View in EDS HTML Full Text PDF Full Text

Impact of COVID-19 and Financial Factors on First-Year Student Retention: A Comparative Study of Pre- and Post-Pandemic Cohorts

Saved in:
Bibliographic Details
Title: Impact of COVID-19 and Financial Factors on First-Year Student Retention: A Comparative Study of Pre- and Post-Pandemic Cohorts
Language: English
Authors: Kwideok Han (ORCID 0000-0002-5491-4364), Ranjit Pandey, Kimberly Meints, Larry Burns, Chanjin Chung
Source: Higher Education: The International Journal of Higher Education Research. 2026 91(2):723-739.
Availability: Springer. Available from: Springer Nature. One New York Plaza, Suite 4600, New York, NY 10004. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-460-1700; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/
Peer Reviewed: Y
Page Count: 17
Publication Date: 2026
Document Type: Journal Articles
Reports - Research
Education Level: Higher Education
Postsecondary Education
Descriptors: COVID-19, Pandemics, College Freshmen, Academic Persistence, Public Colleges, Student Financial Aid, Parent Financial Contribution, Student Loan Programs, Low Income Students, At Risk Students, Prediction
DOI: 10.1007/s10734-025-01439-7
ISSN: 0018-1560
1573-174X
Abstract: This study examines factors influencing first-year student retention, with a particular focus on the impact of the COVID-19 pandemic. We employ logistic regression and machine learning algorithms to identify key predictors of retention and assess variations in retention probabilities across cohorts. The analysis is based on a dataset of 8,320 first-time freshmen at a public university in the central United States during the fall semesters of 2018 (pre-COVID-19) and 2020 (post-COVID-19). Results indicate that students enrolled in the post-COVID-19 period were less likely to persist into their second year. Higher tuition assistance--such as grants, tuition waivers, and scholarships--and greater family financial contributions positively influenced retention, particularly among the post-COVID-19 cohort. Federal and private loans were also positively associated with retention, reflecting student's financial commitments. However, those with higher financial need were less likely to persist, underscoring the challenges faced by financially vulnerable students. Machine learning models demonstrated strong predictive accuracy, with the Random Forest algorithm achieving the highest performance in effectively identifying students at risk of dropping out. These findings provide actionable insights for targeted institutional interventions. By analysing retention patterns across cohorts and financial factors, this study offers valuable recommendations for addressing financial disparities and enhancing student persistence during and beyond periods of crisis.
Abstractor: As Provided
Entry Date: 2026
Accession Number: EJ1506589
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwGOdnv4K--24yUSoEsAl5XWAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDDtW2d8t6VL-fCe1JgIBEICBm2NNDY8otzDQFr_oUX6SlL3l8HRwODCqz5mmM1HY8aPS2WILeFt2dPlzeN7S6zL57JNtAJNM4mjtOsf1OX4o97F-s437Ozml2ZPgHc2LMt5gdF10s_sVEkmwxwLeaXb9SvQIu9FLdCX-4uo93n6KQyg7ns1Tl7qhkWHIev6lrDRMaRrlSrpELIQdWq47QtRHZ95iif-K67oIdMQJ
Text:
  Availability: 1
  Value: <anid>AN0191500347;hie01feb.26;2026Feb13.06:44;v2.2.500</anid> <title id="AN0191500347-1">Impact of COVID-19 and financial factors on first-year student retention: a comparative study of pre- and post-pandemic cohorts </title> <p>This study examines factors influencing first-year student retention, with a particular focus on the impact of the COVID-19 pandemic. We employ logistic regression and machine learning algorithms to identify key predictors of retention and assess variations in retention probabilities across cohorts. The analysis is based on a dataset of 8,320 first-time freshmen at a public university in the central United States during the fall semesters of 2018 (pre-COVID-19) and 2020 (post-COVID-19). Results indicate that students enrolled in the post-COVID-19 period were less likely to persist into their second year. Higher tuition assistance—such as grants, tuition waivers, and scholarships—and greater family financial contributions positively influenced retention, particularly among the post-COVID-19 cohort. Federal and private loans were also positively associated with retention, reflecting student's financial commitments. However, those with higher financial need were less likely to persist, underscoring the challenges faced by financially vulnerable students. Machine learning models demonstrated strong predictive accuracy, with the Random Forest algorithm achieving the highest performance in effectively identifying students at risk of dropping out. These findings provide actionable insights for targeted institutional interventions. By analysing retention patterns across cohorts and financial factors, this study offers valuable recommendations for addressing financial disparities and enhancing student persistence during and beyond periods of crisis.</p> <p>Keywords: First-year student retention; COVID-19 pandemic; Financial aid; Machine learning; Higher education; Student persistence</p> <p>Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.</p> <hd id="AN0191500347-2">Introduction</hd> <p>Student retention, defined as the continuation of enrolment from the first to the second year, is a critical measure of both institutional effectiveness and student success (Hagedorn, [<reflink idref="bib20" id="ref1">20</reflink>]; Tinto, [<reflink idref="bib54" id="ref2">54</reflink>]). High retention rates enhance an institution's reputation and financial stability, while also serving as key indicators of positive student outcomes (Gansemer-Topf & Schuh, [<reflink idref="bib18" id="ref3">18</reflink>]). Retention, in turn, has significant implications for students' educational trajectories, career opportunities, and lifelong earning potential (Hagedorn, [<reflink idref="bib20" id="ref4">20</reflink>]).</p> <p>Despite its importance, retention rates in the United States (U.S.) remain an ongoing concern. Over the past decade, approximately 27% of full-time, first-year students at two-year and four-year institutions have not returned for their second year. Although attrition is most pronounced at two-year and private institutions, public four-year institutions report an average first-year retention rate of about 80% (National Student Clearinghouse Research Center, [<reflink idref="bib37" id="ref5">37</reflink>], [<reflink idref="bib38" id="ref6">38</reflink>]). This persistent challenge carries substantial consequences for individuals, institutions, and society at large (Astin, [<reflink idref="bib1" id="ref7">1</reflink>]; Hagedorn, [<reflink idref="bib20" id="ref8">20</reflink>]; Tinto, [<reflink idref="bib54" id="ref9">54</reflink>]).</p> <p>Retention trends vary globally due to differences in educational systems and funding structures. In the U.S., higher education is widely accessible, with admissions criteria determined by individual institutions rather than national entrance exams. Unlike in many other countries, U.S. students are not required to declare a major upon entry; instead, they typically complete general education requirements during their first two years before transitioning into a declared major (U.S. Department of Education, [<reflink idref="bib56" id="ref10">56</reflink>]). Academic years usually follow a two-semesters, credit-based system; to maintain full-time status, students must register for a minimum of 12 credit hours in both the fall and spring semesters. An abbreviated summer semester is also available allowing enrolment in 6 to 9 credits, though this is generally not required to retain full-time status.</p> <p>The U.S. higher education system introduces unique challenges for student retention particularly due to its heavy reliance on tuition revenue. Compared to heavily subsidized systems in many European nations (Organisation for Economic Co-operation and Development, [<reflink idref="bib41" id="ref11">41</reflink>]), U.S. students often face substantial financial burdens. In the 2019 − 20 academic year, approximately 61% of undergraduate students received loans to fund their educational costs (National Center for Education Statistics, [<reflink idref="bib36" id="ref12">36</reflink>]). While financial aid is available through grants, loans, scholarships, and work-study programs, many households continue to bear a significant share of educational expenses. This financial strain remains a critical factor influencing students' decisions to persist into their second year (Webster & Showers, [<reflink idref="bib57" id="ref13">57</reflink>]).</p> <hd id="AN0191500347-3">Theoretical framework</hd> <p>According to Tinto's ([<reflink idref="bib54" id="ref14">54</reflink>]) Institutional Departure Model, a student's decision to persist or leave an institution is shaped by their degree of academic and social integration within the institutional environment. Academic integration encompasses performance, engagement with coursework, and interactions with faculty, while social integration reflects a sense of belonging fostered through peer relationships and campus activities. Institutional support plays a pivotal role in fostering both forms of integration, particularly during the first year, when students are most vulnerable to departure (Tinto, [<reflink idref="bib55" id="ref15">55</reflink>]). Tinto's model also highlights the influence of pre-entry attributes—such as demographic characteristics, family background, and prior educational experiences—on students' initial goals and institutional commitments. These attributes, in conjunction with institutional experiences, continuously shape academic and social integration, providing a comprehensive perspective on factors influencing retention.</p> <p>While Tinto's model provides valuable insights into academic and social integration, Bean's Student Attrition Model ([<reflink idref="bib6" id="ref16">6</reflink>], [<reflink idref="bib7" id="ref17">7</reflink>]) offers a complementary framework that emphasizes economic factors—such as financial costs, perceived utility, and the balance between investing in education and other life obligations—in shaping students' decisions to persist. Bean posits that students' persistence decisions mirror employee turnover in organizations, driven partly by satisfaction and organizational (institutional) fit, but strongly influenced by financial considerations. These dimensions—ranging from tuition costs and student loans to scholarships and family contributions—can be decisive in determining whether a student continues. Beans's perspective is thus particularly pertinent to this study, where financial variables play a central role in explaining retention patterns.</p> <hd id="AN0191500347-4">Factors influencing retention</hd> <p>Retention is shaped by a combination of demographic, academic, social, and financial factors. Among demographic characteristics, age, gender, and residency status are frequently linked to retention outcomes (Aulck et al., [<reflink idref="bib2" id="ref18">2</reflink>]; Delen, [<reflink idref="bib17" id="ref19">17</reflink>]; Lorenzo-Quiles et al., [<reflink idref="bib29" id="ref20">29</reflink>]; Millea et al., [<reflink idref="bib30" id="ref21">30</reflink>]; Murtaugh et al., [<reflink idref="bib32" id="ref22">32</reflink>]; Reason, [<reflink idref="bib45" id="ref23">45</reflink>]). For example, younger students and female students tend to persist at higher rates (Millea et al., [<reflink idref="bib30" id="ref24">30</reflink>]; Reason, [<reflink idref="bib45" id="ref25">45</reflink>]), whereas out-of-state and international students often encounter additional financial burdens and social adjustment challenges that can hinder their retention (Aulck et al., [<reflink idref="bib2" id="ref26">2</reflink>]; Delen, [<reflink idref="bib17" id="ref27">17</reflink>]; Murtaugh et al., [<reflink idref="bib32" id="ref28">32</reflink>]).</p> <p>Pre-entry attributes—such as transfer or dual enrolment credits—can enhance academic readiness by alleviating financial pressures and fostering time management skills (Karp et al., [<reflink idref="bib23" id="ref29">23</reflink>]). Additionally, declaring a major upon entry is associated with higher retention rates (Leppel, [<reflink idref="bib27" id="ref30">27</reflink>]), as students who have clear academic goals are typically more engaged and motivated. In contrast, undeclared students may experience uncertainty that undermines their persistence (Leppel, [<reflink idref="bib27" id="ref31">27</reflink>]).</p> <p>Academic integration, often measured through first-term GPA and credit load, is also critical for retention. Higher GPAs and heavier credit loads tend to correlate with increased persistence, highlighting the importance of early academic success (Ortiz-Lozano et al., [<reflink idref="bib42" id="ref32">42</reflink>]; Stewart et al., [<reflink idref="bib51" id="ref33">51</reflink>]). Social integration further enhances retention: participation in student organizations—such as fraternities or sororities—fosters a sense of belonging, thereby enriching the college experience (Biddix et al., [<reflink idref="bib9" id="ref34">9</reflink>]; Bowman & Holmes, [<reflink idref="bib10" id="ref35">10</reflink>]). However, financial barriers can limit low-income students' access to these opportunities, thereby widening retention gaps (Bowman & Holmes, [<reflink idref="bib10" id="ref36">10</reflink>]). Recent reviews (e.g., Lorenzo-Quiles et al., [<reflink idref="bib29" id="ref37">29</reflink>]) underscore the interplay between social integration and economic constraints, reinforcing that financial stress can impede the very campus engagement that supports persistence. According to Hoyt ([<reflink idref="bib22" id="ref38">22</reflink>]), student affairs and academic support services play a crucial role in building these connections underscoring the need for campus-wide support to promote student success.</p> <p>In the U.S., financial considerations are especially central to retention, as many students bear significant tuition expenses. Scholarships, grants, and tuition waivers help alleviate financial stress and encourage persistence (Millea et al., [<reflink idref="bib30" id="ref39">30</reflink>]; Moores & Burgess, [<reflink idref="bib31" id="ref40">31</reflink>]). Similarly, increased family contributions and institutional merit-based aid reduce financial pressures and thus support retention (Olbrecht et al., [<reflink idref="bib40" id="ref41">40</reflink>]). Conversely, large unmet financial needs or heavy reliance on loans can divert students' attention from their studies, undermining their ability to persist (Baker & Montalto, [<reflink idref="bib3" id="ref42">3</reflink>]; Britt et al., [<reflink idref="bib11" id="ref43">11</reflink>]; Kuh et al., [<reflink idref="bib26" id="ref44">26</reflink>]; Nandeshwar et al., [<reflink idref="bib33" id="ref45">33</reflink>]; Nieuwoudt & Pedler, [<reflink idref="bib39" id="ref46">39</reflink>]). External commitments—such as paid work, commuting, and caregiving responsibilities—can also limit students' time for academic and social engagement Leveson et al. ([<reflink idref="bib28" id="ref47">28</reflink>]). Indeed, while academic involvement positively affects persistence, outside responsibilities can impede full integration into the campus community (Lorenzo-Quiles et al., [<reflink idref="bib29" id="ref48">29</reflink>]).</p> <p>Despite these challenges, financial aid programs can serve as equalizers. By supporting both academic and social integration (Cabrera et al., [<reflink idref="bib13" id="ref49">13</reflink>]), these programs increase the likelihood that students will remain enrolled. Institutions that proactively address students' financial difficulties through targeted interventions often report improved retention outcomes (Chen, [<reflink idref="bib14" id="ref50">14</reflink>]). Bean's ([<reflink idref="bib7" id="ref51">7</reflink>]) emphasis on these economic levers aligns with Tinto's ([<reflink idref="bib54" id="ref52">54</reflink>]) notion that institutional support—especially in the form of financial relief—can enhance both academic and social integration, ultimately boosting persistence.</p> <hd id="AN0191500347-5">Pandemic effects on enrolment and retention</hd> <p>The COVID-19 pandemic caused unprecedented disruptions in higher education, affecting enrolment, retention, and student well-being. U.S. colleges experienced about a 13.1% decline in freshman enrolment in 2020, driven largely by health concerns, financial instability, and scepticism about the value of online learning (National Student Clearinghouse Research Center, [<reflink idref="bib37" id="ref53">37</reflink>], [<reflink idref="bib38" id="ref54">38</reflink>]). However, for some individuals, job losses and economic uncertainty prompted enrolment in higher education to upskill during a challenging job market (National Center for Education Statistics, [<reflink idref="bib35" id="ref55">35</reflink>]). Despite these varied motivations, financial stress—already a significant barrier to persistence—intensified due to the economic downturn, compounding mental health challenges and reducing academic engagement (Bennett et al., [<reflink idref="bib8" id="ref56">8</reflink>]; Goldrick-Rab et al., [<reflink idref="bib19" id="ref57">19</reflink>]; Rahlin et al., [<reflink idref="bib44" id="ref58">44</reflink>]; Russell et al., [<reflink idref="bib48" id="ref59">48</reflink>]).</p> <p>The pandemic also weakened social and academic integration, which are central to Tinto's retention model. Remote learning disrupted peer and faculty interactions, diminishing students' sense of belonging and connection to their institutions (Barringer et al., [<reflink idref="bib4" id="ref60">4</reflink>]; Resch et al., [<reflink idref="bib46" id="ref61">46</reflink>]; Sigurdardottir et al., [<reflink idref="bib50" id="ref62">50</reflink>]). Studies show that marginalized groups, including racial minorities and low-income students, were disproportionately affected by these disruptions (Brown et al., [<reflink idref="bib12" id="ref63">12</reflink>]; Sharaievska et al., [<reflink idref="bib49" id="ref64">49</reflink>]). Limited access to suitable study spaces and reliable technology further hindered academic engagement and performance in a distance-learning environment (Chiodaroli et al., [<reflink idref="bib15" id="ref65">15</reflink>]; Hagedorn et al., [<reflink idref="bib21" id="ref66">21</reflink>]).</p> <p>Despite these obstacles, targeted institutional interventions helped mitigate some of the pandemic's adverse effects. Expanded financial aid programs and structured support services improved retention outcomes (Swani et al., [<reflink idref="bib52" id="ref67">52</reflink>]; Parvez et al., [<reflink idref="bib43" id="ref68">43</reflink>]). These findings underscore the importance of comprehensive strategies that address financial, academic, and social barriers to support student success during and beyond periods of disruption. Moreover, recent machine learning research highlights how predictive analytics can guide such interventions by identifying at-risk students early (Colpo et al., [<reflink idref="bib16" id="ref69">16</reflink>]), particularly in contexts where financial strain is a key determinant of dropout risk (Bean, [<reflink idref="bib5" id="ref70">5</reflink>]).</p> <hd id="AN0191500347-6">Aim and research questions</hd> <p>This study investigates the factors influencing first-year student retention, focusing on demographic, academic, and social, and financial dimensions. It also explores how the COVID-19 pandemic has shaped retention patterns and employs machine learning models to identify students at risk of dropping out. Specifically, this study addresses the following research questions:</p> <p></p> <ulist> <item> What demographic, academic, social, and financial factors influence first-year student retention?</item> <p></p> <item> What impact does the COVID-19 pandemic have on first-year retention?</item> <p></p> <item> How do retention outcomes differ between pre-COVID-19 and post-COVID-19 cohorts, particularly regarding financial need, family financial contributions, and financial aid?</item> <p></p> <item> What role do different types of financial aid (e.g., grants, loans, scholarships) play in influencing retention outcomes across cohorts?</item> <p></p> <item> How accurately can machine learning models predict first-year retention, and which model best identifies students at risk of dropping out?</item> </ulist> <hd id="AN0191500347-7">Materials and methods</hd> <p></p> <hd id="AN0191500347-8">Data</hd> <p>The data for this study were obtained from institutional records of 8,320 first-time freshmen enrolled at a public university in the central United States. The students were divided into two cohorts: pre-COVID-19 (Fall 2018, <emph>N</emph> = 4,166) and post-COVID-19 (Fall 2020, <emph>N</emph> = 4,154). The outcome variable, retention rate, was defined as continued enrolment in the fall semester of the students' second year. Retention for the pre-COVID-19 cohort was measured in Fall 2019, and for the post-COVID-19 cohort, it was measured in Fall 2021.</p> <p>Key demographic factors considered included age at first enrolment, gender, residency status (in-state or out-of-state), and legacy status (whether a parent or grandparent graduated from the institution). Residency status also had financial implications, as out-of-state tuition including for international students, can exceed in-state tuition by more than twice the amount.</p> <p>Academic factors included whether a student earned college credits at another institution after secondary school and whether they earned college credits prior to matriculation from secondary school. Students with fewer than 12 h of transfer credit were classified as first-year freshmen. Other factors included whether the student declared a major upon entry, the number of credits attempted in the first semester, and first-term GPA. Social engagement was measured by membership in a fraternity or sorority (Greek Organization). Greek membership reflects social integration and may also indirectly indicate financial capacity, as associated costs can amount to thousands of dollars per semester.</p> <p>Financial factors included the Total Family Contribution (FAFSA_TFC) derived from the Free Application for Federal Student Aid, which estimates a family's financial capacity to contribute toward educational costs. An institutional need rank (Need Rank) categorized students' financial need based on FAFA_TFC and residency status. Additional financial variables included the types and amounts of aid received during the first year, such as federal and private loans, grants, tuition waivers, and scholarships.</p> <p>To address missing data, multiple imputation was employed (Rubin, [<reflink idref="bib47" id="ref71">47</reflink>]) to preserve a complete dataset and reduce potential bias associated with missing values. Log transformations were applied to financial variables, including FAFSA_TFC, loan amounts, grants, tuition waivers, and scholarships to address skewed distributions and scaling issues. Detailed variable definitions and summary statistics for the sub-samples are presented in Table 1.</p> <p>Table 1 Variable definitions and summary statistics (<emph>N</emph> = 8,320)</p> <p> <ephtml> <table rules="groups"><thead><tr><th align="left"><p>Variable</p></th><th align="left"><p>Definition</p></th><th align="left" colspan="2"><p>Pre-COVID-19</p></th><th align="left" colspan="2"><p>Post-COVID-19</p></th></tr></thead><tbody><tr><td align="left" colspan="6"><p>Outcome variable</p></td></tr><tr><td align="left"><p> First-year retention</p></td><td align="left"><p>% of students who enrolled during the subsequent fall semester</p></td><td align="left"><p>0.805</p></td><td align="left"><p>(0.397)</p></td><td align="left"><p>0.811</p></td><td align="left"><p>(0.391)</p></td></tr><tr><td align="left" colspan="6"><p>Explanatory variable</p></td></tr><tr><td align="left"><p> Age</p></td><td align="left"><p>Age in years at first enrolment</p></td><td align="left"><p>18.546</p></td><td align="left"><p>(0.841)<sup>a</sup></p></td><td align="left"><p>18.541</p></td><td align="left"><p>(0.913)</p></td></tr><tr><td align="left"><p> Male</p></td><td align="left"><p>1 if male, 0 if female</p></td><td align="left"><p>0.476</p></td><td align="left"><p>(0.499)</p></td><td align="left"><p>0.462</p></td><td align="left"><p>(0.499)</p></td></tr><tr><td align="left"><p> Out-of-state resident</p></td><td align="left"><p>1 if out-of-state or international resident, 0 in-state resident</p></td><td align="left"><p>0.323</p></td><td align="left"><p>(0.468)</p></td><td align="left"><p>0.337</p></td><td align="left"><p>(0.473)</p></td></tr><tr><td align="left"><p> Legacy</p></td><td align="left"><p>1 if student's parent (grandparent) graduated from the institution, 0 if not</p></td><td align="left"><p>0.294</p></td><td align="left"><p>(0.456)</p></td><td align="left"><p>0.296</p></td><td align="left"><p>(0.456)</p></td></tr><tr><td align="left"><p> Transfer credit</p></td><td align="left"><p>1 if student with credit hours from another institutions, 0 if not</p></td><td align="left"><p>0.386</p></td><td align="left"><p>(0.487)</p></td><td align="left"><p>0.393</p></td><td align="left"><p>(0.489)</p></td></tr><tr><td align="left"><p> Concurrent credit</p></td><td align="left"><p>1 if student with credit hours prior to high school graduation, 0 if not</p></td><td align="left"><p>0.462</p></td><td align="left"><p>(0.499)</p></td><td align="left"><p>0.501</p></td><td align="left"><p>(0.500)</p></td></tr><tr><td align="left"><p> Undeclared major</p></td><td align="left"><p>1 if freshman undeclared major and fully admissible, 0 if not</p></td><td align="left"><p>0.036</p></td><td align="left"><p>(0.185)</p></td><td align="left"><p>0.029</p></td><td align="left"><p>(0.169)</p></td></tr><tr><td align="left"><p> First-term credits attempted</p></td><td align="left"><p>Number of hours registered for the first term</p></td><td align="left"><p>13.498</p></td><td align="left"><p>(2.344)</p></td><td align="left"><p>14.017</p></td><td align="left"><p>(2.160)</p></td></tr><tr><td align="left"><p> First-term GPA</p></td><td align="left"><p>First-term grade point average</p></td><td align="left"><p>3.015</p></td><td align="left"><p>(0.967)</p></td><td align="left"><p>3.095</p></td><td align="left"><p>(1.013)</p></td></tr><tr><td align="left"><p> Greek organization</p></td><td align="left"><p>1 if participating in Greek Organization, 0 if not</p></td><td align="left"><p>0.289</p></td><td align="left"><p>(0.453)</p></td><td align="left"><p>0.255</p></td><td align="left"><p>(0.436)</p></td></tr><tr><td align="left"><p> FAFSA_TFC</p></td><td align="left"><p>Total amount of family contribution according to FAFSA ($)</p></td><td align="left"><p>50,745</p></td><td align="left"><p>(94,421)</p></td><td align="left"><p>53,060</p></td><td align="left"><p>(92,267)</p></td></tr><tr><td align="left"><p> Need rank</p></td><td align="left"><p>A scale of the level of financial need (0 = no need, 5 = very needy)</p></td><td align="left"><p>2.108</p></td><td align="left"><p>(1.865)</p></td><td align="left"><p>2.035</p></td><td align="left"><p>(1.811)</p></td></tr><tr><td align="left"><p> Federal loan amount</p></td><td align="left"><p>Total amount of federal loans received ($)</p></td><td align="left"><p>2,906</p></td><td align="left"><p>(5,150)</p></td><td align="left"><p>2,844</p></td><td align="left"><p>(5,130)</p></td></tr><tr><td align="left"><p> Private loan amount</p></td><td align="left"><p>Total amount of private loans received ($)</p></td><td align="left"><p>3,235</p></td><td align="left"><p>(7,667)</p></td><td align="left"><p>3,353</p></td><td align="left"><p>(8,052)</p></td></tr><tr><td align="left"><p> Grant amount</p></td><td align="left"><p>Total amount of grants received ($)</p></td><td align="left"><p>1,299</p></td><td align="left"><p>(2,324)</p></td><td align="left"><p>1,681</p></td><td align="left"><p>(2,799)</p></td></tr><tr><td align="left"><p> Tuition waiver amount</p></td><td align="left"><p>Total amount of tuition waivers received ($)</p></td><td align="left"><p>3,624</p></td><td align="left"><p>(4,387)</p></td><td align="left"><p>4,352</p></td><td align="left"><p>(4,785)</p></td></tr><tr><td align="left"><p> Scholarship amount</p></td><td align="left"><p>Total amount of scholarships received ($)</p></td><td align="left"><p>2,698</p></td><td align="left"><p>(4,337)</p></td><td align="left"><p>3,346</p></td><td align="left"><p>(4,615)</p></td></tr><tr><td align="left"><p> Number of observations</p></td><td align="left" /><td align="left" colspan="2"><p>4,166</p></td><td align="left" colspan="2"><p>4,154</p></td></tr></tbody></table> </ephtml> </p> <p> <sups>a</sups> Numbers in parentheses are standard deviations</p> <hd id="AN0191500347-9">Method</hd> <p></p> <hd id="AN0191500347-10">Logistic regression analysis</hd> <p>To investigate the factors influencing first-year student retention, with a particular focus on the impact of the COVID-19, a logistic regression analysis was employed. This method is well-suited for modelling a binary dependent variable, where 1 represents students who continue enrolment into their second year, and 0 represents those who do not. The logit model is specified as:</p> <olist> <item> <ephtml> <math display="block" xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mi>l</mi><mi>o</mi><mi>g</mi><mi>i</mi><mi>t</mi><mfenced close=")" open="("><mi mathvariant="italic">Pr</mi><mfenced close=")" open="("><msub><mi>Y</mi><mi>i</mi></msub><mo>=</mo><mn>1</mn></mfenced></mfenced><mo>=</mo><msub><mi>β</mi><mrow><mn>0</mn><mo>+</mo></mrow></msub><msub><mi>β</mi><mn>1</mn></msub><msub><mi>x</mi><mn>1</mn></msub><mo>+</mo><mo>⋯</mo><mo>+</mo><msub><mi>β</mi><mi>k</mi></msub><msub><mi>x</mi><mi>k</mi></msub><mo>+</mo><mi>γ</mi><mi>P</mi><mi>o</mi><mi>s</mi><mi>t</mi><mi>_</mi><mi>C</mi><mi>O</mi><mi>V</mi><mi>I</mi><mi>D</mi><mi>_</mi><mn>19</mn><mo>,</mo></mrow></math> </ephtml> </item> </olist> <p>where <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mi mathvariant="italic">Pr</mi><mfenced close=")" open="("><msub><mi>Y</mi><mi>i</mi></msub><mo>=</mo><mn>1</mn></mfenced></mrow></math> </ephtml> denotes the probability of a student <emph>i</emph> being retained in the fall semester of their second year. The independent variables <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mfenced close=")" open="("><msub><mi>x</mi><mn>1</mn></msub><msub><mrow><mo>,</mo><mo>⋯</mo><mo>,</mo><mi>x</mi></mrow><mi>k</mi></msub></mfenced></math> </ephtml> include demographic, academic, social, and financial attributes, capturing various dimensions of retention. The variable <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mrow><mi>P</mi><mi>o</mi><mi>s</mi><mi>t</mi><mi>_</mi><mi>C</mi><mi>O</mi><mi>V</mi><mi>I</mi><mi>D</mi><mi>_</mi><mn>19</mn></mrow></math> </ephtml> is a binary indicator, taking a value of 1 if the student first enrolled during Fall 2020 (post-COVID-19 cohort) and 0 otherwise. The coefficients <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mfenced close=")" open="("><msub><mi>β</mi><mn>0</mn></msub><mo>,</mo><mspace width="0.277778em" /><msub><mi>β</mi><mn>1</mn></msub><mo>,</mo><mo>⋯</mo><mo>,</mo><msub><mi>β</mi><mi>k</mi></msub><mo>,</mo><mspace width="0.277778em" /><mi>γ</mi></mfenced></math> </ephtml> are estimated through the analysis, quantifying the effects of each independent variable on the log-odds of retention. This approach enables a detailed examination of retention predictors while accounting for the distinct impact of the COVID-19 pandemic on first-year retention outcomes.</p> <hd id="AN0191500347-11">Predictive analysis</hd> <p>To identify students at risk of dropping out after their first year, machine learning algorithms were applied through the following steps:</p> <p></p> <ulist> <item> Data Splitting: The dataset was randomly divided into training and testing subsets. Approximately 80% of the data was allocated to the training set for model development and parameter tuning, while the remaining 20% was reserved for testing and model evaluation.</item> <p></p> <item> Model Building: Three commonly used classification algorithms were employed: logistic regression, Random Forest, and Support Vector Machine (SVM), following the methodology outlined by Delen ([<reflink idref="bib17" id="ref72">17</reflink>]).</item> <p></p> <item> Model Evaluation: The predictive performance of each model was evaluated using confusion matrices, which provided key metrics such as overall accuracy, sensitivity, and specificity.</item> </ulist> <p>After evaluating all three models, the one with the highest predictive accuracy was selected to calculate likelihood scores for individual students, estimating their probability of continuing enrolment into the fall semester of their second year. These likelihood scores were then analysed to assess the effects of students' financial need, family financial contributions toward educational costs, and financial aid on retention probabilities comparing pre-COVID-19 and post-COVID-19 cohorts. Additionally, the scores were used to examine variations in retention probabilities across financial factors and cohorts, offering a framework for understanding how financial stability and institutional support shape student persistence.</p> <hd id="AN0191500347-12">Results</hd> <p></p> <hd id="AN0191500347-13">Factors influencing first-year student retention</hd> <p>Table 2 presents the estimated coefficients, odds ratios, and 95% confidence intervals of the odds ratios for factors influencing first-year student retention. After controlling for other relevant variables, the results indicate that students who initially enrolled during the COVID-19 pandemic were approximately 13.4% less likely to persist into their second year. This finding suggests that the pandemic negatively impacted students' persistence, likely due to disruptions in learning and financial challenges experienced during that period.</p> <p>Table 2 Effect of the post-COVID-19 on first-year student retention</p> <p> <ephtml> <table rules="groups"><thead><tr><th align="left"><p>Variable</p></th><th align="left"><p>Estimate</p></th><th align="left"><p>SE</p></th><th align="left"><p>Odds ratio</p></th><th align="left"><p>95% <italic>CI</italic></p></th></tr></thead><tbody><tr><td align="left"><p>Intercept</p></td><td char="." align="char"><p> − 1.738<sup>***</sup></p></td><td align="left"><p>(0.670)</p></td><td char="." align="char"><p>0.176</p></td><td char="–" align="char"><p>[0.046 – 0.652]</p></td></tr><tr><td align="left"><p>Post-COVID-19</p></td><td char="." align="char"><p> − 0.144<sup>**</sup></p></td><td align="left"><p>(0.070)</p></td><td char="." align="char"><p>0.866</p></td><td char="–" align="char"><p>[0.755 – 0.993]</p></td></tr><tr><td align="left"><p>Age</p></td><td char="." align="char"><p> − 0.061<sup>*</sup></p></td><td align="left"><p>(0.033)</p></td><td char="." align="char"><p>0.941</p></td><td char="–" align="char"><p>[0.881 – 1.006]</p></td></tr><tr><td align="left"><p>Male</p></td><td char="." align="char"><p>0.194<sup>***</sup></p></td><td align="left"><p>(0.070)</p></td><td char="." align="char"><p>1.214</p></td><td char="–" align="char"><p>[1.058 – 1.393]</p></td></tr><tr><td align="left"><p>Out-of-state resident</p></td><td char="." align="char"><p> − 0.138<sup>*</sup></p></td><td align="left"><p>(0.080)</p></td><td char="." align="char"><p>0.871</p></td><td char="–" align="char"><p>[0.745 – 1.018]</p></td></tr><tr><td align="left"><p>Legacy</p></td><td char="." align="char"><p>0.280<sup>***</sup></p></td><td align="left"><p>(0.083)</p></td><td char="." align="char"><p>1.323</p></td><td char="–" align="char"><p>[1.125 – 1.558]</p></td></tr><tr><td align="left"><p>Transfer credit</p></td><td char="." align="char"><p>0.276<sup>***</sup></p></td><td align="left"><p>(0.080)</p></td><td char="." align="char"><p>1.318</p></td><td char="–" align="char"><p>[1.127 – 1.544]</p></td></tr><tr><td align="left"><p>Concurrent credit</p></td><td char="." align="char"><p>0.324<sup>***</sup></p></td><td align="left"><p>(0.072)</p></td><td char="." align="char"><p>1.383</p></td><td char="–" align="char"><p>[1.201 – 1.593]</p></td></tr><tr><td align="left"><p>Undeclared major</p></td><td char="." align="char"><p> − 0.392<sup>**</sup></p></td><td align="left"><p>(0.173)</p></td><td char="." align="char"><p>0.676</p></td><td char="–" align="char"><p>[0.485 – 0.954]</p></td></tr><tr><td align="left"><p>First-term credits attempted</p></td><td char="." align="char"><p>0.040<sup>***</sup></p></td><td align="left"><p>(0.015)</p></td><td char="." align="char"><p>1.040</p></td><td char="–" align="char"><p>[1.010 – 1.072]</p></td></tr><tr><td align="left"><p>First-term GPA</p></td><td char="." align="char"><p>1.105<sup>***</sup></p></td><td align="left"><p>(0.036)</p></td><td char="." align="char"><p>3.021</p></td><td char="–" align="char"><p>[2.815 – 3.246]</p></td></tr><tr><td align="left"><p>Greek organization</p></td><td char="." align="char"><p>0.993<sup>***</sup></p></td><td align="left"><p>(0.096)</p></td><td char="." align="char"><p>2.698</p></td><td char="–" align="char"><p>[2.239 – 3.267]</p></td></tr><tr><td align="left"><p>ln FAFSA_TFC</p></td><td char="." align="char"><p>0.012<sup>**</sup></p></td><td align="left"><p>(0.005)</p></td><td char="." align="char"><p>1.012</p></td><td char="–" align="char"><p>[1.002 – 1.023]</p></td></tr><tr><td align="left"><p>Need rank</p></td><td char="." align="char"><p> − 0.231<sup>***</sup></p></td><td align="left"><p>(0.035)</p></td><td char="." align="char"><p>0.794</p></td><td char="–" align="char"><p>[0.741 – 0.850]</p></td></tr><tr><td align="left"><p>ln federal loan amount</p></td><td char="." align="char"><p>0.017<sup>**</sup></p></td><td align="left"><p>(0.009)</p></td><td char="." align="char"><p>1.018</p></td><td char="–" align="char"><p>[1.000 – 1.036]</p></td></tr><tr><td align="left"><p>ln private loan amount</p></td><td char="." align="char"><p>0.018<sup>*</sup></p></td><td align="left"><p>(0.010)</p></td><td char="." align="char"><p>1.018</p></td><td char="–" align="char"><p>[0.999 – 1.037]</p></td></tr><tr><td align="left"><p>ln grant amount</p></td><td char="." align="char"><p>0.066<sup>***</sup></p></td><td align="left"><p>(0.016)</p></td><td char="." align="char"><p>1.069</p></td><td char="–" align="char"><p>[1.036 – 1.102]</p></td></tr><tr><td align="left"><p>ln tuition waiver amount</p></td><td char="." align="char"><p>0.020<sup>**</sup></p></td><td align="left"><p>(0.009)</p></td><td char="." align="char"><p>1.020</p></td><td char="–" align="char"><p>[1.002 – 1.039]</p></td></tr><tr><td align="left"><p>ln scholarship amount</p></td><td char="." align="char"><p>0.076<sup>***</sup></p></td><td align="left"><p>(0.011)</p></td><td char="." align="char"><p>1.079</p></td><td char="–" align="char"><p>[1.057 – 1.101]</p></td></tr></tbody></table> </ephtml> </p> <p>Number of observations = 8,320 Single, double, and triple asterisks (*, **, ***) represent statistical significance at the 10%, 5%, and 1% levels, respectively</p> <p>After accounting for the impact of the pandemic, younger students demonstrated a slightly higher likelihood of retention, though this effect was only marginally significant. This finding suggests that older students may face additional challenges related to life circumstances or competing responsibilities. Male students were more likely to persist compared to female students, which may indicate that female students encounter specific barriers to continued enrolment. Out-of-state students, including international students, had a lower likelihood of retention compared to in-state residents, likely due to the financial burden of higher tuition costs and limited access to local support networks. In contrast, legacy students exhibited a higher likelihood of retention, suggesting that a family connection to the institution fosters a stronger sense of belonging and engagement.</p> <p>Academic factors influenced student retention. Students with prior college credits—whether earned after high school or through concurrent enrolment in high school—were more likely to be retained, suggesting that early exposure to college coursework enhances academic preparedness and persistence. Conversely, students who entered with undeclared majors were less likely to be retained likely due to challenges with academic focus, engagement, and motivation. Additionally, students who carried a heavier academic load and/or achieved a higher GPA in their first semester were more likely to persist. This underscores the critical importance of academic success in the first term as a key predictor of continued enrolment. Students who join a fraternity or sorority (Greek Organization) are more likely to be retained. This suggests that these organizations enhance retention by fostering social engagement and a sense of belonging. Additionally, the financial capacity to afford membership may indicate a level of financial stability that supports continued enrolment.</p> <p>Financial factors played a critical role in student retention. A higher expected family contribution (FAFSA_TFC) was associated with a slightly higher likelihood of retention, suggesting that greater financial stability reduces financial stress and supports continued enrolment. Conversely, students with a higher financial need rank were less likely to be retained, indicating that greater financial need poses challenges that hinder persistence. Federal loans were modestly linked to higher retention, potentially because student taking out these loans may feel more committed to their education due to their financial investment. Similarly, an increase in private loan amount was associated with higher retention, reflecting students' commitment to their education. Increases in grants, tuition waivers, and scholarships were also associated with slightly higher retention rates. These forms of financial aid not only alleviate financial burdens but also provide recognition and motivation, contributing to students' persistence.</p> <hd id="AN0191500347-14">Performance of predictive models</hd> <p>We evaluated three machine learning classification algorithms—logistic regression, Random Forest, and SVM—to assess their predictive accuracy for first-year student retention. All three models outperformed the baseline retention rate of 80.77% (Table 3). Among these, the Random Forest model demonstrated the highest accuracy at 89.18%, along with a sensitivity of 98.07% for identifying students likely to be retained and the highest specificity, correctly identifying 51.88% of students not retained. This balance between sensitivity and specificity is crucial for minimizing false positives.</p> <p>Table 3 Performance metrics of predictive models for first-year student retention</p> <p> <ephtml> <table rules="groups"><thead><tr><th align="left"><p>Confusion matrix</p></th><th align="left" colspan="2"><p>Logistic regression</p></th><th align="left" colspan="2"><p>Random forest</p></th><th align="left" colspan="2"><p>SVM<sup>a</sup></p></th></tr><tr><th align="left" /><th align="left" colspan="2"><p>Prediction</p></th><th align="left" colspan="2"><p>Prediction</p></th><th align="left" colspan="2"><p>Prediction</p></th></tr><tr><th align="left"><p>Actual</p></th><th align="left"><p>Yes</p></th><th align="left"><p>No</p></th><th align="left"><p>Yes</p></th><th align="left"><p>No</p></th><th align="left"><p>Yes</p></th><th align="left"><p>No</p></th></tr></thead><tbody><tr><td align="left"><p>Yes</p></td><td align="left"><p>1,307</p></td><td align="left"><p>37</p></td><td align="left"><p>1,318</p></td><td align="left"><p>26</p></td><td align="left"><p>1,324</p></td><td align="left"><p>20</p></td></tr><tr><td align="left"><p>No</p></td><td align="left"><p>175</p></td><td align="left"><p>145</p></td><td align="left"><p>154</p></td><td align="left"><p>166</p></td><td align="left"><p>185</p></td><td align="left"><p>135</p></td></tr><tr><td align="left"><p>Overall accuracy</p></td><td align="left" colspan="2"><p>87.26%</p></td><td align="left" colspan="2"><p>89.18%</p></td><td align="left" colspan="2"><p>87.68%</p></td></tr><tr><td align="left"><p>No information rate</p></td><td align="left" colspan="2"><p>80.77%</p></td><td align="left" colspan="2"><p>80.77%</p></td><td align="left" colspan="2"><p>80.77%</p></td></tr><tr><td align="left"><p>Sensitivity</p></td><td align="left" colspan="2"><p>97.25%</p></td><td align="left" colspan="2"><p>98.07%</p></td><td align="left" colspan="2"><p>98.51%</p></td></tr><tr><td align="left"><p>Specificity</p></td><td align="left" colspan="2"><p>45.31%</p></td><td align="left" colspan="2"><p>51.88%</p></td><td align="left" colspan="2"><p>42.19%</p></td></tr></tbody></table> </ephtml> </p> <p>Number of observations = 1,664 <sups>a</sups> SVM is Support Vector Machine</p> <p>In comparison, the logistic regression and SVM models achieved slightly lower accuracies, at 87.26% and 87.68%, respectively. The SVM model displayed the highest sensitivity at 98.51%, indicting strong performance in predicting students likely to continue, but its specificity was lower at 42.19%. The logistic regression model exhibited both lower sensitivity and specificity compared to the Random Forest model. While all three models effectively predicted first-year student retention, the Random Forest offered the best balance between accuracy and sensitivity, making it the most suitable for our analysis. Nevertheless, further refinement is necessary to improve the identification of students at risk of not being retained.</p> <hd id="AN0191500347-15">First-year student retention pre- and post-COVID-19</hd> <p>We analysed first-year student retention pre- and post-COVID-19 applying the Random Forest model to the complete dataset (<emph>N</emph> = 8,320). The model computed likelihood scores for each student, estimating the probability of continuing enrolment into their second year. It achieved an accuracy of 97.84%, with a sensitivity of 99.61% and a specificity of 90.36% (Fig. 1).</p> <p>Graph: Fig. 1 Confusion matrix of Random Forest model predicting first-year student retention (N = 8,320)</p> <p>We examined the relationship between students' financial need and first-year retention (Fig. 2). Retention likelihood was higher in lower need rank and decreased as need rank increased. The post-COVID-19 cohort showed similar or higher retention rates in the lower need ranks (0–2) but exhibited comparable (rank 3) or slightly lower retention rates in the higher need ranks (4–5) compared to the pre-COVID-19 cohort. Students in higher need ranks faced persistent challenges, underscoring the necessity for enhanced interventions to support financially vulnerable students in the aftermath of the pandemic.</p> <p>Graph: Fig. 2 First-year student retention by need rank between pre- and post-COVID-19</p> <p>To further explore the impact of financial variables on student retention, we categorized expected family financial contributions (FAFSA-TFC), total federal and private loan amounts, and total tuition assistance (including grants, tuition waivers, and scholarships) into practical ranges. This approach aimed to identify variations in retention likelihood based on different levels of financial contributions, debt, or aid received (Fig. 3).</p> <p>Graph: Fig. 3 Distributions of first-year student retention between pre- and post-COVID-19 based on varying financial factors</p> <p>Higher FAFSA-TFC values were strongly associated with increased retention, with the post-COVID-19 cohort showing slightly higher retention rates and reduced variability, particularly in the highest contribution category (> $30,000). Loan amounts had minimal overall impact on retention. However, in the middle loan range ($10,000 − $20,000], the pre-COVID-19 cohort displayed a slight advantage in retention with less variability. Conversely, in the high loan range (> $20,000), the post-COVID-19 cohort demonstrated higher retention rates and reduced variability. These results suggest that students in the post-COVID-19 cohort with higher family contributions or substantial loans may have benefited from increased financial stability or pandemic-related support measures contributing to improved retention outcomes.</p> <p>Similarly, higher levels of total financial assistance amounts were positively associated with retention rates, with both cohorts showing a clear trend of increased retention as financial aid levels rose. In the "no aid" category, retention rates were comparable between cohorts. However, in the high financial assistance category, the post-COVID-19 cohort demonstrated higher retention rates but exhibited slightly increased variability. These findings underscore the critical role of financial assistance in promoting student retention and suggest that the COVID-19 pandemic influenced retention patterns, particularly among students receiving substantial financial aid. Overall, these results emphasize the importance of sustained or expanded financial aid programs in the post-pandemic period. Such support appears to mitigate financial challenges and enhance retention outcomes, particularly for students with significant financial need.</p> <hd id="AN0191500347-16">Discussion and conclusion</hd> <p>This study examined how demographic, academic, social, and financial factors influenced first-year student retention, focusing on the impact of the COVID-19 pandemic. The findings indicate that the pandemic negatively affected student persistence, with students who enrolled during this period being less likely to continue into their second year. This underscores challenges such as learning disruptions, limited social integration, and heightened financial pressures. However, job losses and economic uncertainty may have prompted some individuals to pursue higher education to upskill, potentially offsetting attrition rates. Institutional recruitment strategies, such as test-optional admissions and virtual events, also likely influenced enrolment patterns, complicating the net effects of the pandemic on retention. While our data identify key trends, they do not capture the nuanced motivations behind individual enrolment decisions, limiting the ability to fully isolate these factors.</p> <p>In response to the pandemic, universities in the U.S. implemented various measures to support students, including emergency financial aid programs like the CARES Act (HEERF I), CRRSAA (HEERF II), and ARP (HEERF III) (National Association of Student Financial Aid Administrators, [<reflink idref="bib34" id="ref73">34</reflink>]). At this institution, these grants helped students cover essential expenses such as tuition, housing, and technology, highlighting the importance of tailored financial aid programs in addressing the needs of vulnerable populations during crises.</p> <p>The study also revealed gender-based differences in retention. Female students were less likely to persist, consistent with research suggesting the pandemic disrupted the formation of social tie, particularly for women. Women lost their relative advantage in building connections (Sigurdardottir et al., [<reflink idref="bib50" id="ref74">50</reflink>]) and may have taken on a larger share of caregiving responsibilities during this time. Legacy students exhibited higher retention rates, reflecting the importance of strong institutional ties. However, the transition to online learning limited opportunities for meaningful integration, weakening students' sense of belonging, particularly among first-year students. Institutions should prioritize fostering social connections and a sense of belonging during disruptive periods.</p> <p>Academic factors, such as prior college credits, declared majors, and strong first-term GPAs emerged as significant predictors of retention, emphasizing the importance of early academic success. Participation in Greek Organizations was also positively associated with retention suggesting that structured social engagement fosters belonging and commitment. However, as Wilcox et al. ([<reflink idref="bib58" id="ref75">58</reflink>]) noted, informal social networks and supportive communities are equally vital for retention and institutions should offer diverse opportunities for engagement.</p> <p>Financial factors played a pivotal role in retention. Higher family contributions, loans, and financial aid were positively associated with persistence, underscoring the importance of financial stability in reducing stress and supporting continued enrolment (Cabrera et al., [<reflink idref="bib13" id="ref76">13</reflink>]; Millea et al., [<reflink idref="bib30" id="ref77">30</reflink>]). Notably, students in the post-COVID-19 cohort showed improved retention rates in higher financial aid categories suggesting that pandemic-related interventions effectively mitigated some challenges.</p> <p>The evaluation of predictive models showed that the Random Forest model achieved the highest accuracy and balance between sensitivity and specificity. However, further refinement is needed to improve the identification of students at risk of not persisting particularly those facing social and financial barriers. Early warning systems informed by predictive models could help institutions proactively support at-risk students.</p> <p>This study highlights the need to expand how social factors are measured in higher education. As Tight ([<reflink idref="bib53" id="ref78">53</reflink>]) emphasizes, retention and engagement are deeply interconnected requiring institutions to adopt holistic strategies that address both academic and social aspects of the student experience. Institutions that actively foster environments promoting academic success, social belonging, and adequate financial support will be better equipped to enhance retention outcomes.</p> <p>This study has several limitations that may have influenced retention outcomes. Differences in institutional support systems, teaching methods, class sizes, and evaluation processes during the pandemic were not captured in the datasets. Institutions varied in their online transitions using strategies like asynchronous learning and online assessment (Koh & Daniel, [<reflink idref="bib25" id="ref79">25</reflink>]). Challenges such as reduced engagement and altered teaching methods during emergency remote teaching likely contributed to retention disparities (Wilhelm et al., [<reflink idref="bib59" id="ref80">59</reflink>]). Digital inequality and limited engagement also played significant roles (Khlaif et al., [<reflink idref="bib24" id="ref81">24</reflink>]). Additionally, while this study includes residency status, it does not indicate whether students lived on campus during online learning. Further research should examine how variations in institutional practices, teaching methods, and support systems influenced retention to better understand the interplay between these factors and student persistence during disruptions.</p> <p>By analysing retention patterns across cohorts and examining financial factors, this study provides valuable insights into addressing disparities and fostering persistence during and beyond periods of crisis. Future research should examine institutional differences and explore the motivations driving students to remain enrolled or leave an institution. A deeper understanding of these factors will enable the development of evidenced-based strategies to improve retention outcomes and support diverse student populations.</p> <hd id="AN0191500347-17">Authors' contributions</hd> <p>• K. Han: Conceptualization, Methodology, Statistical Analysis, Visualization, Writing – Original Draft, Writing – Review & Editing.</p> <p>• R. Pandey: Conceptualization, Methodology, Writing – Original Draft, Writing – Review & Editing.</p> <p>• K. Meints: Conceptualization, Methodology, Writing – Original Draft, Writing – Review & Editing.</p> <p>• L. Burns: Supervision, Conceptualization, Data Collection, Methodology, Writing – Original Draft, Writing – Review & Editing.</p> <p>• C. Chung: Methodology, Writing – Review & Editing.</p> <hd id="AN0191500347-18">Funding</hd> <p>None.</p> <hd id="AN0191500347-19">Data Availability</hd> <p>The data used in this study is available upon reasonable request.</p> <hd id="AN0191500347-20">Declarations</hd> <p></p> <hd id="AN0191500347-21">Ethics approval and consent to participate</hd> <p>Not applicable.</p> <hd id="AN0191500347-22">Competing interests</hd> <p>Not applicable.</p> <hd id="AN0191500347-23">Publisher's Note</hd> <p>Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.</p> <ref id="AN0191500347-24"> <title> References </title> <blist> <bibl id="bib1" idref="ref7" type="bt">1</bibl> <bibtext> Astin AW. What matters in college?. 1993; Jossey-Bass</bibtext> </blist> <blist> <bibl id="bib2" idref="ref18" type="bt">2</bibl> <bibtext> Aulck, L, Velagapudi, N, Blumenstock, J, & West, J. (2016). Predicting student dropout in higher education [Workshop] (v. 4). 2016 International Conference on Machine Learning, New York, NY, United States. https://doi.org/10.48550/arXiv.1606.06364</bibtext> </blist> <blist> <bibl id="bib3" idref="ref42" type="bt">3</bibl> <bibtext> Baker AR, Montalto CP. Student loan debt and financial stress: Implications for academic performance. Journal of College Student Development. 2019; 60; 1: 115-120. 10.1353/csd.2019.0008</bibtext> </blist> <blist> <bibl id="bib4" idref="ref60" type="bt">4</bibl> <bibtext> Barringer A, Papp LA, Gu P. College students' sense of belonging in times of disruption: Prospective changes from before to during the COVID-19 pandemic. Higher Education Research & Development. 2023; 42; 6: 1309-1322. 10.1080/07294360.2022.2138275</bibtext> </blist> <blist> <bibl id="bib5" idref="ref70" type="bt">5</bibl> <bibtext> Bean, J. P. (1990). Using retention research in enrollment management. In D. Hossler, J. P. Bean, & Associates (Eds.), The strategic management of enrollment (pp. 170–185). Jossey-Bass.</bibtext> </blist> <blist> <bibl id="bib6" idref="ref16" type="bt">6</bibl> <bibtext> Bean JP. Dropouts and turnover: The synthesis and test of a causal model of student attrition. Research in Higher Education. 1980; 12: 155-187. 10.1007/BF00976194</bibtext> </blist> <blist> <bibl id="bib7" idref="ref17" type="bt">7</bibl> <bibtext> Bean JP. Conceptual models of student attrition: How theory can help the institutional researcher. New Directions for Institutional Research. 1982; 1982; 36: 17-33. 10.1002/ir.37019823604</bibtext> </blist> <blist> <bibl id="bib8" idref="ref56" type="bt">8</bibl> <bibtext> Bennett J, Heron J, Kidger J, Linton M. Investigating change in student financial stress at a UK university: Multi-year survey analysis across a global pandemic and recession. Education Sciences. 2023; 13; 12: 1175. 10.3390/educsci13121175</bibtext> </blist> <blist> <bibl id="bib9" idref="ref34" type="bt">9</bibl> <bibtext> Biddix JP, Singer KI, Aslinger E. First-year retention and National Panhellenic Conference sorority membership: A multi-institutional study. Journal of College Student Retention: Research, Theory & Practice. 2018; 20; 2: 236-252. 10.1177/1521025116656633</bibtext> </blist> <blist> <bibtext> Bowman NA, Holmes JM. A quasi-experimental analysis of fraternity or sorority membership and college student success. Journal of College Student Development. 2017; 58; 7: 1018-1034. 10.1353/csd.2017.0081</bibtext> </blist> <blist> <bibtext> Britt, S. L, Ammerman, D. A, Barrett, S. F, & Jones, S. (2017). Student loans, financial stress, and college student retention. Journal of Student Financial Aid, 47(1), Article 3. https://eric.ed.gov/?id=EJ1141137. Accessed 15 Jan 2024.</bibtext> </blist> <blist> <bibtext> Brown JT, Kush JM, Volk FA. Centering the marginalized: The impact of the pandemic on online student retention. Journal of Student Financial Aid. 2022; 51; 1: Article 3. 10.55504/0884-9153.1777</bibtext> </blist> <blist> <bibtext> Cabrera AF, Nora A, Castaneda MB. The role of finances in the persistence process: A structural model. Research in Higher Education. 1992; 33; 5: 571-593. 10.1007/BF00973759</bibtext> </blist> <blist> <bibtext> Chen R. Institutional characteristics and college student dropout risks: A multilevel event history analysis. Research in Higher Education. 2012; 53; 5: 487-505. 10.1007/s11162-011-9241-4</bibtext> </blist> <blist> <bibtext> Chiodaroli M, Freyhult L, Solders A. "Every morning I take two steps to my desk...": Students' perspectives on distance learning during the COVID-19 pandemic. Higher Education. 2024; 88; 8: 1483-1502. 10.1007/s10734-023-01179-6</bibtext> </blist> <blist> <bibtext> Colpo MP, Primo TT, de Aguiar MS. Lessons learned from the student dropout patterns on COVID-19 pandemic: An analysis supported by machine learning. British Journal of Educational Technology. 2024; 55; 2: 560-585. 10.1111/bjet.13380</bibtext> </blist> <blist> <bibtext> Delen, D. (2010). A comparative analysis of machine learning techniques for student retention management. Decision Support Systems, 49(4), 498–506. https://doi.org/10.1016/j.dss.2010.06.003</bibtext> </blist> <blist> <bibtext> Gansemer-Topf AM, Schuh JH. Institutional selectivity and institutional expenditures: Examining organizational factors that contribute to retention and graduation. Research in Higher Education. 2006; 47; 6: 613-642. 10.1007/s11162-006-9009-4</bibtext> </blist> <blist> <bibtext> Goldrick-Rab, S, Coca, V, Kienzl, G, Welton, C. R, Dahl, S, Magnelia, S. (2020). #RealCollege during the pandemic: New evidence on basic needs insecurity and student well-being. Rebuilding the Launchpad: Serving Students During Covid Resource Library, (5). https://scholarworks.boisestate.edu/covid/5. Accesed 20 Oct 2024.</bibtext> </blist> <blist> <bibtext> Hagedorn LSSeidman A. How to define retention: A new look at an old problem. College student retention: Formula for student success. 2005; Praeger: 89-106</bibtext> </blist> <blist> <bibtext> Hagedorn RL, Wattick RA, Olfert MD. "My entire world stopped": College students' psychosocial and academic frustrations during the COVID-19 pandemic. Applied Research in Quality of Life. 2022; 17; 2: 1069-1090. 10.1007/s11482-021-09948-0</bibtext> </blist> <blist> <bibtext> Hoyt JE. Student connections: The critical role of student affairs and academic support services in retention efforts. Journal of College Student Retention: Research, Theory & Practice. 2023; 25; 3: 480-491. 10.1177/1521025121991502</bibtext> </blist> <blist> <bibtext> Karp MM, Calcagno JC, Hughes KL, Jeong DW, Bailey TR. The postsecondary achievement of participants in dual enrollment: An analysis of student outcomes in two states. 2007; Community College Research Center, Columbia University</bibtext> </blist> <blist> <bibtext> Khlaif ZN, Salha S, Kouraichi B. Emergency remote learning during COVID-19 crisis: Students' engagement. Education and Information Technologies. 2021; 26; 6: 7033-7055. 10.1007/s10639-021-10566-4</bibtext> </blist> <blist> <bibtext> Koh JHL, Daniel BK. Shifting online during COVID-19: A systematic review of teaching and learning strategies and their outcomes. International Journal of Educational Technology in Higher Education. 2022; 19; 1: 56. 10.1186/s41239-022-00361-7</bibtext> </blist> <blist> <bibtext> Kuh, G. D, Kinzie, J. L, Buckley, J. A, Bridges, B. K, & Hayek, J. C. (2006). What matters to student success: A review of the literature. National Postsecondary Education Cooperative. https://nces.ed.gov/npec/pdf/kuh%5fteam%5freport.pdf. Accessed 10 Jan 2024.</bibtext> </blist> <blist> <bibtext> Leppel K. The impact of major on college persistence among freshmen. Higher Education. 2001; 41; 3: 327-342. 10.1023/A:1004189906367</bibtext> </blist> <blist> <bibtext> Leveson L, McNeil N, Joiner T. Persist or withdraw: The importance of external factors in students' departure intentions. Higher Education Research & Development. 2013; 32; 6: 932-945. 10.1080/07294360.2013.806442</bibtext> </blist> <blist> <bibtext> Lorenzo-Quiles O, Galdón-López S, Lendínez-Turón A. Factors contributing to university dropout: A review. Frontiers in Education. 2023; 8: 1159864. 10.3389/feduc.2023.1159864Frontiers Media SA</bibtext> </blist> <blist> <bibtext> Millea, M, Wills, R, Elder, A, & Molina, D. (2018). What matters in college student success? Determinants of college retention and graduation rates. Education, Summer 2018, 138(4), 309–322. https://<ulink href="http://www.ingentaconnect.com/content/prin/ed/2018/00000138/00000004/art00003">www.ingentaconnect.com/content/prin/ed/2018/00000138/00000004/art00003</ulink>. Accessed 10 Jan 2024.</bibtext> </blist> <blist> <bibtext> Moores E, Burgess AP. Financial support differentially aids retention of students from households with lower incomes: A UK case study. Studies in Higher Education. 2023; 48; 1: 220-231. 10.1080/03075079.2022.2125950</bibtext> </blist> <blist> <bibtext> Murtaugh PA, Burns LD, Schuster J. Predicting the retention of university students. Research in Higher Education. 1999; 40; 3: 355-371. 10.1023/A:1018755201899</bibtext> </blist> <blist> <bibtext> Nandeshwar A, Menzies T, Nelson A. Learning patterns of university student retention. Expert Systems with Applications. 2011; 38; 12: 14984-14996. 10.1016/j.eswa.2011.05.048</bibtext> </blist> <blist> <bibtext> National Association of Student Financial Aid Administrators. (2022). Evaluating student and institutional experiences with HEERF: Insights from the COVID-19 emergency relief program. Retrieved December 8, 2024, from https://<ulink href="http://www.nasfaa.org/uploads/documents/Evaluating%5fStudent%5fand%5fInstitutional%5fExperiences%5fwith%5fHEERF.pdf">www.nasfaa.org/uploads/documents/Evaluating%5fStudent%5fand%5fInstitutional%5fExperiences%5fwith%5fHEERF.pdf</ulink></bibtext> </blist> <blist> <bibtext> National Center for Education Statistics. (2022). The Condition of Education 2022 (NCES 2022–144). U.S. Department of Education. Retrieved December 8, 2024, from https://nces.ed.gov/pubs2022/2022144.pdf</bibtext> </blist> <blist> <bibtext> National Center for Education Statistics. (2023). Table 331.95. Number and percentage of students enrolled at degree-granting postsecondary institutions, by level of institution and control of institution, residency and attendance status of student, and state or jurisdiction: Fall 2021. U.S. Department of Education. Retrieved December 19, 2024, from https://nces.ed.gov/programs/digest/d23/tables/dt23_331.95.asp</bibtext> </blist> <blist> <bibtext> National Student Clearinghouse Research Center. (2024a). Persistence and retention: Fall 2023 report. Retrieved December 8, 2024, from https://nscresearchcenter.org/persistence-retention</bibtext> </blist> <blist> <bibtext> National Student Clearinghouse Research Center. (2024b). Current term enrollment estimates: Spring 2024. Retrieved December 8, 2024, from https://nscresearchcenter.org/current-term-enrollment-estimates/</bibtext> </blist> <blist> <bibtext> Nieuwoudt JE, Pedler ML. Student retention in higher education: Why students choose to remain at university. Journal of College Student Retention: Research, Theory & Practice. 2023; 25; 2: 326-349. 10.1177/1521025120985228</bibtext> </blist> <blist> <bibtext> Olbrecht AM, Romano C, Teigen J. How money helps keep students in college: The relationship between family finances, merit-based aid, and retention in higher education. Journal of Student Financial Aid. 2016; 46; 1: Article 2. 10.55504/0884-9153.1548</bibtext> </blist> <blist> <bibtext> Organisation for Economic Co-operation and Development. (2022). Education at a glance 2022: OECD indicators. OECD Publishing. https://doi.org/10.1787/69096873-en</bibtext> </blist> <blist> <bibtext> Ortiz-Lozano JM, Rua-Vieites A, Bilbao-Calabuig P, Casadesús-Fa M. University student retention: Best time and data to identify undergraduate students at risk of dropout. Innovations in Education and Teaching International. 2018; 57; 1: 74-85. 10.1080/14703297.2018.1502090</bibtext> </blist> <blist> <bibtext> Parvez, R, Tarantino, A, & Meerza, S. I. A. (2023). Understanding the prediction of student retention behavior during covid-19 using effective data mining techniques. ResearchGate. https://doi.org/10.21203/rs.3.rs-2948727/v1</bibtext> </blist> <blist> <bibtext> Rahlin, N. A, Suriawaty, A, Bahkiar Bahkiar, S. A, Turan, S, spsampsps Fauzi, S. N. A. (2024). Global crucial risk factors associated stress among university students during post Covid-19 pandemic: Empirical evidence from Asian country. In Studies in systems, decision, and control (Vol. 487, pp. 135–327). https://doi.org/10.1007/978-3-031-35828-9_27</bibtext> </blist> <blist> <bibtext> Reason RD. Student variables that predict retention: Recent research and new developments. NASPA Journal. 2009; 46; 3: 482-501. 10.2202/1949-6605.5022</bibtext> </blist> <blist> <bibtext> Resch K, Alnahdi G, Schwab S. Exploring the effects of the COVID-19 emergency remote education on students' social and academic integration in higher education in Austria. Higher Education Research & Development. 2023; 42; 1: 215-229. 10.1080/07294360.2022.2040446</bibtext> </blist> <blist> <bibtext> Rubin DB. Multiple imputation for nonresponse in surveys. 2004; Wiley; 81</bibtext> </blist> <blist> <bibtext> Russell, M. B, Head, L. S. W, Wolfe-Enslow, K, Holland, J, & Zimmerman, N. (2022). The COVID-19 effect: How student financial well-being, needs satisfaction, and college persistence has changed. Journal of College Student Retention: Research, Theory & Practice. https://doi.org/10.1177/15210251221133767</bibtext> </blist> <blist> <bibtext> Sharaievska I, McAnirlin O, Browning MHEM, Larson LR, Mullenbach L, Rigolon A, D'Antonio A, Cloutier S, Thomsen J, Metcalf EC, Reigner N. "Messy transitions": Students' perspectives on the impacts of the COVID-19 pandemic on higher education. Higher Education. 2022. 10.1007/s10734-022-00843-7</bibtext> </blist> <blist> <bibtext> Sigurdardottir MS, Torfason MT, Jonsdottir AH. Social tie formation of COVID-19 students: Evidence from a two-cohort longitudinal study. Higher Education. 2023; 86; 3: 333-351. 10.1007/s10734-022-00935-4</bibtext> </blist> <blist> <bibtext> Stewart, S, Lim, D. H, & Kim, J. (2015). Factors influencing college persistence for first-time students. Journal of Developmental Education, 38(3), 12–20. <ulink href="http://www.jstor.org/stable/24614019">http://www.jstor.org/stable/24614019</ulink>. Accessed 20 Oct 2024.</bibtext> </blist> <blist> <bibtext> Swani K, Wamwara W, Goodrich K, Schiller S, Dinsmore J. Understanding business student retention during Covid-19: Roles of service quality, college brand, and academic satisfaction, and stress. Services Marketing Quarterly. 2022; 43; 3: 329-352. 10.1080/15332969.2021.1993559</bibtext> </blist> <blist> <bibtext> Tight M. Student retention and engagement in higher education. Journal of Further and Higher Education. 2019; 44; 5: 689-704. 10.1080/0309877X.2019.1576860</bibtext> </blist> <blist> <bibtext> Tinto V. Leaving college: Rethinking the causes and cures of student attrition. 19932; University of Chicago Press</bibtext> </blist> <blist> <bibtext> Tinto V. Completing college: Rethinking institutional action. 2012; The University of Chicago Press. 10.7208/chicago/9780226804545.001.0001</bibtext> </blist> <blist> <bibtext> U.S. Department of Education. (2022). The condition of education. National Center for Education Statistics. Retrieved December 8, 2024, from https://nces.ed.gov/programs/coe/</bibtext> </blist> <blist> <bibtext> Webster AL, Showers VE. Measuring predictors of student retention rates. American Journal of Economics and Business Administration. 2011; 3; 2: 301-311. 10.3844/ajebasp.2011.296.306</bibtext> </blist> <blist> <bibtext> Wilcox P, Winn S, Fyvie-Gauld M. 'It was nothing to do with the university, it was just the people': The role of social support in the first-year experience of higher education. Studies in Higher Education. 2005; 30; 6: 707-722. 10.1080/03075070500340036</bibtext> </blist> <blist> <bibtext> Wilhelm J, Mattingly S, Gonzalez VH. Perceptions, satisfactions, and performance of undergraduate students during Covid-19 emergency remote teaching. Anatomical Sciences Education. 2022; 15; 1: 42-56. 10.1002/ase.2161</bibtext> </blist> </ref> <aug> <p>By Kwideok Han; Ranjit Pandey; Kimberly Meints; Larry Burns and Chanjin Chung</p> <p>Reported by Author; Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib20" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib54" firstref="ref2"></nolink> <nolink nlid="nl3" bibid="bib18" firstref="ref3"></nolink> <nolink nlid="nl4" bibid="bib37" firstref="ref5"></nolink> <nolink nlid="nl5" bibid="bib38" firstref="ref6"></nolink> <nolink nlid="nl6" bibid="bib56" firstref="ref10"></nolink> <nolink nlid="nl7" bibid="bib41" firstref="ref11"></nolink> <nolink nlid="nl8" bibid="bib36" firstref="ref12"></nolink> <nolink nlid="nl9" bibid="bib57" firstref="ref13"></nolink> <nolink nlid="nl10" bibid="bib55" firstref="ref15"></nolink> <nolink nlid="nl11" bibid="bib17" firstref="ref19"></nolink> <nolink nlid="nl12" bibid="bib29" firstref="ref20"></nolink> <nolink nlid="nl13" bibid="bib30" firstref="ref21"></nolink> <nolink nlid="nl14" bibid="bib32" firstref="ref22"></nolink> <nolink nlid="nl15" bibid="bib45" firstref="ref23"></nolink> <nolink nlid="nl16" bibid="bib23" firstref="ref29"></nolink> <nolink nlid="nl17" bibid="bib27" firstref="ref30"></nolink> <nolink nlid="nl18" bibid="bib42" firstref="ref32"></nolink> <nolink nlid="nl19" bibid="bib51" firstref="ref33"></nolink> <nolink nlid="nl20" bibid="bib10" firstref="ref35"></nolink> <nolink nlid="nl21" bibid="bib22" firstref="ref38"></nolink> <nolink nlid="nl22" bibid="bib31" firstref="ref40"></nolink> <nolink nlid="nl23" bibid="bib40" firstref="ref41"></nolink> <nolink nlid="nl24" bibid="bib11" firstref="ref43"></nolink> <nolink nlid="nl25" bibid="bib26" firstref="ref44"></nolink> <nolink nlid="nl26" bibid="bib33" firstref="ref45"></nolink> <nolink nlid="nl27" bibid="bib39" firstref="ref46"></nolink> <nolink nlid="nl28" bibid="bib28" firstref="ref47"></nolink> <nolink nlid="nl29" bibid="bib13" firstref="ref49"></nolink> <nolink nlid="nl30" bibid="bib14" firstref="ref50"></nolink> <nolink nlid="nl31" bibid="bib35" firstref="ref55"></nolink> <nolink nlid="nl32" bibid="bib19" firstref="ref57"></nolink> <nolink nlid="nl33" bibid="bib44" firstref="ref58"></nolink> <nolink nlid="nl34" bibid="bib48" firstref="ref59"></nolink> <nolink nlid="nl35" bibid="bib46" firstref="ref61"></nolink> <nolink nlid="nl36" bibid="bib50" firstref="ref62"></nolink> <nolink nlid="nl37" bibid="bib12" firstref="ref63"></nolink> <nolink nlid="nl38" bibid="bib49" firstref="ref64"></nolink> <nolink nlid="nl39" bibid="bib15" firstref="ref65"></nolink> <nolink nlid="nl40" bibid="bib21" firstref="ref66"></nolink> <nolink nlid="nl41" bibid="bib52" firstref="ref67"></nolink> <nolink nlid="nl42" bibid="bib43" firstref="ref68"></nolink> <nolink nlid="nl43" bibid="bib16" firstref="ref69"></nolink> <nolink nlid="nl44" bibid="bib47" firstref="ref71"></nolink> <nolink nlid="nl45" bibid="bib34" firstref="ref73"></nolink> <nolink nlid="nl46" bibid="bib58" firstref="ref75"></nolink> <nolink nlid="nl47" bibid="bib53" firstref="ref78"></nolink> <nolink nlid="nl48" bibid="bib25" firstref="ref79"></nolink> <nolink nlid="nl49" bibid="bib59" firstref="ref80"></nolink> <nolink nlid="nl50" bibid="bib24" firstref="ref81"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1506589
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Impact of COVID-19 and Financial Factors on First-Year Student Retention: A Comparative Study of Pre- and Post-Pandemic Cohorts
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Kwideok+Han%22">Kwideok Han</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0002-5491-4364">0000-0002-5491-4364</externalLink>)<br /><searchLink fieldCode="AR" term="%22Ranjit+Pandey%22">Ranjit Pandey</searchLink><br /><searchLink fieldCode="AR" term="%22Kimberly+Meints%22">Kimberly Meints</searchLink><br /><searchLink fieldCode="AR" term="%22Larry+Burns%22">Larry Burns</searchLink><br /><searchLink fieldCode="AR" term="%22Chanjin+Chung%22">Chanjin Chung</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Higher+Education%3A+The+International+Journal+of+Higher+Education+Research%22"><i>Higher Education: The International Journal of Higher Education Research</i></searchLink>. 2026 91(2):723-739.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Springer. Available from: Springer Nature. One New York Plaza, Suite 4600, New York, NY 10004. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-460-1700; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 17
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2026
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Audience
  Label: Education Level
  Group: Audnce
  Data: <searchLink fieldCode="EL" term="%22Higher+Education%22">Higher Education</searchLink><br /><searchLink fieldCode="EL" term="%22Postsecondary+Education%22">Postsecondary Education</searchLink>
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22COVID-19%22">COVID-19</searchLink><br /><searchLink fieldCode="DE" term="%22Pandemics%22">Pandemics</searchLink><br /><searchLink fieldCode="DE" term="%22College+Freshmen%22">College Freshmen</searchLink><br /><searchLink fieldCode="DE" term="%22Academic+Persistence%22">Academic Persistence</searchLink><br /><searchLink fieldCode="DE" term="%22Public+Colleges%22">Public Colleges</searchLink><br /><searchLink fieldCode="DE" term="%22Student+Financial+Aid%22">Student Financial Aid</searchLink><br /><searchLink fieldCode="DE" term="%22Parent+Financial+Contribution%22">Parent Financial Contribution</searchLink><br /><searchLink fieldCode="DE" term="%22Student+Loan+Programs%22">Student Loan Programs</searchLink><br /><searchLink fieldCode="DE" term="%22Low+Income+Students%22">Low Income Students</searchLink><br /><searchLink fieldCode="DE" term="%22At+Risk+Students%22">At Risk Students</searchLink><br /><searchLink fieldCode="DE" term="%22Prediction%22">Prediction</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1007/s10734-025-01439-7
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0018-1560<br />1573-174X
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: This study examines factors influencing first-year student retention, with a particular focus on the impact of the COVID-19 pandemic. We employ logistic regression and machine learning algorithms to identify key predictors of retention and assess variations in retention probabilities across cohorts. The analysis is based on a dataset of 8,320 first-time freshmen at a public university in the central United States during the fall semesters of 2018 (pre-COVID-19) and 2020 (post-COVID-19). Results indicate that students enrolled in the post-COVID-19 period were less likely to persist into their second year. Higher tuition assistance--such as grants, tuition waivers, and scholarships--and greater family financial contributions positively influenced retention, particularly among the post-COVID-19 cohort. Federal and private loans were also positively associated with retention, reflecting student's financial commitments. However, those with higher financial need were less likely to persist, underscoring the challenges faced by financially vulnerable students. Machine learning models demonstrated strong predictive accuracy, with the Random Forest algorithm achieving the highest performance in effectively identifying students at risk of dropping out. These findings provide actionable insights for targeted institutional interventions. By analysing retention patterns across cohorts and financial factors, this study offers valuable recommendations for addressing financial disparities and enhancing student persistence during and beyond periods of crisis.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2026
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1506589
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1506589
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s10734-025-01439-7
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 17
        StartPage: 723
    Subjects:
      – SubjectFull: COVID-19
        Type: general
      – SubjectFull: Pandemics
        Type: general
      – SubjectFull: College Freshmen
        Type: general
      – SubjectFull: Academic Persistence
        Type: general
      – SubjectFull: Public Colleges
        Type: general
      – SubjectFull: Student Financial Aid
        Type: general
      – SubjectFull: Parent Financial Contribution
        Type: general
      – SubjectFull: Student Loan Programs
        Type: general
      – SubjectFull: Low Income Students
        Type: general
      – SubjectFull: At Risk Students
        Type: general
      – SubjectFull: Prediction
        Type: general
    Titles:
      – TitleFull: Impact of COVID-19 and Financial Factors on First-Year Student Retention: A Comparative Study of Pre- and Post-Pandemic Cohorts
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Kwideok Han
      – PersonEntity:
          Name:
            NameFull: Ranjit Pandey
      – PersonEntity:
          Name:
            NameFull: Kimberly Meints
      – PersonEntity:
          Name:
            NameFull: Larry Burns
      – PersonEntity:
          Name:
            NameFull: Chanjin Chung
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 0018-1560
            – Type: issn-electronic
              Value: 1573-174X
          Numbering:
            – Type: volume
              Value: 91
            – Type: issue
              Value: 2
          Titles:
            – TitleFull: Higher Education: The International Journal of Higher Education Research
              Type: main
ResultId 1