STEM workforce retention and preparation: Landscape analysis of STEM students in the workforce across Texas

Abstract

There is a pressing need to increase the number of STEM graduates within the U.S. and to ensure those graduating with STEM degree are retained in workforce. The shortages within STEM fields impact global competitiveness, economic growth, and employability of the population. Academic institutions have varying initiatives to expand the pipeline of STEM students but it has been difficult to measure how these initiatives translate into workforce outcomes. This article explores the ties academic experience with workforce perceptions and outcomes of STEM graduates across the State of Texas to understand the landscape of STEM student lived experiences in industry, focusing on both retention and preparedness. A sub-focus of this article was to understand differences in outcomes and experiences of STEM students participating in a specific initiative, the Houston-Louis Stokes Alliance for Minority Participation programming, an NSF-funded STEM retention program, graduates and their non-LSAMP peers in the STEM workforce. This landscape analysis utilized a survey (n = 1743) of Texas graduates to understand undergraduate experiences, graduation metrics, job sector representation, salary data, retention trends, and workforce preparedness. Our recommendations and findings help identify areas in which industry and higher education can better collaborate to potentially improve STEM retention and workforce preparedness.

Keywords

career perceptions workforce preparedness STEM education STEM retention education/workforce connection

Introduction

The U.S. labor market has seen a great disruption post the COVID-19 pandemic, being referred to as ‘The Great Resignation’ and ‘The Great Reshuffle’, in which millions of workers have quit their jobs while simultaneously moving to different fields/roles (Ferguson, 2023). In the fields of Science, Technology, Engineering, and Mathematics (STEM), workforce demand remains high and the retention of both STEM students and workforce participants are integral components of such. Globally, there is a pressing shortage in the STEM talent pool, and the U.S. is no exception, where there is an expected shortage of STEM workers of up to 3.5 million people by 2025 (Lazio and Ford, 2019). STEM fields are major drivers of economic growth and global competitiveness, and STEM workforce participants support this through technological advancements, scientific progress, innovation, research, and alike (National Science Board, 2021). The STEM retention gap is a critical factor to be address in both achieving workforce diversity and focusing on the skills gap in the STEM sector (Estrada et al., 2016; Okrent and Burke, 2021; Patrick et al., 2022).

STEM retention in higher education relates to both academic and non-academic factors. Academic factors related to STEM retention include foundational course-work preparedness/access, critical thinking skills, problem-solving abilities, and a strong foundation of STEM concepts, among other facets (Leong et al., 2021; Wang, 2015). Non-academic factors include engagement in STEM, a student’s sense of belonging in STEM, mentorship opportunities, among other factors (Bonsangue et al., 2018; Strayhorn, 2018; Treisman, 1992).

STEM retention in the workforce relates to various issues, some of which are general retention issues in the workplace and others being centered directly on the specific STEM fields (Chen, 2013; Patrick et al., 2022). The most prevalent issues related to STEM workforce retention include lack of representation, through the lack of diversity across the fields, and unfair work environments (Brown, 2016; Kahn and Ginther, 2017).

Other issues impacting the STEM pipeline include the declining interest in science by young people and the aging scientific workforce (Riegle-Crumb et al., 2019). These gaps create challenges for industries due to the potential loss of knowledge and are exacerbated by the aforementioned retention issues. This study focuses on the lived experiences of STEM graduates across the State of Texas to better understand the landscape of the STEM workforce and the workforce perceptions and experiences in hopes to uncover how IHE’s can best support students entering the STEM workforce.

H-LSAMP description

The retention and persistence rates among underrepresented minority (URM) students in STEM has prompted STEM intervention programs at the collegiate level (Clewell et al., 2005; Estrada et al., 2016). There are various barriers that URM students face when it comes to collegiate success, including inadequate academic preparation, financial struggles, unwelcoming campus climates, among other facets (Hurtado et al., 2010; National Science Foundation, 2019; Patrick et al., 2022; Strayhorn, 2018). As part of a national intervention to reduce racial inequalities in STEM degree achievement, the Houston-Louis Stokes Alliance for Minority Participation (H-LSAMP) was established in 1998 across numerous universities within Texas. Nationally, LSAMP has had positive results for URM students in STEM over the past several decades with regards to graduation and retention (Hicks, 2007; National Science Foundation, 2017).

The H-LSAMP alliance currently consists of five institutions of higher education that focus on collaborative learning communities (CLC), academic supports such as peer-led tutoring and learning, faculty mentorship, dedicated communal space, student research experiences, and other retention-based program components found to be beneficial to minority STEM students (Ghazzawi et al., 2022; Preuss et al., 2021). The H-LSAMP institutions use the CLC model to promote academic success, increase persistence levels, and improve the sense of belonging among students (Bonsangue et al., 2018; Drew, 2011; Treisman, 1992). The intention of the H-LSAMP is to improve graduation and persistence rates of underrepresented minority students in STEM, which prevails through efforts to support students academically, through both engagement and support systems. In this study we aim to take a deeper look at how H-LSAMP participants are faring in the workforce compared to their non H-LSAMP STEM peers to understand the longer term patterns of STEM retention programs.

Despite the research on H-LSAMP students and their course and graduation outcomes, there is limited research on workforce outcomes and workforce preparation perceptions of H-LSAMP graduates in the workforce in addition to STEM graduates generally. As such within this study we intend to explore the following overarching research questions:

(1) What is the current landscape of STEM retention pipeline from education to industry for H-LSAMP graduates and non H-LSAMP graduates?

(2) What are the primary factors that affect retention in an academic STEM pathway and in the STEM workforce for H-LSAMP graduates and non H-LSAMP graduates?

(3) What are primary facets of the workforce for which H-LSAMP and non H-LSAMP graduates feel unprepared?

Literature review

With a pressing need to understand STEM retention and preparedness through the lens of student outcomes and perceptions (Sithole et al., 2017), this literature review aims to examine the existing research on the workforce retention and climate of STEM professions, with a goal of improving the alignment between STEM education and demands of the modern workforce. Literature depicts the importance of surveying students’ perceptions, highlighting that these perceptions can influence academic decisions and change (Nguyen et al., 2021). Students have indicated that career preparation is the primary factor for attending college (Cass et al., 2011). Which corresponds with the increasing college-degree requirements for many jobs (Carnevale and Rose, 2015). Research indicates that although students are drawn to study STEM, they also face challenges in their STEM education (Chen et al., 2018; Gasiewski et al., 2012).

Retention in STEM pathways is based on both both higher education retention and workforce retention. The rigorous coursework in STEM can be daunting, leading to academic pressures/stress, in addition to some students struggling to see the practical applications and real-world relevance of their studies in coursework that focuses more on theory. Coursework preparation in foundational mathematics and science course specifically contribute to early attrition (Aulck et al., 2017; Bressoud, 2021). Another prevailing STEM retention issue in collegiate programs is the lack of clear pathways within specific STEM degree plans (Maltese and Cooper, 2017; Seymour and Hewitt, 1997). Moreover, an academic program may have a very clear pathway – such as biology pre-medicine – but the low attrition rates and low medical school acceptances push students to have to discover non-medical disciplines and pathways (Chang et al., 2020).

Professionally, STEM fields are presently underrepresented by both underrepresented minorities (URM) and women, with women making up 28% of the STEM workforce and black and Latinos making up eight and 10% respectively (U.S. Bureau of Labor Statistics, 2022), albeit these groups each making up a larger portion of the overall workforce. URM and female students are faced with stereotypes, biases, and limited access to resources that contribute to the gender and ethnic disparities in both STEM education and the workforce (Chetty et al., 2014; Grosmman and Porche, 2013; Lazio and Ford, 2019). Women and students of color have experienced negative social experiences, through sexism and racism respectively, even during internship experiences which in turn influenced their postgraduation trajectories (Park et al., 2022).

Deming and Noray (2019) theorized that the retention challenges in the STEM workforce are in part due to ‘unstable career footing’ for many STEM graduates. In a national study, researchers found that nearly 40% of STEM degree holders never enter a STEM profession and over 30% who enter the STEM field leave before they turn 40 (Deming and Noray, 2019). Analyzing why employees left the technology industry specifically, researchers found that unfair work conditions were touted as the primary reason, which created issues of discrimination and lack of career advancement opportunities (Kapor Center, 2017). These issues in the workforce itself are particularly key for women and people of color leaving the field. Women, specifically, have experienced intimidation from male superiors and harassment from male peers, which is exacerbated by the lack of representation of women in STEM fields (Kahn and Ginther, 2017; U.S. Bureau of Labor Statistics, 2022).

Scholars claim that STEM workforce shortages are dependent on sector (Camilli and Hira, 2019; Xue and Larson, 2015) and that the rapid pace of technological change has created skills gaps in the workforce resulting in shortages (Deming and Noray, 2019; IBM, 2019). The STEM skills gap has been reported globally, particularly that the technological advancements within STEM require that STEM workers be constantly trained and developed as their current skills become increasingly superseded (Christo-Baker et al., 2017; Zaza et al., 2020). Beyond STEM, employers have generally criticized workforce readiness and skill deficiencies of university graduates entering the job market (Sayed, 2023; Waite and McDonald, 2018).

This study focuses on both the retention landscape and workforce preparedness in STEM from the perspective of those in the workforce specifically. The intent is to identify areas in which industry and higher education can better collaborate to potentially improve STEM retention and workforce preparedness. Through this landscape analysis we will also take a more intensive look at differences in retention and preparedness of H-LSAMP participants, who are primarily URM graduates, compared to their peers in the workforce.

Methodology

Instrumentation

This study utilized a mixed-methods approach through the use of an anonymous, self administered on-line survey. The quantitative aspect provided numerical insights while the qualitative aspect provided depth and context to the findings, allowing for an enriched overall interpretation of the data. Surveys were disseminated based on records of STEM graduates across five universities in the State of Texas. A survey was deemed the most appropriate mechanism for data collection as it allowed for efficient and systematic collection of quantitative and qualitative responses. The survey instrument questions were reviewed by eight researchers and/or program staff across the five universities of interest. Slight modifications were made based on the review. The survey was organized into seven topic areas, consisting of questions related to undergraduate experiences, graduation outcomes, workforce outcomes and STEM retention, workforce preparedness, graduate education pursuit, LSAMP connection to workforce/education outcomes, and demographics. The survey instrument included Likert-scale questions, ranking questions, selection questions, and open response questions. In most questions there was an opportunity for respondents to add notes/context to their response or identify an alternative response. Due to the extensive size of the survey the full result tables are available upon request.

Analytical strategy

Data from the survey instrument were analyzed through descriptive statistics and inferential analysis. Statistical measures such as averages, range of responses, percentage of respondents, cross-tabulations, etc. Were computed to describe and summarize the data and make inferences on the survey sample. Open-ended questions allowed respondents to provide more detail through qualitive responses, which were coded using grounded theory in which themes were extracted from the responses (Xu and Zammit, 2020), allowing themes to emerge organically from the responses. These were then cross-referenced with the emergent themes from the literature review comparing responses with the established knowledge in the field, to provide an inclusive understanding of the findings.

Descriptive statistics and study sample

Approximately 38,000 email requests were made to past STEM students across the five universities of interest to complete the instrument, in which we received 1890 responses. However, a number of respondents did not complete any of the content questions, resulting in a sample size of 1743 responses. Survey questions were not required thus the response rate for each question varies. Among the sample, 140 respondents indicated they participated in H-LSAMP and the remaining respondents were classified as non-participants. The demographic characteristics of the sample are depicted in Table 1, including sex, race, and highest level of parental education, broken out by H-LSAMP respondents and non-H-LSAMP respondents.

Table 1.

Demographic characteristics of survey respondents.

	Total respondents (percent)	H-LSAMP (percent)	Non H-LSAMP (percent)
Sex
Male	698 (57)	68 (56)	630 (57)
Female	494 (40)	51 (42)	443 (40)
Other	36 (3)	2 (2)	34 (3)
	1228	121	1107
Race/Ethnicity
White	509 (43)	14 (12)	495 (47)
Hispanic/Latino	287 (24)	34 (29)	253 (24)
Black	141 (12)	46 (40)	95 (9)
Asian	105 (9)	10 (9)	95 (9)
Other	17 (1)	1 (1)	16 (2)
Two or more races/ethnicities	118 (10)	11 (9)	107 (10)
	1177	116	1061
Underrepresented minority (URM)
URM respondent	563 (48)	92 (79)	471 (44)
Non URM respondent	614 (52)	24 (21)	590 (56)
	1177	116	1061
Highest level of parental education
Less than high school	76 (6)	8 (7)	68 (6)
High school diploma or GED	204 (16)	19 (15)	185 (17)
Occupational certificate or associates degree	110 (9)	13 (11)	97 (9)
Bachelor’s degree	463 (37)	40 (33)	423 (38)
Master’s degree	265 (21)	30 (24)	235 (21)
Doctorate or other professional degree (Ph.D., JD, MD, etc.	124 (10)	13 (11)	111 (10)
	1242	123	1119

In terms of demographics, 40% of the entire sample was female, albeit the women only representing 28% of the STEM workforce, as per the U.S. Bureau of Labor Statistics (2022). The H-LSAMP specific demographics were very similar to the non H-LSAMP in terms of gender but differed substantially based on race. In terms of underrepresented minorities (URM), we included Hispanic/Latino, Black, Other, and those who indicated two or more races as part of this count. Of the H-LSAMP sample, 79% were URM compared to only 44% of the non H-LSAMP sample. In terms of highest level of parental education, we classified responses based on the highest level – thus if respondents indicated one parent has a high school diploma and the other had a master’s degree, we counted this as master’s degree in our dataset. These results depicted that 22% of the respondents indicated their parents’ highest level of education was a high school diploma or less and another nine percent indicated their parents’ had an occupational certificate or associates degree, thus over a quarter of the respondents were first-generation 4-year college students.

Results

To understand the landscape, we explored the trajectory of students from their undergraduate experiences, such as their declared majors, GPA, hours and jobs worked during school, etc. In addition to their graduation outcomes, beginning experiences in the workforce, current jobs, sectors, and salaries, time in the STEM workforce, graduate studies, among other facets. We analyzed these components by disaggregating H-LSAMP respondents from non H-LSAMP respondents.

RQ1: What is the current landscape of STEM retention pipeline from education to industry for H-LSAMP graduates and non H-LSAMP graduates?

To best understand the retention pipeline of STEM students into the workforce, we first evaluate characteristics of students’ differing undergraduate experiences through understanding how much students worked during their baccalaureate pursuits. Figure 1 depicts the number of hours worked per month, on average, during undergrad. When looking at those students who worked over 80 hours per month, equating to an average of 20 hours per week or more, 18% of H-LSAMP students were in this category compared to 27% of non H-LSAMP students. When analyzing the hours worked during undergrad by race, we found considerably similar results for URM and non URM respondents with regards to the time spent working during undergrad.

Figure 1.

Hours worked while in undergrad.

We then evaluated the primary jobs students held during their undergraduate experiences, which were categorized based both the NSF Science and Engineering crosswalk of occupations (NCES, 2020) and nuances within this dataset, as shown in Table 2. The most prevailing trends between H-LSAMP and non H-LSAMP students were in the job categories of library/tutoring, research, and retail/service. Of the H-LSAMP respondents, 44% worked in the library, tutored, or worked on some sort of research during their time in undergrad. Comparatively, only 17% of the non H-LSAMP respondents worked in these sorts of environments. In terms of retail and service jobs, 15% of H-LSAMP respondents indicated they worked primarily in these roles during undergrad, compared to 37% of non H-LSAMP respondents. These differences depict the experiential distinctions among students. H-LSAMP students working during undergrad, but their experiences were more academic or adjacent in some way to their baccalaureate pursuit compared to the non H-LSAMP STEM students.

Table 2.

Jobs worked while in undergrad.

	Total respondents (percent of respondents)	H-LSAMP (percent of H-LSAMP)	Non H-LSAMP (percent of non H-LSAMP)
Job category
Administrative	71 (7)	8 (8)	63 (7)
Engineering	38 (4)	4 (4)	34 (4)
Instruction	19 (2)	4 (4)	15 (2)
Internship^a	40 (4)	3 (3)	37 (4)
It tech support	20 (2)	2 (2)	18 (2)
Laboratory	61 (6)	8 (8)	53 (6)
Library/Tutoring	126 (13)	27 (26)	99 (11)
Manual	22 (2)	0 (0)	22 (3)
Mentoring	3 (0)	1 (1)	2 (0)
Research	73 (8)	19 (18)	54 (6)
Retail	165 (17)	10 (10)	155 (18)
Sales	20 (2)	0 (0)	20 (2)
Service	168 (17)	5 (5)	163 (19)
Teaching assistant	62 (6)	5 (5)	57 (7)
Technician^b	48 (5)	4 (4)	44 (5)
Work study	30 (3)	4 (4)	26 (3)
Total respondents	966	104	862

^aRespondents that indicated they were in an internship but not the specific field of said internship.

^bTechnician was classified as a job that involved a “technical skill” such as vet techs, pharmacy techs, medical scribes, those who worked in finance, etc.

In terms of academic retention, the majority of our respondents were graduates, as per Table 3. This finding is anticipated as the survey itself was disseminated to past STEM students, in which approximately 98% were alumni of the respective institutions in this study. We then assessed the differences between students’ graduating majors and the majors they had declared the majority of the time in college to understand retention of STEM students during school.

Table 3.

Graduation.

	Number of respondents	Number of graduates	Graduation percentage (%)
H-LSAMP students	140	138	98.6
Non H-LSAMP students	1602	1569	97.9′

For ease of analysis, we separated these results into H-LSAMP specific and non H-LSAMP specific tables, Tables 4 and 5 respectively. The largest representations by major were in biology, engineering, and computer engineering/science, for both H-LSAMP and non H-LSAMP respondents, with over 66% of the sample declaring and/or graduating with one of these majors. However, it’s important to note that H-LSAMP recruitment is more focused on physical sciences, such as mathematics, chemistry, physics, and engineering, rather than life sciences or natural sciences, such as biological studies. This can be in part seen through the differences in the percentages of math and chemistry graduates, where 30% of H-LSAMP respondents were math or chemistry graduates, cumulatively, while only 17% of non H-LSAMP graduated with math or chemistry degrees.

Table 4.

Primary declared major, graduation major, and differences - non H-LSAMP only.

	Primary declared major (percentage)	Graduating major (Percentage)	Difference – Switched major or did not graduate
Biochemistry	14 (1)	12 (1)	−2
Biology	478 (33)	437 (32)	−41
Chemistry	120 (8)	116 (9)	−4
Computer Engineering/Science	242 (17)	236 (17)	−6
Construction/Concrete Management	40 (3)	42 (3)	2
Engineering	310 (22)	285 (21)	−25
Environmental Science	15 (1)	14 (1)	−1
Geology	14 (1)	12 (1)	−2
Industrial Technology	7 (0)	7 (1)	0
Information Technology	8 (1)	6 (0)	−2
Mathematics	109 (8)	107 (8)	−2
Other	26 (2)	22 (2)	−4
Physics	38 (3)	40 (3)	2
Wildlife/Marine Biology/Ecology	13 (1)	18 (1)	5
Total	1434	1354	−80

Table 5.

Primary declared major, graduation major, and differences - H-LSAMP only.

	Primary declared major (percentage)	Graduating major (percentage)	Differences – Switched major or did not graduate
Biochemistry	0 (0)	0 (0)	0
Biology	29 (22)	26 (20)	−3
Chemistry	18 (14)	19 (15)	1
Computer Engineering/Science	26 (20)	23 (18)	−3
Construction/Concrete Management	1 (1)	1 (1)	0
Engineering	35 (27)	35 (28)	0
Environmental Science	1 (1)	0 (0)	−1
Geology	1 (1)	1 (1)	0
Industrial Technology	1 (1)	1 (1)	0
Information Technology	0 (0)	0 (0)	0
Mathematics	18 (14)	19 (15)	1
Other	0 (0)	1 (1)	1
Physics	1 (1)	1 (1)	0
Wildlife/Marine Biology/Ecology	0 (0)	0 (0)	0
Total	131	127	−4

The most prominent changes or drops in major were in the realms of biology and engineering, specifically for non H-LSAMP students, where 80 students switched majors out of STEM or did not graduate. More specifically, 478 non H-LSAMP students were declared biology majors during the majority of undergrad, however only 437 graduated with a biology degree, which shows that 41 non H-LSAMP students switched from biology to another non-STEM major or did not graduate. In exploring the data further, we found that 14 of the 41 students switched majors and graduated, while 27 did not graduate. Similarly, this sample included 310 non H-LSAMP students who were primarily declared as engineering majors during undergrad, while only 285 graduated, we found that 16 of the 25 students switched majors into a non-STEM field, while nine did not graduate. We will explore reasons in which respondents did not graduate in research question 2.

We then shifted to understanding retention in the workforce, by asking a series of questions related to time in the overall workforce, time in the STEM workforce specifically, and the types of jobs respondents had. To understand retention in the STEM workforce we calculated time in the STEM workforce by using the number of years respondents had been in the STEM workforce specifically and divided that by the total number of years they had been in the overall workforce. Thus, if a respondent was in the STEM workforce for 15 years and the overall workforce for 20 years, we calculated their time in the STEM workforce as 75%. The categorizations for these calculations are depicted in Table 6. In conducting a t-test to compare the means of time spent in the STEM workforce between H-LSAMP and non H-LSAMP respondents, we found that the mean time spent in the STEM workforce was 95% for H-LSAMP respondents which was significantly higher than non H-LSAMP respondents which had a mean time of 85% (p < .001). These findings suggest that H-LSAMP respondents are either retained in the STEM workforce longer than their non H-LSAMP peers and/or enter the STEM market at a faster rate.

Table 6.

Time in the STEM workforce.

Time in the STEM workforce	Total respondents (percentage)	H-LSAMP (percentage)	Non H-LSAMP (Percentage)
No time in STEM workforce	68 (6)	1 (1)	67 (6)
25% or less time in the STEM workforce	23 (2)	1 (1)	22 (2)
Between 25 and 50% time in STEM workforce	89 (7)	4 (3)	85 (8)
Between 50 and 75% time in STEM workforce	39 (3)	1 (1)	38 (4)
Between 75 and 100% time in STEM workforce	74 (6)	7 (6)	67 (6)
100% of time in STEM workforce	908 (76)	106 (88)	802 (74)
Total respondents	1201	120	1081

To understand potential disparities in salary we solicited responses related to current salary, in which we had a lower response rate of 782 respondents. We then analyzed average salaries for H-LSAMP and non H-LSAMP respondents and intersected that with GPA to understand potential differences, shown in Table 7. We conducted two t-tests to estimate whether the differences in salaries were significant and found that the difference in salary means for H-LSAMP (mean = $117,655) and non H-LSAMP (mean = $110,992) respondents was not statistically different, albeit the average salary for H-LSAMP respondents being higher. However, the t-test comparing the difference in salary means for respondents based on GPA showed that respondents with over a 3.3 GPA has a statistically different salary (p = .05) with an average salary of $118,041, while respondents with a 3.2 and below GPA has an average salary of $103,236. Both groups, graduates with a 3.2 GPA and below and graduates with a 3.3 GOA and higher, had between 7.1 and 7.2 years of workforce experience.

Table 7.

Average salary and GPA cross tabulation.

	All respondents average salary	H-LSAMP average salary	Non H-LSAMP Average salary	Average years in the workforce
Category
3.2 and below GPA	$103,236	$101,288	$99,195	7.2 years
3.3 and above GPA	$118,041	$135,221	$117,060	7.1 years
All	$111,696	$117,655	$110,992	7.1 years

As part of understanding the full landscape of STEM retention from education through the workforce, we wanted to assess the trajectory of students who obtained or are currently pursuing graduate education. In terms of H-LSAMP participants, 90 of the 140 (64%) respondents pursued or obtained a graduate degree compared to 613 of the 1602 (38%) non H-LSAMP participants. Table 8 focuses on the different types of graduate degrees obtained specifically, as depicted 62% of the entire H-LSAMP sample has obtained a graduate degree, half being master’s degrees. Comparatively, 34% of the non H-LSAMP sample has completed a graduate degree. From a national context, the U.S. Census Bureau reports that approximately 40% of college-educated STEM workers have a graduate degree (2021). In terms of medical degrees (MD’s) and PhD’s, 22% of the H-LSAMP respondents have completed one of these terminal degrees, compared to 10% of the non H-LSAMP respondents.

Table 8.

Types of graduate degrees obtained.

	H-LSAMP	Non H-LSAMP
Response
MD	15 (11)	93 (6)
PhD	15 (11)	61 (4)
JD	0 (0)	8 (0)
MBA	8 (6)	32 (2)
DDS	1 (1)	10 (1)
Masters	43 (31)	279 (17)
Other	5 (4)	54 (3)
Total	87 (62)	537 (34)

RQ2:What are the primary factors that affect retention in an academic STEM pathway and in the STEM workforce for H-LSAMP graduates and non H-LSAMP graduates?

Retention in school

To evaluate the primary factors that affect retention in both the STEM academic pipeline and the STEM workforce pipeline, we explored the responses from respondents who did not graduate and the reasons associated in addition to the respondents who left or never entered the STEM workforce and why. Albeit the majority of the respondents were university graduates and this study focuses more on retention in the workforce, 35 respondents did not graduate and indicated ‘taking care of siblings’, followed by ‘major not being a good fit’, ‘academic struggles’, and the ‘financial burdens of staying in school’ were the primary reasons for not graduating.

Retention in the workforce

Graduates who diverted from the STEM workforce were asked to rank their reasons for leaving the field, with a rating of one being the most integral reason. Table 9 depicts the varying reasons for not staying or entering the field, associated ratings for both H-LSAMP and non H-LSAMP and the total number of respondents for each reason. We assessed both the average rating and the number of respondents for each of the depicted reasons. H-LSAMP participants that did not enter a STEM profession indicated that their career interests changing and that it was difficult to find employment in their specific STEM fields as the two most influential factors. Non H-LSAMP participants that did not enter a STEM profession indicated that it was difficult to find employment in their specific STEM fields and barriers to entry such as certification and/or experience required were the two most influential factors for not being retained in a STEM profession. Despite 77% of the respondents indicating they did enter a STEM profession, a subset of these respondents either did not stay in a STEM career or switched careers and may have reverted back to STEM, based on the response rates for the reasons for leaving STEM.

Table 9.

Retention in STEM career.

	H-LSAMP respondent average rating (number of respondents)	Non-HLSAMP respondents average rating (number of respondents)	Total number of respondents (percentage of respondents)
Reason for leaving STEM
Difficult to find employment in the specific STEM field for which I got a degree	2.83 (24)	2.68 (371)	395 (44)
Barriers to entry such as certification and/or experience required	2.96 (24)	2.79 (363)	387 (43)
Lost interest in the field	3.13 (24)	3.44 (325)	349 (39)
Change in career or professional interests	2.89 (28)	3.1 (321)	349 (39)
Lack of inclusivity in the field	2.61 (18)	3.38 (299)	317 (35)
Found an opportunity outside of STEM that paid better	2.79 (14)	3.22 (312)	326 (36)
Started a family	3.71 (14)	3.68 (288)	302 (34)
Childcare responsibilities	3.71 (14)	3.66 (280)	294 (33)
Eldercare	3.71 (14)	3.8 (271)	285 (32)
Taking care of siblings	3.62 (13)	3.91 (269)	282 (31)
Availability of affordable childcare	3.64 (14)	3.78 (271)	285 (32)
Lack of degree	3.79 (14)	3.73 (266)	280 (31)
Other^a	3.4 (5)	3 (101)	106 (12)
N/A, I did enter a STEM profession	1.53 (72)	2.05 (621)	693 (77)

^aIndicated STEM employers required more experience, higher degrees, higher GPA’s, etc., financial barriers, degree was not helpful, entered adjacent field, needing more advising/counseling, and it being hard to find/keep a job as “Other” reasons.

Note.The lower the average rating, the more respondents indicated this was of importance in their decision to enter and/or stay in the field of STEM.

Workforce retention by subgroup

Taking a deeper look at how family responsibilities related to retention in the STEM workforce, we assessed how men and women and URM and non-URM respondents fared with this regard and found that 23-25% of women did not enter or left STEM at some point due to family responsibilities, such as starting a family, childcare responsibilities, eldercare, taking care of siblings, and/or the availability of affordable childcare, compared to 20-22% of men. Comparatively, the differential was wider between URM and non-URM respondents. Between 26 and 28% of URM respondents left STEM at some point or did not enter due to the aforementioned family responsibilities, compared to between 20 and 22% of non-URM respondents.

RQ3: What are primary facets of the workforce for which H-LSAMP and non H-LSAMP graduates feel unprepared?

In addressing unpreparedness of STEM graduates, we assessed the categories in which respondents selected that they felt unprepared for in addition to classifying the qualitative responses participants specified. Participants were able to select multiple components in which they felt unprepared for, depicted in Table 10. As shown, both supervising personnel and budget management were the highest selected categories for both H-LSAMP and non H-LSAMP participants, with between 36 and 41% of the subgroups feeling unprepared to supervise personnel and between 31 and 36% of the subgroups feeling unprepared for budget management. Both of these can be seen more as ‘non technical’ skills and are typically not part of a STEM college sequence of courses, as compared to student studying business that may have more exposure in these realms (University of the People, 2023).

Table 10.

Aspects of the workforce graduates felt unprepared for.

	Number of H-LSAMP (percent of H-LSAMP)	Number of non H-LSAMP (percent of non H-LSAMP)
Response
Budget management	50 (36)	501 (31)
Project management	43 (31)	453 (28)
Passing professional certifications (ex. PE for engineering)	25 (18)	260 (16)
Supervising personnel	58 (41)	570 (36)
Interacting at academic conferences	16 (11)	257 (16)
Professional presentations	29 (21)	272 (17)
Other^a	18 (13)	144 (9)
Total responses	239	2457

^aIndicated alignment of current industry standards, practical experience, job processes, technical skills (software use, Excel, GIS, etc.), soft skills in the workplace, personal finance, among other reasons of workforce preparedness.

Some specific comments from participants related to workforce preparedness included: “not enough computer science in biology, everything needs programming skills nowadays,” and “use of scientific equipment that wasn’t outdated,” and “needing practical coding,” and “field work consistent with industry standards,” and “current programming languages at the time, C++ and assembly, were dated even before graduation,” and “education was mostly theoretical. I understand the calculus behind electrical engineering, but coming out of school, I did not know anything about the equipment and devices that are actually used for electrical distribution.” Other comments included “coursework was not sufficient for the fields I worked in post graduation. Science today is not simply biology, everything requires bioinformatics skills that were not offered or required for my major,” and “real world application, also use of computer programs such as Excel and Blue Beam would have been extremely helpful.” These responses depict the need for academic experiences to incorporate more up-to-date equipment/software and practical experiences to mimic real-world dilemmas, which may require more collaboration with industry to determine those and/or ensure there are real-world exposure to the changes in the field.

Various graduates commented they felt unprepared related to soft skills, primarily including networking, navigating workplace politics, negotiating, stakeholder engagement, communications, etc. Specific comments were made on needing “basic life skills such as tax, personal finance, how to achieve emotional balance in work, navigating workplace disputes,” “career exploration and finding work that aligns with my degree,” and “being knowledgeable about the variety of jobs in the field as I feel many students just hyper focused on the developer jobs, and being knowledgeable about just the current tools that are used.” These responses depicts two differing needs that span beyond STEM, (1) being that students need basic and soft skills as part of their academic training and (2) students need more career exploration opportunities. One respondent summarized a broad issue within the education to workforce pipeline that corresponded with several other responses:

There are very few opportunities for students who do not pursue a Master's degree, and internships are not a required part of graduating. In general, you learn a lot about subjects but are entirely unprepared for the working world unless you have been personally coached on what to do. Naturally, most students (especially those who don't come from families with professional backgrounds) have no idea what to expect. You are supposed to get a job, but nobody will hire you without prior experience.

Further research with regard to the differing workforce tracks and graduate sentiments will be important to understand the nuances among different sectors and their needs. One respondent in the field of oil and gas specified that they felt “burnout, low career prospects, monotony, constant layoffs, high stress, low pay, [and] no advancement opportunities.” These sentiments may or may not be specific to these fields but exploring the intersections between unpreparedness and job sector will be an important follow up to this study to understand the disparities.

Discussion & recommendations

The STEM workforce continues to become a larger component of the United States workforce. From 2011 to 2021, the percentage of workers in STEM occupations increased from 22% to 24% of the total workforce (National Science Board, 2024). STEM workers have reported lower rates of unemployment and higher wages than their educational peers (National Science Board, 2024). Despite this growth, disparities in racial and ethnic groups among STEM employees across the nation persist (National Science Board, 2021, 2024; Zhou et al., 2023).

Within this study we assessed both the distinctions between H-LSAMP graduates and non-H-LSAMP graduates, in terms of STEM workforce outcomes, and general trends in the STEM retention pipeline. We found that several differences between H-LSAMP respondents and non H-LSAMP respondents. H-LSAMP respondents generally worked less hours per month during college, likely due to the emphasis on financial support within the H-LSAMP in hopes to ensure students focus on their academic pursuits without facing too many financial burdens and hardship. Furthermore, H-LSAMP respondents were more likely to participate in work experiences that were more aligned with their coursework and STEM pathways. H-LSAMP graduates generally had better cumulative GPAs, suggesting potential advantages from collaborative learning experiences and academic supports. With regards to workforce retention and higher education, we found that H-LSAMP graduates were retained in the STEM workforce at a statistically significant and higher rate, and had nearly double the percent of graduate degrees compared to non H-LSAMP graduates.

Generally, we found that graduates have difficulty finding employment in the specific STEM fields in which they had a degree for, in which one graduate specified that “most companies don’t actually want to hire a physics major, [I] was only able to find a job because of my computer science classes.” We also found that barriers to entry such as certification and/or experience requirements were obstacles for many respondents, as articulated by one respondent that “a lot of the jobs require a certain amount of experience to sit for certification for jobs in labs.” When analyzing differences between URM and non-URM graduates, we found that a higher rate of URM respondents left STEM at some point or did not enter due to the family responsibilities.

We also explored how graduates felt unprepared for certain aspects of the workforce, in which the most predominant aspects were supervising personnel, budget management, technical skills, soft skills, and alignment with industry standards. Some graduates also expressed the need for practical skills, updated equipment/software, and real-world experiences in their education. Graduates expressed a need for basic life skills such as personal finance, tax management, emotional balance, and navigating workplace disputes. They also highlighted the importance of career exploration, understanding various job opportunities, and staying up to date with industry tools and trends. Based on our findings we developed specific recommendations for academic institutions and industry partners in the workforce:

Recommendations for academic institutions

• Coursework and pedagogy should incorporate experiential learning and embeds industry-relevant skills to bridge the gap between education and workforce readiness. This requires collaborative curriculum refinement on an ongoing basis by industry representatives and university faculty.

• Surveying and partnering with industry representatives to understand what software’s, tools, and skills need to the most focus in the academic settings. As one graduate in this study depicted “the degree needs to be more focused on practice. Add one project based course on the basics of HVAC, machine design, and systems engineering.” As industry requirements and needs change, it will be important for even freshman students to mentally prepare for specific changes in the field. An example would be if specific coding languages are becoming more and more obsolete in the workforce, it’s important for universities and students to know and prepare for these industry shifts.

• Developing and/or providing access to workforce readiness course on STEM technical skills such as budget/grant management, project management, understanding and passing professional certifications, interacting/presenting at conferences, etc. This course could be an elective or a workshop series, and could be a potential event to collaborate with other collegiate departments, such as business, communications, etc.

• Creating more bilateral research shadowing and/or conducting opportunities. Universities with large enrollments cannot typically offer every STEM student lab research experience, thus it’s integral to support this pipeline through the connections with industry. Research exposure and experience enhances students’ practical skills and readiness for the STEM workforce (Linn et al., 2015; Lopatto, 2010).

Recommendations for industry partners in the workforce

• Increasing the number of entry-level rotational programs that do not require 3-years of experience, as several respondents indicated. Through a rotational program and scaled onboarding approach, companies can allow entre-level STEM workers to learn about systems, processes, and technical expertise within differing high-need departments. These programs should ensure subject matter expert (SME) knowledge sharing among tenured employees occurs with new staff, to aid in the dissemination of institutional knowledge.

• Offering more internships, co-ops, apprenticeships, shadowing opportunities, etc. To increase exposure for students pursuing STEM pathways. Supportive learning environments that expose students to STEM fields and strengthen connections between academia and industry, which increase STEM student retention rates (Estrada et al., 2016; Maltese and Cooper, 2017). As one graduate reflected “in hindsight I now know that my degree alone holds no weight, and had I been able to take advantage of internships or research opportunities I may have gained relative knowledge to obtain a job in the STEM field.”

• Increasing mentorship and role model opportunities on college campuses. Effective mentorship can help shape student perceptions, experiences, and skills in STEM, and can foster a sense of belonging/confidence for URM students (Sandrone, 2022; Strayhorn, 2018)

• Offering pre-employee bootcamps/trainings that can serve as both an avenue to attract talent and a way to create company-specific ‘certifications’ that expose students to industry knowledge that their universities may not teach specifically.

This study compared workforce outcomes and perceptions of H-LSAMP and non-HLSAMP STEM graduates to best understand the current STEM workforce landscape and how industry/education can better align. Despite the positive findings of H-LSAMP graduates, survey respondents who depicted challenges in finding STEM employment attributed this to certification and experience requirements. STEM graduates expressed feeling unprepared for certain workforce aspects, highlighting the need for more practical experiences in college and updated equipment/software based on industry standards. This study’s recommendations pertain to how academic institutions can better align with industry requirements/standards and how industry partners can be more involved in the STEM pipeline through earlier exposure and integration.

Footnotes

ORCID iD

Toni Templeton

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Division of Equity for Excellence in STEM (2409022) and Texas Southern University.

Declaration of conflicting interests

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

References

Aulck

Aras

, et al. (2017) STEM-ming the tide: predicting STEM attrition using student transcript data. Proceedings of Machine Learning for Education Workshop. https://jevinwest.org/papers/Aulck2017kdd.pdf

Bonsangue

Drew

Gilmore

(2018) The impact of a supportive community experience on African American students in chemistry and chemical engineering. Learning Assistance Review 23(2): 9–40.

Bressoud

(2021) Decades Later, Problematic Role of Calculus as Gatekeeper to Opportunity Persists. University of Texas at Austin. https://www.utdanacenter.org/blog/decades-later-problematic-role-calculus-gatekeeper-opportunity-persists

Brown

(2016) Inclusion: Diversity, the New Workplace & the Will to Change. Purpose Driven Publishing.

Camilli

Hira

(2019) STEM workforce: STEM education and the post-scientific society. Journal of Science Education and Technology 28: 1–8.

Carnevale

Rose

(2015) The Economy Goes to College: The Hidden Promise of Higher Education in the Post-industrial Service Economy. Georgetown University. https://cew.georgetown.edu/wp-content/uploads/IO-Executive-Summary-web.pdf

Cass

Hazari

Cribbs

, et al. (2011) Examining the impact of mathematics identity on the choice of engineering careers for male and female students. ASEE/IEEE Frontiers in Education Conference Paper.

Chang

Kuncel

Sackett

(2020) The process of attrition in pre-medical studies: a large-scale analysis across 102 schools. PloS one 15(12): e0243546.

Chen

(2013) STEM attrition: college students' paths into and out of STEM fields. In: Statistical Analysis Report. National Center for Education Statistics. https://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2014001rev

10.

Chen

Johri

Rangwala

(2018) Running out of STEM: A Comparative Study Across STEM Majors of College Students at-risk of Dropping out Early. Association for Computing Machinery, 270–279.

11.

Chetty

Hendren

Kline

, et al. (2014) Is the United States still a land of opportunity? Recent trends in intergenerational mobility. American Economic Review 104(5): 141–147.

12.

Christo-Baker

Sindone

Roper

(2017) Addressing the skills gap: a regional analysis. Journal of Applied Business and Economics 19(8): 10–12.

13.

Clewell

Cohen

Tsui

, et al. (2005) Final Report: Evaluation of the National Science Foundation Louis Stokes Alliances for Minority Participation Program. The Urban Institute. https://webarchive.urban.org/UploadedPDF/411301_LSAMP_report_appen.pdf

14.

Deming

Noray

(2019) STEM Careers and the Changing Skill Requirements of Work. National Bureau of Economic Research.

15.

Drew

(2011) STEM the Tide: Reforming Science, Technology, Engineering, and Math Education in America. Johns Hopkins University Press.

16.

Estrada

Burnett

Campbell

, et al. (2016) Improving underrepresented minority student persistence in STEM. CBE-Life Sciences Education 15(3): es5.

17.

Ferguson

(2023) A National Cris by the Numbers. U.S. Chamber of Commerce. https://www.uschamber.com/workforce/america-works-data-center

18.

Gasiewski

Eagan

Garcia

, et al. (2012) From gatekeeping to engagement: a multicontextual, mixed method study of student academic engagement in introductory STEM courses. Research in Higher Education 53: 229–261.

19.

Ghazzawi

Pattison

Horn

, et al. (2022) Houston-Louis Stokes alliance for minority participation: findings from 17 years of a multi-institutional consortium focused on building minority student success in STEM. Electronic Journal for Research in Science and Mathematics 26(3): 1–18.

20.

Grosmman

Porche

(2013) Perceive gender and racial/ethnic barriers to STEM success. Urban Education 49(6): 698–727.

21.

Hicks

(2007) Alliances for success. The FASEB Journal 21: A40.

22.

Hurtado

Newman

Tran

, et al. (2010) Improving the rate of success for underrepresented racial minorities in STEM fields: insights from a national project. New Directions for Institutional Research 2010(148): 5–15.

23.

IBM (2019) IBM Study: The Skills Gap Is Not a Myth, but Can Be Addressed with Real Solutions. IBM. https://newsroom.ibm.com/2019-09-06-IBM-Study-The-Skills-Gap-is-Not-a-Myth-But-Can-Be-Addressed-with-Real-Solutions

24.

Kahn

Ginther

(2017) Women and STEM. Annual Review of Economics 9: 307–331.

25.

Kapor Center (2017) Tech leavers study: a first of it’s kind analysis of why people voluntarily left jobs in tech. Kapor Center. https://www.kaporcenter.org/wp-content/uploads/2017/05/KAPOR_Tech-Leavers-17-0514.pdf

26.

Lazio

Ford

(2019) The U.S. needs to prepare workers for STEM jobs. Society for Human Resource Management. https://www.shrm.org/hr-today/news/hr-magazine/summer2019/pages/the-u.s.-needs-to-prepare-workers-for-stem-jobs.aspx

27.

Leong

Mercer

Danczak

, et al. (2021) Transition to first year chemistry: student, secondary and tertiary educator's perceptions of student preparedness. Chemistry Education: Research and Practice 2021(41): 923–947.

28.

Linn

Palmer

Baranger

, et al. (2015) Undergraduate research experiences: impacts and opportunities. Science 347(6222): 1261757.

29.

Lopatto

(2010) Undergraduate research experiences support science career decisions and active learning. CBE-Life Sciences Education 9(4): 297–306.

30.

Maltese

Cooper

(2017) STEM pathways. Do men and women differ in why they enter and exit? AERA Open 3(3): 1–16.

31.

National Center for Science & Engineering (2020) Crosswalk of occupations used in the National Survey of College Graduates. https://www.nsf.gov/statistics/2020/nsf20300/pdf/nsf20300-taba-001.pdf

32.

National Science Board (2021) The STEM labor force of today: scientists, engineers, and skilled technical workers. Science & Engineering Indicators 2021. (NSB-2021-1). https://ncses.nsf.gov/pubs/nsb20212

33.

National Science Board (2024) The STEM Labor Force: Scientists, Engineers, and Skilled Technical Workers. Science & Engineering Indicators 2024. (NSB-2024-5). https://ncses.nsf.gov/pubs/nsb20245/assets/nsb20245.pdf

34.

National Science Foundation (2017) Louis stokes alliances for minority participation (LSAMP). https://www.nsf.gov/funding/pgm_summ.jsp?pims_id=13646

35.

National Science Foundation (2019) Women, Minorities, and Persons with Disabilities in Science and Engineering (NSF 19-304). https://ncses.nsf.gov/pubs/nsf19304/

36.

Nguyen

Netto

CLM

Wilkins

, et al. (2021) Insights into students’ experiences and perceptions of remote learning methods: from the COVID-19 pandemic to best practice for the future. Frontiers in Education 6: 647986.

37.

Okrent

Burke

(2021) The STEM labor force of today: scientists, engineers, and skilled technical workers. National Science Board. https://ncses.nsf.gov/pubs/nsb20212

38.

Park

Zheng

Lue

, et al. (2022) Looking beyond college: STEM college seniors on entering the workforce and the impact of race and gender. Journal of Diversity in Higher Education 17: 527–540.

39.

Patrick

Smith

Estes

(2022) An exploratory study of STEM persistence programs for underrepresented minority students. Journal of Higher Education Theory and Practice 22(1): 49–61.

40.

Preuss

Merriweather

Avila

, et al. (2021) The impacts of global research and international educational experiences on Texas A&M university system LSAMP participants. Frontiers in Education 6: 674772.

41.

Riegle-Crumb

King

Irizarry

(2019) Does STEM stand out? Examining racial/ethnic gaps in persistence across postsecondary fields. Educational Researcher 48(3): 133–144.

42.

Sandrone

(2022) Science identity and its “identity crisis”: on science identity and strategies to foster self-efficacy and sense of belonging in STEM. Frontiers in Education 7: 871869.

43.

Sayed

(2023) The STEM skills gap: a growing challenge for countries to overcome. HR Forecast. https://hrforecast.com/the-stem-skills-gap-a-growing-challenge-for-countries-to-overcome/

44.

Seymour

Hewitt

(1997) Talking About Leaving: Why Undergraduates Leave the Sciences. Westview Press.

45.

Sithole

Chiyaka

McCarthy

, et al. (2017) Student attraction, persistence and retention in STEM programs: successes and continuing challenges. Higher Education Studies 7(1): 46.

46.

Strayhorn

(2018) College Students’ Sense of Belonging: A Key to Educational Success for all Students. Routledge.

47.

Treisman

(1992) Studying students studying calculus: a look at the lives of minority mathematics students in college. The College Mathematics Journal 23(5): 362–372.

48.

University of the People (2023) What Do You Learn in Business School? University of the People. https://www.uopeople.edu/blog/what-do-you-learn-in-business-school/

49.

U.S. Bureau of Labor Statistics (2022) Table 11: employed persons by detailed occupation, sex, race, and Hispanic or Latino ethnicity. https://www.bls.gov/cps/cpsaat11.htm

50.

U.S. Census Bureau (2021) STEM Majors Earned More than Other Stem Workers. U.S. Census Bureau. https://www.census.gov/library/stories/2021/06/does-majoring-in-stem-lead-to-stem-job-after-graduation.html#:∼:text=About40%25ofcollege-educatedSTEM,workershadagraduatedegree

51.

Waite

McDonald

(2018) Exploring challenges and solutions facing stem careers in the 21st century: a human resource development perspective. Advances in Developing Human Resources 21(1): 3–15.

52.

Wang

(2015) Pathway to a baccalaureate in STEM fields: are community colleges a viable route and does early STEM momentum matter? Educational Evaluation and Policy Analysis 37(3): 376–393.

53.

Zammit

(2020) Applying thematic analysis to education: a hybrid approach to interpreting data in practitioner research. International Journal of Qualitative Methods 19: 1609406920918810.

54.

Xue

Larson

(2015) STEM crisis or STEM surplus? Yes and yes. U.S Bureau of Labor Statistics. https://www.bls.gov/opub/mlr/2015/article/stem-crisis-or-stem-surplus-yes-and-yes.htm

55.

Zaza

Abston

Arik

, et al. (2020) What CEOs have to say: insights on the STEM workforce. American Business Review 23: 155.

56.

Zhou

Stratton

(2023) Retention in STEM: factors influencing student persistence and employment The Proceeding of the 19th Annual National Symposium on Student Retention, 205–217.