Machine Learning Algorithms},
	pages = {128--148},
	number = {3},
	journaltitle = {Asian Journal of Research in Computer Science},
	author = {Oppong, Stephen Opoku},
	urldate = {2024-11-14},
	date = {2023-07-20},
	langid = {english},
	keywords = {data generated, educational data mining, machine learning, Predicting},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\HX3BECL7\\Oppong - 2023 - Predicting Students’ Performance Using Machine Learning Algorithms A Review.pdf:application/pdf},
}

@article{shanker_effect_1996,
	title = {Effect of data standardization on neural network training},
	volume = {24},
	issn = {0305-0483},
	url = {https://www.sciencedirect.com/science/article/pii/0305048396000102},
	doi = {10.1016/0305-0483(96)00010-2},
	abstract = {Data transformation is a popular option in training neural networks. This study evaluates the effectiveness of two well-known transformation methods: linear transformation and statistical standardization. These two are referred to as data standardization. A carefully designed experiment is used in which data from two-group classification problems were trained by feedforward networks. Different kinds of classification problems, from relatively simple to hard, were generated. Other experimental factors include network architecture, sample size, and sample proportion of group 1 members. Three performance measurements for the effect of data standardization are employed. The results suggest that networks trained on standardized data yield better results in general, but the advantage diminishes as network and sample size become large. In other words, neural networks exhibit a self-scaling capability. In addition, impact of data standardization on the performance of training algorithm in terms of computation time and number of iterations is evaluated. The results indicate that, overall, data standardization slows down training. Finally, these results are illustrated with a data set obtained from the American Telephone and Telegraph Company.},
	pages = {385--397},
	number = {4},
	journaltitle = {Omega},
	shortjournal = {Omega},
	author = {Shanker, M. and Hu, M. Y. and Hung, M. S.},
	urldate = {2024-11-15},
	date = {1996-08-01},
	keywords = {modelling, neural networks},
	file = {ScienceDirect Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\XJM5NJ5R\\Shanker et al. - 1996 - Effect of data standardization on neural network training.pdf:application/pdf;ScienceDirect Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\ERJPLHER\\0305048396000102.html:text/html},
}

@inproceedings{ozsahin_impact_2022,
	title = {Impact of feature scaling on machine learning models for the diagnosis of diabetes},
	url = {https://ieeexplore.ieee.org/document/9898687/?arnumber=9898687},
	doi = {10.1109/AIE57029.2022.00024},
	abstract = {Due to its high prevalence and incidence, diabetes is considered significant public health. Since diabetes has no known cure, early diagnosis plays a vital role in effectively managing the disease. Feature scaling is a vital step in pre-processing data before building a model using machine learning. The datasets used for model training in machine learning often contain unpredictable values that may have varying scales. This can result in inequalities in comparing these values. Feature scaling techniques can address these challenges by adjusting the values and promoting easy and fair comparisons among values. This study aims to evaluate the impact of normalization, standardization, and no feature scaling on the performance of five machine learning models in diagnosing diabetes. The machine learning algorithms implemented for this study include random forest, naive Bayes, k-nearest neighbor ({KNN}), logistic regression, and support vector machine ({SVM}). These algorithms support supervised learning. Furthermore, several open-source frameworks and libraries were implemented. They include; Jupyter notebook, {SkLearn}, Pandas, {NumPy}, Matplotlib, and seaborn. The result obtained from the study indicates that the random forest model performed significantly well without implementing any feature scaling technique. This contrasts with the {KNN} and {SVM} model, which performed better when the normalization technique was implemented. Also, the naive Bayes model shows no changes when either standardization, normalization, or no feature scaling was implemented. This study concludes that not all model requires feature scaling techniques to be applied to the dataset to achieve optimal performance. Furthermore, distance-based and gradient descent algorithms previously thought to be sensitive to feature scaling may not necessarily be true, as indicated by the outcome of this study. Finally, feature scaling techniques significantly impact some models while others do not.},
	eventtitle = {2022 International Conference on Artificial Intelligence in Everything ({AIE})},
	pages = {87--94},
	booktitle = {2022 International Conference on Artificial Intelligence in Everything ({AIE})},
	author = {Ozsahin, Dilber Uzun and Taiwo Mustapha, Mubarak and Mubarak, Auwalu Saleh and Said Ameen, Zubaida and Uzun, Berna},
	urldate = {2024-11-15},
	date = {2022-08},
	keywords = {Machine learning algorithms, machine learning, diabetes, Diabetes, diagnosis, feature scaling, Libraries, normalization, Random forests, scaling, standardization, Supervised learning, Support vector machines, Training},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\9K3DSM6X\\Ozsahin et al. - 2022 - Impact of feature scaling on machine learning models for the diagnosis of diabetes.pdf:application/pdf;IEEE Xplore Abstract Record:C\:\\Users\\mlchr\\Zotero\\storage\\23GCQXI5\\9898687.html:text/html},
}

@article{lieberman_how_2023,
	title = {How Much Money Do Public Schools Get? The Latest Numbers},
	issn = {0277-4232},
	url = {https://www.edweek.org/leadership/how-much-money-do-public-schools-get-the-latest-numbers/2023/06},
	shorttitle = {How Much Money Do Public Schools Get?},
	abstract = {America collectively spent more than \$800 billion on K-12 education in 2021, according to newly available federal data.},
	journaltitle = {Education Week},
	author = {Lieberman, Mark},
	urldate = {2025-01-23},
	date = {2023-06-08},
	langid = {english},
	keywords = {Federal Budget, School Funding, State Budgets},
	file = {Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\Z4WMZD7S\\06.html:text/html},
}

@inproceedings{chen_xgboost_2016,
	title = {{XGBoost}: A Scalable Tree Boosting System},
	url = {http://arxiv.org/abs/1603.02754},
	doi = {10.1145/2939672.2939785},
	shorttitle = {{XGBoost}},
	abstract = {Tree boosting is a highly effective and widely used machine learning method. In this paper, we describe a scalable end-to-end tree boosting system called {XGBoost}, which is used widely by data scientists to achieve state-of-the-art results on many machine learning challenges. We propose a novel sparsity-aware algorithm for sparse data and weighted quantile sketch for approximate tree learning. More importantly, we provide insights on cache access patterns, data compression and sharding to build a scalable tree boosting system. By combining these insights, {XGBoost} scales beyond billions of examples using far fewer resources than existing systems.},
	pages = {785--794},
	booktitle = {Proceedings of the 22nd {ACM} {SIGKDD} International Conference on Knowledge Discovery and Data Mining},
	author = {Chen, Tianqi and Guestrin, Carlos},
	urldate = {2025-02-16},
	date = {2016-08-13},
	eprinttype = {arxiv},
	eprint = {1603.02754 [cs]},
	keywords = {Computer Science - Machine Learning},
	file = {Preprint PDF:C\:\\Users\\mlchr\\Zotero\\storage\\P8EMPGL2\\Chen and Guestrin - 2016 - XGBoost A Scalable Tree Boosting System.pdf:application/pdf;Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\A4ZRSWIW\\1603.html:text/html},
}

@article{white_segregation_1986,
	title = {Segregation and Diversity Measures in Population Distribution},
	volume = {52},
	issn = {0032-4701},
	url = {https://www.jstor.org/stable/3644339},
	doi = {10.2307/3644339},
	pages = {198--221},
	number = {2},
	journaltitle = {Population Index},
	author = {White, Michael J.},
	urldate = {2025-02-16},
	date = {1986},
	note = {Publisher: Office of Population Research},
	file = {JSTOR Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\DHNXVY6Y\\White - 1986 - Segregation and Diversity Measures in Population Distribution.pdf:application/pdf},
}

@article{cao_review_2022,
	title = {Review and evaluation of imputation methods for multivariate longitudinal data with mixed-type incomplete variables},
	volume = {41},
	issn = {1097-0258},
	doi = {10.1002/sim.9592},
	abstract = {Estimating relationships between multiple incomplete patient measurements requires methods to cope with missing values. Multiple imputation is one approach to address missing data by filling in plausible values for those that are missing. Multiple imputation procedures can be classified into two broad types: joint modeling ({JM}) and fully conditional specification ({FCS}). {JM} fits a multivariate distribution for the entire set of variables, but it may be complex to define and implement. {FCS} imputes missing data variable-by-variable from a set of conditional distributions. In many studies, {FCS} is easier to define and implement than {JM}, but it may be based on incompatible conditional models. Imputation methods based on multilevel modeling show improved operating characteristics when imputing longitudinal data, but they can be computationally intensive, especially when imputing multiple variables simultaneously. We review current {MI} methods for incomplete longitudinal data and their implementation on widely accessible software. Using simulated data from the National Health and Aging Trends Study, we compare their performance for monotone and intermittent missing data patterns. Our simulations demonstrate that in a longitudinal study with a limited number of repeated observations and time-varying variables, {FCS}-Standard is a computationally efficient imputation method that is accurate and precise for univariate single-level and multilevel regression models. When the analyses comprise multivariate multilevel models, {FCS}-{LMM}-latent is a statistically valid procedure with overall more accurate estimates, but it requires more intensive computations. Imputation methods based on generalized linear multilevel models can lead to biased subject-level variance estimates when the statistical analyses involve hierarchical models.},
	pages = {5844--5876},
	number = {30},
	journaltitle = {Statistics in Medicine},
	shortjournal = {Stat Med},
	author = {Cao, Yi and Allore, Heather and Vander Wyk, Brent and Gutman, Roee},
	date = {2022-12-30},
	pmid = {36220138},
	pmcid = {PMC9771917},
	keywords = {Biometry, chained equations, Computer Simulation, Humans, joint modeling, longitudinal analysis, Longitudinal Studies, Models, Statistical, multiple imputation, Research Design, Software},
}

@article{perez_miguez_smartcytoflow_2024,
	title = {Smartcytoflow: A Machine Learning Decision Support System for Flow Cytometry Analysis in B Cell Acute Lymphoblastic Leukemia Diagnosis and Monitoring},
	volume = {144},
	issn = {0006-4971},
	url = {https://doi.org/10.1182/blood-2024-201439},
	doi = {10.1182/blood-2024-201439},
	shorttitle = {Smartcytoflow},
	abstract = {{IntroductionB}-cell acute lymphoblastic leukemia (B-{ALL}) presents unique diagnostic challenges due to its complex cellular makeup. Although flow cytometry, especially with {EuroFlow} standards, offers detailed insights, it is still a time-consuming process susceptible to variations between observers. The integration of machine learning ({ML}) can enhance this workflow, ensuring consistent and accurate diagnostic outcomes. This study examines how {ML} can be applied to flow cytometry data for identifying measurable residual disease ({MRD}) in B-{ALL}.{ObjectivesThe} objective of this project was to develop and validate a {ML} model for the accurate detection of {MRD} in B-{ALL} using standardized flow cytometry data from a real-world diagnostic unit in Spain.{MethodologyWe} collected 595 samples processed according to {EuroFlow} standards for {MRD} detection in B-{ALL}. These samples included two different antibody panels: the first panel targeted {CD}81, {CD}304+{CD}73, {CD}34, {CD}10, {CD}20, {CD}45, {CD}38, and {CD}19, while the second panel targeted {CD}81, {CD}66C+{CD}73, {CD}34, {CD}10, {CD}20, {CD}45, {CD}38, and {CD}19. All studies were conducted on follow-up samples analyzed for {MRD} detection or due to suspicion of relapse.Preprocessing involved using the Bioconductor package {flowAI} to remove doublets, margins, and artifacts. A gating strategy was then applied to enrich the analysis for B cells, focusing on extracting predominantly positive events from each sample. To address dataset imbalances, we employed the Synthetic Minority Over-sampling Technique ({SMOTE}) in the training set.For clustering, we used {flowSOM} to extract clusters and metaclusters from each tube, which were subsequently fed into a random forest classifier. The model was trained and cross-validated on the training set, followed by independent validation on the test set.{ResultsThe} training set comprised 800 samples with a slightly unbalanced distribution of class labels (500 positive, 300 negative), due to the initial dataset lacking sufficient positive samples. The random forest model exhibited robust performance during the training phase, achieving an out-of-bag ({OOB}) area under the curve ({AUC}) of 99.44\%, a precision-recall ({PR}) {AUC} of 99.05\%, and a Brier score of 0.123 indicating high accuracy. The {OOB} G-mean was 0.96, with a misclassification rate of 3.12\%. The confusion matrix indicated a class error of 2.90\% for the negative class and 3.27\% for the positive class.The test set consisted of 92 samples. During the validation phase, the model maintained strong performance, with an {AUC} of 92.93\%, a {PR}-{AUC} of 0.92, and a Brier score of 0.08. The G-mean for the test set was 0.83, with a misclassification rate of 10.87\%. The confusion matrix for the test set showed a class error of 0\% for the negative class and 30.3\% for the positive class. Further analysis of the test set revealed that 56 samples were extracted due to suspicion of disease relapse, with a misclassification rate of 5\%, while 82 samples were obtained for {MRD} detection, where the rate increased to 13.58\%.In addition to the robust performance metrics, the implementation of our {ML} model within the {SmartCytoFlow} platform has significantly streamlined the diagnostic workflow. {SmartCytoFlow} automates data preprocessing, gating, clustering, and classification, providing real-time diagnostic support. The integration has resulted in a substantial reduction in analysis time and improved diagnostic consistency.{ConclusionOur} study highlights the effectiveness of integrating {ML} with flow cytometry for monitoring B-{ALL}. This approach complements traditional diagnostic methods, providing consistent and accurate results. Further validation in diverse cohorts is needed to establish its broader applicability in routine practice.},
	pages = {7483},
	issue = {Supplement 1},
	journaltitle = {Blood},
	shortjournal = {Blood},
	author = {Pérez Míguez, Carlos and Diaz Arias, Jose Angel and Crucitti, Davide and Gómez Fernández, Jesús and Piñeiro Fiel, Manuel and Yanez San Segundo, Lucrecia and Mosquera Orgueira, Adrian},
	urldate = {2025-03-01},
	date = {2024-11-05},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\FBMICWGN\\Pérez Míguez et al. - 2024 - Smartcytoflow A Machine Learning Decision Support System for Flow Cytometry Analysis in B Cell Acut.pdf:application/pdf;Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\G7ENAKKP\\Smartcytoflow-A-Machine-Learning-Decision-Support.html:text/html},
}

@article{subramaniam_individualized_2023,
	title = {Individualized Prediction of Outcomes of Hematopoietic Cell Transplantation for Sickle Cell Disease: A Machine Learning Approach},
	volume = {142},
	issn = {0006-4971},
	url = {https://doi.org/10.1182/blood-2023-178352},
	doi = {10.1182/blood-2023-178352},
	shorttitle = {Individualized Prediction of Outcomes of Hematopoietic Cell Transplantation for Sickle Cell Disease},
	abstract = {Complications of Sickle cell disease ({SCD}) are ameliorated by modifying therapies but, hematopoietic cell transplantation ({HCT}) remains the only therapeutic option with curative intent. The therapeutic dilemma faced by families regarding {HCT} is partly related to the uncertainty regarding outcomes and adverse events. Computational machine-learning ({ML}) methods add to population-level outcomes by helping determine generalizable predictive patterns and quantifying uncertainty in those estimates. The incorporation of uncertainty as well as predictions allows clinicians to assess the confidence level of the predictions, avoid over-reliance on potentially incorrect results, allows for risk management and mitigation by considering the potential range of outcomes. Thus, such models may permit better decision-making by considering different scenarios and their probabilities, and increase transparency and accountability, by providing insight into how the model arrived at its predictions. Our objective was to develop a personalized, predictive machine learning model derived from center for international bone marrow transplant research ({CIBMTR})datasets which meets acceptable {AUC}, calibration slope and discrimination.{MethodsWe} applied {ML} methods to clinical parameters and both categorical and time-to-event outcomes in a de-identified {CIBMTR} dataset of patients undergoing {HCT}. A supervised random forest model was created with baseline covariates as independent variables. Model selection was performed by both the clinician and data scientists to create a model using prognostically relevant variables. Since the number and percentage of negative outcomes in {HCT} for {SCD} is smaller than the positive outcomes, the model is imbalanced and biased towards predicting positive outcomes. To counter the imbalance, we constructed a training dataset taking each outcome variable of interest, and included randomly sampled positive outcomes, typically 1.5-3 times the total instances of the variable of interest. We ran the test dataset, and used a random forest on 20 such trials. To account for the effect of the undersampling, we propose a positive threshold δ, and assigned a final prediction of a negative outcome if the average sum for an element is greater than delta. We performed a 2 times Repeated 10 Fold Cross Validation, to demonstrate our model's versatility and response to unknown data. The accuracy score may be misleading as it may result from the model being able to correctly predict the numerical majority of the positive outcomes, and does not indicate the ability to detect negative outcomes. We therefore estimated balanced accuracy which is the arithmetic mean of sensitivity and specificity. Thus, a higher balanced accuracy results only when both the negative and positive outcomes are predicted with a high accuracy. We also measured Area under the Receiver Operator Characteristic Curve ({ROC} {AUC}), a measure of the ability of a binary classifier to distinguish between classes. We define model confidence as the average probability across 20 trials, as well as our confidence in the model. We describe predictive probability percentage as : model confidence*our confidence in the model*100.{ResultsWe} examined de-identified records of 1641 patients who underwent {HCT} and were reported to the {CIBMTR}. Patients were followed for a median of 47.8 months (0.3-312.9) Patient characteristics included 73.4\% patient's age at {HCT} \<18 years, Karnofsky-Lansky ({KPS}) score ≥ 90 in 74.7\%, overall survival 91.2\%, event-free survival 75.5\%, graft failure ({GF}) 17.9\%, {AGVHD} 18.3\%, and {CGVHD} 22.3\%. Predictive model performance is described in Table 1. Predictive variables that made a significant contribution, and predicted outcomes in three hypothetical scenarios are described in Figure 1. Overall, the predictive model provided acceptable {AUC}, Balanced accuracy, positive predictive value and sensitivity.{ConclusionsWe} report the development, testing, and validation of an {ML} model for individualized prediction of outcomes of {HCT} for {SCD}. The model provides acceptable {AUC}, accuracy, balanced accuracy, positive predictive value and sensitivity. This predictive model has the potential to aid clinicians in making shared decisions with their patients regarding {HCT} for {SCD}.},
	pages = {1058},
	issue = {Supplement 1},
	journaltitle = {Blood},
	shortjournal = {Blood},
	author = {Subramaniam, Rajagopalan and Kane, Michael and Krishnamurti, Lakshmanan},
	urldate = {2025-03-01},
	date = {2023-11-02},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\6ZNYR4D2\\Subramaniam et al. - 2023 - Individualized Prediction of Outcomes of Hematopoietic Cell Transplantation for Sickle Cell Disease.pdf:application/pdf;Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\PCWMTS6P\\Individualized-Prediction-of-Outcomes-of.html:text/html},
}

@article{dabek_evaluation_2022,
	title = {Evaluation of Machine Learning Techniques to Predict the Likelihood of Mental Health Conditions Following a First {mTBI}},
	volume = {12},
	url = {https://pmc.ncbi.nlm.nih.gov/articles/PMC8847217/},
	doi = {10.3389/fneur.2021.769819},
	abstract = {Limited research has evaluated the utility of machine learning models and longitudinal data from electronic health records ({EHR}) to forecast mental health outcomes following a traumatic brain injury ({TBI}). The objective of this study is to assess ...},
	pages = {769819},
	journaltitle = {Frontiers in Neurology},
	author = {Dabek, Filip and Hoover, Peter and Jorgensen-Wagers, Kendra and Wu, Tim and Caban, Jesus J.},
	urldate = {2025-03-01},
	date = {2022-02-02},
	langid = {english},
	pmid = {35185749},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\8JGVV2RZ\\Dabek et al. - 2022 - Evaluation of Machine Learning Techniques to Predict the Likelihood of Mental Health Conditions Foll.pdf:application/pdf},
}

@article{baldo_stiffness_2019,
	title = {Stiffness Modulus and Marshall Parameters of Hot Mix Asphalts: Laboratory Data Modeling by Artificial Neural Networks Characterized by Cross-Validation},
	volume = {9},
	rights = {https://creativecommons.org/licenses/by/4.0/},
	issn = {2076-3417},
	url = {https://www.mdpi.com/2076-3417/9/17/3502},
	doi = {10.3390/app9173502},
	shorttitle = {Stiffness Modulus and Marshall Parameters of Hot Mix Asphalts},
	abstract = {The present paper discusses the analysis and modeling of laboratory data regarding the mechanical characterization of hot mix asphalt ({HMA}) mixtures for road pavements, by means of artificial neural networks ({ANNs}). The {HMAs} investigated were produced using aggregate and bitumen of different types. Stiffness modulus ({ITSM}) and Marshall stability ({MS}) and quotient ({MQ}) were assumed as mechanical parameters to analyze and predict. The {ANN} modeling approach was characterized by multiple layers, the k-fold cross validation ({CV}) method, and the positive linear transfer function. The effectiveness of such an approach was verified in terms of the coefficients of correlation (   R   ) and mean square errors; in particular,    R    values were within the range     0.965 – 0.919     in the training phase and     0.881 – 0.834     in the {CV} testing phase, depending on the predicted parameters.},
	pages = {3502},
	number = {17},
	journaltitle = {Applied Sciences},
	shortjournal = {Applied Sciences},
	author = {Baldo, Nicola and Manthos, Evangelos and Miani, Matteo},
	urldate = {2025-03-01},
	date = {2019-08-25},
	langid = {english},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\S3TTHHA7\\Baldo et al. - 2019 - Stiffness Modulus and Marshall Parameters of Hot Mix Asphalts Laboratory Data Modeling by Artificia.pdf:application/pdf},
}

@article{admojo_estimating_2024,
	title = {Estimating Obesity Levels Using Decision Trees and K-Fold Cross-Validation: A Study on Eating Habits and Physical Conditions},
	volume = {5},
	rights = {https://creativecommons.org/licenses/by-nc/4.0},
	issn = {2715-9930},
	url = {https://jurnal.yoctobrain.org/index.php/ijodas/article/view/126},
	doi = {10.56705/ijodas.v5i1.126},
	shorttitle = {Estimating Obesity Levels Using Decision Trees and K-Fold Cross-Validation},
	abstract = {This study harnesses the predictive capabilities of machine learning to explore the determinants of obesity within populations from Mexico, Peru, and Colombia, using a Decision Tree algorithm bolstered by 5-fold cross-validation. Our comprehensive analysis of 2111 individuals' lifestyle and physical condition data yielded accuracy, precision, recall, and F1-scores that notably peaked in the third and fifth folds. The findings affirmed the significance of dietary habits and physical activity as substantial predictors of obesity levels. The variability in model performance across the folds underscored the importance of robust cross-validation in enhancing the model's generalizability. This research contributes to the burgeoning field of data science in public health by providing a viable model for obesity prediction and laying the groundwork for targeted health interventions. Our study's insights are pivotal for public health officials and policymakers, serving as a stepping stone towards more sophisticated, data-driven approaches to combating obesity. The study, however, recognizes the inherent limitations of self-reported data and the need for broader datasets that encompass more diverse variables. Future research directions include the analysis of longitudinal data to establish causal relationships and the comparison of various machine learning models to optimize predictive performance},
	pages = {37--44},
	number = {1},
	journaltitle = {Indonesian Journal of Data and Science},
	shortjournal = {ijodas},
	author = {Admojo, Fadhila Tangguh and {Nurul Rismayanti}},
	urldate = {2025-03-01},
	date = {2024-03-31},
}

@article{sriliana_truncated_2022,
	title = {A Truncated Spline and Local Linear Mixed Estimator in Nonparametric Regression for Longitudinal Data and Its Application},
	volume = {14},
	rights = {https://creativecommons.org/licenses/by/4.0/},
	issn = {2073-8994},
	url = {https://www.mdpi.com/2073-8994/14/12/2687},
	doi = {10.3390/sym14122687},
	abstract = {Longitudinal data modeling is widely carried out using parametric methods. However, when the parametric model is misspecified, the obtained estimator might be severely biased and lead to erroneous conclusions. In this study, we propose a new estimation method for longitudinal data modeling using a mixed estimator in nonparametric regression. The objective of this study was to estimate the nonparametric regression curve for longitudinal data using two combined estimators: truncated spline and local linear. The weighted least square method with a two-stage estimation procedure was used to obtain the regression curve estimation of the proposed model. To account for within-subject correlations in the longitudinal data, a symmetric weight matrix was given in the regression curve estimation. The best model was determined by minimizing the generalized cross-validation value. Furthermore, an application to a longitudinal dataset of the poverty gap index in Bengkulu Province, Indonesia, was conducted to illustrate the performance of the proposed mixed estimator. Compared to the single estimator, the truncated spline and local linear mixed estimator had better performance in longitudinal data modeling based on the {GCV} value. Additionally, the empirical results of the best model indicated that the proposed model could explain the data variation exceptionally well.},
	pages = {2687},
	number = {12},
	journaltitle = {Symmetry},
	shortjournal = {Symmetry},
	author = {Sriliana, Idhia and Budiantara, I Nyoman and Ratnasari, Vita},
	urldate = {2025-03-01},
	date = {2022-12-19},
	langid = {english},
	file = {Full Text:C\:\\Users\\mlchr\\Zotero\\storage\\VXGKP4HQ\\Sriliana et al. - 2022 - A Truncated Spline and Local Linear Mixed Estimator in Nonparametric Regression for Longitudinal Dat.pdf:application/pdf},
}

@article{guesne_mind_2024,
	title = {Mind your prevalence!},
	volume = {16},
	issn = {1758-2946},
	url = {https://doi.org/10.1186/s13321-024-00837-w},
	doi = {10.1186/s13321-024-00837-w},
	abstract = {Multiple metrics are used when assessing and validating the performance of quantitative structure–activity relationship ({QSAR}) models. In the case of binary classification, balanced accuracy is a metric to assess the global performance of such models. In contrast to accuracy, balanced accuracy does not depend on the respective prevalence of the two categories in the test set that is used to validate a {QSAR} classifier. As such, balanced accuracy is used to overcome the effect of imbalanced test sets on the model’s perceived accuracy. Matthews' correlation coefficient ({MCC}), an alternative global performance metric, is also known to mitigate the imbalance of the test set. However, in contrast to the balanced accuracy, {MCC} remains dependent on the respective prevalence of the predicted categories. For simplicity, the rest of this work is based on the positive prevalence. The {MCC} value may be underestimated at high or extremely low positive prevalence. It contributes to more challenging comparisons between experiments using test sets with different positive prevalences and may lead to incorrect interpretations. The concept of balanced metrics beyond balanced accuracy is, to the best of our knowledge, not yet described in the cheminformatic literature. Therefore, after describing the relevant literature, this manuscript will first formally define a confusion matrix, sensitivity and specificity and then present, with synthetic data, the danger of comparing performance metrics under nonconstant prevalence. Second, it will demonstrate that balanced accuracy is the performance metric accuracy calibrated to a test set with a positive prevalence of 50\% (i.e., balanced test set). This concept of balanced accuracy will then be extended to the {MCC} after showing its dependency on the positive prevalence. Applying the same concept to any other performance metric and widening it to the concept of calibrated metrics will then be briefly discussed. We will show that, like balanced accuracy, any balanced performance metric may be expressed as a function of the well-known values of sensitivity and specificity. Finally, a tale of two {MCCs} will exemplify the use of this concept of balanced {MCC} versus {MCC} with four use cases using synthetic data.},
	pages = {43},
	number = {1},
	journaltitle = {Journal of Cheminformatics},
	shortjournal = {J Cheminform},
	author = {Guesné, Sébastien J. J. and Hanser, Thierry and Werner, Stéphane and Boobier, Samuel and Scott, Shaylyn},
	urldate = {2025-03-02},
	date = {2024-04-15},
	langid = {english},
	keywords = {Balanced Matthews’ correlation coefficient, Balanced metrics, Calibrated metrics, Classification models, Imbalanced, Model validation, Prevalence, Prevalence shift, {QSAR}},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\K93E3A5P\\Guesné et al. - 2024 - Mind your prevalence!.pdf:application/pdf},
}

@article{akosa_predictive_nodate,
	title = {Predictive Accuracy: A Misleading Performance Measure for Highly Imbalanced Data},
	abstract = {The most commonly reported model evaluation metric is the accuracy. This metric can be misleading when the data are imbalanced. In such cases, other evaluation metrics should be considered in addition to the accuracy. This study reviews alternative evaluation metrics for assessing the effectiveness of a model in highly imbalanced data. We used credit card clients in Taiwan as a case study. The data set contains 30,000 instances (22.12\% risky and 77.88\% non-risky) assessing the likeliness of a customer defaulting on a payment. Three different techniques were used during the model building process. The first technique involved down-sampling the majority class in the training subset. The second used the original imbalanced data whereas prior probabilities were set to account for oversampling in the third technique. The same sets of predictive models were then built for each technique after which the evaluation metrics were computed. The results suggest that model evaluation metrics might reveal more about distribution of classes than they do about the actual performance of models when the data are imbalanced. Moreover, some of the predictive models were identified to be very sensitive to imbalance. The final decision in model selection should consider a combination of different measures instead of relying on one measure. To minimize imbalance-biased estimates of performance, we recommend reporting both the obtained metric values and the degree of imbalance in the data.},
	author = {Akosa, Josephine},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\XM5BT8XZ\\Akosa - Predictive Accuracy A Misleading Performance Measure for Highly Imbalanced Data.pdf:application/pdf},
}

@misc{domingue_intermodel_2025,
	title = {The {InterModel} Vigorish ({IMV}) as a flexible and portable approach for quantifying predictive accuracy with binary outcomes},
	url = {https://osf.io/gu3ap_v2},
	doi = {10.31235/osf.io/gu3ap_v2},
	abstract = {Understanding the ``fit'' of models designed to predict binary outcomes has been a long-standing problem. We propose a flexible, portable, and intuitive metric for quantifying the change in accuracy between two predictive systems in the case of a binary outcome: the {InterModel} Vigorish ({IMV}). The {IMV} is based on an analogy to weighted coins, well-characterized physical systems with tractable probabilities. The {IMV} is always a statement about the change in fit relative to some baseline model---which can be as simple as the prevalence---whereas other metrics are stand-alone measures that need to be further manipulated to yield indices related to differences in fit across models. Moreover, the {IMV} is consistently interpretable independent of baseline prevalence. We contrast this metric with alternatives in numerous simulations. The {IMV} is more sensitive to estimation error than many alternatives and also shows distinctive sensitivity to prevalence. We then showcase its flexibility across examples spanning the social, biomedical, and physical sciences. We also demonstrate how it can be used to provide straightforward interpretation of logistic regression coefficients. The {IMV} allows for precise answers to questions about changes in model fit in a variety of settings in a manner that will be useful for furthering research with binary outcomes.},
	publisher = {{OSF}},
	author = {Domingue, Ben and Rahal, Charles and Faul, Jessica and Freese, Jeremy and Kanopka, Klint and Rigos, Alexandros and Stenhaug, Ben and Tripathi, Ajay},
	urldate = {2025-03-02},
	date = {2025-02-19},
	langid = {english},
	keywords = {Binary Outcomes, Fit Index, Logistic Regression, Prediction},
	file = {OSF Preprint:C\:\\Users\\mlchr\\Zotero\\storage\\ENJAHSQG\\Domingue et al. - 2025 - The InterModel Vigorish (IMV) as a flexible and portable approach for quantifying predictive accurac.pdf:application/pdf},
}

@article{bates_fitting_2015,
	title = {Fitting Linear Mixed-Effects Models Using lme4},
	volume = {67},
	rights = {Copyright (c) 2015 Douglas Bates, Martin Mächler, Ben Bolker, Steve Walker},
	issn = {1548-7660},
	url = {https://doi.org/10.18637/jss.v067.i01},
	doi = {10.18637/jss.v067.i01},
	abstract = {Maximum likelihood or restricted maximum likelihood ({REML}) estimates of the parameters in linear mixed-effects models can be determined using the lmer function in the lme4 package for R. As for most model-fitting functions in R, the model is described in an lmer call by a formula, in this case including both fixed- and random-effects terms. The formula and data together determine a numerical representation of the model from which the profiled deviance or the profiled {REML} criterion can be evaluated as a function of some of the model parameters. The appropriate criterion is optimized, using one of the constrained optimization functions in R, to provide the parameter estimates. We describe the structure of the model, the steps in evaluating the profiled deviance or {REML} criterion, and the structure of classes or types that represents such a model. Sufficient detail is included to allow specialization of these structures by users who wish to write functions to fit specialized linear mixed models, such as models incorporating pedigrees or smoothing splines, that are not easily expressible in the formula language used by lmer.},
	pages = {1--48},
	journaltitle = {Journal of Statistical Software},
	author = {Bates, Douglas and Mächler, Martin and Bolker, Ben and Walker, Steve},
	urldate = {2025-03-05},
	date = {2015-10-07},
	langid = {english},
	keywords = {Cholesky decomposition, linear mixed models, penalized least squares, sparse matrix methods},
	file = {Full Text:C\:\\Users\\mlchr\\Zotero\\storage\\X8C7S8EZ\\Bates et al. - 2015 - Fitting Linear Mixed-Effects Models Using lme4.pdf:application/pdf},
}

@article{nelder_simplex_1965,
	title = {A Simplex Method for Function Minimization},
	volume = {7},
	issn = {0010-4620},
	url = {https://doi.org/10.1093/comjnl/7.4.308},
	doi = {10.1093/comjnl/7.4.308},
	abstract = {A method is described for the minimization of a function of n variables, which depends on the comparison of function values at the (n + 1) vertices of a general simplex, followed by the replacement of the vertex with the highest value by another point. The simplex adapts itself to the local landscape, and contracts on to the final minimum. The method is shown to be effective and computationally compact. A procedure is given for the estimation of the Hessian matrix in the neighbourhood of the minimum, needed in statistical estimation problems.},
	pages = {308--313},
	number = {4},
	journaltitle = {The Computer Journal},
	shortjournal = {The Computer Journal},
	author = {Nelder, J. A. and Mead, R.},
	urldate = {2025-03-05},
	date = {1965-01-01},
	file = {Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\LWXPRE3Y\\354237.html:text/html},
}

@software{wright_ranger_2024,
	title = {ranger: A Fast Implementation of Random Forests},
	rights = {{GPL}-3},
	url = {https://cran.r-project.org/web/packages/ranger/index.html},
	shorttitle = {ranger},
	abstract = {A fast implementation of Random Forests, particularly suited for high dimensional data. Ensembles of classification, regression, survival and probability prediction trees are supported. Data from genome-wide association studies can be analyzed efficiently. In addition to data frames, datasets of class 'gwaa.data' (R package '{GenABEL}') and '{dgCMatrix}' (R package 'Matrix') can be directly analyzed.},
	version = {0.17.0},
	author = {Wright, Marvin N. and Wager, Stefan and Probst, Philipp},
	urldate = {2025-03-05},
	date = {2024-11-08},
	keywords = {{MachineLearning}, Survival},
}

@software{kuhn__aut_caret_2024,
	title = {caret: Classification and Regression Training},
	rights = {{GPL}-2 {\textbar} {GPL}-3 [expanded from: {GPL} (≥ 2)]},
	url = {https://cran.r-project.org/web/packages/caret/index.html},
	shorttitle = {caret},
	abstract = {Misc functions for training and plotting classification and regression models.},
	version = {7.0-1},
	author = {Kuhn  [aut, Max and cre and Wing, Jed and Weston, Steve and Williams, Andre and Keefer, Chris and Engelhardt, Allan and Cooper, Tony and Mayer, Zachary and Kenkel, Brenton and R Core Team and Benesty, Michael and Lescarbeau, Reynald and Ziem, Andrew and Scrucca, Luca and Tang, Yuan and Candan, Can and Hunt, Tyler},
	urldate = {2025-03-05},
	date = {2024-12-10},
	keywords = {{MachineLearning}, {HighPerformanceComputing}},
}

@software{friedman_glmnet_2023,
	title = {glmnet: Lasso and Elastic-Net Regularized Generalized Linear Models},
	rights = {{GPL}-2},
	url = {https://cran.r-project.org/web/packages/glmnet/index.html},
	shorttitle = {glmnet},
	abstract = {Extremely efficient procedures for fitting the entire lasso or elastic-net regularization path for linear regression, logistic and multinomial regression models, Poisson regression, Cox model, multiple-response Gaussian, and the grouped multinomial regression; see {\textless}doi:10.18637/jss.v033.i01{\textgreater} and {\textless}doi:10.18637/jss.v039.i05{\textgreater}. There are two new and important additions. The family argument can be a {GLM} family object, which opens the door to any programmed family ({\textless}doi:10.18637/jss.v106.i01{\textgreater}). This comes with a modest computational cost, so when the built-in families suffice, they should be used instead. The other novelty is the relax option, which refits each of the active sets in the path unpenalized. The algorithm uses cyclical coordinate descent in a path-wise fashion, as described in the papers cited.},
	version = {4.1-8},
	author = {Friedman, Jerome and Hastie, Trevor and Tibshirani, Rob and Narasimhan, Balasubramanian and Tay, Kenneth and Simon, Noah and Qian, Junyang and Yang, James},
	urldate = {2025-03-05},
	date = {2023-08-22},
	keywords = {{MachineLearning}, Survival},
}

@article{abadi_tensorflow_nodate,
	title = {{TensorFlow}: Large-Scale Machine Learning on Heterogeneous Distributed Systems},
	abstract = {{TensorFlow} [1] is an interface for expressing machine learning algorithms, and an implementation for executing such algorithms. A computation expressed using {TensorFlow} can be executed with little or no change on a wide variety of heterogeneous systems, ranging from mobile devices such as phones and tablets up to large-scale distributed systems of hundreds of machines and thousands of computational devices such as {GPU} cards. The system is flexible and can be used to express a wide variety of algorithms, including training and inference algorithms for deep neural network models, and it has been used for conducting research and for deploying machine learning systems into production across more than a dozen areas of computer science and other fields, including speech recognition, computer vision, robotics, information retrieval, natural language processing, geographic information extraction, and computational drug discovery. This paper describes the {TensorFlow} interface and an implementation of that interface that we have built at Google. The {TensorFlow} {API} and a reference implementation were released as an open-source package under the Apache 2.0 license in November, 2015 and are available at www.tensorflow.org.},
	author = {Abadi, Martın and Agarwal, Ashish and Barham, Paul and Brevdo, Eugene and Chen, Zhifeng and Citro, Craig and Corrado, Greg S and Davis, Andy and Dean, Jeffrey and Devin, Matthieu and Ghemawat, Sanjay and Goodfellow, Ian and Harp, Andrew and Irving, Geoffrey and Isard, Michael and Jia, Yangqing and Jozefowicz, Rafal and Kaiser, Lukasz and Kudlur, Manjunath and Levenberg, Josh and Mane, Dan and Monga, Rajat and Moore, Sherry and Murray, Derek and Olah, Chris and Schuster, Mike and Shlens, Jonathon and Steiner, Benoit and Sutskever, Ilya and Talwar, Kunal and Tucker, Paul and Vanhoucke, Vincent and Vasudevan, Vijay and Viegas, Fernanda and Vinyals, Oriol and Warden, Pete and Wattenberg, Martin and Wicke, Martin and Yu, Yuan and Zheng, Xiaoqiang},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\UJ55FKS9\\Abadi et al. - TensorFlow Large-Scale Machine Learning on Heterogeneous Distributed Systems.pdf:application/pdf},
}

@software{chen_xgboost_2024,
	title = {xgboost: Extreme Gradient Boosting},
	rights = {Apache License (== 2.0) {\textbar} file {LICENSE}},
	url = {https://cran.r-project.org/web/packages/xgboost/index.html},
	shorttitle = {xgboost},
	abstract = {Extreme Gradient Boosting, which is an efficient implementation of the gradient boosting framework from Chen \& Guestrin (2016) {\textless}doi:10.1145/2939672.2939785{\textgreater}. This package is its R interface. The package includes efficient linear model solver and tree learning algorithms. The package can automatically do parallel computation on a single machine which could be more than 10 times faster than existing gradient boosting packages. It supports various objective functions, including regression, classification and ranking. The package is made to be extensible, so that users are also allowed to define their own objectives easily.},
	version = {1.7.8.1},
	author = {Chen, Tianqi and He, Tong and Benesty, Michael and Khotilovich, Vadim and Tang, Yuan and Cho, Hyunsu and Chen, Kailong and Mitchell, Rory and Cano, Ignacio and Zhou, Tianyi and Li, Mu and Xie, Junyuan and Lin, Min and Geng, Yifeng and Li, Yutian and Yuan, Jiaming and implementation), {XGBoost} contributors (base {XGBoost}},
	urldate = {2025-03-06},
	date = {2024-07-24},
	keywords = {{MachineLearning}, Survival, {HighPerformanceComputing}, {ModelDeployment}},
}

@article{hofman_prediction_2017,
	title = {Prediction and explanation in social systems},
	volume = {355},
	issn = {0036-8075, 1095-9203},
	url = {https://www.science.org/doi/10.1126/science.aal3856},
	doi = {10.1126/science.aal3856},
	abstract = {Historically, social scientists have sought out explanations of human and social phenomena that provide interpretable causal mechanisms, while often ignoring their predictive accuracy. We argue that the increasingly computational nature of social science is beginning to reverse this traditional bias against prediction; however, it has also highlighted three important issues that require resolution. First, current practices for evaluating predictions must be better standardized. Second, theoretical limits to predictive accuracy in complex social systems must be better characterized, thereby setting expectations for what can be predicted or explained. Third, predictive accuracy and interpretability must be recognized as complements, not substitutes, when evaluating explanations. Resolving these three issues will lead to better, more replicable, and more useful social science.},
	pages = {486--488},
	number = {6324},
	journaltitle = {Science},
	shortjournal = {Science},
	author = {Hofman, Jake M. and Sharma, Amit and Watts, Duncan J.},
	urldate = {2025-03-06},
	date = {2017-02-03},
	langid = {english},
	file = {Full Text PDF:C\:\\Users\\mlchr\\Zotero\\storage\\IRHKRJ5L\\Hofman et al. - 2017 - Prediction and explanation in social systems.pdf:application/pdf},
}

@misc{baker_using_2022-1,
	title = {Using Demographic Data as Predictor Variables: a Questionable Choice},
	rights = {http://opensource.org/licenses/{MIT}},
	url = {https://osf.io/y4wvj},
	doi = {10.35542/osf.io/y4wvj},
	shorttitle = {Using Demographic Data as Predictor Variables},
	abstract = {Predictive analytics methods in education are seeing widespread use and are producing increasingly accurate predictions of students’ outcomes. With the increased use of predictive analytics comes increasing concern about fairness for specific subgroups of the population. One approach that has been proposed to increase fairness is using demographic variables directly in models, as predictors. In this paper we explore issues of fairness in the use of demographic variables as predictors of long-term student outcomes, studying the arguments for and against this practice in the contexts where this literature has been published. We analyze arguments for the inclusion of demographic variables, specifically claims that this approach improves model performance and charges that excluding such variables amounts to a form of ‘color-blind’ racism. We also consider arguments against including demographic variables as predictors, including reduced actionability of predictions, risk of reinforcing bias, and limits of categorization. We then discuss how contextual factors of predictive models should influence case-specific decisions for the inclusion or exclusion of demographic variables and discuss the role of proxy variables. We conclude that, on balance, there are greater benefits to fairness if demographic variables are used to validate fairness rather than as predictors within models.},
	publisher = {{EdArXiv}},
	author = {Baker, Ryan Shaun and Esbenshade, Lief and Vitale, Jon and Karumbaiah, Shamya},
	urldate = {2025-03-06},
	date = {2022-12-19},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\VW2WHHSI\\Baker et al. - 2022 - Using Demographic Data as Predictor Variables a Questionable Choice.pdf:application/pdf},
}

@article{cornman_documentation_nodate,
	title = {Documentation for the {NCES} Common Core of Data National Public Education Financial Survey ({NPEFS}), School Year 2020–21 (Fiscal Year 2021): Provisional File Version 1a},
	author = {Cornman, Stephen Q},
	langid = {english},
	file = {PDF:C\:\\Users\\mlchr\\Zotero\\storage\\KKYXXJES\\Cornman - Documentation for the NCES Common Core of Data National Public Education Financial Survey (NPEFS), S.pdf:application/pdf},
}

@online{noauthor_what_2024,
	title = {What is downsampling? {\textbar} {IBM}},
	url = {https://www.ibm.com/think/topics/downsampling},
	shorttitle = {What is downsampling?},
	abstract = {Downsampling decreases the number of data samples in a dataset. In doing so, it aims to correct imbalanced data and thereby improve model performance.},
	urldate = {2025-04-17},
	date = {2024-06-15},
	langid = {english},
	file = {Snapshot:C\:\\Users\\mlchr\\Zotero\\storage\\NZ3CHEHL\\downsampling.html:text/html},
}

@article{polley_super_2010-1,
	title = {Super Learner In Prediction},
	url = {https://biostats.bepress.com/ucbbiostat/paper266},
	journaltitle = {U.C. Berkeley Division of Biostatistics Working Paper Series},
	author = {Polley, Eric and Laan, Mark van der},
	date = {2010-05-03},
	file = {text/html Attachment:C\:\\Users\\mlchr\\Zotero\\storage\\2YMN3RJ7\\paper266.html:text/html},
}

@software{polley_superlearner_2024,
	title = {{SuperLearner}: Super Learner Prediction},
	rights = {{GPL}-3},
	url = {https://cran.r-project.org/web/packages/SuperLearner/index.html},
	shorttitle = {{SuperLearner}},
	abstract = {Implements the super learner prediction method and contains a library of prediction algorithms to be used in the super learner.},
	version = {2.0-29},
	author = {Polley, Eric and {LeDell}, Erin and Kennedy, Chris and Lendle, Sam and Laan, Mark van der},
	urldate = {2025-04-17},
	date = {2024-02-20},
	keywords = {Bayesian, {MachineLearning}},
}

@article{abadi,
	title = {TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems},
	author = {Abadi, {Mart{\i}n} and Agarwal, Ashish and Barham, Paul and Brevdo, Eugene and Chen, Zhifeng and Citro, Craig and Corrado, Greg S and Davis, Andy and Dean, Jeffrey and Devin, Matthieu and Ghemawat, Sanjay and Goodfellow, Ian and Harp, Andrew and Irving, Geoffrey and Isard, Michael and Jia, Yangqing and Jozefowicz, Rafal and Kaiser, Lukasz and Kudlur, Manjunath and Levenberg, Josh and Mane, Dan and Monga, Rajat and Moore, Sherry and Murray, Derek and Olah, Chris and Schuster, Mike and Shlens, Jonathon and Steiner, Benoit and Sutskever, Ilya and Talwar, Kunal and Tucker, Paul and Vanhoucke, Vincent and Vasudevan, Vijay and Viegas, Fernanda and Vinyals, Oriol and Warden, Pete and Wattenberg, Martin and Wicke, Martin and Yu, Yuan and Zheng, Xiaoqiang},
	langid = {en}
}

@article{akosa,
	title = {Predictive Accuracy: A Misleading Performance Measure for Highly Imbalanced Data},
	author = {Akosa, Josephine},
	langid = {en}
}

@misc{alejo2024,
	title = {Denver Public Schools seeks community input ahead of possible school closures - CBS Colorado},
	author = {Alejo, Anna},
	year = {2024},
	month = {06},
	date = {2024-06-07},
	url = {https://www.cbsnews.com/colorado/news/denver-public-schools-seeks-community-input-ahead-of-possible-school-closures/},
	langid = {canadian}
}

@misc{commonc2024,
	title = {Common Core of Data (CCD)},
	year = {2024},
	date = {2024},
	url = {https://nces.ed.gov/ccd/aboutccd.asp},
	note = {Publisher: National Center for Education Statistics},
	langid = {EN}
}

@techreport{delatorre2009,
	title = {When Schools Close  Effects on Displaced Students in Chicago Public Schools},
	author = {de la Torre, Marisa and Gwynne, Julia},
	year = {2009},
	month = {10},
	date = {2009-10},
	langid = {canadian}
}

@book{friedman2023,
	title = {glmnet: Lasso and Elastic-Net Regularized Generalized Linear Models},
	author = {Friedman, Jerome and Hastie, Trevor and Tibshirani, Rob and Narasimhan, Balasubramanian and Tay, Kenneth and Simon, Noah and Qian, Junyang and Yang, James},
	year = {2023},
	month = {08},
	date = {2023-08-22},
	url = {https://cran.r-project.org/web/packages/glmnet/index.html}
}

@misc{whatis2024,
	title = {What is downsampling? | IBM},
	year = {2024},
	month = {06},
	date = {2024-06-15},
	url = {https://www.ibm.com/think/topics/downsampling},
	langid = {en}
}

@misc{cavalotti,
	title = {Sherlock documentation - Sherlock},
	author = {Cavalotti, Kilian},
	url = {https://www.sherlock.stanford.edu/docs/},
	langid = {en}
}

@article{cornman,
	title = {Documentation for the NCES Common Core of Data National Public Education Financial Survey (NPEFS), School Year 2020{\textendash}21 (Fiscal Year 2021): Provisional File Version 1a},
	author = {Cornman, Stephen Q},
	langid = {en}
}

@inproceedings{lundberg2017,
	title = {A Unified Approach to Interpreting Model Predictions},
	author = {Lundberg, Scott M and Lee, Su-In},
	year = {2017},
	date = {2017},
	publisher = {Curran Associates, Inc.},
	volume = {30},
	url = {https://proceedings.neurips.cc/paper_files/paper/2017/hash/8a20a8621978632d76c43dfd28b67767-Abstract.html}
}

@article{jack,
	title = {School Closings in Philadelphia},
	author = {Jack, James and Sludden, John},
	langid = {en}
}

@misc{thedeva,
	title = {The Devastating Impact of School Closures on Students and Communities},
	url = {https://populardemocracy.org/news-article/news-and-publications-devastating-impact-school-closures-students-and-communities/},
	langid = {canadian}
}

@misc{whathap,
	title = {What happened when Philly closed 30 schools? New study offers answers},
	url = {https://whyy.org/articles/what-happened-when-philly-closed-30-schools-new-study-offers-answers/},
	langid = {canadian}
}

@misc{theimpa2023,
	title = {The impact of the 2013 school closures},
	year = {2023},
	month = {05},
	date = {2023-05-11},
	url = {https://philadelphianeighborhoods.com/2023/05/11/education-the-lasting-effects-of-the-2013-school-closures/},
	langid = {canadian}
}

@misc{admin2022,
	title = {Closing schools in a time of enrollment decline {\textendash} Center on Reinventing Public Education},
	author = {Admin, CRPE},
	year = {2022},
	month = {10},
	date = {2022-10-31},
	url = {https://crpe.org/closing-schools-in-a-time-of-enrollment-decline/},
	note = {Section: Publications},
	langid = {canadian}
}

@misc{(re)defi,
	title = {(Re)Defining Urban Education: A Conceptual Review and Empirical Exploration of the Definition of Urban Education - Richard O. Welsh, Walker A. Swain, 2020},
	url = {https://journals.sagepub.com/doi/full/10.3102/0013189X20902822}
}

@misc{welsh,
	title = {(Re)Defining Urban Education: A Conceptual Review and Empirical Exploration of the Definition of Urban Education - Richard O. Welsh, Walker A. Swain, 2020},
	author = {Welsh, Richard O. and Swain, Walker A.},
	url = {https://journals.sagepub.com/doi/full/10.3102/0013189X20902822}
}