Fostering Persistence in Science,Technology,Engineering,and Mathematics (STEM): Creating an Equitable Environment That Addresses the Needs of Undergraduate Students

Abstract

Student retention is a critical issue for universities, and nearly half of the students who start degree programs in science, technology, engineering, and mathematics (STEM) do not complete them. The current study tracks the progress of STEM students taking part in an entry-to-graduation program designed to build community, provide academic and social support, and promote engagement in academically purposeful activities. Although it had no effect on the number of students who changed their major, the program more than doubled the number of students who graduated in their original major. Black or Hispanic students taking part in the program also graduated at twice the rate of comparator students, largely attributable to the success of women in these groups. The results provide needed real-world insights into how to create an equitable environment that promotes the persistence and graduation of students, including those from groups historically underrepresented in STEM.

Keywords

student success retention graduation STEM persistence underrepresented groups women

Introduction

Student retention is a critical issue for academic institutions. Student attrition directly translates into a loss of tuition income, fees, and alumni contributions (DeBerard et al., 2004), and government funding for universities is increasingly based on performance metrics such as retention and graduation. Moreover, student departure without a degree represents the failure of a university to fulfill its educational mission (Bean, 1990) and greater responsibility to society (Norton, 2012; Watts, 2001).

For the student, leaving college has lasting effects. It lowers self-esteem and impacts the lifelong economic prospects of the individual (Stillman, 2009; Torpey, 2018). Education is a critical determinant of upward socioeconomic mobility for many students from groups historically underserved by higher education, including low-income, first-generation, Black, or Hispanic students. This is particularly true for students in science, technology, engineering, and mathematics (STEM) fields, who can expect to earn more than twice the annual wage of those in non-STEM occupations (United States Bureau of Labor Statistics, 2021). STEM fields drive innovation and global economies. While STEM employment is projected to grow faster than non-STEM employment, the demand for workers in STEM fields exceeds the supply of STEM graduates (Langdon et al., 2011; United States Bureau of Labor Statistics, 2021).

And yet, nearly half of the students who begin STEM baccalaureate degree programs do not complete them (Chen, 2013; President's Council of Advisors on Science & Technology, 2012). This is particularly true for women and students of races or ethnicities underrepresented in STEM, including Black, multiracial, or Hispanic backgrounds. Although 57% of all bachelor's degrees were awarded to women in 2018–2019, women received only 36% of the bachelor's degrees given in STEM fields (National Center for Education Statistics [NCES], 2020a, 2020b). During this period, Black, Hispanic, and multiracial individuals received only 7.1%, 12.8%, and 4.2% of all STEM degrees, respectively (NCES, 2020c), despite accounting for 12.4%, 18.7%, and 10.2% of the United States population (Jones et al., 2021). To fill the demand for STEM jobs and compete economically on a global scale, the number of STEM students who complete their degree needs to increase. Furthermore, institutions of higher learning need to support inclusion and foster equity to improve the retention and graduation of underrepresented STEM students.

Why Students Leave

Student departure from a university can be classified into two categories: students leaving one university to attend another and students discontinuing higher education entirely. A National Student Clearinghouse Report (2020) found that 82.5% of full-time first-year students entering college in fall 2018 returned to any college for their second year; 67% returned to the same institution in which they started. These numbers varied by race and ethnicity: 66% of African American students and 72% of Hispanic students persisted for their second year. African Americans were the least likely to be retained at their current institution (55.4%), followed by Hispanic students (63.5%). After controlling for factors such as grade point average (GPA) and scholastic aptitude test (SAT) scores, students of color leave higher education more often than White students (Banks & Dohy, 2019; Hoffman & Lowitzki, 2005).

Student attrition is a complex problem with multiple contributions that include academic, financial, practical, psychosocial, and quality factors (Coates, 2014). Prior student academic achievement is a strong predictor of college persistence, measured by variables such as high school GPA or SAT/American college testing (ACT) score (e.g., Astin, 1997; Geiser & Santelices, 2007; Hoffman & Lowitzki, 2005; Johnson, 2008; Kim, 2015; Tracey & Robbins, 2006). Other student attributes have also been shown to affect student retention, including age, gender, and socioeconomic status (Pfeffer, 2018; Seidel & Kutieleh, 2017). Student attributes have received much attention over the years, although the focus has begun shifting to consider the accountability of the university in fostering an environment conducive to student retention and graduation. Psychosocial and noncognitive factors contribute meaningfully to a student's intention to leave a university. For instance, Coates (2014) found that the student's perception of the support they receive from their institution and their relationships with fellow students and faculty were linked to their intention to leave a university. Likewise, a sense of belonging correlates with academic success (Han et al., 2017, 2020), and students from socially stigmatized groups may be more susceptible to doubts about their sense of belonging in a university setting (Walton & Cohen, 2007, 2011). Academic and social support are critical to student satisfaction and can compensate for feelings of underrepresentation and marginalization (see Brown, 2000; Hoffman, 2002; Hoffman & Lowitzki, 2005; Loo & Rolison, 1986).

Students in STEM fields are subject to the same personal and academic challenges as students in non-STEM fields, and yet differences are observed in the persistence and retention of STEM students specifically, especially those who are underrepresented in STEM. Black and Hispanic students declare STEM majors at a similar rate as their White counterparts but change majors or leave college more often, regardless of socioeconomic status (Eagan et al., 2015; Riegle-Crumb et al., 2019; Xie et al., 2015). For instance, Riegle-Crumb et al. (2019) recently found that 40% of Black and 37% of Hispanic STEM students switch majors before earning a degree, compared to 29% of White STEM students. Moreover, they found that STEM is the only field where Black and Hispanic students are significantly more likely to switch majors. Another group underrepresented in many STEM fields, women are less likely to select STEM majors, not because of the focus on math and science but because they perceive STEM fields as being more biased against women (Ganley et al., 2018). Women who enroll in STEM degree programs are much more likely to change majors than men, due more to an aversion to the culture rather than the content or rigor of the programs (Astorne-Figari & Speer, 2018, 2019; Strenta et al., 1994). The attrition of intelligent and well-prepared female, Black, and/or Hispanic students from STEM fields highlights the importance of creating an inclusive environment that fosters a sense of belonging for all individuals. In support of this, Xie et al. (2015) conclude that socioeconomic factors can predict the attainment of general education, but psychosocial factors are more important influences on the participation and achievement of students in STEM fields specifically.

How to Retain Students

Numerous content and process theories focus on the interaction of the student with the academic environment, offering insights into the range of problems encountered by first-year students and how to successfully address them. Sanford's theory of challenge and support (1966) involves the relationship between a student's transition into adulthood and the college environment. Challenges occur when students are not prepared to cope—academically, socially, or psychologically—and the amount of challenge a student can tolerate depends upon the amount and quality of support. If students are not supported, they may look to escape by changing their major or leaving the university. Schlossberg's transition theory reminds us that even positive, anticipated transitions can cause stress, and being able to cope with the transition results in the successful integration of the individual into the new environment (Goodman et al., 2006). A student's personal and psychological characteristics affect the ability to cope, but universities can also facilitate social support mechanisms and strategies for coping. Schlossberg's theory of mattering and marginality (1989) and Rendón's theory of validation (1994) apply to all first-year students but especially to women and underrepresented students, who are more likely to doubt their academic ability or succumb to stereotype threat (Beasley & Fischer, 2012; Murphy et al., 2007; Rendón, 1994; Walton & Cohen, 2007, 2011). Feelings of marginality can lead students to believe they do not matter, and universities should foster environments that help students feel that they do matter. For first-year college students, social support given by friends is one of the most important factors in student mattering (Rayle & Chung, 2007). Validation is an “enabling, confirming, and supportive process initiated by in- and out-of-class agents that foster academic and interpersonal development” (Rendón, 1994, p. 44). It is necessary for student development, and universities should support efforts to actively promote validation in classrooms, in academic social settings, and by faculty and staff, keeping in mind that validation often requires action on the part of the institution, rather than the student (Rendón, 1994). Students who experience challenges but are provided appropriate support (Sanford, 1966), who are given resources to cope with the transition (Goodman et al., 2006), who are not marginalized but believe they matter (Schlossberg, 1989), and who are validated by those within the institution (Rendón, 1994) are more likely to experience personal development and persevere in college. Many of these facets also increase a student's integration into and commitment to the university, foundations of Tinto's interactionalist theory (1993), and useful revisions of it (Braxton, 2016; Braxton et al., 1997, 2011).

Astin's involvement theory (1984) provides a means by which institutions can facilitate engagement and subsequent persistence of their students. It advocates that colleges should provide meaningful ways for students to engage in their education, both inside and outside the classroom, and it has overwhelming support in the literature as a means to retain students (see Kahu & Nelson, 2018; Kinzie et al., 2008; Korobova & Starobin, 2015; Kuh, 2016; Kuh et al., 2007). Kuh (2016) suggests five early college factors that affect student engagement and subsequent persistence: (1) psychosocial fit, connections to the group that lead to social acceptance; (2) academic and social support through study skills and opportunities for social interaction; (3) participation in educationally purposeful activities; (4) consistent academic progress and accumulation of at least 15 credit hours in the first year; and (5) goal realization, the understanding of why one is pursuing a particular educational path. By instituting interventions early after student entry, institutions can support student development and promote persistence. Such interventions can provide the psychosocial support that influences the achievement of students, including STEM students, and can foster the success of underrepresented or underprepared STEM students (Kinzie et al., 2008; Kuh, 2016).

Methods

The current study examines the effectiveness of an entry-to-graduation program designed to reduce the attrition of students at a large suburban public comprehensive university in the southeastern United States. The program began in the first semester the students attended college, fall 2017. To participate in the program, first-year students entering the university had to have selected a major housed within the College of Science and Mathematics, namely biochemistry, biology, chemistry, environmental science, mathematics, or physics. To complement other programs in the college that assist students who are underprepared for their first-year courses, this program focused on reducing the attrition of students entering from high school with adequate academic preparation. To assess prior academic preparation, a proprietary predicted success model was used to determine the odds of a student passing his/her first semester science class, a majors-level general chemistry course. The model incorporated the student's high school GPA, math SAT or ACT score, and whether (s)he had taken any advanced placement courses. Any incoming first-year student with a qualifying major and at least a 0.70 likelihood of passing the majors-level general chemistry course was invited to take part. This corresponded to roughly 40% of applicants to the college. A personal invitation from the dean was sent to the student's mailing address after a qualifying student was accepted for admission to the university. Participation was free and voluntary, and students could leave the program at any time.

One hundred and eleven first-year students chose to take part in the program; eligible students who did not elect to participate in the program were selected as a comparator group (Table 1). The predicted success probability of students in the program (N = 111) averaged 0.80 (SD = .06); students in the comparator group (N = 111) also averaged 0.80 (SD = .06). Comparing these two academically comparable groups allowed us to address three major research questions: (1) Does a comprehensive multiyear engagement program affect the persistence and graduation of students in their originally selected STEM major? (2) Does such a program increase the retention of students at the university, regardless of major? and (3) Does the program provide an equitable environment that better supports students underrepresented in STEM? To answer these questions, we compared the graduation rates of students in the program against those in the comparator group, and we analyzed the rates of students who changed majors or left the university. We also examined if differences existed between the groups pertaining to student GPA and time to graduation.

Table 1.

Attributes of Program Participants or Comparator Students.

	Programn (%)	Comparatorn (%)
Students	111	111
Sex
Female	69 (62.2)	67 (60.4)
Male	42 (37.8)	44 (39.6)
Race/ethnicity
Asian	5 (4.5)	9 (8.1)
Black or African American	31 (27.9)	23 (20.7)
Hispanic	15 (13.5)	9 (8.1)
Multiracial	7 (6.3)	14 (12.6)
White	53 (47.7)	56 (50.5)
Major
Biochemistry	16 (14.4)	8 (7.2)
Biology	68 (61.3)	80 (72.1)
Chemistry	14 (12.6)	13 (11.7)
Environmental sciences	5 (4.5)	5 (4.5)
Math	3 (2.7)	3 (2.7)
Physics	5 (4.5)	2 (1.8)

Program Attributes

Proactive interventions to support student persistence and retention were incorporated based upon the findings and recommendations of the literature and theories summarized above. Program activities were designed to provide support for challenges, supply resources for coping with the transition, reduce student feelings of marginality and provide validation, and encourage in-class and college engagement. Program components have been organized according to Kuh's five factors predicting student engagement and success (2016) to better clarify the overarching goal of each component (Figure 1). A major component of the program involved course sections that were reserved for program participants only, and students joining the program were advised of their course options before their first-year orientation session. For their first semester of college, students selected learning communities that included science, math, and first-year seminar courses. In learning communities, the same group of students enroll in each course to facilitate social interaction and foster community. Students also participate in activities that span the courses and tie the material together. Following their first semester, students continued to take part in course sections for program participants, but these were not within learning communities. Faculty teaching courses for the program had taken part in pedagogical workshops and were encouraged to incorporate active learning, innovative activities, and real-world applications into their classes to foster goal realization. The first-year seminar course provided support for challenges and resources for coping with the transition to college. It included topics such as study skills and metacognition but also introduced students to university offices and ways to get involved in the college.

Figure 1.

Program components classified by their overarching goals.

Two large events gathered all program participants together to create a sense of belonging and provide extra resources. A welcome event was held for students during the first week of each fall semester. The event included free ice cream, and students were given a special program T-shirt. Guests to the event were purposefully selected to introduce the students to people who would be helpful on their academic journey, including faculty teaching their courses, college deans and department chairs, academic advisors, career service representatives, each student organization within the college, and more seasoned students performing undergraduate research. To continue fostering a sense of belonging and highlight progression, an end-of-year barbeque gave an opportunity for students to reflect on their progress while taking part in outdoor games. Faculty, staff, and administrators associated with the program also attended.

To maintain the community students formed through events and their learning community the first semester, course sections specifically for program participants continued to be offered each semester for the first two years of college, and sporadically afterwards. Courses for program participants included the chemistry, biology, statistics, and/or mathematics courses that were part of the recommended sequence for the degree, to encourage students to maintain progress on their four-year academic trajectory. Students were not required to participate in the sections sponsored by the program and could instead choose a “normal” section of the course.

A weekly email digest highlighted educationally purposeful events in which the students could become engaged. Typical activities included research seminars given by college faculty, learning assistant and laboratory assistant openings, opportunities to participate in undergraduate research, scholarship deadlines, relevant conferences, and career workshops. Many of these events were organized by the program, such as science movie nights, lunch-and-learn events with current faculty, research funding competitions, panels with upper-level and graduate students, career workshops, trivia nights, and visitors from graduate and professional schools. Students received the weekly email digest for the entire duration of their enrollment. During their third year, a student activities board was created to plan and carry out events of interest.

Data Acquisition and Statistics

The research study was submitted and approved by the institutional review board. Data concerning student attributes, retention, and graduation were managed by the university's office of institutional research, supplied to the researchers without identifying information. GPAs are calculated on a maximum scale of 4.00. Because students often do not enroll in contiguous semesters after matriculation, the number of semesters to graduate was converted into “years of enrollment,” a measure of the time that the student was in classes. In this conversion, a spring or fall semester corresponds to 0.4 years, and a summer semester corresponds to 0.2 years because it is half the number of weeks as a fall or spring semester (e.g., a student taking two fall semesters, two spring semesters, and one summer semester to complete course requirements would graduate in two calendar years but only 1.8 years of enrollment). Compared to calendar years, the “years of enrollment” measure better represents academically relevant time. It also accounts for the summer semester being a shorter semester in which students generally take fewer courses.

When indicated in the text, a t-test of independent means was used to determine whether results were significantly different when comparing the GPA or time to graduation of students, and the Fisher exact test was used to determine whether significant associations existed between the program and comparator groups in graduation, retention, or departure rates. Statistics were performed using calculators on the social science statistics website (Stangroom, 2021).

Results

Student Graduation Rates

The enrollment or graduation status of the students in the program or comparator groups were analyzed four academic years (12 semesters) after their matriculation, which also corresponds to four enrollment years. As shown in Table 2, students taking part in the program graduated at a significantly higher rate than comparator students (43.2% vs. 27.0%, p = .0166). The increase was due to the retention of students in their original major: program participants graduated in their original major at over twice the rate of comparator students (32.4% vs. 15.3%, p = .0043), while the graduation rate of students who had changed to a different major was the same between the two groups. The increase in the graduation rate of program participants was observed across all science and math majors (Table 3), suggesting that the program affected student retention equally. Notable exceptions were observed in chemistry and physics: no students from the comparator group graduated in chemistry, and no program or comparator students with an initial major in physics graduated from the university. Of the program or comparator students graduating in a different major, only one student (in the comparator group) graduated with a degree that would be considered STEM, as defined by the Department of Homeland Security (2020).

Table 2.

Status of Students Four Academic Years After Entry.

Student status	Programn (%)	Comparatorn (%)
Graduated*	48 (43.2)	30 (27.0)
In original major**	36 (32.4)	17 (15.3)
In different major	12 (10.8)	13 (11.7)
Changed major	44 (39.6)	43 (38.7)
Continuing (in any major)	5 (4.5)	10 (9.0)
Left the university***	28 (25.2)	41 (36.9)

Note. “Continuing” indicates students who were still enrolled in the university in any major.

*p = .0166, **p = .0043, or ***p = .059 between program and comparator groups using Fisher exact test.

Table 3.

Graduation Rates by Major.

Major	Program (%)	Comparator (%)
Biochemistry	25	12.5
Biology	38.2	17.5
Chemistry	14.3	0
Environmental science	40	20
Math	66.7	33.3
Physics	0	0

Although a greater number of program students graduated, there were no differences in the graduating GPAs of the students who did or did not participate in the program (data not shown), regardless of whether they graduated in their original major or a different one. Although not statistically different, the GPAs of students graduating in a different major tended lower than those graduating in their original major. The time to graduate was also similar for all groups, averaging around 3.6 years of enrollment, or 10–11 semesters. This shows that students generally took classes for two or three summer semesters, in addition to the typical fall and spring semesters, in order to graduate within four academic years. Taken together, participation in the program led to a greater number of graduating students, particularly in their original major, but did not affect their graduating GPAs or time to graduation.

Students Changing Majors

Nearly 40% of program or comparator students changed their major to one outside the College of Science and Mathematics (39.6% vs. 38.7%, Table 2). Interestingly, the GPA of students changing their major was much higher in comparator students (M = 3.15, SD = 0.63) than program participants (M = 2.67, SD = 0.73), t(85) = -3.25, p = .00082. This indicates that program students who changed their major were primarily those who were struggling with the content, while the comparator group included students who were performing well in their original science or math major but nonetheless changed their major. This suggests that the program was effective in retaining high-performing students in their original major.

A slight difference was observed between program participants and comparator students when examining the majors to which the students changed. The top major for program participants was nursing (22.7%), a STEM-related profession, while the top majors for comparator students were business (20.9%), followed by nursing (14.0%). Only 6.8% of program participants changed to a business major. Exercise science and psychology were common majors selected to a lesser extent by both groups. The reason for these differences is unknown but warrants further investigation.

Students Leaving the University

Program participants were retained and graduated at a higher rate than comparator students, and both groups changed majors at the same rate. The remaining difference in student status is explained by the proportion of comparator students leaving the university. Compared to 25.2% of program students, 36.9% of comparator students left the university before obtaining their degree (p = .059 using the Fisher exact test). Additionally, there was a marked difference in the GPA of program or comparator students when they left the university (Figure 2). Sixty-four percent of program students had a GPA ≥ 3.0 upon departure, compared to only 36% of comparator students. The departing GPA of program students (M = 2.87, SD = 1.05) was significantly higher than comparator students (M = 2.27, SD = 1.12), t(67) = 2.22, p = .015. Taken together, this suggests that the majority of program students could have been eligible for transfer to higher-caliber universities while most comparator students left the university with poorer academic performance.

Figure 2.

Grade point averages (GPAs) of program or comparator students leaving the university. Note. Each box shows the 2nd and 3rd quartiles; the horizontal line in the box indicates the inclusive median. Outliers are shown.

Program Effect on Groups Underrepresented in STEM Fields

We next analyzed whether the program had any effect on the retention and graduation of female students and/or students from underrepresented races/ethnicities. Fifty-three program students and 46 comparator students identified as members of an underrepresented race or ethnicity in STEM, namely Black, Hispanic, or multiracial. As observed with the entire group, underrepresented students who took part in the program graduated at a higher rate than those who did not, particularly in their original major (41.5% vs. 21.7%, p = .052) (Table 4). Students from Black, Hispanic, or multiracial groups showed no differences in graduating GPA or time to graduation as their White or Asian classmates (data not shown). Similar numbers of underrepresented program or comparator students changed their major, and slightly fewer program participants left the university. Thus, the program supported the graduation of Black, Hispanic, and multiracial students to the same extent as their White or Asian peers.

Table 4.

Status of Students with Races or Ethnicities Underrepresented in Science, Technology, Engineering, and Mathematics (STEM), Four Academic Years After Entry.

Student status	Program(N = 53)n (%)	Comparator(N = 46)n (%)
Graduated	25 (47.2)	17 (37.0)
In original major*	22 (41.5)	10 (21.7)
In different major	3 (5.7)	7 (15.2)
Changed major	19 (35.8)	20 (43.5)
Left the university	11 (20.8)	15 (32.6)

*p = .052 using the Fisher exact test, comparing program and comparator students from Black, Hispanic, or multiracial backgrounds graduating in their original major.

We next examined the retention and graduation of women of any race or ethnicity (Table 5). Sixty-nine program participants and 67 comparator students identified as female. Highly significant differences were observed between women who did or did not take part in the program. Although women from either group changed their major at the same rate, women participating in the program left the university at lower rates (20.3% vs. 41.8%, p = .0091) and with much higher GPAs (M = 3.26, SD = 1.05) than women leaving from the comparator group (M = 2.31, SD = 1.12), t(40) = 2.90, p = .003. Female and male students who participated in the program left at the same rate, while 68.3% of comparator students who left the university were female (data not shown).

Table 5.

Status of Female Students, Four Academic Years After Entry.

Student status	Program(N = 69)n (%)	Comparator(N = 67)n (%)
Graduated*	34 (49.3)	17 (25.4)
In original major**	25 (36.2)	10 (14.9)
In different major	9 (13.0)	7 (10.4)
Changed major	29 (42.0)	26 (38.8)
Left the university***	14 (20.3)	28 (41.8)

*p = .0047, **p = .0058, or ***p = .0091 between women participating in the program or in the comparator group, using the Fisher exact test.

Pertaining to graduation, the graduation rate in any major (49.3% vs. 25.4%, p = .0047) or in their original science/math major (36.2% vs. 14.9%, p = .0058) was higher in women who had participated in the program. Women from races or ethnicities underrepresented in STEM (Black, Hispanic, or multiracial backgrounds) graduated at 47.1% in the major, compared to 17.6% of comparator students, p = .0186. In fact, Black women graduated in their original major at the highest rate of any students in the program (61.9% compared to 15.8% of comparator students, p = .004).

Interesting results were observed when more closely analyzing the graduation success of program students versus comparator students, who were eligible to take part in the program but declined to do so (Table 6). In the comparator group, men and women graduating in their original major did so at roughly the same rates (15.9% vs. 14.9%). In program participants, both sexes graduated at higher rates than comparator students, although women appeared to benefit more from the program: graduation rates rose to 26.2% for men and 36.2% for women.

Table 6.

Rates of Students Graduating with a Science or Math Major, by Sex and Race/Ethnicity.

	Male (%)	Female (%)	Both sexes (%)
Program participants
Asian	100^a	25.0^a	40.0
Black	30.0	61.9**	51.6***
Hispanic	20.0	30.0	26.7
Multiracial	50.0^a	0^a	28.6
White	18.2	25.8	22.6
All races/ethnicities	26.2	36.2*	32.4****
Comparator students
Asian	50.0^a	0	22.2
Black	50.0^a	15.8**	21.7***
Hispanic	0^a	14.3	11.1
Multiracial	33.3	25.0	28.6
White	3.7	14.3	8.9
All races/ethnicities	15.9	14.9*	15.3****

Indicates a category with <5 students.

*p = .0058, **p = .0041, ***p = .047, ****p = .0043 using the Fisher exact test to compare the two groups indicated by corresponding asterisks.

Looking at the race and ethnicity breakdown of students in the comparator group graduating in their original major, White students graduated at the lowest rates of any race or ethnicity (8.9%). This was particularly true for White males, only 3.7% of whom graduated with a degree in science or math. These percentages increased for White students participating in the program, to 22.6% of all White students and 18.2% of White men. Although low numbers in some categories prevent a robust analysis of all the data, the graduation rates indicate that the program was effective in providing an environment that supported the retention of students from all backgrounds within their science or math major, particularly for women.

Discussion and Implications

This study investigated the effects of an enrollment-to-graduation program on the persistence and progress of students with STEM majors, specifically those in science or mathematics. The first research question asked whether the program could affect the persistence and graduation of students in their originally selected major. There was no difference in the number of comparator or program students who changed their major. However, the GPA of comparator students who changed their major was significantly higher, suggesting that the program was able to retain students performing well who may not have kept their original STEM major if they did not participate in the program. Concerning graduation rates, program participants graduated at a 15.2% higher rate than comparator students (43.2% vs. 27.0%, p = .0166 using Fisher exact test), particularly in their originally selected major (32.4% vs. 15.3%, p = .0043 using Fisher exact test), although the average graduating GPA and time to graduation were the same for both groups. Thus, the results suggest that a comprehensive program can indeed lead to better retention of students in their originally selected STEM major.

The second research question asked whether such a program could affect the retention of students at the University, regardless of their major. The study found that significantly more students in the comparator group left the university than program participants (36.9% vs. 25.2%, p = .059 using the Fisher exact test). Additionally, comparator students leaving the university had lower GPAs at departure, suggesting that academic challenges may have been a contributing reason. Considering the rates of students graduating or changing majors (Table 2), the reduction observed in the graduation rate of comparator students is explained by their departure from the university. Taken together, this suggests that the program was effective in retaining students in their original major who would otherwise have left the university.

The third research question asked whether the program succeeded in providing an equitable environment that better supports underrepresented STEM students. Program participants from underrepresented groups graduated in their original STEM major at nearly twice the rate of comparator students (41.5% vs. 21.7%, p = .052). Women, particularly those from underrepresented racial or ethnic backgrounds, were also retained at a higher rate in their original major if they participated in the program. Therefore, it appears that our program was successful in providing an environment that fostered the retention and success of students from groups that leave STEM majors at disproportionally high rates.

Findings from this study provide deeper insights into how to retain students, particularly those from groups underrepresented in STEM. The program was carefully designed to include components to foster a sense of belonging, support student challenges, provide educationally purposeful activities, keep students on a consistent academic trajectory, and allow students to better understand their major of choice. The program increased the retention of students in their STEM major regardless of their specific major, sex, or demographic attributes, highlighting that such programs have the potential to greatly affect the number of students who graduate in STEM fields. Women and students from underrepresented races or ethnicities showed some of the largest benefits, suggesting that the program was able to foster a more inclusive environment than is typical in STEM degree programs. It is unknown which aspects of the program were most effective, but considering that a sense of belonging and academic/social support can compensate for feelings of marginalization (e.g., Brown, 2000; Han et al., 2017, 2020; Hoffman, 2002; Hoffman & Lowitzki, 2005; Loo & Rolison, 1986; Walton & Cohen, 2007, 2011), it is likely that these were important components for promoting inclusion of all individuals.

A few important points emerged from the results of this study that should not be overlooked. First, students in either group took an average of 3.6 enrollment years, corresponding to 10–11 semesters, to graduate. This shows that students often took classes during the summer semesters in order to graduate in four academic years. Moreover, these particular students were classified within the top 40% of students entering the college, based upon high school academic metrics. A takeaway is that it is unreasonable to expect that students will graduate in four academic years by only attending spring and fall semesters, at least not at institutions similar to the large public comprehensive university in which the study took place. Thus, administrators should encourage students to take classes during the summer semesters in order to catch up on the recommended course sequence or to alleviate the load of future semesters with more difficult courses. A second important observation is that doing well academically does not ensure that a student will stay at an institution. Inclusive programs can encourage students to remain, but students transfer out for a variety of reasons. For instance, at the university in which this study was performed, an internal campus climate survey in 2015 revealed that 42% of students in this college intended to transfer or never intended to finish their degree at this institution. Other common reasons that good students leave include financial reasons, personal/family reasons, or unavailability of specialized majors (e.g., biotechnology vs. biology, or chemical engineering vs. chemistry). Although retention and graduation statistics are frequently used as metrics by universities and governments to allocate funds, it is not always realistic to assume that high rates can be achieved, particularly at institutions with lower admission standards or at which students do not intend to stay long-term. And ethically, universities should not discourage students from transferring to institutions of higher ranking. Instead of graduation rates, a fairer funding metric might take into account the success of students while at the institution, or use appropriate comparator universities and past data to gauge progress. Considering the strong correlation of high school GPA and SAT/ACT scores with graduation success, modified funding metrics would ensure a more equitable distribution of funds to institutions that train students with lesser academic preparation. In any case, high historical transfer rates or lower admission standards should not excuse universities from creating inclusive environments that foster the success of all their students.

Limitations and Future Directions

There are a few limitations to this study. First, students elected to participate in the program. Although academic and demographic attributes were highly similar between the program and comparator groups, it is unknown if self-selection bias existed. Additionally, data for other variables shown to affect student retention, such as socioeconomic status, were not available. Concerning students who left the university, it is unknown whether departing students later graduated from other universities (and the type of university) or left higher education entirely.

Because the program incorporated many interventions to address potential barriers to student success, it is not possible to determine from this study which interventions, alone or in concert, were most beneficial to persistence and retention. It is also unknown if a shorter program (e.g., one semester or one year) would have shown similar effectiveness as this program, which spanned the student's entire undergraduate experience. Future studies will examine the contribution of individual program aspects and the required duration of effective components, information that will be useful for prioritizing activities in environments with limited funding or personnel. Future research will also investigate the possible influence of noncognitive student factors, such as perseverance, well-being, and self-confidence, in the likelihood of students persisting until graduation with their original STEM major.

Conclusion

It is well established that nearly half of the students who initially declare STEM majors in college do not graduate with a STEM degree, and recent research has shown that women and students from underrepresented racial/ethnic groups are less likely to persist in STEM fields specifically (e.g., Astorne-Figari & Speer, 2018, 2019; Ganley et al., 2018; Jones et al., 2021; Riegle-Crumb et al., 2019; Xie et al., 2015). Few studies, however, have examined the real-world effectiveness of evidence-based initiatives on the retention and graduation of STEM students. This study contributes significantly to the existing body of knowledge by demonstrating that such comprehensive programs can greatly affect the retention of students in STEM majors, including those historically underrepresented in STEM fields. Students were invited to take part in a program that started at entry and continued until their graduation. The program included components supported by the literature to build community, provide academic and social support, promote engagement in academically purposeful activities, encourage academic progress, and foster goal realization. The program more than doubled the number of students who graduated in their originally declared STEM major. It notably increased the retention of women, particularly Black and Hispanic women. The results suggest a method of creating an equitable environment that promotes the persistence and graduation of STEM students. Such programs and similar initiatives will be necessary to evolve a historically unwelcoming STEM culture into one that supports the growth and success of all students.

Footnotes

Acknowledgments

The author acknowledges the Dean's Office, the departments of the college, and the faculty involved in the program for their participation, support, and assistance. The author would also like to highlight the efforts of Leah Weaver, Vickie Burris, Sara Franka, and the college's advising office, particularly Hannah Stocks and Jaely Cruz. Additional thanks are given to Lewis VanBrackle, III, and Marla Bell for the use of their predicted success model.

Declaration of Conflicting Interests

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

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author received funding for the research project through the Kennesaw State University Center for Excellence in Teaching and Learning. Program activities were funded primarily by the institution with generous community grants from Greystone Power and the Carlyle Fraser Fund.

ORCID iD

Jennifer Louten

Author Biography

Jennifer Louten, PhD, is a professor of biology at Kennesaw State University in the greater Atlanta, Georgia, metropolitan area. In addition to teaching, she has directed multiple transition programs for first-year or transfer students and leads efforts to foster student persistence and graduation.

References

Astin

(1997). How “good” is your institution's retention rate? Research in Higher Education, 38(6), 647–658.

Astin

A. W.

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

Astorne-Figari

Speer

J. D.

(2018). Drop out, switch majors, or persist? The contrasting gender gaps. Economics Letters, 164, 82–85. https://doi.org/10.1016/j.econlet.2018.01.010

Astorne-Figari

Speer

J. D.

(2019). Are changes of major major changes? The roles of grades, gender, and preferences in college major switching. Economics of Education Review, 70, 75–93. https://doi.org/10.1016/j.econedurev.2019.03.005

Banks

Dohy

(2019). Mitigating barriers to persistence: A review of efforts to improve retention and graduation rates for students of color in higher education. Higher Education Studies, 9(1), 118–131. https://doi.org 10.5539/ hes.v9n1p118

Bean

(1990). Using retention management in enrollment management. Jossey-Bass.

Beasley

M. A.

Fischer

M. J.

(2012). Why they leave: The impact of stereotype threat on the attrition of women and minorities from science, math and engineering majors. Social Psychology of Education, 15, 427–448. https://doi.org/10.1007/s11218-012-9185-3

Braxton

J. M.

(2016). Reworking the student departure puzzle. Vanderbilt University Press.

Braxton

J. M.

Hirschy

A. S.

McClendon

S. A.

(2011). Understanding and reducing college student departure: ASHE-ERIC higher education report, volume 30, number 3. John Wiley & Sons.

10.

Braxton

J. M.

Sullivan

A. S.

Johnson

(1997). Appraising Tinto's theory of college student departure. In Smart

(Ed.), Higher education: Handbook of theory and research (Vol. 12, pp. 107–164). Agathon.

11.

Brown

T. L.

(2000). Gender differences in African American students’ satisfaction with college. Journal of College Student Development, 41(5), 479–487.

12.

Chen

(2013). STEM attrition: College students’ paths into and out of STEM fields (NCES 2014–001). National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education.

13.

Coates

(2014). Students’ early departure intentions and the mitigating role of support. Australian Universities’ Review, 56(2), 20–29.

14.

DeBerard

M. S.

Spielmans

G. I.

Julka

D. L.

(2004). Predictors of academic achievement and retention among college freshmen: A longitudinal study. College Student Journal, 38(1), 66–81.

15.

Department of Homeland Security. (2020). DHS STEM designated degree program list. U.S. Immigration and Customs Enforcement. https://www.ice.gov/sites/default/files/documents/stem-list.pdf.

16.

Eagan

Hurtado

Figueroa

Hughes

(2015). Examining STEM pathways among students who begin college at four-year institutions. National Academy of Sciences.

17.

Ganley

C. M.

George

C. E.

Cimpian

J. R.

Maowski

M. B.

(2018). Gender equity in college majors: Looking beyond the STEM/non-STEM dichotomy for answers regarding female participation. American Educational Research Journal, 55(3), 453–487. https://doi.org/10.3102/0002831217740221

18.

Geiser

Santelices

(2007). Validity of high-school grades in predicting student success beyond the freshman year: High school record vs. standardized tests as indicators of four-year college outcomes (occasional paper series, 6.07). Center for Studies in Higher Education.

19.

Goodman

Schlossberg

N. K.

Anderson

M. L.

(2006). Counseling adults in transition (3rd ed.). Springer.

20.

Han

Farruggia

S. P.

Moss

T. P.

(2017). Effects of academic mindsets on college Students’ achievement and retention. Journal of College Student Development, 58(8), 1119–1134. https://doi.org/10.1353/csd.2017.0089

21.

Han

Farruggia

S. P.

Solomon

B. J.

(2020). Effects of high school students’ noncognitive factors on their success at college. Studies in Higher Education, 1–15, https://doi.org/10.1080/03075079.2020.1770715.

22.

Hoffman

J. L.

(2002). The impact of student cocurricular involvement on student success: Racial and religious differences. Journal of College Student Development, 43(5), 712–739.

23.

Hoffman

J. L.

Lowitzki

K. E.

(2005). Predicting college success with high school grades and test scores: Limitations for minority students. The Review of Higher Education, 28(4), 455–474. https://doi.org/10.1353/rhe.2005.0042

24.

Johnson

(2008). Enrollment, persistence, and graduation of in-state students at a public research university: Does high school matter? Research in Higher Education, 49(8), 776–793.

25.

Jones

Marks

Ramirez

Ríos-Vargas

(2021). 2020 census illuminates racial and ethnic composition of the country. U.S. Census Bureau.

26.

Kahu

E. R.

Nelson

(2018). Student engagement in the educational interface: Understanding the mechanism of student success. Higher Education Research and Development, 37(1), 58–71. https://doi.org/10.1080/07294360.2017.1344197

27.

Kim

(2015). Predictors of college retention and performance between regular and special admissions. Journal of Student Affairs Research and Practice, 52(1), 50–63. https://doi.org/10.1080/19496591.2015.995575

28.

Kinzie

Gonyea

Shoup

Kuh

G. D.

(2008). Promoting persistence and success of underrepresented students: Lessons for teaching and learning. New Directions for Teaching and Learning, 2008(115), 21–38. http://doi.org/10.1002/tl.323

29.

Korobova

Starobin

S. S.

(2015). A comparative study of student engagement, satisfaction, and academic success among international and American students. Journal of International Students, 5(1), 72–85.

30.

Kuh

G. D.

(2016). Making learning meaningful: Engaging students in ways that matter to them. New Directions for Teaching and Learning, 2016(145), 49–56. https://doi.org/10.1002/tl.20174

31.

Kuh

G. D.

Kinzie

Cruce

Shoup

Gonyea

R. M.

(2007). Connecting the dots: Multi-faceted analyses of the relationships between student engagement results from the NSSE, and the institutional practices and conditions that foster student success. Indiana University Center for Postsecondary Research.

32.

Langdon

McKittrick

Beede

Khan

Doms

(2011). STEM: Good jobs now and for the future. ESA issue brief #03–11. U.S. Department of Commerce.

33.

Loo

C. M.

Rolison

(1986). Alienation of ethnic minority students at a predominantly white university. Journal of Higher Education, 57(1), 58–77.

34.

Murphy

M. C.

Steele

C. M.

Gross

J. J.

(2007). Signaling threat: How situational cues affect women in math, science, and engineering settings. Psychological Science, 18(10), 879–885.

35.

National Center for Education Statistics (202°c). Table 318.45, number and percentage distribution of science, technology, engineering, and mathematics (STEM) degrees/certificates conferred by postsecondary institutions, by race/ethnicity, level of degree/certificate, and sex of student: 2009–2010 through 2018–2019. U.S. Department of Education.

36.

National Center for Education Statistics (2020a). Table 322.40, bachelor's degrees conferred to males by postsecondary institutions, by race/ethnicity and field of study: 2017–2018 and 2018–2019. U.S. Department of Education.

37.

National Center for Education Statistics (2020b). Table 322.50, bachelor's degrees conferred to females by postsecondary institutions, by race/ethnicity and field of study: 2017–2018 and 2018–2019. U.S. Department of Education.

38.

National Student Clearinghouse (2020). First-year persistence and retention: 2018 beginning cohort. National Student Clearinghouse Research Center.

39.

Norton

(2012). Graduate winners: Assessing the public and private benefits of higher education. Grattan Institute.

40.

Pfeffer

F. T.

(2018). Growing wealth gaps in education. Demography, 55(3), 1033–1068. https://doi.org/10.1007/s13524-018-0666-7

41.

President's Council of Advisors on Science and Technology (2012). Engage to excel: Producing one million additional college graduates with degrees in science, technology, engineering, and mathematics. United States White House.

42.

Rayle

A. D.

Chung

K. Y.

(2007). Revisiting first-year college students’ mattering: Social support, academic stress, and the mattering experience. Journal of College Student Retention: Research, Theory & Practice, 9(1), 21–37.

43.

Rendón

L. I.

(1994). Validating culturally diverse students: Toward a new model of learning and student development. Innovative Higher Education, 19(1), 33–51.

44.

Riegle-Crumb

King

Irizarry

(2019). Does STEM stand out? Examining racial/ethnic gaps in persistence across postsecondary fields. Educational Researcher, 48(3), 133–144. https://doi.org/10.3102/0013189X19831006

45.

Sanford

(1966). Self and society: Social change and individual development. Atherton Press.

46.

Schlossberg

N. K.

(1989). Marginality and mattering: Key issues in building community. New Directions for Student Services, 48(1), 5–15.

47.

Seidel

Kutieleh

(2017). Using predictive analytics to target and improve first year student retention. Australian Journal of Education, 61(2), 200–218. https://doi.org/10.1177/0004944117712310

48.

Stangroom

(2021). Social science statistics. https://www.socscistatistics.com/

49.

Stillman

(2009). Making the case for the importance of student retention. Enrollment Management Journal, 3(2), 76–91.

50.

Strenta

A. C.

Elliott

Adair

Matier

Scott

(1994). Choosing and leaving science in highly selective institutions. Research in Higher Education, 35(5), 513–547.

51.

Tinto

(1993). Leaving college: Rethinking the causes and cures of student attrition (2nd ed.). University of Chicago Press.

52.

Torpey

(2018). Measuring the value of education. Career Outlook, U.S. Bureau of Labor Statistics. https://www.bls.gov/careeroutlook/2018/data-on-display/education-pays.htm.

53.

Tracey

Robbins

(2006). The interest-major congruence and college success relation: A longitudinal study. Journal of Vocational Behavior, 69(1), 64–89. https://doi.org/10.1016/j.jvb.2005.11.003

54.

United States Bureau of Labor Statistics (2021). Employment in STEM occupations: Table 1.11, Employment in STEM occupations, 2020 and projected 2030 . https://www.bls.gov/emp/tables/stem-employment.htm

55.

Walton

G. M.

Cohen

G. L.

(2007). A question of belonging: Race, social fit, and achievement. Journal of Personality and Social Psychology, 92(1), 82–96. https://doi.org/10.1037/0022-3514.92.1.82

56.

Walton

G. M.

Cohen

G. L.

(2011). A brief social-belonging intervention improves academic and health outcomes of minority students. Science, 331(6023), 1447–1451.

57.

Watts

(2001). Education and the common good: Social benefits of higher education in Kentucky. Kentucky Long Term Policy Research Center.

58.

Xie

Fang

Shauman

(2015). STEM education. Annual Review of Sociology, 41, 331–357. https://doi.org/10.1146/annurev-soc-071312-145659