Search

Colleges and universities are increasingly turning to algorithms that predict college-student success to inform various decisions, including those related to admissions, budgeting, and student-success interventions. Because predictive algorithms rely on historical data, they capture societal injustices, including racism. In this study, we examine how the accuracy of college student success predictions differs between racialized groups, signaling algorithmic bias. We also evaluate the utility of leading bias-mitigating techniques in addressing this bias. Using nationally representative data from the Education Longitudinal Study of 2002 and various machine learning modeling approaches, we demonstrate how models incorporating commonly used features to predict college-student success are less accurate when predicting success for racially minoritized students. Common approaches to mitigating algorithmic bias are generally ineffective at eliminating disparities in prediction outcomes and accuracy between racialized groups.

The education sector has been quick to recognize the power of predictive analytics to enhance student success rates. However, there are challenges to widespread adoption, including the lack of accessibility and the potential perpetuation of inequalities. These challenges present in different stages of modeling, including data preparation, model development, and evaluation. These steps can introduce additional bias to the system if not appropriately performed. Substantial incompleteness in responses is a common problem in nationally representative education data at a large scale. This can lead to missing data and can potentially impact the representativeness and accuracy of the results. While many education-related studies address the challenges of missing data, little is known about the impact of handling missing values on the fairness of predictive outcomes in practice. In this paper, we aim to assess the disparities in predictive modeling outcomes for college student success and investigate the impact of imputation techniques on model performance and fairness using various notions. We conduct a prospective evaluation to provide a less biased estimation of future performance and fairness than an evaluation of historical data. Our comprehensive analysis of a real large-scale education dataset reveals key insights on modeling disparities and the impact of imputation techniques on the fairness of the predictive outcome under different testing scenarios. Our results indicate that imputation introduces bias if the testing set follows the historical distribution. However, if the injustice in society is addressed and, consequently, the upcoming batch of observations is equalized, the model would be less biased.

Prior studies have documented the pattern of decreased state funding for higher education in periods of economic contraction (i.e., the balance wheel phenomenon). This qualitative case study examines how policymakers in California and Texas made decisions about funding higher education at the onset of the COVID-19 pandemic, when policymakers faced an economic downturn. Data comprise 28 interviews with key state actors and 69 documents. The analysis expands prior understandings of how state policymakers make budgeting decisions that affect higher education by exploring how they perceive certain target populations as deserving or undeserving of state support. The study also sheds light on the tenuous relationship between policymakers' views of higher education and their funding decisions.

During economic recessions, state funding for higher education contracts (Delaney & Doyle, 2011; Hovey, 1999; SHEEO, 2022). Despite this reality, public higher education officials need to offer insights and explanations to state legislators about the current status of their institutions and their needs when discussing their budget requests. We use a multiple case-study design, framed by the narrative policy framework, to examine how campus officials in California and Texas justify their budget requests to the state legislature during the COVID-19 pandemic. Drawing on 131 h of transcribed legislative budget meetings and 62 documents, our findings suggest that higher education leaders emphasize the economic functions of higher education and center their ability to successfully manage during these uncertain and difficult times by highlighting improved or stable accountability measures such as enrollment, persistence, graduation, and job placement rates. During these budget requests, there are commonalities between the states regarding the structure, justifications, and narrative strategies used. However, higher education leaders evoked different narrative objects depending on the perceived values, beliefs, and norms of their state legislators.

Background: Nearly two decades have passed since the last successful reauthorization of the Higher Education Act. Since then, student loan debt and the accumulation patterns based on race have become a pressing issue to address in U.S. society. Purpose: Student debt is one of the key issues on the federal higher education policy agenda. The purpose of this paper is to examine how race is addressed in a congressional hearing held to discuss the reauthorization of the Higher Education Act. Specifically, we examined one congressional policy markup hearing to understand how members frame student debt and the racialized dynamics embedded within. Research Design: We combined critical race theory and racial frames to discursively analyze 14 hours of congressional hearings on the reauthorization of the Higher Education Act. Through critical discourse analysis, we interrogated the racialized discourse among policymakers as they proposed solutions and alternatives to address the issue of student debt during the policy markup process. Conclusions/Recommendations: Our findings highlight four types of discourse within a policy markup hearing: "All Students" Matter, Paternalistic, Race-Evasive, and Explicit Racial Discourse. We offer recommendations for policymakers and researchers to contend with ahistoricism and race-evasiveness prevalent in policy markup hearings and ways for future policy proposals to be more explicit in naming the groups facing disproportionate negative impact, the mechanisms that produce such inequities, and interventions that can address them.

Many policies in higher education are intended to improve college access and degree completion, yet often those policies fall short of their aims by making it difficult for prospective or current college students to access benefits for which they are eligible. Barriers that inhibit access to policy benefits, such as cumbersome paperwork, can weigh more heavily on members of marginalized communities, including racially minoritized students. Such administrative burdens can thus reinforce patterns of inequity. In this paper, we present a conceptual framework for examining administrative burdens embedded in higher education policies that can negatively affect prospective and current college students, especially those who are racially minoritized. With the use of our proposed framework for addressing racialized administrative burdens, researchers can improve the understanding of ethnoracial disparities in higher education, inform policymakers' design of racially equitable policies for higher education, and enable practitioners to implement those policies to promote racial equity.