Beyond the Clock: The Influence of College Delay on Male Student Persistence

Abstract

Male students are more likely than female students to delay college entry. In male culture that widely expects stoicism, the delay decision among male students and its risks often go unnoticed. This study investigated the link between college delay—a “silent input” in such culture—and male student persistence through the third year of study. Drawing from the 2012/17 Beginning Postsecondary Students Longitudinal study (BPS:12/17), the authors analyzed a sample of 4,440 male students at four-year higher education institutions (HEIs). Using descriptive and inferential statistics (propensity score matching [PSM] and quasi-maximum likelihood estimation [QMLE]), the authors found a high dropout rate among male delayers and a negative relationship between college delay and persistence. High school grade point average (GPA), household income, degree expectations, and academic integration quality positively predicted persistence, whereas attending an open admission institution had the opposite influence. These findings highlight the critical need to understand the academic challenges faced by male students and to identify practice and research directions for enhancing their college enrollment and success.

Keywords

male students college delay persistence male culture

The United States has experienced a decreasing rate of male high school graduates going directly to college since the 1980s (Reeves & Smith, 2021). The National Center for Education Statistics (NCES, 2024) recently reported that the immediate enrollment rate of male high school graduates is nine percentage points lower than that of their female peers. This gender gap is particularly concerning because delaying college entry often worsens students’ long-term postsecondary trajectories. Research has demonstrated that college delay negatively predicts persistence (Gururaj, 2011; Horn et al., 2005), attainment (Andrews, 2018; Bozick & DeLuca, 2005; Horn et al., 2005; Lin & Liu, 2019; Niu & Tienda, 2013; Perna, 2000; Roksa & Velez, 2012), and earnings (Lin & Liu, 2019). Research has also demonstrated that male students are less likely than their female counterparts to persist in college (Barron, 2016; Conger & Long, 2010; Severiens & Ten Dam, 2012). Together, research points to a negative association between college delay and male student persistence. However, the magnitude of this association among male students remains unclear.

This study employed male students as a distinct sample to investigate the influence of college delay on persistence. This focus reflects the data disaggregation approach to better understanding the characteristics of target student populations (Chen, 2008; Polit & Beck, 2010). Prior research indicates that male students’ academic issues often go unnoticed as they struggle silently (Connell & Messerschmidt, 2005; Schwab & Dupuis, 2022; Vytniorgu et al., 2023). This tendency is particularly common when male students enter emerging adulthood and are socialized into higher education where concerns about conflicting with expected norms shape their stoic behaviors (Pascoe, 2012; Vytniorgu et al., 2023). The result is that male students’ educational trajectories are vulnerable within male culture, warranting empirical attention.

Situating male students in the male culture context, the authors employed Astin's (1993) Input-Environment-Output (I-E-O) model to frame the influence of college delay on their persistence. The I-E-O model provides a useful lens for examining how students’ precollege characteristics and late college entry interact with the college environment in shaping academic outcomes. In this study, college delay is conceptualized as an input, college experiences and institutional characteristics as the environment, and persistence through third year as the output. The third year marks a transition to the graduating year (i.e., the fourth year in most academic programs) and a critical stage in academic life. Framing persistence within a three-year model provides important insights for early interventions. The necessity of early interventions makes this model more realistic than longer ones, as students who pass the third year often enroll until graduation (Bui, 2024a, 2025).

The authors employed propensity score matching (PSM) and quasi-maximum likelihood estimation (QMLE) to test the hypothesis that college delay negatively predicts male student persistence. The sample in the study was 4,440 male students attending four-year higher education institutions (HEIs), selected from the 2012/2017 Beginning Postsecondary Students Longitudinal Study (BPS:12/17). The study was guided by two key questions:
What are the characteristics of male students who attend four-year HEIs after college delay?

In what way does college delay influence the persistence of male students at four-year HEIs?

This study underscores the importance of understanding the often-overlooked academic challenges that male students face. While the sample consists of male students at four-year HEIs, the findings have broader relevance for the whole K-16 education system, encouraging a re-evaluation of how male students are educated and supported. The gendered role of “being a man,” shaped by cultural norms that often emphasize self-reliance, emotional restraint, and risk-taking, undermines their efforts to improve persistence and graduation compared to their female peers (Barron, 2016; Conger & Long, 2010; Reeves & Smith, 2021; Severiens & Ten Dam, 2012). Persistence is a key predictor of degree completion and long-term postsecondary outcomes, including employment and economic stability (Andrade et al., 2022; Kuh et al., 2006). Therefore, male students who do not persist in college may forgo the economic and social benefits typically associated with earning a degree. By providing empirical evidence on the relationship between their college delay and persistence, the study encourages strategies and further research aimed at promoting timely college enrollment and academic performance among them.

Theoretical Background

College Delay

The college enrollment rate of male students has consistently been lower than that of female students for decades (NCES, 2024; Reeves & Smith, 2021). Before the 1970s, male students consistently enrolled at a higher rate than their female peers. However, the gap closed during the 1980s, and male students’ enrollment rate has lagged since then. For example, 38% of male students enrolled in four-year HEIs, while 51% of female students did in 2022 (NCES, 2024). Across two-year and four-year HEIs, 57% of male high school graduates were immediate enrollees , whereas 66% of female high school graduates were (NCES, 2024). Research has demonstrated that if these high school graduates do not go directly to college, they typically wait two years (Rowan-Kenyon, 2007). These years are referred to as the “gap years” or the transition period (Goldrick-Rab & Han, 2011; Wells & Lynch, 2012).

College delay is associated with various student characteristics and factors. Delayers tend to be male rather than female (Bozick & DeLuca, 2005; Jepson & Tobolowsky, 2020; Niu & Tienda, 2013) and come from racially minoritized backgrounds (Bozick & DeLuca, 2005; Goldrick-Rab & Han, 2011; Perna, 2000; Rowan-Kenyon, 2007). In particular, delayers are most likely to come from families with low socioeconomic status (SES) (Bozick & DeLuca, 2005; Cabrera & La Nasa, 2001; Goldrick-Rab & Han, 2011; Hearn, 1992; Wells & Lynch, 2012). Many are less academically prepared for college than their peers are, start families, have children, and join the military, among other activities (Goldrick-Rab & Han, 2011; Horn et al., 2005; Roksa & Velez, 2012). Male students, on average, show slower gains in measured intellectual development, lower levels of academic achievement, and greater doubts about their academic abilities and the value of a college degree than their female peers (Chambers & Crumb, 2022; Fischer et al., 2013; Severiens & Ten Dam, 2012). In some instances, male students are prepared but join missions that take years, which causes them to delay college entry (Jepson & Tobolowsky, 2020).

College delay matters, given its implications for students. Bozick and DeLuca (2005) found that delayers are less likely than new high school graduates to enroll in college. Only 24% of these delayers attend four-year HEIs, whereas 62% of new high school graduates do. Moreover, Andrews (2018) found that delayers are less likely to participate in academic and social life once they enroll in college. For example, they are 41% and 58% less likely to participate in internships and study abroad programs, respectively. Bozick and DeLuca (2005) also found that one year of delay reduces the likelihood of graduation by 78%. Bozick and DeLuca (2005) explained that in college, delayers do not have close connections with on-time, smarter cohorts. The lack of connections may demotivate them from engaging in high-impact activities such as internships and study abroad programs, which coshape their grade point average (GPA), persistence, and attainment (Andrews, 2018; Brower & Inkelas, 2010). The influence of college delay on postsecondary outcomes underscores the necessity of higher education practice and research that supports students' college enrollment and success.

Persistence

Persistence is one of the most crucial academic outcomes in college (Andrade et al., 2022; Kuh et al., 2006). While persistence is often understood as synonymous with retention, researchers have indicated a difference. Retention is persistence within an institution, whereas persistence refers to the success of transfer and nontransfer students in achieving the final goal (Mamiseishvili, 2012; St. John et al., 2018). Persistence has also been studied under the term “dropout” (Andrade et al., 2022). Research has demonstrated that persistence is linked to academic outcomes such as attainment and postcollege outcomes, including employment rates, salaries, mobility, tax revenues, better health, life expectancy, civic engagement, and social harmony (Andrade et al., 2022). Therefore, whether students persist has critical implications for their personal and professional life.

Given its association with postsecondary outcomes, persistence has been a focal topic in higher education research. There is a significant relationship between persistence and SES, academic preparation, and students’ behaviors (e.g., taking the Scholastic Aptitude Test [SAT] or American College Testing [ACT]) (Kim, 2007; Kuh et al., 2006; Stephan et al., 2015), high school and college GPA (Conger & Long, 2010; Dowd & Coury, 2006; Musoba & Krichevskiy, 2013), and financial aid (Carales, 2020; McKinney & Burridge, 2015; Musoba & Krichevskiy, 2013). Horn et al. (2005) and Gururaj (2011) found that persistence is significantly associated with college delay. With respect to male students, the relationship between persistence and college delay is the core issue, as they are more likely than female students to delay college entry and less likely than these peers to persist in college (Barron, 2016; Conger & Long, 2010; Severiens & Ten Dam, 2012).

Male Student Persistence

Although college delay and persistence have been widely studied, these processes are often theorized without explicit attention to gendered socialization. Male students’ academic challenges are often rendered muted within the dominant construction of masculinity, shaping how they experience and respond to academic difficulty. According to Schwab and Dupuis (2022), male students’ academic issues often go unnoticed because male culture discourages the open expression of vulnerability or academic difficulty. Male students are reluctant to report their struggles or ask for help from peers, faculty, or institutional resources (e.g., Connell & Messerschmidt, 2005; Schwab & Dupuis, 2022; Vytniorgu et al., 2023). Their reluctance is stronger when they reach adulthood (Pascoe, 2012) and enter the college environment (Vytniorgu et al., 2023). They are often concerned that sharing academic struggles with other people or asking for help would show a lack of masculinity. Connell and Messerschmidt (2005) considered this social anxiety a form of oppression where masculine norms constrain men's behaviors and emotional expressions. Their academic struggles and reluctance to seek help can be understood as response to socially enforced expectations surrounding masculinity.

Male student persistence represents a neglected issue within U.S. male culture (Schwab & Dupuis, 2022). Pressure to conform to masculinity begins at a young age for male students in various family and social environments (Connell & Messerschmidt, 2005). Male students, while sometimes benefiting from socially rewarded traits, such as aggression and dominance over their counterparts often neglect self-care and silently struggle to address their problems (Schwab & Dupuis, 2022). A study involving 15 American college students found that many male students feel compelled to hide their emotions and refrain from seeking academic help, fearing that others would not be receptive to their struggles (Schwab & Dupuis, 2022). This reluctance stems from adherence to the “Male Code,” which dictates that they must act masculinely, avoid expressing emotions or weaknesses, take risks, and succeed (Brannon, as cited in Musser, 2012). These hegemonic codes create a normative divide between men and women in discussions about academic matters. In such culture, the negative relationship between college delay and persistence is often unheard, which prevents male students from appropriate support (Schwab & Dupuis, 2022; Vytniorgu et al., 2023).

College Delay as the “Silent Input” in Male Student Persistence

The authors grounded this study in Astin's (1993) I-E-O model with the support of male culture scholarship. Prior research suggests that male students tend not to reveal their academic struggles or seek help from others due to the policing of masculinity (Schwab & Dupuis, 2022). They are afraid that doing so makes them look weak in other people's eyes. Guided by these insights, the authors conceptualized college delay in male culture as the “silent input,” which negatively influences male student persistence (output). To examine this relationship, the authors employed Astin's (1993) I-E-O model.

Astin's (1993) I-E-O model explains persistence as the function of what students are (input, including college delay and background factors) and the college environment that students enter (environment, including college experiences and institutional factors). This association can be illustrated by Koricich et al.'s (2018) observation that students who are not college-ready do not have many college options. Therefore, they attend nonselective or open admission, rather than selective, HEIs. Unprepared students may concentrate at such institutions, and the peer-effect makes the nonselective higher education environment less academically beneficial (Winston & Zimmerman, 2004). This explains why the environmental factors are important in understanding the influence of input, such as college delay, on the output, such as persistence. The mediation role of environmental factors in the I-E-O model suggests covariate adjustments in modeling the output (Herzog, 2018).

Through Astin's (1993) I-E-O model, the authors frame college delay and persistence among male students as interconnected yet neglected issues within U.S. male culture. This culture spans high school to college: while delay functions as an input to persistence (Gururaj, 2011; Horn et al., 2005), environments dominated by male norms may further exacerbate it. The I-E-O framework suits examining these dynamics. Notably, the authors did not measure the “Male Code” or masculinity policing (Schwab & Dupuis, 2022). Instead, they drew on extant scholarship to illuminate male students’ silent academic struggles. The “silent input” makes a strong case for examining college delay and its influence on male student persistence.

Methodology

In this study, the authors used data from the 2012/2017 Beginning Postsecondary Students Longitudinal Study (BPS:12/17), focusing on students at four-year HEIs. The BPS:12/17 dataset is well suited for examining college delay and persistence, as it provides detailed information on students’ background characteristics, college experiences, institutional features, and the persistence outcome, as suggested in prior research (e.g., Andrews, 2018; Bozick, 2007; Carales, 2020; Horn et al., 2005; McKinney & Burridge, 2015). To address the first research question, the authors employed descriptive statistics. For the second research question, the authors employed PSM to reduce selection bias (Austin, 2011; McBee, 2022; Rosenbaum & Rubin, 1983; Rubin, 2001). Students may differ systematically from their counterparts, and fitting regression models alone to predict the influence of college delay on persistence may produce unreliable results. For this reason, PSM has often been employed to have a matched sample before modeling outcomes of interest (e.g., Herzog, 2018; McKinney & Burridge, 2015; Santelices et al., 2015). With the matched sample obtained from PSM $(n = 1, 900)$ , the authors examined the hypothesized influence, fitting the outcome model via QMLE (Faraway, 2016; Paul & Islam, 1995). Variable selection for the outcome model was guided by Astin's (1993) I-E-O model and existing scholarship.

Data and Sampling

The BPS:12/17 dataset combines survey and administrative records for a nationally representative cohort of approximately four million undergraduates attending U.S. Title IV–eligible institutions (Bryan et al., 2019). Title IV institutions participate in federal student aid programs and are required to report standardized information on enrollment, financial aid, student progress, and degree attainment to the U.S. Department of Education. These federal reporting requirements ensure consistent, high-quality data across institutions and make Title IV schools the most analytically appropriate context for studying the influence of college delay on persistence. Prior research on delayed enrollment and persistence has relied on the BPS datasets for similar reasons (e.g., Andrews, 2018; Bozick, 2007; Carales, 2020; Horn et al., 2005; McKinney & Burridge, 2015).

After data wrangling, the dataset included 10,640 students enrolled in four-year HEIs. The authors selected 4,440 male students out of this number for the analytic sample, using the “male” category from the gender [sik] variable. Although labeled as gender, the variable only contains “male” and “female,” which reflect self-reported sex assigned at birth. All students in the selected sample were first-time beginning (FTB) students who had not previously attended any college. The sample selection aligns with current scholarship emphasizing the academic challenges and trajectories among students at four-year HEIs (Bui, 2024a, 2024b, 2025; Crisp et al., 2019).

Statistical Procedures

For research question 1, the authors employed WTA000, a cross-sectional weight, and the survey package (R 4.3.1) to conduct a descriptive analysis of 4,440 male students (the entire sample in this study). The incorporation of cross-sectional weights aimed to address oversampling and nonresponse, among other issues (Bui, 2025). The descriptive analysis focused on the characteristics of male delayers. In this article, the authors compare solely dropout rates between male delayers and the entire sample to provide descriptive evidence for the hypothesized association between college delay and male student persistence.

For Research Question 2, the authors selected covariates to estimate a propensity score model (see Table A1). The propensity score—the probability of delaying college entry—is a balancing score used to reduce selection bias inherent in observational data such as the BPS:12/17 dataset (Austin, 2011; McBee, 2022; Rosenbaum & Rubin, 1983; Rubin, 2001). If D represents college delay as a binary status, the propensity score of D can be defined as follows:
$P S = p r o b (D = 1 | X) = E (D | X)$
where $p r o b (D = 1 | X)$ is the conditional probability of delaying college entry given the selected covariates, and $E$ (D $| X$ ) is the expected probability of delaying college entry conditional on those covariates.

Since college delay was a binary variable, the authors estimated the propensity score using logistic regression. After obtaining a propensity score for each observation, nearest-neighbor matching with replacement was implemented to pair students who delayed college entry with comparable students who did not. The authors tested different caliper widths on the raw propensity score and determined 0.10 as the matching threshold. This was to ensure that the matching would not severely hurt the selected sample size while treated and control cases were sufficiently close. Using the MatchIt package, the authors retained 99.9% of treated cases and obtained a matched sample of 1,900 male students. Then, the quality of the matches was assessed using standard diagnostics. The standardized mean differences between treated and untreated groups were below 0.10, the variance ratios were close to 1.00, and the empirical cumulative distribution function (eCDF) differences were near zero (see Table A1). Together, these diagnostics indicated that the matched sample exhibited minimal standardized bias (Rubin, 2001).

The authors fitted a quasibinomial logistic regression model with the matched sample (n = 1,900). Using QMLE, they obtained a dispersion (scale) parameter to leave logit coefficients unchanged while adjusting the standard errors, confidence intervals, and p-values accordingly (Faraway, 2016; Paul & Islam, 1995). As a result, the inference was properly scaled to reflect the observed dispersion, reducing the chance of overstating statistical significance in the study. The quasibinomial model was compared with a null model using an F test (see Table A2). The results indicated that the model fit the data significantly better than the intercept-only specification.

Variables in the Outcome Model

Main variables: The outcome of interest in this study was persistence, measured by the dropout status (0 = No and 1 = Yes) of male students at four-year HEIs from the 2011–2012 academic year through June 2014. Any students who still enrolled without a degree in January 2014 were considered nondropouts. Persistence in this study was examined through a three-year timeframe, which provides an early update on male students’ academic progress toward graduation (Bui, 2024a, 2025). Meanwhile, the predictor of interest was college delay, a binary variable (0 = No and 1 = Yes). It indicates whether male students spent at least one gap year before attending four-year HEIs as FTB students in the 2011–2012 academic year. The use of college delay as a binary predictor is compatible with the PSM approach to addressing selection bias.

To isolate the influence of college delay on persistence, the authors employed background factors, college experiences, and institutional factors as controls.

Background factors: Background factors included race/ethnicity (Castle et al., 2024), high school GPA (Musoba, 2010; Musoba & Krichevskiy, 2013), and precolleges experiences (household income and a BA or an advanced degree expected) (Carales, 2020; McKinney & Burridge, 2015). It should be noted that: (1) Race/ethnicity was used as a binary variable to indicate whether the students were White or racially minoritized (Castle et al., 2024). Disaggregating races and ethnicities into a White group and racially minoritized groups (e.g., African Americans, Hispanics/Latinos, and Asians) might be an option. However, the authors chose this binary coding, as Castle et al. (2024) proposed, to reflect the historical exclusion of racially minoritized students from college education. (2) High school GPA was a continuous variable, measured on a seven-point scale. This variable was included as a control, given its long-lasting influence on later academic success (Musoba, 2010; Musoba & Krichevskiy, 2013). (3) Household income was used as a continuous variable to indicate students’ economic conditions one year before they entered college. (4) Degree expectations were used as a binary variable to indicate whether students expected a BA⁺ degree upon college entry.

College experiences: Included in this cluster were enrollment intensity, student loans (Carales, 2020; McKinney & Burridge, 2015), academic integration quality (Astin, 1984, 1993), and first-year college GPA (Chen & St. John, 2011; Musoba & Krichevskiy, 2013). It should also be noted that: (1) enrollment intensity (part-time/mixed or full-time) and student loans (Yes/No) were binary variables; (2) academic integration quality and the first-year college GPA were continuous variables. Academic integration quality was measured on a five-point scale, reflecting the quality that students rated on their interactions with faculty and their involvement in studies (Bui, 2024a, 2024b, 2025), while the first-year college GPA was measured on a four-point scale.

Institutional factors: Together with college experiences, institutions constitute the college environment to mediate the association between the input and the output (Herzog, 2018). They included control and selectivity in this study (Chen & St. John, 2011). Institutional control was categorized into public, private nonprofit, and private for-profit, whereas institutional selectivity was categorized into highly selective, moderately selective, minimally selective, and open admission.

Results

Descriptive Statistics

The characteristics of 4,440 male students at four-year HEIs are presented in Table 1. In this entire sample, 21% delayed college entry. While 21% of the entire sample dropped out, 36% of the delayers did. These descriptive findings provide initial evidence for the hypothesis that delaying college negatively influences male student persistence. Within the delayer group, the results indicate that 55% were racially minoritized students. These delayers had a school GPA of 5.430 (SE = 0.05). When entering college, their reported income was $64,618 (SE = 256.48), and less than half of them, 42%, expected to complete a BA⁺ degree. Regarding college experiences, the results reveal that 73% of the delayers enrolled full-time; they borrowed $5,370 (SE = 137.15) each; the quality of their academic integration was 4.339 (SE = 0.03) on a five-point scale; their first-year college GPA was 3.023 (SE = 0.03) on a four-point scale. Concerning institutional characteristics, the results show a larger proportion, 62%, of the delayers attending private for-profit HEIs, followed by 26% attending public HEIs and 12% attending private nonprofit HEIs. The delayers also tended to attend less selective HEIs. Up to 48% and 32% attended minimally selective and open admission HEIs, respectively, while only 12% and 7% attended moderately selective and highly selective HEIs, respectively.

Table 1.
Characteristics of Male Students (n = 4,440).

Variables Entire Sample Delayers

Unweighted count Weighted % Weighted mean (SE) Unweighted count Weighted % Weighted mean (SE)

College delay

No 3,220 79

Yes 1,210 21

Dropout

No 3,440 79 770 64

Yes 1,000 21 440 36

Race/ethnicity

White 2,510 58 620 45

Raciallyminoritized 1,920 42 590 55

High school GPA 4,440 5.654 (0.02) 1,210 5.430 (0.05)

Household income 4,440 76,791 (123.18) 1,210 64,618 (256.48)

BA⁺ degree expected

No 2,260 47 790 58

Yes 2,170 53 430 42

Enrollment intensity

Part-time/mixed 900 21 300 27

Full-time 3,540 79 910 73

Student loans 4,440 4,796 (49.79) 1,210 5,370 (137.15)

Academic integration quality 4,440 4.304 (0.01) 1,210 4.339 (0.03)

First-year college GPA 4,440 2.981 (0.01) 1,210 3.023 (0.03)

Institutional control

Public 1,540 42 190 26

Private nonprofit 1,290 32 150 12

Private for-profit 1,600 27 880 62

Institutional selectivity

Highly selective 940 25 60 7

Moderately selective 1,270 32 100 12

Minimally selective 1,480 27 660 48

Open admission 750 15 400 32

Note. Due to required rounding, unweighted counts and percentages may not be used to calculate totals. SE = standard error.

The Influence of College Delay on Persistence

Quasibinomial logistic regression was conducted with a matched sample of 1,900 male students. As shown in Table 2, college delay positively, significantly predicted the persistence outcome. Compared to male nondelayers, male delayers had a 98% increase in the odds of dropping out (OR = 1.983, p < .001), controlling for other variables. Conversely, if male students did not delay enrollment, the odds of dropping out decreased by 50% (1/1.983 ≈ 0.504), controlling for other variables.

Table 2.
Logistic Regression Results: The Influence of College Delay on Dropout.

Variables Odds Ratio Estimate SE t Value p (>|t|) 95% CI

LL UL

College delay 1.983 0.685 0.193 3.553 * 1.359 2.893

Race/ethnicity 0.956 −0.045 0.180 −0.250 .803 0.671 1.361

High school GPA 0.294 −1.226 0.309 −3.970 * 0.160 0.538

Household income 0.075 −2.589 1.011 −2.561 * 0.010 0.545

BA⁺ degree expected 0.466 −0.764 0.181 −4.230 * 0.327 0.664

Enrollment intensity 0.890 −0.116 0.209 −0.558 .577 0.591 1.340

Student loans 1.209 0.189 0.176 1.078 .281 0.857 1.705

Academic integration quality 0.670 −0.400 0.121 −3.303 * 0.529 0.850

First-year college GPA 1.284 0.250 0.389 0.642 .521 0.599 2.750

Institutional control

Private nonprofit 0.698 −0.360 0.257 −1.398 .162 0.422 1.156

Private for-profit 0.677 −0.391 0.292 −1.337 .181 0.381 1.200

Institutional selectivity

Moderately selective 1.479 0.391 0.335 1.167 .243 0.766 2.854

Minimally selective 1.712 0.538 0.372 1.444 .149 0.825 3.551

Open admission 2.234 0.804 0.360 2.235 * 1.104 4.523

Note. Unweighted n = 1,900; *** p < .001, * p < .05; CI = confidence interval; LL = lower limit; UL = upper limit.

The results in Table 2 also demonstrate that when other variables were controlled for, high school GPA, 2010 household income, degree expectations, academic integration quality, or attending open admission institutions significantly predicted the odds of dropping out among male students. The association between high school GPA and dropping out was negative. If high school GPA (log-transformed) increased by one unit, the odds of dropping out decreased by 71% (OR = 0.294, p < .001). Regarding household income, the results also indicate a significantly negative association. For a one-unit increase in household income (log-transformed), male students had 92% lower odds of dropping out (OR = 0.075, p < .05). Compared to male students who did not expect to complete a BA⁺ degree, those who expected to complete a BA⁺ degree had 53% lower odds of dropping out (OR = 0.466, p < .001). There was also a 33% decrease in the odds of dropping out for a one-unit increase in academic integration quality (OR = 0.670, p < .001). The greatest change in the odds of dropping out among male students was also linked to institutional selectivity. If they attended open admission HEIs, the odds of dropping out increased by 123% (OR = 2.234, p < .05).

Discussion

To understand how college delay influences the persistence of male students at four-year HEIs, the authors conducted a study with 4,440 male students, using descriptive and inferential methods. The authors selected covariates for the PSM and outcome models to reduce selection bias. The PSM model provided evidence on the similar likelihood of delaying college entry among male delayers and nondelayers, while the outcome model provided significant findings on the association between college delay and male student persistence. Descriptively, the authors found that a sizable proportion of male delayers, 36%, did not persist through the third year of study, which aligns with prior research (Gururaj, 2011; Horn et al., 2005). Inferentially, the authors found that the delayers were 98% more likely than their peers to drop out. This is one of the most important findings in the study, highlighting the academic disruptions among male delayers.

The authors also found in the study that delayers’ high school GPA significantly predicted the odds of dropping out. This suggests that academic preparation in high school has a lasting impact on college success (Musoba, 2010; Musoba & Krichevskiy, 2013). Moreover, more than half of the delayers did not expect to complete a BA⁺ degree although degree expectations predicted persistence. This lack of credential expectations resonates with the findings of previous studies that delayers do not often hold high educational expectations (Hearn, 1992; Lin & Liu, 2019; Niu & Tienda, 2013; Perna, 2000). Male students are often skeptical about themselves and question the value of academic success (Barron, 2016; Chambers & Crumb, 2022; Kerwin, 2022). They may expect to pursue a realistic endeavor rather than devoting time and energy to achieving a credential. Without a strong motivation to complete a BA⁺ degree, male students do not always integrate themselves academically to a level that optimally counters the negative influence of college delay on their persistence.

Just as important, the authors found that male delayers at open admission HEIs had high odds of dropping out compared to peers at other institutions. Approximately one-third of the delayers, 32%, attended open admission HEIs. This finding resonates with prior research that their college choices are limited when students are not academically prepared (Koricich et al., 2018). It signals the struggle to find a conducive learning environment for male students after gap years. It also signals that open admission HEIs face challenges to address students’ diverse learning needs. In such a context, the lack of academic preparedness may create a negative peer effect over time (Winston & Zimmerman, 2004). This negative effect may increase dropout rates, reinforcing the link between delayed college entry and persistence.

Consistent with Astin's I-E-O framework, the authors found that the input (e.g., high school GPA) and environmental factors (e.g., academic integration quality and attending open admission institutions) both predicted male student persistence (output). These findings reveal how students’ precollege characteristics may relate to college experiences, which mediate the college delay and persistence relationship. Predicated on the findings, the authors recommend the following practice and research directions.

Recommendations for Practice

HEIs, particularly those that are open admission, should provide targeted support for male delayers. HEIs should survey first-year male students who delay college. This survey aims to understand the students’ motivations and expectations in attending college and to incorporate those factors into first-year seminar programs. These programs are vital for male delayers because they positively influence persistence (Das et al., 2021). In addition to the programs, the role of student affairs professionals is vital in complementing and fostering male delayers’ learning outcomes. Specifically, student affairs professionals should strive to create a college environment where male delayers increase their educational expectations. They should also address these nontraditional students’ unique challenges and integrate academic and nonacademic activities to increase their learning momentum (Blake, 2007). Above all, monitoring and measuring the impact of these strategies on expected outcomes should be an integral part of student affairs professionals’ agenda (Kuh, 2009).

Meanwhile, male students should take the support provided as opportunities to counter the negative implications that college delay has on their persistence. Once in college, male delayers should fully utilize available resources and engage in academic programs (Astin, 1993), such as internships, study abroad, first-year seminars, learning communities, collaborative assignments, and writing-intensive courses (Andrews, 2018; Brower & Inkelas, 2010). They should establish male ally groups to discuss their unique situation and potential peer support (Musser, 2012). In colleges where male culture persists, male students should recognize these norms and step beyond their comfort zones to seek help and voice academic challenges. While peer culture may pressure them toward stoicism, reframing help-seeking as strength—not weakness—can enhance their persistence and overall success in college.

Recommendations for Research

To inform practice, future research should expand the scholarship on male delayers and their persistence through multiple dimensions. First, prior learning significantly influences college outcomes (Martin et al., 2013). Therefore, future studies should examine whether male delayers engage in academic activities during their gap years (Jepson & Tobolowsky, 2020; Niu & Tienda, 2013). Understanding these activities would clarify how the association between college delay and persistence changes, guiding delayers toward a productive transition period (Lin & Liu, 2019). Next, while the male delayers in this study reported a high level of academic integration quality, the relationship between college delay and persistence remained negative. Future research should explore at what quality level academic integration can optimally counter the influence of college delay on persistence and whether academic integration should be strategic (Astin, 1984; Bui, 2024a). Such research would clarify how a factor can support the persistence of students who delay college entry, distinguishing them from other delayers. Finally, exploring male delayers’ narratives regarding their college enrollment decisions and persistence is essential. Future research should explore whether the pressure to avoid seeking academic support and advice evolves as they transition to college. If that is the case, this pressure may compound existing academic challenges, leading to the decision to drop out (Barron, 2016; Conger & Long, 2010). Addressing these aspects would offer a vivid picture of male students′ struggles in transition to and persistence in college.

Limitations

The authors acknowledge multiple limitations in this study. First, the study lacked information on two-year college delayers who later applied to and attended a four-year HEIs (i.e., they were not transfer students, as the BPS: 12/17 students were only FTB students). Due to the absence of this information, the authors did not observe changes in male delayers’ college enrollment decisions. Second, an average U.S. student has two gap years before eventual enrollment in college (Rowan-Kenyon, 2007), but this study did not address how long male students postpone enrollment. The timing of college delay reveals students’ determination to enter college, and its absence limited insights into their psychological characteristics. Third, PSM does not address unobservable bias (Lin & Liu, 2019). Confounders related to both college delay and persistence might exist although they were not measured. Therefore, any causal interpretations of the findings in this study should be made with caution. Finally, data for covariates (e.g., enrollment intensity, student loans, and academic integration quality) were obtained from the first-year data. These data might have changed when the sampled students reached the second and third years of study. Unfortunately, the nuanced differences that these changes might have introduced into the persistence outcome were not captured.

Conclusion

Research on college delay as a determinant of male student success remains limited, particularly in relation to the norms embedded within male culture. This study examined male students attending four-year HEIs as a distinct analytic sample and investigated the relationship between college delay and persistence among male students. Using descriptive and inferential methods (PSM and QMLE), the authors found that delaying college enrollment was negatively associated with the persistence outcome. This central finding indicates that the link between college delay and male student persistence warrants greater institutional and scholarly attention. Because male culture often emphasizes stoicism, this negative association poses meaningful challenges for higher education practice and research. HEIs should provide targeted support for male students who delay college, and male delayers should be encouraged to take advantage of such resources as part of exercising their own agency. Future research should further examine the academic experiences of male students to inform the recommended practice. This study does not argue for prioritizing male students over other underserved groups. Instead, it underscores the importance of understanding and addressing the academic challenges that male students face, particularly when male culture may render such disruptions unheard.

Variables	Entire Sample	Delayers
College delay
No	3,220	79
Yes	1,210	21
Dropout
No	3,440	79		770	64
Yes	1,000	21		440	36
Race/ethnicity
White	2,510	58		620	45
Raciallyminoritized	1,920	42		590	55
High school GPA	4,440		5.654 (0.02)	1,210		5.430 (0.05)
Household income	4,440		76,791 (123.18)	1,210		64,618 (256.48)
BA⁺ degree expected
No	2,260	47		790	58
Yes	2,170	53		430	42
Enrollment intensity
Part-time/mixed	900	21		300	27
Full-time	3,540	79		910	73
Student loans	4,440		4,796 (49.79)	1,210		5,370 (137.15)
Academic integration quality	4,440		4.304 (0.01)	1,210		4.339 (0.03)
First-year college GPA	4,440		2.981 (0.01)	1,210		3.023 (0.03)
Institutional control
Public	1,540	42		190	26
Private nonprofit	1,290	32		150	12
Private for-profit	1,600	27		880	62
Institutional selectivity
Highly selective	940	25		60	7
Moderately selective	1,270	32		100	12
Minimally selective	1,480	27		660	48
Open admission	750	15		400	32

Variables	Odds Ratio	Estimate	SE	t Value	p (>\|t\|)	95% CI
College delay	1.983	0.685	0.193	3.553	***	1.359	2.893
Race/ethnicity	0.956	−0.045	0.180	−0.250	.803	0.671	1.361
High school GPA	0.294	−1.226	0.309	−3.970	***	0.160	0.538
Household income	0.075	−2.589	1.011	−2.561	*	0.010	0.545
BA⁺ degree expected	0.466	−0.764	0.181	−4.230	***	0.327	0.664
Enrollment intensity	0.890	−0.116	0.209	−0.558	.577	0.591	1.340
Student loans	1.209	0.189	0.176	1.078	.281	0.857	1.705
Academic integration quality	0.670	−0.400	0.121	−3.303	***	0.529	0.850
First-year college GPA	1.284	0.250	0.389	0.642	.521	0.599	2.750
Institutional control
Private nonprofit	0.698	−0.360	0.257	−1.398	.162	0.422	1.156
Private for-profit	0.677	−0.391	0.292	−1.337	.181	0.381	1.200
Institutional selectivity
Moderately selective	1.479	0.391	0.335	1.167	.243	0.766	2.854
Minimally selective	1.712	0.538	0.372	1.444	.149	0.825	3.551
Open admission	2.234	0.804	0.360	2.235	*	1.104	4.523

Footnotes

Acknowledgments

The authors thank Lyle McKinney for assisting in securing the data license from the Institute of Education Sciences while he was at the University of Houston.

ORCID iDs

Binh C. Bui

Catherine L. Horn

Man C. Wong

Alice E. Lee

Funding

The authors received no financial support for the research, authorship, or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, or publication of this article.

Author biographies

Binh C. Bui is an Assistant Professor of Research at the University of Texas at El Paso's Center for Education Research and Policy Studies (CERPS). His research focuses on students' enrollment and success in college, T-shaped skills, and (self)employment outcomes with particular attention to STEM fields.

Catherine L. Horn is the Dean and Elizabeth D. Rockwell Chair, College of Education, University of Houston. Her work focuses on the influence of local, state, and federal policy on the academic trajectories of students traditionally underserved by educational and social systems.

Man C. Wong is Professor and Associate Dean for Graduate Studies at the Hobby School of Public Affairs at the University of Houston. His research focuses on monetary policy, business-cycle theory, foreign direct investment, and economic policy.

Alice E. Lee is a Postdoctoral Fellow at the Center for Education Research and Policy Studies (CERPS), the University of Texas at El Paso. Her scholarship examines how institutional policies, organizational structures, and pedagogical practices influence student learning, persistence, identity development, and career pathways.

Appendix

Table A2.

Analysis of Deviance.

Variables	Df	Deviance	Residual Deviance	F	p (>F)
NULL			2224.4
College delay	1	67.823	2156.6	69.399	***
Race/ethnicity	1	0.198	2156.4	0.203	.652
High school GPA	1	71.474	2084.9	73.135	***
Household income	1	24.374	2060.5	24.940	***
BA⁺ degree expected	1	38.212	2022.3	39.100	***
Enrollment intensity	1	4.383	2017.9	4.485	*
Student loans	1	0.920	2017.0	0.941	.332
Academic integration quality	1	27.273	1989.7	27.906	***
First-year GPA	1	0.126	1989.6	0.129	.720
Institutional control	2	7.301	1982.3	3.735	*
Institutional selectivity	3	13.577	1968.7	4.631	**
F	18.686***
Reduction of deviance	255.66

Note. Residual degrees of freedom are not reported to comply with data security protocols; *** p < .001; ** p < .01; * p < .05.

References

Andrade

M. S.

Miller

R. M.

McArthur

Ogden

(2022). The impact of learning on student persistence in higher education. Journal of College Student Retention: Research, Theory & Practice, 24(2), 316–336. https://doi.org/10·1177/1521025120915576

Andrews

B. D.

(2018). Delayed enrollment and student involvement: Linkages to college degree attainment. The Journal of Higher Education, 89(3), 368–396. https://doi.org/10·1080/00221546·2017·1390972

Astin

A. W.

(1984). Student involvement: A developmental theory for higher education. Journal of College Student Personnel, 25(4), 297–308.

Astin

A. W.

(1993). What matters in college? Four critical years revisited. Jossey-Bass.

Austin

P. C.

(2011). An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behavioral Research, 46(3), 399–424. https://doi.org/10·1080/00273171·2011·568786

Barron

. (2016). The Silent Epidemic: Young men dropping out of college. Psychology Today. https://www.psychologytoday.com/us/blog/the-creativity-cure/201604/the-silent-epidemic-young-men-dropping-out-of-college

Blake

J. H.

(2007). The crucial role of student affairs professionals in the learning process. New Directions for Student Services, 2007(117), 65–72. https://doi.org/10·1002/ss.234

Bozick

(2007). Making it through the first year of college: The role of students’ economic resources, employment, and living arrangements. Sociology of Education, 80(3), 261–285. https://doi.org/10·1177/003804070708000304

Bozick

DeLuca

(2005). Better late than never? Delayed enrollment in the high school to college transition. Social Forces, 84(1), 531–554. https://doi.org/10·1353/sof.2005·0089

10.

Brower

A. M.

Inkelas

K. K.

(2010). Living-learning programs: One high-impact educational practice we now know a lot about. Liberal Education, 96(2), 36–43.

11.

Bryan

Cooney

Elliott

(2019). 2012/17 Beginning Postsecondary Students Longitudinal Study (BPS:12/17) (NCES 2020522). https://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2020522

12.

Bui

B. C

. (2024a). Delayed enrollment among minority and male students: Factors and effects on college outcomes [Doctoral dissertation, University of Houston]. https://hdl.handle.net/10657/16145

13.

Bui

B. C.

(2024b). How debt and attainment relate through the GPA of non-White college students. Social Sciences, 13(8), 407. https://doi.org/10·3390/socsci13080407

14.

Bui

B. C.

(2025). Striving for college success: The persistence of students who work off-campus. Innovative Higher Education, 50(6), 1945–1968. https://doi.org/10·1007/s10755-025-09834-9

15.

Cabrera

A. F.

La Nasa

S. M.

(2001). On the path to college: Three critical tasks facing America's disadvantaged. Research in Higher Education, 42(2), 119–149. https://doi.org/10·1023/A:1026520002362

16.

Carales

V. D.

(2020). Examining educational attainment outcomes: A focus on Latina/o community college students. Community College Review, 48(2), 195–219. https://doi.org/10·1177/0091552120903087

17.

Castle

S. D.

Byrd

W. C.

Koester

B. P.

Pearson

M. I.

Bonem

Caporale

Cwik

Denaro

Fiorini

(2024). Systemic advantage has a meaningful relationship with grade outcomes in students’ early STEM courses at six research universities. International Journal of STEM Education, 11(1), 1–20. https://doi.org/10·1186/s40594-024-00474-7

18.

Chambers

C. R.

Crumb

(2022). College knowledge is power: Supporting the postsecondary aspirations of Black male ninth graders. Spectrum: A Journal on Black Men, 10(1), 57–78. https://doi.org/10.2979/spectrum.10.1.04

19.

Chen

(2008). Financial aid and student dropout in higher education: A heterogeneous research approach. In Smart

J. C.

(Ed.), Higher education: Handbook of theory and research (Vol. 23, pp. 209–239). Springer.

20.

Chen

St. John

E. P. S.

(2011). State financial policies and college student persistence: A national study. The Journal of Higher Education, 82(5), 629–660. https://doi.org/10·1080/00221546·2011·11777220

21.

Conger

Long

M. C.

(2010). Why are men falling behind? Gender gaps in college performance and persistence. The ANNALS of the American Academy of Political and Social Science, 627(1), 184–214. https://doi.org/10·1177/0002716209348751

22.

Connell

R. W.

Messerschmidt

J. W.

(2005). Hegemonic masculinity: Rethinking the concept. Gender & Society, 19(6), 829–859. https://doi.org/10.1177/0891243205278639

23.

Crisp

Horn

C. L.

Kuczynski

Zhou

Cook

(2019). Describing and differentiating four-year broad access institutions: An empirical typology. The Review of Higher Education, 42(4), 1373–1400. https://doi.org/10·1353/rhe.2019·0069

24.

Das

Schmitt

Stephenson

M. T.

(2021). A quasiexperimental analysis of first-year seminar outcomes at a large university. Journal of College Student Retention: Research, Theory & Practice, 25(4), 940–954. https://doi.org/10·1177/15210251211038591

25.

Dowd

A. C.

Coury

(2006). The effect of loans on the persistence and attainment of community college students. Research in Higher Education, 47(1), 33–62. https://doi.org/10·1007/s11162-005-8151-8

26.

Faraway

J. J.

(2016). Extending the linear model with R: Generalized linear, mixed effects and nonparametric regression models (2nd ed.). CRC press.

27.

Fischer

Schult

Hell

(2013). Sex differences in secondary school success: Why female students perform better. European Journal of Psychology of Education, 28(2), 529–543. https://doi.org/10·1007/s10212-012-0127-4

28.

Goldrick-Rab

Han

S. W.

(2011). Accounting for socioeconomic differences in delaying the transition to college. The Review of Higher Education, 34(3), 423–445. https://doi.org/10·1353/rhe.2011·0013

29.

Gururaj

S. V

. (2011). Postsecondary enrollment delay and associated persistence outcomes [Doctoral dissertation, University of Texas at Austin]. Texas Scholar Works. https://repositories.lib.utexas.edu/handle/2152/ETD-UT-2011-05-2721

30.

Hearn

J. C.

(1992). Emerging variations in postsecondary attendance patterns: An investigation of part-time, delayed, and nondegree enrollment. Research in Higher Education, 33(6), 657–687. https://doi.org/10.1007/BF00992053

31.

Herzog

(2018). Financial aid and college persistence: Do student loans help or hurt? Research in Higher Education, 59(3), 273–301. https://doi.org/10·1007/s11162-017-9471-1

32.

Horn

Cataldi

E. F.

Sikora

(2005). Waiting to attend college. Undergraduates who delay their postsecondary enrollment (NCES 2005–152). National Center for Education Statistics. https://nces.ed.gov/pubs2005/2005152.pdf

33.

Jepson

J. A.

Tobolowsky

B. F.

(2020). From delay to degree: The postsecondary experiences of six nontraditional students. Journal of College Student Retention: Research, Theory & Practice, 22(1), 27–48. https://doi.org/10·1177/15210251177243

34.

Kerwin

. (2022). The male enrollment crisis. Chronicle of Higher Education. https://www.chronicle.com/featured/student-success/student-centric-institution/male-enrollment-crisis#:∼:text=Men%20make%20up%20just%20over,been%20building%20for%2040%20years

35.

Kim

(2007). The effect of loans on students’ degree attainment: Differences by student and institutional characteristics. Harvard Educational Review, 77(1), 64–100. https://doi.org/10·17763/haer.77·1.n14t69l0q8292784

36.

Koricich

Chen

Hughes

R. P.

(2018). Understanding the effects of rurality and socioeconomic status on college attendance and institutional choice in the United States. The Review of Higher Education, 41(2), 281–305. https://doi.org/10·1353/rhe.2018·0004

37.

Kuh

G. D.

(2009). What student affairs professionals need to know about student engagement. Journal of College Student Development, 50(6), 683–706. https://doi.org/10·1353/csd.0·0099

38.

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 (Commissioned Report). National Postsecondary Education Cooperative. https://nces.ed.gov/npec/pdf/Kuh_Team_Report.pdf

39.

Lin

Liu

V. Y. T.

(2019). Timing matters: How delaying college enrollment affects earnings trajectories (Working Paper No. 105). Teachers’ College: Community College Research Center.

40.

Mamiseishvili

(2012). International student persistence in US postsecondary institutions. Higher Education, 64(1), 1–17. https://doi.org/10·1007/s10734-011-9477-0

41.

Martin

A. J.

Wilson

Liem

G. A. D.

Ginns

(2013). Academic momentum at university/college: Exploring the roles of prior learning, life experience, and ongoing performance in academic achievement across time. The Journal of Higher Education, 84(5), 640–674. https://doi.org/10·1080/00221546·2013·11777304

42.

McBee

(2022). Statistical approaches to causal analysis. Sage Publications.

43.

McKinney

Burridge

A. B.

(2015). Helping or hindering? The effects of loans on community college student persistence. Research in Higher Education, 56(4), 299–324. https://doi.org/10·1007/s11162-014-9349-4

44.

Musoba

G. D.

(2010). Accountability policies and readiness for college for diverse students. Educational Policy, 25(3), 451–487. https://doi.org/10·1177/0895904810361721

45.

Musoba

G. D.

Krichevskiy

(2013). Early coursework and college experience predictors of persistence at a Hispanic-Serving Institution. Journal of Hispanic Higher Education, 13(1), 48–62. https://doi.org/10·1177/1538192713513463

46.

Musser

. (2012). Are men the new at-risk population of students? Kansas State University. Retrieved October 23, 2024, from https://nacada.ksu.edu/Resources/Clearinghouse/View-Articles/Males-as-an-at-risk-population.aspx

47.

National Center for Education Statistics. (2024). Immediate college enrollment rate. Retrieved March 29, 2026, from https://nces.ed.gov/programs/coe/indicator/cpa/immediate-college-enrollment-rate

48.

Niu

Tienda

(2013). Delayed enrollment and college plans: Is there a postponement penalty? The Journal of Higher Education, 84(1), 1–27. https://doi.org/10·1080/00221546·2013·11777276

49.

Pascoe

C. J.

(2012). Dude, you're a fag: Masculinity and sexuality in high school. University of California Press.

50.

Paul

S. R.

Islam

A. S.

(1995). Analysis of proportions in the presence of over-/under-dispersion. Biometrics, 51(4), 1400–1410. https://doi.org/10.2307/2533270

51.

Perna

L. W.

(2000). Differences in the decision to attend college among African Americans, Hispanics, and Whites. The Journal of Higher Education, 71(2), 117–141. https://doi.org/10·2307/2649245

52.

Polit

D. F.

Beck

C. T.

(2010). Generalization in quantitative and qualitative research: Myths and strategies. International Journal of Nursing Studies, 47(11), 1451–1458. https://doi.org/10·1016/j.ijnurstu.2010·06·004

53.

Reeves

Smith

(2021). The male college crisis is not just in enrollment, but completion. Retrieved October 8, 2022, from https://www.brookings.edu/blog/up-front/2021/10/08/the-male-college-crisis-is-not-just-in-enrollment-but-completion/

54.

Roksa

Velez

(2012). A late start: Delayed entry, life course transitions and bachelor's degree completion. Social Forces, 90(3), 769–794. https://doi.org/10·1093/sf/sor018

55.

Rosenbaum

P. R.

Rubin

D. B.

(1983). The central role of the propensity score in observational studies for causal effects. Biometrika, 70(1), 41–55. https://doi.org/10·1093/biomet/70.1.41

56.

Rowan-Kenyon

H. T.

(2007). Predictors of delayed college enrollment and the impact of socioeconomic status. The Journal of Higher Education, 78(2), 188–214. https://doi.org/10·1080/00221546·2007·11780873

57.

Rubin

D. B.

(2001). Using propensity scores to help design observational studies: Application to the tobacco litigation. Health Services and Outcomes Research Methodology, 2, 169–188. https://doi.org/10·1023/A:1020363010465

58.

Santelices

M. V.

Catalán

Kruger

Horn

(2015). Determinants of persistence and the role of financial aid: Lessons from Chile. Higher Education, 71(3), 323–342. https://doi.org/10·1007/s10734-015-9906-6

59.

Schwab

J. R.

Dupuis

D. L.

(2022). Masculinity and silence in the academic struggles of American male college students. Emerging Adulthood, 10(1), 249–263. https://doi.org/10·1177/2167696820983052

60.

Severiens

Ten Dam

(2012). Leaving college: A gender comparison in male and female-dominated programs. Research in Higher Education, 53(4), 453–470. https://doi.org/10·1007/s11162-011-9237-0

61.

Stephan

J. L.

Davis

Lindsay

Miller

(2015). Who will succeed and who will struggle? Predicting early college success with Indiana‘s Student Information System (REL 2015–078). National Center for Education Evaluation and Regional Assistance and Regional Educational Laboratory Midwest. https://ies.ed.gov/ncee/edlabs/projects/project.asp?projectID=354

62.

St. John

E. P.

Daun-Barnett

Moronski-Chapman

K. M.

(2018). Public policy and higher education: Reframing strategies for preparation, access, and college success. Routledge.

63.

Vytniorgu

Cooper

Barreto

(2023). ‘From a guy’s perspective’? Male students, masculinity and autobiographical loneliness narratives. Gender and Education, 35(5), 505–519. https://doi.org/10·1080/09540253·2023·2213726

64.

Wells

R. S.

Lynch

C. M.

(2012). Delayed college entry and the socioeconomic gap: Examining the roles of student plans, family income, parental education, and parental occupation. The Journal of Higher Education, 83(5), 671–697. https://doi.org/10·1080/00221546·2012·11777262

65.

Winston

Zimmerman

(2004). Peer effects in higher education. In Hoxby

C. M

(Ed.), College choices: The economics of where to go, when to go, and how to pay for it (pp. 395–424). University of Chicago Press.

Variables	Entire Sample			Delayers
Variables	Unweighted count	Weighted %	Weighted mean (SE)	Unweighted count	Weighted %	Weighted mean (SE)
College delay
No	3,220	79
Yes	1,210	21
Dropout
No	3,440	79		770	64
Yes	1,000	21		440	36
Race/ethnicity
White	2,510	58		620	45
Raciallyminoritized	1,920	42		590	55
High school GPA	4,440		5.654 (0.02)	1,210		5.430 (0.05)
Household income	4,440		76,791 (123.18)	1,210		64,618 (256.48)
BA⁺ degree expected
No	2,260	47		790	58
Yes	2,170	53		430	42
Enrollment intensity
Part-time/mixed	900	21		300	27
Full-time	3,540	79		910	73
Student loans	4,440		4,796 (49.79)	1,210		5,370 (137.15)
Academic integration quality	4,440		4.304 (0.01)	1,210		4.339 (0.03)
First-year college GPA	4,440		2.981 (0.01)	1,210		3.023 (0.03)
Institutional control
Public	1,540	42		190	26
Private nonprofit	1,290	32		150	12
Private for-profit	1,600	27		880	62
Institutional selectivity
Highly selective	940	25		60	7
Moderately selective	1,270	32		100	12
Minimally selective	1,480	27		660	48
Open admission	750	15		400	32