Attrition and Persistence in Undergraduate Physics Programs

Executive Summary

This five-year longitudinal study aimed to understand how undergraduate students become interested in a physics bachelor’s degree and persist in that degree until graduation. Furthermore, this study aimed to understand how physics interest and persistence may differ by gender identity and race/ethnicity. Compared to science, technology, engineering, and mathematics (STEM) fields in general, the number of physics bachelor’s degree earners is growing more slowly over time (Mulvey & Nicholson, 2020), and physics students are more likely to switch to another college major during their undergraduate education (National Center for Education Statistics, 2017). Physics students from underrepresented groups, including women, Black or African American students, and Hispanic or Latino students, are even more likely to drop a physics major (Turnbull et al., 2019; Whitcomb & Singh, 2021). We surveyed 3,917 students in the first week of their introductory college physics course at four large, predominately White universities, and asked students whether they intended to major in physics. There were 745 introductory students considering a physics major who were sent follow-up surveys annually for five years. Using data from surveys and one-on-one interviews, we compared the experiences of students who graduated with a physics degree and students who were no longer interested in a physics major.

Who is more likely to become interested in a physics major and why?

Of the 3,917 introductory physics students we surveyed, 19% were considering a physics major. Most students became interested in physics because of their high school physics experiences. They were drawn to the physics field because of the opportunity to learn how the world works, apply their problem-solving and math skills, and pursue interesting career options. Students who were interested in a physics major were more confident in their physics abilities and believed they would receive a higher grade in their current introductory physics class compared to students who were not interested in a physics major.

Who is more likely to leave or persist in a physics major and why?

Of the 745 students in our study who were initially interested in a physics major, 31% persisted in their physics major until graduation. Over 70% of students who lost interest in a physics major did so during their first or second year, demonstrating that introductory college physics courses play a key role in retaining physics majors. Students most commonly left physics because they were interested in another major or career path; however, some students left because of barriers they experienced in physics. Compared to students who persisted, students who lost interest in a physics major were less likely to interact with physics professors outside of class, reported lower self-efficacy in math, rated the physics department climate as less positive, and were less likely to do research. They reported more challenges with lower quality physics teaching and rigid course sequencing in physics.

How do these outcomes differ by gender identity and race/ethnicity?

Introductory physics students who identified as women and Black or African American were significantly less likely to be interested in pursuing physics as a major in the first place compared to students identifying as White men. Underrepresented introductory physics students (identifying as women, Black or African American, and Hispanic or Latino) who were not interested in a physics major reported lower self-efficacy in math, expected to receive a lower grade in their physics course, and were less likely to have taken high school physics or calculus. These findings were not the same for women who were interested in a physics major. These women did not show any significant differences in their self-efficacy or prior course experiences compared to men, whereas we still found significantly lower self-efficacy for Black or African American and Hispanic or Latino students, even if they were interested in a physics major.

We did not find any significant differences when we looked longitudinally at the students who wanted to major in physics and the likelihood of underrepresented students leaving a physics major. Even if underrepresented students were not more likely to leave physics, these students still had different experiences in physics than students who identified as White men. Students who identified as women or an underrepresented race/ethnicity group member felt their physics courses were less interactive, their physics professors were less encouraging, and believed their physics peers were more skilled than themselves. They were also more likely to hear about, witness, or experience discrimination in their physics courses and departments.

Asian or Asian American introductory physics students were more likely to have an initial interest in a physics major. Despite also reporting lower self-efficacy in math and a lower likelihood of taking high school physics than White students, these students still felt more confident they would receive a higher grade in their current physics course and were less discouraged in physics.

Acknowledgments

We want to personally thank all the individuals who helped make this project possible over the last five years. Most importantly, we want to thank the students at the participating physics departments for sharing their experiences in our student surveys and interviews, and the department faculty and staff for helping recruit physics students into our study.

We thank Dr. Evalyn Gates for input on the original project proposal and the conceptual design.

We thank Dr. Susan White, Dr. Laura Merner, and Patrick Mulvey for their help with designing the surveys and recruiting participating departments.

We thank Dr. Laura Merner, Dr. Courtney Walsh, and John Tyler for their work in distributing the surveys, performing student interviews, and assisting with data analysis.

We thank all the reviewers who helped us write the conclusions and recommendations in this report, including Dr. Laurie McNeil (University of North Carolina at Chapel Hill), Dr. Laura McCullough (University of Wisconsin-Stout), Debbie Andres (Paramus High School), Arlene Modeste Knowles (AIP, TEAM-UP), Dr. Trevor Owens (AIP), and the additional reviewers who wish to remain anonymous.

Table of Contents

• Introduction
• Interest in Undergraduate Physics
  • Section Summary: How Can We Recruit More Undergraduate Physics Students?
• Persistence of Undergraduate Physics Students
  • Section Summary: How Can We Help Students Persist in Undergraduate Physics Programs?
• Limitations
• Conclusions
• Appendix A: Methodology
• Appendix B: Literature Review
• References

Introduction

Physics is a field that allows us to have a better understanding of the world around us and pursue scientific breakthroughs that can increase the quality of life for people worldwide (Shahid, 2020). For us to continue growing the field further, it’s essential to implement high-quality, evidence-based physics teaching practices at schools and universities, and provide support and encouragement to all students interested in becoming physicists.

Bachelor’s degree attainment and retention in physics

Recruitment in physics has room for improvement. Physics only represents around 2% of all STEM bachelor’s degrees earned (National Center for Education Statistics, 2020). As illustrated in Figure 1, although the number of students earning physics bachelor’s degrees has been increasing over the past two decades, the rate is slower than students earning STEM bachelor’s degrees in general (Mulvey & Nicholson, 2022).

Figure 1

Even after students declare a physics major, changing majors during undergraduate education is common. The National Center for Education Statistics (2017) showed that 33% of students switch majors during college. Furthermore, physics students may switch more often compared to other disciplines. A total of 40% of natural science undergraduate students (including physics and biological sciences) switched to another major, which is higher than undergraduate students in computer science (28%), engineering (32%), or STEM degrees in general (35%; National Center for Education Statistics, 2017).

Women and underrepresented race/ethnicity groups among physics bachelor’s degree earners

Data shows that women and students from certain race/ethnicity groups are underrepresented in physics. In 2021, women earned only 24% of physics bachelor’s degrees, which differs greatly from STEM fields such as biomedical science, where women earned 66% of bachelor’s degrees (National Center for Education Statistics, 2021). As illustrated in Figure 2, the rate of women earning degrees in physics has not increased at the same rates as women in biomedical science, chemistry, or astronomy. Previous data has shown that women are among 46% of high school physics students (Porter and Ivie, 2019). Because the percentage of women is higher among high school physics students than physics bachelor’s degree earners, it is important to understand whether, during introductory physics, women are less interested in majoring in physics than men are, or if women initially are just as interested in a physics major but are more likely to leave the physics major later in their undergraduate careers. To do this, we must follow students through their college experience to understand when the gender gap among physics bachelors occurs.

Figure 2

Students identifying as Black or African American and/or Hispanic or Latino are also less represented among physics bachelor’s degree earners compared to other STEM fields (National Center for Education Statistics, 2021). For example, as illustrated in Table 1, in 2021, Black or African American students earned 9% of all bachelor’s degrees but only earned 3% of the bachelor’s degrees in physics.

Table 1

Although both groups are underrepresented in physics, the proportion of Hispanic or Latino physics students who earn a physics bachelor’s degree has been increasing, but the same increase has not occurred for Black or African American physics students. Figure 3 provides an illustration of these trends based on data from the National Center for Education Statistics.

Figure 3

When examining retention rates in physics, studies have shown that women were 1.8 times more likely to switch from physics to a life science major after taking the required math courses for physics (Turnbull et al., 2019), and 65% of students from underrepresented race/ethnicity groups (Black, Hispanic, or “another race/ethnicity”) dropped a physics major compared to 34% of White students (Whitcomb & Singh, 2021). However, this is not unique to physics. Women were also less likely to declare a STEM major, and women, as well as students who identified as Black or Hispanic, were more likely to leave a STEM major during their undergraduate education (Beasley & Fischer, 2012). For example, 32% of women switched out of a STEM degree compared to 26% of men (National Center for Education Statistics, 2013), and 58% of White students persisted in a STEM degree compared to only 43% of Hispanic or Latino students and 34% of Black or African American students, who either switched majors or left their institution entirely (Riegle-Crumb et al., 2019).

Why students leave or persist in a college major

As part of the physics field’s effort to improve the current growth rate among physics degree earners, all educators and education researchers need to better understand why students persist in or leave a physics major during their undergraduate education. The social cognitive career theory (SCCT; Lent et al., 1994) provides a well-validated and helpful framework in understanding physics academic and career interests through four key factors: 1) self-efficacy, or confidence in the ability to successfully perform required actions (Bandura, 1977); 2) outcome expectations for a career choice; 3) personal goals and interests to pursue a career path; and 4) environmental influences including social support, financial support, and professional development opportunities.

A detailed literature review of the factors in the SCCT model and how they apply to STEM students in general and underrepresented students in STEM can be found in Appendix B. In summary, the literature supports the applicability of the SCCT model. STEM students who persisted were more likely to have higher self-efficacy in science research, more positive perceptions about science job opportunities, more experience with science and math courses in high school, more positive course experiences with skilled STEM teachers and smaller class sizes, more financial support, and more social support from faculty members, family, and peers. Women in STEM were also more likely to persist when they felt more social belonging and interacted with female faculty members and peers in their programs. Previous studies utilizing the SCCT model have yet to apply this framework to physics interest and retention, and it is likely that all of these factors also play a role in the experience of undergraduate physics majors.

The current project

The current report presents our findings and recommendations based on three research questions. For these questions, we examined the role of self-efficacy, career expectations, academic background, course experiences, social support, and financial support on our study outcomes.

1. Who is more likely to be interested in a physics major and why?
2. Who is more likely to leave or persist in a physics major and why?
3. How do these outcomes differ by gender identity and race/ethnicity?

The current study aimed to study the interest and persistence of students in undergraduate physics majors in two unique ways. First, we applied the social cognitive career theory (SCCT) to undergraduate physics, which has not been done in the SCCT literature before. Previous research applied the model to STEM in general or nonphysics disciplines such as engineering. Using this theory allowed us to comprehensively examine all the known factors related to persisting in a physics major. Second, we conducted this study longitudinally over five years and followed undergraduate students from enrollment in their first introductory college physics course, and if they persisted, until graduation with a physics bachelor’s degree. This enabled us to more objectively determine whether a physics student persisted. Previous studies used only a measure of a student’s intention to persist and did not explore whether a student actually graduated. Third, we were able to conduct this longitudinal analysis by gender identity and race/ethnicity, and examine if underrepresented students who identify as women, Black or African American, and Hispanic or Latino may leave a physics major at different rates or for different reasons.

This longitudinal study was conducted at four large, predominately White universities in the United States between 2018 and 2023. We surveyed 3,917 physics students across 44 introductory physics courses in their first week and asked if they were interested in pursuing a physics major. Introductory physics courses included traditional lectures, seminars, or small group formats. In some physics courses, only honors students or students who had declared a physics major upon entering college were enrolled, while other physics courses enrolled all STEM majors in general. Additional details about the methodology can be found in Appendix A.

We sent students who indicated they were interested in a physics major follow-up surveys at the end of their introductory physics course and at the beginning of each fall semester for the next five years. In these follow-up surveys, we asked students if they were still interested in majoring in physics. This allowed us to identify students who were no longer interested in a physics major. In addition, to include any physics graduates who may have transferred or switched into the major during the course of the study, we surveyed all students who were graduating with a physics degree whether they were part of the initial cohort or not across the five years. Both students leaving their physics major and students graduating in physics participated in in-depth interviews about their experiences.

back to top

Interest in Undergraduate Physics

Our first study goal was to learn more about when and why students become interested in physics. We started by surveying 3,917 students in introductory college-level physics courses. During the first week of class, we asked them, “Are you interested in pursuing a physics major?” Figure 4 shows all the majors of interest for introductory physics students in the study. We found that 19% of the 3,917 introductory physics students in our study were considering a physics major. The most popular chosen majors for introductory physics students were engineering or computer science. Students in the study were enrolled in a wide variety of introductory physics courses, including lecture halls, flipped/reverse classrooms, or SCALE-UP classrooms, and were enrolled in classes with general STEM students, honors students, or physics majors.

Figure 4

Why graduating physics students became interested in physics

We wanted to understand why students feel drawn to physics in the first place. We asked 75 graduating physics students to tell us how they became interested in physics. These students felt physics was a unique and beautiful way to understand the world around them.

Physics presented students with opportunities for interesting and satisfying problem solving.

Physics was a unique incorporation of science and math.

Physics gave students a feeling of accomplishment by succeeding in a challenging major.

And lastly, physics presented these students with a satisfying and flexible career path.

When graduating physics students became interested in physics

In our interviews with graduating physics students, we asked when they became interested in physics. Table 2 shows the percentage of interviewed students who became interested in physics at different time points.

Table 2

High school courses and programs. The majority of graduating physics students (72%) became interested in physics because of their high school experiences. They had excellent experiences with high school physics teachers and courses.

Outside of high school courses, these students also had positive experiences in physics while participating in summer high school programs and internships. These opportunities were beneficial in showing students what a career in physics would look like and building their confidence.

Childhood science media exposure. Some students (12%) became interested in physics even earlier, by watching and reading science media:

College courses. Not all graduating physics students became inspired to pursue physics before college. Some became interested through positive experiences with their college physics professors and courses (16%).

Furthermore, a small group (38) of physics majors in our study previously pursued a major in another field and switched to a physics major during college. As illustrated in Figure 5, our survey results indicated that most students who switched to physics came from engineering or computer science.

Figure 5

These students elaborated on why they switched to a physics major. Some felt their initial major did not give them the experience they wanted. Others avoided physics at first due to a lack of self-efficacy, an individual’s belief in their ability to reach and achieve specific goals (Bandura, 1977). It is important to note that one’s self-efficacy in physics might not reflect their actual performance in physics. For example, studies have found that women report lower self-efficacy in physics and math than men, regardless of their actual performance level (Marshman et al., 2018: Zander et al., 2020).

Pre-college background of introductory physics students

Over 90% of our introductory physics students had taken a previous calculus or physics course at the high school level. The most common high school calculus-based course taken by students was AP Calculus AB and the most common high school physics course taken by students was outside of an AP/IB program. A greater percentage of students had previously taken AP-level calculus courses than physics courses. A breakdown of study participants’ high school physics course backgrounds is provided in Table 3.

Table 3

To assess their physics and math self-efficacy, which is an individual’s belief in their ability to reach and achieve specific goals in physics and math (Bandura, 1977), we asked students to guess their expected grade in their current introductory physics course and share their perceived abilities in math. In the first week of introductory college physics, over 90% of students felt they were good at math and would achieve an A or B in the physics course (see Table 4).

Table 4

We also assessed whether students’ parents or guardians earned a bachelor’s degree (in STEM or in general). Most students had at least one parent or guardian who graduated from college, and over half of these parents or guardians graduated with a degree in STEM (see Table 5).

Table 5

Differences between students initially interested and not interested in majoring in physics

We hypothesized that the factors listed above (self-efficacy in math, expected grade in their current introductory physics course, whether they had taken a calculus or physics course, and whether their parent/guardian earned a STEM degree) might differ between introductory physics students who were interested and introductory physics students who were not interested in a physics major. Controlling for gender identity, race/ethnicity, and institution in a logistic regression, we only found one significant difference in interest in a physics major. As illustrated in Figure 6, students who were not interested in a physics major believed they would earn a lower grade in their introductory physics course. This could be because they were less confident in their physics skills or less invested in their grade because of a lack of interest in physics as a major. High school course experience, parent or guardian education, and math self-efficacy did not show any statistical significance in the analysis.

Figure 6

Differences among students from underrepresented groups in introductory physics

Interest in a physics major was not the same across introductory physics students of different gender identities and races/ethnicities. In our study, we found that women, Black or African American students, and Middle Eastern students during their introductory physics courses were less likely to be interested in a physics major, while Asian or Asian American students and men were more likely to be interested in a physics major. Table 6 shows this by comparing the percentage of students interested in a physics major to the percentages of all students enrolled in introductory physics courses by gender identity and race/ethnicity.

Table 6

To further understand why students from certain underrepresented groups were not interested in a physics major, we used regression models to determine whether there were gender identity and race/ethnicity differences in students’ background factors (self-efficacy in math, expected grade in their current introductory physics course, whether they had taken a calculus or physics course, and whether their parent/guardian earned a STEM degree). We controlled for the students’ institution in all regression models. To see if background experiences varied for students interested in a physics major and students not interested in a physics major, we performed two sets of five regression models. Due to smaller group sizes, we were unable to see any statistically significant results for students identifying as Middle Eastern, Native American or American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, or “another race/ethnicity.”

When comparing the background experiences of men and women, the regressions showed very different results based on physics major interest. Women who were interested in a physics major had no statistically significant differences in their self-efficacy or course background compared to men. However, women who were not interested in a physics major showed a completely different pattern. Compared to men, these women were less likely to have taken high school calculus and physics, were less confident in their math skills, and expected a lower grade in their current introductory physics course. This shows that upon entering an introductory college physics course, women with less interest in a physics major have different perceptions about their self-efficacy and less prior physics course experience compared to men with less interest in a physics major. A summary of these findings is presented in Table 7.

Table 7

The regression findings were similar for people from different races and ethnicities regardless of interest in a physics major. Compared to White students, all students identifying as Black or African American were less likely to have taken high school calculus and were less confident in their math skills. Only Black or African American students who were not interested in a physics major were less likely to have taken physics in high school. This was not the case for Black or African American students who were interested in a physics major, demonstrating that access to high school physics likely plays an important role for this group of students.

Students identifying as Hispanic or Latino, regardless of whether they were interested in a physics major or not, expected a lower grade in their current introductory physics course and were less confident in their math skills, compared to White students. Only Hispanic or Latino students who were not interested in a physics major were less likely to have taken high school calculus or physics, showing, once again, the importance of access to these high school subjects for this group of students.

Although Asian or Asian American students also reported lower self-efficacy in their math skills and a lack of high school physics experience, these students still expected to get a higher grade in their current introductory physics course than White students. This was the case for all Asian or Asian American students regardless of their interest in a physics major. A summary of these findings is presented in Table 8.

Table 8

Section summary: How can we recruit more undergraduate physics students?

Our interviews showed that there is a lot for students to appreciate about physics. Physics students enjoy learning about the universe, applying their problem-solving and math skills, succeeding in a challenging subject, and pursuing a satisfying career. Most graduating physics students (72%) we interviewed were inspired to pursue physics before college and had discovered their interest in physics during high school. Our results show that high school is a critical time to introduce students to physics and inspire their interest in the field.

Below we provide some recommendations based on our findings and the expertise of our reviewers working in high schools and college physics departments. Educators and administrators working in high schools may be able to increase students’ physics interest by

• Offering high school physics courses, or if standalone physics courses are unable to be offered, introducing students to physics in other science courses, programs, or events.
• Within high school physics courses, applying physics concepts to everyday, real-world phenomena with interesting models and demonstrations.
• Including more information about current areas of research in several fields of physics and astronomy (often stated as encouraging modern physics topics in the high school curriculum).
• Sharing stories and information about the breadth of physics careers in high schools.
• Connecting high school physics course content with popular scientific figures and science fiction media.
• Providing multiple opportunities for success in the high school physics classroom besides exam and test grades.
• Making the high school physics classroom more engaging and less didactic, which would include integrating more projects or group work.

Individuals working in institutions of academia, government, or industry may be able to help increase high school students’ interest by

• Offering campus and physics research lab tours and come-to-campus days for high school students.
• Partnering with high school internship and summer programs.
• Encouraging high school physics students to volunteer in physics departments, workplaces, and research labs.
• Performing outreach and science communication in high schools, for instance, at high school physics classes or science fairs as guest speakers.

Increasing diversity when recruiting physics students is another challenge. In our study, women and Black or African American students in introductory physics were less interested in pursuing a physics degree. The percentage of women (20%) and Black or African American students (5%) who were interested in a physics major was similar to the percentage of women (24%) and Black or African American students (3%) earning a physics bachelor’s degree (National Center for Education Statistics, 2021), further suggesting that experiences in high school, or even earlier in life, may be highly influential in developing an interest in physics.

When examining the self-efficacy of women in introductory physics courses, we found several significant differences by physics interest. Women who were interested in a physics major showed no differences compared to men who were interested in a physics major. However, women who were not interested in a physics major reported lower self-efficacy in math and less confidence in getting a higher grade in their physics course, similar to previous literature on women and self-efficacy (Turnbull et al., 2019; Hazari et al., 2007).

When examining gender identity and prior course experiences, only the women who were not interested in a physics major were less likely to have taken high school physics. This is surprising, considering that studies show women are equally represented in high school physics courses overall (Riegle-Crumb & Moore, 2015; Porter & Ivie, 2019). It is possible that the women who were not interested in a physics major may have come from schools that did not offer physics or may have elected not to take physics in high school even when courses were offered.

In our analyses by race/ethnicity and self-efficacy, we found that Black or African American students and Hispanic or Latino students had lower self-efficacy in math and expected lower grades in their introductory physics course than White students, regardless of their interest in a physics major. Unlike our study, some past literature has not shown lower math self-efficacy for Black or African American and Hispanic or Latino students (Kotok, 2017). This may be because this research did not directly measure socioeconomic status, which may explain the gap in math self-efficacy.

When looking at race/ethnicity and high school course background, students identifying as Black or African American and Hispanic or Latino were less likely to have taken a high school physics course, but only for those not interested in a physics major. There were no differences in physics course background for those interested in a physics major. Socioeconomic status may also play a role here, as Black or African American students and Hispanic or Latino students from low-income high schools may have less access to physics and calculus courses and resources (US Department of Education, 2018).

Asian or Asian American students showed significantly different experiences than White students in several ways. Whether they were interested in a physics major or not, they were less likely to have taken high school physics courses and reported lower self-efficacy in math; however, they also expected to receive a higher grade in their current physics course than White students, showing greater self-efficacy in their physics achievement. It is unclear why Asian or Asian American students in the introductory physics courses were less likely to have taken high school physics, and more research is needed to understand this finding.

Although some students in this study reported lower math and physics self-efficacy than others, research shows that self-efficacy is often not reflective of students’ actual performance in math and physics. For example, women have reported lower self-efficacy in math than men, even with similar math performance (Marshman et al., 2018; Zander et al., 2020). Previous studies interviewing women and students of different race/ethnicity groups who left STEM majors showed that this lack of self-efficacy can come from internalized stereotypes and discouraging interactions with high school teachers, counselors, or members of students’ own race/ethnic community (Seymour & Hewitt, 1994). Data shows that Asian or Asian American students were still more likely to be interested in a physics major despite lower self-efficacy in math. It is possible that these students receive positive feedback about physics majors from teachers or race/ethnic community members that help mitigate those effects, but more research is needed.

Below are some recommendations for high schools and college departments to help encourage physics interest among underrepresented groups, based on our findings and reviewer expertise.

• Showing encouragement to all high school and college physics students, regardless of their gender identity or race/ethnicity.
• Increasing accessibility of resources to increase math confidence, such as office hours or tutoring.
• Applying best practices in physics and math teaching in high school and introductory physics courses.

Additional guidelines on high school physics curriculum improvement and recruiting undergraduate physics students can be found at the following resources:

• The AAPT K12 Teacher Portal: https://www.aapt.org/k12/
• The AAPT ComPADRE resource database: https://www.aapt.org/ComPADRE/
• The APS Resources for K-12 Educators and Students page: https://www.aps.org/initiatives/physics-education/k-12
• The STEP-UP program to inspire high school age women to pursue physics in college: https://engage.aps.org/stepup/home
• Effective Practices for Physics Programs (EP3): https://ep3guide.org
  

back to top

Persistence of Undergraduate Physics Students

Once students declare a college physics major, graduating in physics is not a guarantee. Student experiences in college can affect whether they persist in a major or not. Our second study goal was to learn why students may persist or lose interest in a physics major during college. We did this by following up with introductory physics students who were considering a physics major for five years. We annually surveyed these students asking if they still were interested in a physics major, and checked department records of graduating physics students.

Within our study, 31% of the 745 students initially interested in a physics major persisted in physics until graduation and 18% of students directly informed us that they were no longer interested in a physics major. See Table 9 for a full breakdown of this data. Many students stopped responding within the five years of the study. We checked the list of graduating physics students each semester to identify those who persisted, but we could not conclusively identify the outcome for other nonresponding students. Some students may have taken more than five years to graduate and were not included on the graduating lists, or students may have left the physics major or the institution without informing us on the surveys. Because we are unsure of the outcomes for these students, they were not included in any analyses on persistence in physics programs. This left us with a total of 456 students included in the analyses.

Table 9

Most students who left their physics major intended to pursue engineering, mathematics, computer science, or astronomy. Figure 7 shows the new majors those students reported pursuing. We know anecdotally from interviews that some students who left physics continued to pursue a physics minor; however, we do not have the data to give an exact percentage.

Figure 7

When students lost interest in a physics major

By asking students if they were still interested in a physics major each year, we were able to learn when students lost interest in physics during college. Table 10 shows the percentage of students who left a physics major at which survey time point. Typically, decisions to leave a physics major happened early in a student’s college career. Over 70% of students left physics by the end of their introductory physics course or the start of their second year.

Table 10

Why students left physics majors

We performed 39 one-on-one interviews with students who said they lost interest in a physics major and asked why they left. The table below shows a summary of the reasons why students lost interest in physics majors. We’ve categorized these reasons as “pull” factors, positive experiences or interests outside of physics, and “push” factors, negative experiences within physics. This framing of “push” and “pull” has been historically used in previous research studying student and faculty retention (Matier, 1990; Seymour & Hewitt, 1994). Table 11 provides an overview of the percentage of interviewed students who reported each of these push and pull factors.

Table 11

Pull factors (Leaving physics majors due to positive experiences elsewhere)

Some students had positive experiences that drew them away from physics majors, called “pull” factors.

Interest in another subject. The most common reason for leaving their physics major was interest in another subject (74% of interviewed students), such as engineering, mathematics, computer science, neuroscience, finance, or statistics.

As noted in the section above, most students decided to leave a physics major before the start of their second year. Our conversations revealed that for students who wanted to pursue a different major, many of them became interested in this other major during high school courses or earlier. Furthermore, our survey results showed that students having interest in multiple majors when beginning college is common. In our study, 61% of all introductory physics students were considering several majors, with engineering, computer science, and mathematics being popular majors of consideration.

Positive course experiences in another subject. Students also became interested in another major through other college courses and seminars in nonphysics subjects (29%).

Positive department climate in another subject. Furthermore, students felt they were able to more easily connect with professors and peers in other subjects and departments (5%) because they shared a similar interest in that subject.

Interest in a different career path. Another common reason for leaving was students’ interest in another career path (32%).

These students were looking for careers that were better suited for industry and work outside of teaching and research labs.

Students also believed that other majors would offer them more stable, guaranteed, and higher-paying job options.

Push factors (Leaving physics majors due to negative experiences)

Unfortunately, some students lost interest in a physics major due to challenges and negative experiences in physics programs, called “push” factors.

Negative course sequencing experiences in physics. The most common negative reason for leaving a physics major was rigid and long course sequencing (38%). Students who were double-majoring or transferring could not finish their physics program within four years.

The physics programs at their institutions either did not allow enough flexibility in the course sequence or the institution’s tuition was too expensive for students to continue past four years.

Negative course experiences in physics. Students also left physics majors due to barriers in their physics courses (28%). Because our findings show that over 70% of students lost interest in a physics major by the start of their second year, these experiences likely happened during introductory physics courses. For some students, it was the difficulty of the course or the lack of real-world applications when teaching physics concepts.

Other students disliked the structure and teaching style of the courses, for example, the large class sizes, flipped/reverse classrooms, and physics lab sections.

Negative department climate in physics. A small percentage of female physics students lost interest because they did not feel welcome in the physics program (3%). They noticed a lack of diversity and experienced unsupportive attitudes from male peers.

Differences in college experiences between students who persist in or leave physics

Not all students directly told us why they left a physics major. To further examine why students may graduate with a physics degree or leave a physics major, we compared the survey answers of these two groups using a logistic regression. In our surveys, students were asked about several factors regarding their physics experience, including their confidence in math, their physics career expectations, their perception of the physics department climate, how they compare to their peers, their study group experience, their interactions with professors, their research experience, and their discrimination experiences.

We found several statistically significant differences presented in Table 12 below. Some differences were likely due to a student’s length of time within the physics department, such as graduating physics students being more likely to perform research or encounter harassment (hearing about it, witnessing it, or experiencing it). Most students who left physics majors did so early, within their first year or by the start of their second year, while graduating physics students were present in the department for three-five years.

However, other differences would not be sufficiently explained by time in the department. The students who lost interest in a physics major reported lower self-efficacy in math, lower ratings of the physics department climate, and less likelihood to have ever interacted with their physics professors outside of class.

We did not find any statistically significant differences between graduating physics students and students who left a physics major regarding their career expectations, financial concerns, discouragement in physics, study group interactions, and feelings of community among their peers.

Data collection for this study occurred between 2018 and 2023, which included the switch to remote learning in 2020-21 during the pandemic. It is unclear what impact remote learning during the pandemic may have had on our findings. The pandemic was an unprecedented event, and we have no previous events to help us better understand how persistence and attrition were affected during that time.

Table 12

Differences among students from underrepresented groups who leave physics majors

Although our earlier findings showed that members of underrepresented groups were less likely to be interested in a physics major, our findings also showed that members of underrepresented groups, including women, pursuing a physics major were not significantly more likely to lose interest in physics during college. However, we still examined the physics experiences of underrepresented students who left physics majors, because their reasons for leaving may be unique compared to White male students. Using factors related to students’ undergraduate physics experience as dependent variables, we did a series of logistic regressions by gender identity and race/ethnicity, controlling for institution, for students who left a physics major.

Our regression analysis results by gender identity are summarized in Table 13. Compared to men, students who identified as women and left a physics major reported encountering more discrimination in physics, by hearing about it, witnessing it, or experiencing it. Through our one-on-one interviews with students who left physics majors, we have two examples of discrimination experienced by women in introductory physics classrooms, specifically, discrimination from their male classmates and teaching assistants (TAs). This also may provide insight into why women were more likely to feel lower self-efficacy on physics assignments than their peers, because their peers treated them as less competent.

Table 13

Our regression analysis results by race/ethnicity are summarized in Table 14. For the analysis by race/ethnicity, we combined any students identifying as members of an underrepresented race/ethnicity group in physics into a single category due to small group sizes among students for whom we have data. This underrepresented category includes students who identified as Black or African American, Hispanic or Latino, Middle Eastern, Native Hawaiian or other Pacific Islander, Native American or American Indian or Alaskan Native, and “another race/ethnicity.” Since Asian or Asian American students are not underrepresented in physics compared to the general population (National Center for Education Statistics, 2021), they were not combined into the underrepresented race/ethnicity group.

Students from underrepresented race/ethnicity groups who left a physics major were less likely to feel encouraged by their physics professors and believed their physics courses were less interactive (less group work, demonstrations, presentations, and real-world physics applications), compared to White students. Asian or Asian American students were less likely to feel discouraged in their physics studies than White students. Although students from an underrepresented race/ethnicity who left their physics major were more likely to encounter discrimination, none of our interviewed students shared specific examples of discrimination due to their race/ethnicity.

Table 14

Section summary: How can we help students persist in undergraduate physics programs?

In our five-year longitudinal study, only 31% of the 745 students who were initially interested in pursuing a physics major persisted until graduation with a physics bachelor’s degree. Students who lost interest in physics majors were mainly drawn to other subjects such as engineering, mathematics, and computer science. Over 70% of the students decided to leave their physics major by their second year of college, which is consistent with other studies of STEM student retention, even going back several decades (Griffith, 2010; Hilton & Lee, 1988). We can conclude that introductory physics courses are a critical time to retain physics students, and the earlier in college that interventions can be applied, the better.

We asked students why they lost interest in a physics major to help inform departments on how to better retain physics students. We found that “pull” factors, positive experiences outside of physics, were most influential in students’ decisions to leave a physics major. While many students enjoyed their college physics experience, they enjoyed the activities, courses, peer interactions, and potential employment in other subjects more.

However, other students left because of “push” factors — challenges and negative experiences within their physics majors. Students left because they were unable to complete their physics courses within four years (especially for transfers or double majors), experienced issues with poor teaching quality and large class sizes during their physics courses, and had negative perceptions that physics employment consists only of academic positions and desk jobs. Overall, students who left physics majors had lower self-efficacy in their math skills and less positive ratings of the physics department climate and were less likely to have interacted with professors outside of class (for example, in office hours/student hours). This is consistent with the literature on social cognitive career theory that shows the importance of self-efficacy and social support on STEM persistence (Adedokun et al., 2013; Jelks & Crain, 2020; Xu, 2016; Hazari et al., 2007, Navarro et al., 2007).

These push and pull reasons for leaving are consistent with previous studies of STEM student retention (Seymour & Hewitt, 1994), where interviewed students stated they left STEM majors due to interest in other majors or careers, poor teaching quality, competitive peer environments, and course sequencing that would take more than four years. As stated in this previous study, while push and pull factors are presented separately, they are also interrelated. Negative experiences in physics courses can make experiences in other majors more attractive and positive. Therefore, improving the teaching quality and climate of physics departments could help retain all students, regardless of why students leave.

Below are several recommendations that physics departments can implement to encourage student persistence, based on our findings and the expertise of our report reviewers working in physics.

Physics career preparation recommendations:

• Involve undergraduate majors in research and teaching experiences in the department as soon as possible. Support undergraduate research experiences (REUs) and TA/LA programs.
• Discuss physics careers during introductory physics courses and seminars, and disseminate physics career information outside of class, including in career talks, expos, emails, or informational flyers like “Physics Trends” from AIP. Counter the beliefs that physicists can only work in academic research jobs, and show more examples of physicists in industry positions (which can include positions in related fields such as engineering).

Physics program recommendations:

• Adapt and tailor physics course sequencing and curriculum to accommodate students pursuing industry careers. Avoid the assumption that all students will attend graduate school.
• Design more flexible or shorter physics course sequences so students can complete the degree within four years, and provide support for transfer students and double majors where possible.

Physics teaching recommendations:

• Design introductory physics courses that involve more real-world demonstrations for physics concepts and smaller class sizes where possible. When implementing a flipped or reverse classroom format, ensure professors and TAs are applying this teaching approach correctly with evidence-based, active learning practices.
• Encourage professors and TAs to host and advertise office hours, student hours, and study sessions outside of class.
• Provide and advertise resources for extra support with math skills, such as tutoring.

Physics department climate recommendations:

• Promote and provide opportunities for students to participate in peer groups with other physics majors to collaborate on homework or laboratory projects.
• Promote and provide opportunities for students to participate in organizations like the Society of Physics Students (SPS) at the local and national level.
• Promote and provide opportunities for students to engage in and lead social activities of the department, particularly outreach activities.
• Prioritize the inclusion and retention of all students in all aspects of physics program planning.

Additional guidelines on the improvement of physics departments can be found at the following resources:

• Effective Practices for Physics Programs (EP3): https://ep3guide.org
• The AAPT ComPADRE resource database: https://www.aapt.org/ComPADRE/
• The APS Resources for Physics Departments page: https://www.aps.org/initiatives/physics-education/departments
• The SPS programs and resources page: https://www.spsnational.org

Our analyses showed that compared to White students and male students, women and students from underrepresented race/ethnicity groups (which includes students who identified as Black or African American, Hispanic or Latino, or “another race/ethnicity”) felt their physics courses were less interactive and their physics professors were less encouraging, believed they performed worse on physics assignments than their peers, and encountered more instances of discrimination in physics. Most notably, women told us about their experiences of discrimination involving peers and TAs in their introductory physics courses. This is also consistent with previous literature that indicates the importance of social support and self-efficacy in STEM persistence for underrepresented groups (Cech et al., 2011; Sawtelle et al., 2012; Espinosa, 2011; Shapiro & Sax, 2011; Xu & Lastrapes, 2022; Jehangir et al., 2023; MacPhee et al., 2013). In a previous study interviewing students who left STEM majors, women also reported experiencing challenges with their male peers and lower confidence in their science and math skills due to their discouraging interactions with faculty members and male peers (Seymour & Hewitt, 1994). Furthermore, Seymour and Hewitt (1994) found that students identifying as underrepresented race/ethnicity group members reported lower confidence in science and less satisfaction in their courses due to internalized stereotypes, feelings of ethnic isolation, and discouragement from professors and members of their own race/ethnic community.

Asian or Asian American students who left their physics major did not show any significantly differences compared to White students, besides feeling less discouraged and more confident in their physics experience. According to our results, Asian or Asian American students did not seem to leave the major due to any negative experiences in physics.

Below are several recommendations to encourage the persistence of underrepresented students in physics, based on our findings and reviewer expertise.

Physics teaching recommendations:

• Provide appropriate training to professors and TAs on how to teach all students, regardless of their gender identity and race/ethnicity, in an encouraging and empathetic manner. Encourage professors and TAs to model inclusive behavior and create a more welcoming department environment.
• Hire diverse faculty members and staff when possible.

Physics mentorship recommendations:

• Invite diverse speakers of different genders and races/ethnicities for talks and events, and give students opportunities to talk and interact with them.
• Develop and provide mentorship opportunities for students to connect with peers and faculty members, particularly those who identify as the same gender identity or race/ethnicity. Mentors can be found in physics classrooms, research labs, mentorship programs, or student organizations such as the Society of Physics Students (SPS).
• Promote and support student participation in clubs, organizations, or conferences in physics designed to assist those in underrepresented groups, such as the National Society of Black Physicists (NSBP), the Conference for Undergraduate Women in Physics (CUWiP) and other Women in Physics clubs, the National Society of Hispanic Physicists (NSHP), and the Society for the Advancement of Chicanos/Hispanics and Native Americans in Science (SACNAS).

Physics department climate recommendations:

• Create and enforce physics department policies, codes of conduct, and social norms for an inclusive environment for faculty, staff, administrators, and students in classrooms and collaborative department spaces.
• Increase awareness of physics department policies and procedures for reporting harassment.
• Encourage students to engage in and lead efforts aimed at increasing diversity and inclusion within a department.
• Read the AIP TEAM-UP Report and implement its recommendations.

Additional guidelines on the improvement of physics department climates for women and underrepresented minorities can be found at the following resources:

• The AIP TEAM-UP project page: https://www.aip.org/diversity-initiatives/team-up-task-force
• The SEA Change Project: https://www.aapt.org/programs/Sea_Change/index.cfm
• The APS Inclusive Physics page: https://www.aps.org/initiatives/inclusion
• Conferences for Undergraduate Women in Physics (CUWiP): https://www.aps.org/programs/women/cuwip/index.cfm
• The National Society for Black Physicists (NSBP): https://nsbp.org/
• The National Society for Hispanic Physicists (NSHP): https://hispanicphysicists.org/
• The Society for the Advancement of Chicanos / Hispanics and Native Americans in Science (SACNAS): https://www.sacnas.org/

back to top

Limitations

Our study is limited in that we surveyed students at only four predominantly White colleges and universities, and these experiences do not necessarily represent the experiences of all university students or students at two-year colleges, women’s colleges, or historically Black colleges and universities. We also did not collect information on certain demographic variables such as perceived socioeconomic status, which may have impacted race and ethnicity findings. We know from our one-on-one interviews that students in our study came from a wide variety of physics courses, but we did not ask about physics course characteristics on our surveys and could not include them as variables in our analysis. It is likely that class size, class composition (e.g., honors students, general STEM students, or physics majors), and teaching approaches (e.g., lecture hall, SCALE-UP, or seminar) have an important role in generating a student’s interest in physics and encouraging their persistence in the major.

During the course of the study, the participating universities switched to remote learning due to COVID-19 between March of spring 2020 to fall 2021. All the introductory physics students were recruited before the campus closings. Therefore, the effects surrounding COVID-19 would not have impacted students’ initial interest in a physics major in this study. Most students in the study lost interest in a physics major early, by the start of their second year in physics (over 70%), and many students would have made this decision to leave the major in 2018 or 2019 before the campus closures. However, it is possible that challenges due to COVID-19 may have impacted decisions for some students who left their physics major after spring 2020. We are unable to draw any conclusions about the effect of the pandemic with the data in this study.

Conclusions

This five-year longitudinal study followed undergraduate students in introductory college physics who were considering a physics major. Using factors drawn from the social cognitive career theory (Lent et al., 1994), we compared introductory physics students who were interested and those who were not interested in a physics major, as well as students who lost interest in a physics major and those who graduated with a physics bachelor’s degree. In addition, we compared the experiences of physics students by gender identity and race/ethnicity. This study addressed previous limitations in the literature by examining student persistence over time and measuring actual persistence in a physics degree program (graduation) rather than students’ intention to graduate.

At the beginning of their first introductory physics courses, most students who were interested in a physics major were inspired to pursue physics during their high school class or program. Besides these students expecting to earn a higher grade in their current physics course, there were no differences between them and students who were not interested in a physics major based on prior course experience, parent/guardian education, or self-efficacy in math.

When we followed the students interested in majoring in physics over time, we identified several relevant factors in whether they persisted until graduation. Regression analyses showed that students who lost interest in a physics major had lower self-efficacy in math, were less likely to interact with professors outside of class, and rated the physics department climate lower. Interviewed students who left a physics major further elaborated on how rigid physics course sequencing, a perceived lack of career opportunities in physics, and negative introductory physics course experiences impacted their decision to leave. Most students left their physics majors by the start of their second year. This finding underscores how critical students’ first- and second-semester experiences in college physics are.

Another key finding from this study is that women and students identifying as Black or African American were less likely to be interested in a physics major at the beginning of their introductory physics courses. Women, Black or African American students, and Hispanic or Latino students who were not interested in a physics major were significantly less likely to have prior high school calculus and physics course experience, reported lower self-efficacy in math, and expected to achieve a lower grade in their current physics course.

Among women and other underrepresented students who were planning to pursue a physics major, the data show a different story. At the start of their introductory physics courses, women who were interested in a physics major did not show any significant differences from men in their prior course experience or self-efficacy. Black or African American students and Hispanic or Latino students who were interested in physics majors were equally likely to have taken high school physics as White students but still experienced lower self-efficacy in math.

When we followed the students who were pursuing a physics major over time, women and students from underrepresented race/ethnicity groups were not more likely to lose interest in a physics major. They did, however, show potentially different reasons for leaving. Among students who decided to leave their physics major, women, Black or African American students, and Hispanic or Latino students reported encountering more discrimination and experiencing fewer encouraging interactions with peers and professors.

To encourage more physics students to succeed and persist in the major, physics departments and programs will need to consider students’ academic, career, social, and personal needs and create a more welcoming and successful department environment for students of any gender identity or race/ethnicity. Our findings highlight that early intervention is essential. Programs should focus on physics outreach activities with high school students or younger students and implement positive changes within students’ first year in introductory physics.

back to top

Appendix A- Methodology

Survey response rates

Undergraduate students were recruited during introductory college physics courses at four universities in the United States. The project was approved by the universities’ Institutional Review Boards (IRBs). Recruitment was done across four semesters between 2018 and 2020. During the first week of their introductory physics courses, students completed the first survey. Methods of survey distribution differed at each institution. Students were given physical survey copies in person during class, assigned the online survey as an optional extra credit assignment during class, or emailed the online survey outside of class. The first survey was distributed to 6,074 physics students, and 3,917 students completed the survey. Students in the first survey were asked whether they were planning to major in physics: 322 were interested in a physics major, 423 were possibly considering a physics major, and 3,172 were not interested. The 745 students who indicated that they were interested in a physics major or possibly considering a physics major were sent follow-up surveys over the next five years.

Table 15 shows the survey schedule and response rates from 2018 to 2023. All follow-up surveys were sent directly to the students’ emails, which were voluntarily provided in the first survey. Students were sent a follow-up survey at the end of their introductory physics course and then sent a follow-up survey at the beginning of each fall semester until 2022. If students indicated they were no longer interested in physics, they completed a final survey and no longer received any future surveys. A total of 171 participating physics students left the physics major between 2018 and 2023.

In addition, all graduating physics majors at the four universities were recruited to complete a survey each semester between 2018 and 2023. Graduates were recruited in two ways. First, physics department staff provided emails for graduating students each semester, and second, participants indicated they were graduating with a physics degree in one of the follow-up surveys. A total of 285 graduating physics students participated.

Table 15

Interview protocol

The students who left their physics major and the students graduating with a physics major were invited to participate in a single 30–60-minute phone interview. In the semi-structured interview, students discussed their experience in the physics program, and their reasons for persisting in or leaving the major. Both groups were asked about what helped them succeed in their physics program and classes, what made it harder for them to succeed, and what they planned to do after graduation. Graduating students were asked why they were motivated to major in physics, and students who left a physics majors were asked why they took their initial introductory physics course, as well as what nonphysics major they declared and why. Interviewed students were compensated with a $20 gift card. A total of 39 students who left a physics major and 75 physics graduates participated in interviews.

Data collection during the COVID-19 pandemic

Because this study occurred between 2018 and 2023, all four participating institutions switched to remote learning in spring 2020 due to COVID-19 protocols and did not return to in-person learning until fall 2021. These institutional closings did not alter our data collection methods. All introductory students were recruited and initially surveyed before the switch to remote learning. Surveys continued to be distributed over email, and interviews continued to be conducted over the phone.

Survey measures

Interest in physics. On the first survey, students were asked if they were interested in pursuing a physics major (yes/maybe/no). Responses for yes and maybe were combined to indicate anyone who was considering a physics major.

Persistence in physics. On all the surveys over five years, students were asked whether they were still considering a physics major or if they had declared a physics major (yes/no).

Switching to physics. On the survey for graduates, students were asked if they switched to declaring physics from a nonphysics major (yes/no).

Nonphysics majors of interest. Students who were not interested in physics and students who left a physics major were asked if they were interested in 17 other nonphysics majors (yes/maybe/no), including astronomy or astrophysics, business, biology, chemistry, computer science, economics, education, engineering, English, fine arts, foreign language, geological science, history, mathematics, music, social science, or another major. Responses for yes and maybe were combined into a single category.

Self-efficacy in physics and math. Students were asked whether they expected to earn an A, B, C, or lower in their first physics course. On a four-point Likert scale (1=strongly agree to 4=strongly disagree), students were asked four items about how confident they were in their math abilities, which were combined into an average score. Students were also asked if they felt they performed well in physics assignments compared to their peers (1=strongly agree to 4=strongly disagree). Lastly, students were asked if they ever felt discouraged in physics (yes/no).

Career outcome expectations. On a four-point Likert scale (1=strongly agree to 4=strongly disagree), students reported their physics career beliefs, including four items about whether they believed they would make money in a physics career, have the necessary skills to work in a physics career, would enjoy a physics career, and would be able to find physics job opportunities. These items were combined into an average score.

Physics department climate. On a four-point Likert scale (1=strongly agree to 4=strongly disagree), students reported their opinion on the physics department climate including six items asking if the department was welcoming, friendly, threatening, rigid or inflexible, encouraged self-confidence, or made them feel like an outsider.

Interactions with physics professors and peers. On a four-point Likert scale (1=often to 4=never), students indicated how often they interacted with professors and study groups. We recoded this to indicate whether students ever interacted with professors and study groups or not (yes/no). On a four-point Likert scale (1=strongly agree to 4=strongly disagree), students were asked if they felt a sense of community with their physics peers.

Physics professor encouragement. On a four-point Likert scale (1=strongly agree to 4=strongly disagree), students reported the level of support they received from their physics professor across two items: whether their professor helped them when they were stuck and whether they encouraged them to persist through challenging physics problems. These two items were combined into an average score.

Harassment and discrimination. Participants indicated whether they were aware of harassment or discrimination in the department (yes/no), which included an incident happening to them or someone else.

Physics course activities. On a four-point Likert scale (1=often to 4=never), students reported the frequency with which they did interactive course activities across seven items: group assignments, presentations, demonstrations, peer-to-peer feedback, explaining answers to physics questions, limiting lecture time, and learning real-world physics applications. All seven course activities were combined into an average score.

Financial concern. On a four-point Likert scale (1=strongly agree to 4=strongly disagree), students indicated their level of concern about affording college.

Research experience. Students indicated whether they participated in research (yes/no).

High school course experience. Students also reported which physics and calculus courses they took during high school. They could make multiple course selections of regular high school, AP, and college courses, or indicate that they never took a physics or calculus course before.

Parent or guardian education. Students were asked whether their parents/guardians earned a bachelor’s degree or higher (yes/no) and if that degree was in a STEM field (yes/no/unsure).

Gender identity. Students selected man, woman, another gender identity, or prefer not to respond. There were too few respondents with another gender identity to include in any analyses. Therefore, only men and women were compared.

Race and ethnicity. Students could select multiple options including Asian or Asian American, Black or African American, Hispanic or Latino, Middle Eastern, Native American or American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, White, “Another race/ethnicity,” or prefer not to respond. Some groups had too few respondents to include in analyses and these categories were combined into Another race/ethnicity. The final variable used in analyses consisted of Asian or Asian American, Black or African American, Hispanic or Latino, White, and Another race/ethnicity.

Institution. We coded which of the four institutions the student was attending during the study, which was used as a control variable in the analyses. To protect the anonymity of these institutions, we are unable to provide any information on them.

Data analysis approach

We performed two main binary logistic regression analyses. First, we conducted a binary logistic regression with interest in a physics major as the dependent variable (0=not interested in physics; 1=interested in physics). Independent variables included gender identity, race/ethnicity, institution, whether they took high school calculus, whether they took high school physics, whether their parent/guardian had a STEM degree, their self-efficacy in math, and their expected grade in their introductory physics course.

Second, we conducted a binary logistic regression with persistence in physics as the dependent variable (0=lost interest in a physics major and 1=graduated with physics bachelor’s degree). Independent variables included gender identity, race/ethnicity, institution, self-efficacy in math, performance on physics assignments compared to their peers, level of financial concern, whether they participated in research, whether they encountered harassment in the physics department, their rating of the physics department climate, their rating of the sense of community among peers, their level of confidence in a physics career, whether they interacted with professors outside of class, and whether they participated in study groups.

Lastly, we conducted a series of follow-up linear or binary logistic regressions to compare gender identity and race/ethnicity differences on all the variables discussed. Independent variables in these regressions included gender identity, race/ethnicity, and institution as a control variable.

back to top

Appendix B- Literature Review

Survey and interview questions in this study were drawn from social cognitive career theory (SCCT; Lent et al., 1994), based on the theory’s choice model of academic and career interest development. The choice model identifies four key factors: 1) self-efficacy, or the confidence in the ability to successfully perform required actions (Bandura, 1977); 2) outcome expectations for a career choice; 3) personal goals and interests to pursue a career path; and 4) environmental influences including social support, financial support, and professional development opportunities.

The role of self-efficacy, career interest, and career outcome expectations in STEM persistence is well-supported in the literature. Higher self-efficacy in science research skills led to more positive expectations of a research career and increased persistence in science careers (Adedokun et al., 2013). Expecting a better outcome in early STEM courses, such as a B or higher, predicted higher retention (Espinosa, 2011). Perceiving that STEM careers are not aligned with long-term career interests or that there is a lack of job opportunities in STEM predicted less anticipated persistence in STEM fields (Jelks & Crain, 2020). In biology programs, higher personal motivation to pursue biology as well as more perceived potential job opportunities increased the likelihood of students remaining in the program (Ashford-Hanserd et al., 2020).

STEM self-efficacy can further be influenced by previous experiences in science. Studies have shown that better grades in past math and science courses predicted higher science and math self-efficacy (Navarro et al., 2007). This may explain why students who took more math and physics courses in high school were more likely to persist in physics majors during college (Espinosa, 2011; Hazari et al., 2007; Norvilitis et al., 2002).

Environmental supports and barriers consist of social, financial, and academic factors. Increased social support from faculty members, teachers, family members, parents, peers, and friends improved persistence and retention in science (Xu, 2016; Hazari et al., 2007, Nvarro et al., 2007). Financial support is another important factor in retention, as students struggling financially were more likely to leave STEM majors (Xu, 2016; National Center for Education Statistics, 2013). Positive academic experiences play another role. Students were more likely to persist in a STEM major when they attended a higher-quality academic program and experienced smaller class sizes, better teaching, and more accessible teachers (Xu, 2016). Furthermore, students who participated in faculty research were 14% more likely to anticipate persisting in STEM careers (Jelks & Crain, 2020).

Although previous studies have applied the SCCT to STEM degree persistence, most research is correlational or cross-sectional, according to a meta-analytic review of 143 SCCT studies (Lent et al., 2018). The few longitudinal SCCT studies were typically conducted within a 5-to-12-month time frame, and measure STEM persistence intention rather than actual persistence and enrollment in STEM programs over time (Navarro et al., 2014; Lent et al., 2008; Rogers & Creed, 2011). The longest longitudinal SCCT study (Lent et al., 2016) examined engineering major persistence across three years and collected data on future major enrollment rather than persistence intention. The current study extends the work of Lent and colleagues (2016) by examining actual degree completion over five years and applying the SCCT model to persistence in the physics field. Previous studies have used this framework in the study of engineering, mathematics, biology, and computer science students (Ashford-Hanserd, 2020; Lent et al., 2003; Lent et al., 2008; Martin et al., 2012). This is the first major study to apply the SCCT model in physics.

Persistence of women in STEM

Previous social cognitive career theory research has also examined the moderating role of gender in STEM persistence. In previous studies with the SCCT model, social support was significantly related to women choosing to major in engineering but not for men, while men’s interest in engineering topics was significantly more likely to influence choosing engineering as a major (Lent et al., 2005). Men were more motivated by interest in engineering than women. This is consistent with other studies examining gender differences and social support in science. Women were more likely to persist in science when they felt a sense of belonging, studied collaboratively with other students, joined science clubs, and had more female friends in science (Espinosa, 2011; Shapiro & Sax, 2011; Xu & Lastrapes, 2022). Furthermore, women felt more supported when their department had more female faculty members and they experienced more positive faculty interactions, in which faculty did not show favoritism to male students (Blickenstaff, 2005; Shapiro & Sax, 2011). Conversely, women are more likely to leave STEM if they experience negative interactions, particularly sexual or gender-based biases and harassment. According to the National Academies of Sciences, Engineering, and Medicine (2018), women are more likely to experience harassment in STEM, which leads to increased psychological stress, anxiety, depression, and eventual withdrawal from one’s school or field. When women felt questioned by others about their competence and belonging in STEM, they were more likely to leave majors like physics and switch to other fields like biology (Turnbull et al., 2019). In summary, the social environment in a department plays an important role in STEM retention for women.

There are mixed results in the literature when examining gender, self-efficacy, and outcome expectations. Previous literature and meta-analyses with the SCCT model have not shown any significant gender differences in self-efficacy or outcome expectations when predicting STEM interests or major choices (Navarro et al., 2014; Lent et al., 2018; Lent et al., 2005). However, other studies have shown that women and girls are more likely to have lower self-confidence about their abilities and academic performance in science, independent of objective success in a course (Blickenstaff, 2005; MacPhee et al., 2013; Shapiro, 2011), and women were more likely to leave STEM due to a lack of confidence in understanding calculus (Ellis et al., 2016). Women having higher self-confidence and self-efficacy in science led to greater persistence in STEM courses, majors, and careers, as well as higher GPAs and learning outcomes (Cech et al., 2011; Espinosa, 2011; Sawtelle et al., 2012). Women were also more likely to persist when STEM careers perceptions aligned with their desire for personal achievement and helping others (Henderson et al., 2022). Another study showed that higher STEM career self-efficacy predicted better STEM career outcome expectations for Native American women, as predicted in the SCCT model (Turner et al., 2022). Therefore, it is still worth examining how self-efficacy and outcome expectations influence gender differences in physics major persistence.

Persistence of underrepresented race/ethnicity groups in STEM

When studies have applied the SCCT to students from underrepresented race/ethnicity groups in STEM, studies have supported the model, showing that academic self-efficacy, STEM career interest, and social support were important factors in the persistence of underrepresented STEM students (Dutta et al., 2015; Lent et al., 2018). Although limited, some research has shown that students from underrepresented race/ethnicity groups reported lower academic, science, and math self-efficacy in STEM and lack of high school preparation for college STEM courses (MacPhee et al., 2013; Turner et al., 2022), and Black / African American high school students were less likely to want to pursue a STEM career (US Department of Education, 2020). However, other studies have not shown any effect of race/ethnicity on math or STEM self-efficacy, especially once the study controlled for socioeconomic status (Kuchynka et al., 2021; Kotok, 2017). Other studies have shown students from underrepresented race/ethnicity groups have reported higher general self-efficacy (Wilson et al., 2015). The overall evidence on the relationship between students’ purported self-efficacy and race/ethnicity is mixed.

Social support plays an important role in the persistence of students from underrepresented race/ethnicity groups. Students at predominantly White institutions struggled to persist because they felt less belonging among peers and faculty members due to a lack of others with the same race/ethnicity, being overlooked, or feeling questioned about their place in the program (Jehangir et al., 2023). These findings were corroborated in other studies which showed that Black / African American physics students were less likely to feel social belonging among their peers, less likely to feel their physics department supported them, and more likely to experience negative and prejudiced treatment because of their race/ethnicity in their physics courses (American Institute of Physics, 2020). These students experienced more academic success when they had affirmation and support from their teachers and departments.

Students from underrepresented race/ethnicity groups in STEM reported difficulty with financial challenges and being a first-generation college student, which can also impact their career goals and self-efficacy in STEM fields (Jehangir et al., 2023; Turner et al., 2022). For example, Black / African American physics students reported finances as the most common barrier in college and many were required to work an outside job during their undergraduate education (American Institute of Physics, 2020). This led physics students to consider pursuing careers in other, higher-paying STEM fields and limited opportunities for their professional and career development, such as research and conference travel.

back to top

References

Adedokun, O. A., Bessenbacher, A. B., Parker, L. C., Kirkham, L. L., & Burgess, W. D. (2013). Research skills and STEM undergraduate research students’ aspirations for research careers: Mediating effects of research self-efficacy. Journal of Research in Science Teaching, 50(8), 940–951. DOI 10.1002/tea.21102

American Institute of Physics, Statistical Research Center (2020). The time is now: Systemic changes to increase African Americans with bachelor’s degrees in physics and astronomy. Retrieved at https://www.aip.org/diversity-initiatives/team-up-task-force .

Ashford-Hanserd, S., Daniel, K. L., Garcia, D. M., & Idema, J. L. (2020). Factors that influence persistence of biology majors at a Hispanic-serving institution. Journal of Research in Technical Careers, 4(1), 47–60.

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215.

Beasley, M., & Fischer, M. (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(4), 427–448. https://doi.org/10.1007/s11218-012-9185-3

Blickenstaff, J. C. (2005). Women and science careers: Leaky pipeline or gender filter? Gender and Education, 17(4), 369–386. DOI: 10.1080/09540250500145072

Cech, E., Rubineau, B., Silbey, S., & Seron, C. (2011). Professional role confidence and gendered persistence in engineering. American Sociological Review, 76(5), 641–666. 10.1177/0003122411420815

Dutta, A., Kang, H. J., Kaya, C., Benton, S. F., Sharp, S. E., Chan, F., da Silva Cardoso, E., & Kundu, M. (2015). Social-Cognitive Career Theory predictors of STEM career interests and goal persistence in minority college students with disabilities: A path analysis. Journal of Vocational Rehabilitation, 43(2), 159–167.

Ellis, J., Fosdick, B. K., & Rasmussen, C. (2016). Women 1.5 times more Likely to leave STEM pipeline after calculus compared to men: Lack of mathematical confidence a potential culprit. PLOS ONE, 11(7), e0157447. https://doi.org/10.1371/journal.pone.0157447

Espinosa, L. L. (2011). Pipelines and pathways: Women of color in undergraduate STEM majors and the college experiences that contribute to persistence. Harvard Education Review, 81(2), 209–240.

Griffith, A. L. (2010). Persistence of women and minorities in STEM field majors: Is it the school that matters? Economics of Education Review, 29(6), 911–922. https://doi.org/10.1016/j.econedurev.2010.06.010

Hazari, Z., Tai, R. H., & Sadler, P. M. (2007). Gender differences in introductory university physics performance: The influence of high school physics preparation and affective factors. Science Education, 847–876. 10.1002/sce

Henderson, H. L., Bloodhart, B., Adams, A. S., Barnes, R. T., Burt, M., Clinton, S., Godfrey, E., Pollack, I., Fischer, E. V., & Hernandez, P. R. (2022). Seeking congruity for communal and agentic goals: A longitudinal examination of U.S. college women’s persistence in STEM. Social Psychology of Education, 25(2/3), 649–674. https://doi.org/10.1007/s11218-021-09679-y

Hilton, T. L., & Lee, V. E. (1988). Student interest and persistence in science: Changes in the educational pipeline in the last decade. Journal of Higher Education, 59, 510–526.

Jehangir, R. R., Stebleton, M. J., & Collins, K. (2023). STEM stories: Fostering STEM persistence for underrepresented minority students attending predominantly White institutions. Journal of Career Development, 50(1), 87–103. https://doi.org/10.1177/08948453211073706

Jelks, S. M., & Crain, A. M. (2020). Sticking with STEM: Understanding STEM career persistence among STEM bachelor’s degree holders. The Journal of Higher Education, 91(5), 805–831. https://doi.org/10.1080/00221546.2019.1700477

Kotok, S. (2017). Unfulfilled potential: High-achieving minority students and the high school achievement gap in math. The High School Journal, 100(3), 183–202. https://www.jstor.org/stable/90024211 .

Kuchynka, S., Reifsteck, T., Gates, A., & Rivera, L. (2021). Developing self-efficacy and behavioral intentions among underrepresented students in STEM: The role of active learning. Frontiers in Education, 9(6). https://doi.org/10.3389/feduc.2021.668239

Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying social cognitive theory of career and academic interest, choice, and performance. Journal of Vocational Behavior, 45, 79–122.

Lent, R. W., Brown, S. D., Schmidt, J., Brenner, B., Lyons, H., & Treistman, D. (2003). Relation of contextual supports and barriers to choice behavior in engineering majors: Test of alternative social cognitive models. Journal of Counseling Psychology, 50, 458–465.

Lent, R. W., Brown, S. D., Sheu, H. B., Schmidt, J., Brenner, B. R., Gloster, C. S., & Treistman, D. (2005). Social cognitive predictors of academic interests and goals in engineering: Utility for women and students at historically Black universities. Journal of Counseling Psychology, 52(1), 84–92. 10.1037/0022-0167.52.1.84

Lent, R. W., Lopez Jr., A. M., Lopez, F. G., & Sheu, H. B. (2008). Social cognitive career theory and the prediction of interests and choice goals in the computing disciplines. Journal of Vocational Behavior, 73(1), 52–62.

Lent, R. W., Miller, M. J., Smith, P. E., Watford, B. A., Lim, R. H., & Hui, K. (2016). Social cognitive predictors of academic persistence and performance in engineering: Applicability across gender and race/ethnicity. Journal of Vocational Behavior, 94, 79–88.

Lent, R. W., Sheu, H. B., Miller, M. J., Cusick, M. E., Penn, L. T., & Truong, N. N. (2018). Predictors of science, technology, engineering, and mathematics choice options: A meta-analytic path analysis of the social cognitive choice model by gender and race/ethnicity. Journal of Counseling Psychology, 65(1), 17–35. http://dx.doi.org/10.1037/cou0000243

MacPhee, D., Farro, S., & Canetto, S. S. (2013). Academic self-efficacy and performance of underrepresented STEM majors: Gender, ethnic, and social class patterns. Analyses of Social Issues & Public Policy, 13(1), 347–369. https://doi.org/10.1111/asap.12033

Marshman, E. M., Kalendar, Y., Nokes-Malach, T., Schuun, C., & Singh, C. (2018). Female students with A’s have similar physics self-efficacy as male students with C’s in introductory courses: A cause for alarm? Physical Review: Physics Education Research, 14. https://doi.org/10.1103/PhysRevPhysEducRes.14.020123

Martin, A. J., Anderson, J., Bobis, J., Way, J., & Vellar, R. (2012). Switching on and switching off in mathematics: An ecological study of future intent and disengagement among middle school students. Journal of Educational Psychology, 104, 1–18. http://dx.doi.org/10.1037/a0025988

Matier, M. W. (1990). Retaining faculty: A tale of two campuses. Research in Higher Education, 31, 39–60.

Mulvey, P. J., & Nicholson, S. (2020). Physics bachelor’s degrees: 2018. Retrieved at https://www.aip.org/statistics/reports/physics-bachelors-degrees-2018.

National Academies of Sciences, Engineering, and Medicine (2018). Sexual Harassment of Women: Climate, Culture, and Consequences in Academic Sciences, Engineering, and Medicine (F. F. Benya, S. E. Widnall, & P. A. Johnson, Eds.). National Academies Press, Washington, DC.

National Center for Education Statistics (2013). STEM Attrition: College students’ paths into and out of STEM fields. US Department of Education, Washington, DC.

National Center for Education Statistics (2017). Beginning college students who change their majors within three years of enrollment. US Department of Education, Washington, DC.

National Center for Education Statistics (2020). Digest of Education Statistics. Retrieved at https://nces.ed.gov/programs/digest/current_tables.asp .

National Center for Education Statistics (2021). Integrated Postsecondary Education Data System. [Data set]. https://nces.ed.gov/ipeds/use-the-data .

Navarro, R. L., Flores, L. Y., & Worthington, R. L. (2007). Mexican American middle school students’ goal intentions in mathematics and science: A test of social cognitive career theory. Journal of Counseling Psychology, 54(3), 320–335.

Navarro, R. L., Flores, L. Y., Lee, H. S., & Gonzalez, R. (2014). Testing a longitudinal social cognitive model of intended persistence with engineering students across gender and race/ethnicity. Journal of Vocational Behavior, 85, 146–155. http://dx.doi.org/10.1016/j.jvb.2014.05.007

Norvilitis, J. M., Reid, H. M., & Norvilitis, B. M. (2002). Success in everyday physics: The role of personality and academic variables. Journal of Research in Science Teaching, 39(5), 394–409. DOI 10.1002/tea.10028

Porter, A. M., & Ivie, R. (2019). Women in physics and astronomy. Retrieved at https://ww2.aip.org/statistics/women-in-physics-and-astronomy-2019 .

Riegle-Crumb, C., & Moore, C. (2015). The gender gap in high school physics: Considering the context of local communities. Social Science Quarterly, 2014 Mar. 1, 95(1), 253–268. doi: 10.1111/ssqu.12022. PMID: 25605978; PMCID: PMC4297668

Riegle-Crumb, C., King, B., & Irizarry, Y. (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

Rogers, M. E., & Creed, P. A. (2011). A longitudinal examination of adolescent career planning and exploration using a social cognitive career theory framework. Journal of Adolescence, 34(1), 163–172. DOI:10.1016/j.adolescence.2009.12.010

Sawtelle, V., Brewe, E., & Kramer, L. H. (2012). Exploring the relationship between self-efficacy and retention in introductory physics. Journal of Research in Science Teaching, 49(9), 1096–1121. DOI 10.1002/tea.21050

Seymour, E., & Hewitt, N. M. (1994). Talking about leaving: Factors contributing to high attrition rates among science, mathematics, & engineering undergraduate majors. Bureau of Sociological Research, University of Colorado Boulder, Boulder, CO.

Shahid, M. (2020). The importance of physics to man and the society. Retrieved at https://www.linkedin.com/pulse/importance-physics-man-society-dr-muhammad-attique-khan-shahid/ .

Shapiro, C. A., & Sax, L. J. (2011). Major selection and persistence for women in STEM. New Directions for Institutional Research, 152, 5–18. DOI: 10.1002/ir.404

Turnbull, S. M., Locke, K., Vanholsbeeck, F., & O’Neale, D. R. J. (2019). Bourdieu, networks, and movements: Using the concepts of habitus, field and capital to understand a network analysis of gender differences in undergraduate physics. PLoS ONE, 14(9), 1–28. https://doi.org/10.1371/journal.pone.0222357

Turner, S. L., McWhirter, E. H., Lee, H., Mason-Chagil, G., Smith, S., Jacobs, S. C., & Jackson, A. P. (2022). Barriers to STEM efficacy and outcome expectations among Native American college students. The Counseling Psychologist, 50(7), 981–1008. 10.1177/00110000221108454

US Department of Education Office of Civil Rights (2018). 2015–16 Civil rights data collection: STEM course taking. Retrieved at https://www2.ed.gov/about/offices/list/ocr/docs/stem-course-taking.pdf .

Whitcomb, K. M., & Singh, C. (2021). Underrepresented minority students receive lower grades and have higher rates of attrition across STEM disciplines: A sign of inequity? International Journal of Science Education, 43(7), 1054–1089. DOI: 10.1080/09500693.2021.1900623

Wilson, D., Bates, R., Scott, E. P., Painter, S. M., & Shaffer J. (2015). Differences in self-efficacy among women and minorities in STEM. Journal of Women and Minorities in Science and Engineering, 21(1), 27–45. http://dx.doi.org/10.1615/JWomenMinorScienEng.2014005111

Xu, Y. J. (2016). The experience and persistence of college students in STEM majors. Journal of College Student Retention: Research, Theory, & Practice, 1–20. 10.1177/1521025116638344

Xu, C., & Lastrapes, R. E. (2022). Impact of STEM sense of belonging on career interest: The role of STEM attitudes. Journal of Career Development, 49(6), 1215–1229. https://doi.org/10.1177/08948453211033025

Zander, L., Hohne, E., Harms, S., Pfost, M., & Hornsey, M. J. (2020). When grades are high but self-efficacy is low: Unpacking the confidence gap between girls and boys in mathematics. Frontiers in Psychology, 11. https://doi.org/10.3389/fpsyg.2020.552355

back to top