Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Next Article in Journal
Gender Differences in Computational Thinking Skills among Primary and Secondary School Students: A Systematic Review
Next Article in Special Issue
Building Bridges in STEM Education: Minoritized Secondary School Student Computer Science Pathways and Experiences
Previous Article in Journal
Elementary School Teachers’ Enactment of a Content Literacy Curriculum in a Virtual Tutoring Program for Multilingual Students
Previous Article in Special Issue
Alleviating Barriers Facing Students on the Boundaries of STEM Makerspaces
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Encouraging STEM Careers among Minoritized High School Students: The Interplay between Socio-Environmental Factors and Other Social Cognitive Career Constructs

1
School of Education, Notre Dame of Maryland University, Baltimore, MD 21210, USA
2
Whiting School of Engineering, Johns Hopkins University, Baltimore, MD 21218, USA
3
City Teaching Alliance, American University, Washington, DC 20016, USA
*
Author to whom correspondence should be addressed.
Educ. Sci. 2024, 14(7), 789; https://doi.org/10.3390/educsci14070789
Submission received: 31 May 2024 / Revised: 4 July 2024 / Accepted: 14 July 2024 / Published: 20 July 2024
(This article belongs to the Special Issue STEM Education for All: Breaking Barriers and Building Bridges)

Abstract

:
Performance in math, particularly algebra, is a major barrier to student success and participation in STEM among under-represented minoritized students, particularly Black U.S. high school students. This research applies social cognitive career theory (SCCT) to measure the impacts of an afterschool algebra for engineering program on math self-efficacy and interest in STEM among high school students in a large urban district. To study the program’s effects, a mixed methods research design was used where schools were assigned to either treatment or control conditions. Students in treatment schools accessed algebra for engineering modules, STEM professional role model videos, and field trips, while students in control schools accessed role model videos and field trips only. Surveys measuring math self-efficacy and STEM interest, outcome expectations, and choice goals were completed by participants in both conditions at the beginning and end of two separate program years, 2021–2022 and 2022–2023. Across both years, quantitative results suggest some positive effects of participation, particularly for STEM choice goals, but benefits depend upon student participation levels. Qualitative data offer student voice around prior experiences in math and science and the development of postsecondary plans in STEM. In combination, the results suggest that for students who do not initially identify as STEM career-bound, afterschool programming may not necessarily promote preparation for STEM careers due to an accumulation of weak math and science school experiences and other socio-environmental influences.

1. Introduction

Performance in math, particularly algebra, is a major barrier to student participation, enthusiasm, and success in STEM among minoritized students in U.S. high schools. Furthermore, the transition between middle school and high school is a liminal and tumultuous time for adolescents, and it coincides with the time during which students typically undertake algebra. For too many students, algebra is not the gateway to mathematical literacy but a gatekeeper. Algebra is foundational to formal mathematics, so supporting students in this subject specifically is a key lever to promoting rich postsecondary opportunities [1,2,3,4].
In Baltimore City Public Schools (City Schools), high school math has persistently posed challenges to students’ state assessment performance, on-time graduation, as well as access to credit-bearing courses once enrolled in college. For instance, in 2023, just 5% of City Schools 8th graders were proficient in the math state assessment. Among students taking the algebra state assessment at the end of their Algebra I course in 2023, only 6.5% were deemed proficient [5]. A study of Baltimore graduates who enrolled in college found that approximately 90% of the class of 2011 were assessed to need remedial coursework, most often in math; graduates’ SAT math score was, on average, 380 [6]. Gaps in equitable access, instruction, and resources across the district contribute to lower mathematics achievement. City Schools suffer from chronic underfunding; statewide, as the percentage of minoritized students increases, the gap between funding targets and actual funding gets worse [7]. Moreover, teacher shortages and high turnover impact math instruction. In 2018–2019, only 18% of 6th grade teachers were certified in math [8].
To address such challenges and in recognition of the critical need to support burgeoning STEM career interests among high school students in Baltimore, a diverse project team with expertise in math and engineering instruction and online curriculum design at two local universities developed the Baltimore Online Algebra for Students in Technology (BOAST) program in partnership with Baltimore City Public Schools’ Office of Math Learning. As an extended learning program, several priorities guided its development. First, the program needed to offer opportunities that reinforced fundamental algebra skills, as well as rich experiences in engineering design and exposure to specific STEM and engineering career pathways. Second, the project targeted participant recruitment at high schools that did not host engineering learning pathways (e.g., Project Lead the Way) or STEM-focused afterschool opportunities. BOAST’s development was guided by the literature on supplemental math and afterschool programming, STEM career development, and social learning theory.
The current study not only examines the efficacy of a unique afterschool STEM program but also contributes to the literature on social cognitive career theory concerning how high school students’ socio-environmental context influences self-efficacy and their propensity to enter STEM career pathways. Such research is especially lacking for racially and socioeconomically minoritized students who have not yet entered college. The preponderance of existing research about racial minorities’ experiences navigating STEM career pathways explores undergraduate student populations, so much less is known about how minoritized high school students’ emerging aspirations in STEM can be supported. While research demonstrates that under-represented students are more likely to be ill-equipped by high schools to pursue higher education in STEM subjects [9,10], this research contributes by identifying potential mechanisms by which attending under-resourced high schools leads to that outcome.

1.1. Literature Review

Afterschool programming has been shown to significantly improve student achievement, particularly among low-income, at-risk youth [11]. Concerning STEM-focused afterschool programs specifically, research demonstrates that participation is capable of both encouraging STEM interest and helping maintain those interests, which hypothetically promotes higher aspirations for STEM careers [12,13]. Students may especially benefit from afterschool programs because they are perceived as low stakes and can offer a flexible environment that enhances student learning without the same pressure and expectations as in-school learning [14].
Moreover, research has shown that offering school-day opportunities to remediate algebra skills can have a detrimental effect on higher-ability peers who would otherwise be moving on to new skills [15]. Thus, leveraging afterschool programming to extend learning time on algebra is a less academically costly approach to supporting students, and several studies show that using expanded opportunities for algebra in high school is effective for strengthening math skills [16,17,18,19]. While a strategy of extended learning time in algebra has primarily been implemented for the purpose of reducing course failures, its proven effectiveness in bolstering math course pass rates by adding increased instructional time holds promise for bolstering students with an interest in STEM but who have not mastered algebra skills. For example, students in Chicago Public Schools who received double-dose algebra achieved significantly higher algebra assessment scores relative to students with only a single dose [18]. Further, the long-run effects of double-dose algebra included a higher number of credits earned in high school, a higher probability of graduation, and a higher likelihood of college enrollment [16,17].
Ensuring students’ mastery and confidence in algebra is crucial since math proficiency and self-efficacy have been identified as key moderators of student persistence in STEM pathways [20]. Self-efficacy is belief about one’s capabilities for success on a given outcome [21,22], and self-efficacy in math specifically pertains to individuals’ confidence in their approach to performing mathematical tasks and solving math problems successfully [23]. Research suggests that math self-efficacy is predicted by successful math performance [24] and, in particular, experiences where students have the opportunity to show mastery, which are particularly powerful mechanisms to increase students’ feelings of math efficacy [25]. Notably, math self-efficacy has been shown to mediate the effects of gender and maintain independent effects on long-term achievement outcomes, as well as postsecondary matriculation [23,26,27]. Thus, strengthening students’ math self-efficacy helps ensure they are psychologically equipped to succeed in advanced math coursework and satisfy prerequisites to enter STEM pathways in college, as well as maintain STEM interests and career goals.
Social cognitive career theory (SCCT) offers a framework to explore the mechanisms involved in students forming and maintaining goals for STEM career pathways. SCCT posits that developing a career identity is a long, cumulative process, affected recursively by interests, self-efficacy, barriers, and relevant experiences [28]. Yet, more recent applications of SCCT highlight the importance of socio-environmental factors [29], such as ethnic identity and family influences among youth from lower socioeconomic (SES) backgrounds [30].
Specifically, one study found that sibling and peer support for educational goals accounted for most of the explained variance in low-income ninth graders’ vocational and educational self-efficacy, whereas perceived barriers were not significantly predictive. This suggests that social supports are more impactful to self-efficacy than challenges among this subpopulation [31]. The authors further found that educational/vocational self-efficacy was the only significant predictor of outcome expectations when tested against family and peer supports and perceived barriers [31]. Concerning low-income high school students’ academic engagement and vocational outcome expectations, a study found a positive relationship between ethnic identity and both self-efficacy and outcome expectations [32]; the research also identified a positive association between mothers’ support and academic engagement and a positive association between fathers’ support and self-efficacy. They also found self-efficacy to be strongly predictive of outcome expectations, which supports the SCCT model. Surprisingly, the same study also found a positive association between perceived barriers and self-efficacy among the low-income population [32]. These findings imply that low-income students’ self-efficacy may be sourced in ways that are different from students with more socioeconomic resources.
According to SCCT [33], opportunities to experience mastery, learn with peers from nurturing teachers, and apply math standards to practical demonstrations could promote higher self-efficacy, interest, and goal-setting in related careers. However, research on educational self-efficacy among students of color and low-income populations is still emerging, and thus, the current study will contribute to the literature in considering how a predominantly Black and relatively low-income high school population’s experiences—in the program and in school—impact STEM career development.

1.2. Program Description

BOAST, an eight-month afterschool engineering program for high school students, was offered to two cohorts in a large urban school district in school years 2021–2022 and 2022–2023. The program’s curriculum entailed a series of online modules with math/algebra (reinforcement) lessons as applied to engineering challenges. Most students participated in a hybrid format, attending weekly afterschool meetups with an in-person instructor, and students could work on modules asynchronously if needed. Instructors were also available for regularly scheduled virtual office hours and communicated with students via email. Field trips to the sponsoring university’s campus and high-quality videos of interviews with diverse STEM professionals were also key program components.
Six modules covering a range of relevant and engaging engineering topics were developed. Each module was divided into four sections, including an introductory session (i.e., an icebreaker activity and time to view a video of a professional STEM role model sharing about their career trajectory); a ‘play’ session, in which students could experiment with materials; a ‘learn’ session, where students reviewed and practiced relevant algebra standards; and finally a ‘build’ component, where students built a design using the algebra skills reinforced during the learning component. The engineering topics covered included a general introduction to BOAST, technical rescue, machine learning, soundproofing, business optimization, and urban heat islands.
In-person instructors were hired by the primary sponsoring university and chosen based on their proven ability to work with minoritized students and their general demeanor as caring, nurturing teachers. Ten role model videos 5–10 min in length were created that featured predominantly minoritized professionals describing their work in a range of STEM careers, including electrical, optical, computer, cyber security, mechanical, systems, and civil engineering. Professionals shared stories about how their career interests developed, the challenges they encountered, and their experiences in high school and college, including ways that mentors, friends, and family had helped them persist in the face of social or educational obstacles.

1.3. Program Recruitment

Students were recruited to participate in the BOAST program during August–September before the beginning of the school year in several ways. First, information about the program and how to apply was disseminated on a university website using brochures, small posters, and a local radio station announcement. Materials were shared by university staff with school leaders, with the request that they share it with teachers, parents, and students. University staff also went to the high schools during lunch periods to recruit applicant students directly. Interested students, whether self-nominating or encouraged by another, indicated their interest in participation by completing an online application. The only requirements for participation were having already completed Algebra I with a final report card grade of C- or better and being enrolled in a high school that had agreed to host the program.

2. Materials and Methods

The theory of change for the current study was adapted from Lent et al. [33] and based on SCCT. Our model hypothesizes that the BOAST program components impact career goals through students’ math self-efficacy, STEM interests, and their interdependence with socio-environmental factors. This model is illustrated in Figure 1.

2.1. Data

To measure exposure to algebra for engineering modules, data were collected passively from students from the online learning management system (LMS). These data detail the number of hours students spent in BOAST modules throughout the year. In addition, the LMS data captured the number of role model videos students viewed. Field trip participation data were captured separately. Students’ math self-efficacy was measured both at the beginning (pre) and end of the program year (post) using a previously validated self-administered online survey [34]. Data on students’ interest, outcome expectations, and choice goals in STEM were also collected from participants pre- and post-program year using an instrument that prior research had suggested was valid [35].
Finally, qualitative one-on-one interview data from 14 students were also collected over the course of both program years using a semi-structured protocol to capture the influence of socio-environmental factors. Students were asked about their reasoning for electing to participate in BOAST; barriers to participation (e.g., scheduling challenges, technology issues that prevented module completion); reactions to the role model videos and field trips; their experiences in math and science classes; their post-high school plans, including career goals and potential obstacles to reaching them; and sources of social support (i.e., family members and friends) for preparing for college and/or career.

2.2. Research Design

This study employed a convergent parallel mixed methods research design, where schools were assigned to either treatment or control conditions. As a first step, 28 eligible high schools in the district, excluding alternative schools for students with intense academic needs, were paired a priori based on school characteristics, including whether the school had academic entrance criteria, school enrollment size, percent Black, percent Hispanic, percent English learner, graduation rate, and percent of students proficient on the Algebra I state assessment the prior year. This resulted in 14 high school groupings in which each pair was as similar as possible in terms of key factors. The principal of each school per pair was approached by a member of the BOAST program team to gauge interest in hosting the BOAST program, either as a treatment or a control school. Principals of schools in the treatment condition were informed that their eligible student applicants would receive the algebra for engineering afterschool lesson meetups along with all other program components (i.e., STEM role model videos and field trips). Conversely, students in schools under the control condition would have access to STEM role model videos and field trips only. Quantitative and qualitative data were collected simultaneously during both program years, and the data collected depended upon the school’s experimental condition. Specifically, LMS module participation, pre-/post-survey, and student interview data were collected from students in treatment schools, while pre-/post-survey and role model video views were only collected from students in control schools. Field trip participation data were collected for students in both conditions.
Although random assignment of schools to an experimental condition would offer the most robust method to identify causal effects of program participation, randomization was infeasible. School leaders expressed strong concerns about encouraging their students to apply for the BOAST program without knowing ahead of time what participation would mean for students in terms of time and effort commitment; therefore, randomization of schools with applicants was decided to be ethically questionable. As a result, analysis was performed to create like-school pairs to approximate random assignment to reduce the potential for unobserved variable bias. Of particular concern were differences between treatment and control groups in student interest in STEM; thus, the opportunity was similarly advertised at schools in both conditions as a set of STEM-focused activities occurring outside regular school hours. It is unknown the extent to which student applicants were aware of their school’s experimental condition.

2.3. Analysis

Quantitative analysis entailed statistical bivariate comparisons and multivariate regression estimating changes in math self-efficacy, STEM career interests, outcome expectations, and choice goals based on BOAST participation and other student characteristics. Regressions were estimated in Stata v. 18 with robust standard errors to account for student clustering within schools.
Qualitative data were analyzed using a deductive approach. In particular, the researcher read transcripts several times and then began coding with a set of expected categories based on the theoretical framework of SCCT. These initial codes represented topics including ‘barriers to BOAST participation’, ‘educational and career goals’, ‘interest in STEM’, ‘prior experiences with math’, ‘prior experiences with science’, ‘plans after high school’, and ‘social supports’. Secondarily, the researcher further coded the data within each category using an inductive approach to capture greater nuance. For instance, within social supports, codes were created to identify sources of ‘peer support’ and ‘family support’ for particular domains such as college planning and career planning. The findings are organized according to the themes identified from the secondary set of codes.

3. Results

3.1. Sample Descriptives

Data were collected during two program years (2021–2022 and 2022–2023) for 89 students in ninth through eleventh grade, of whom 60 were in treatment and 27 in control schools (refer to Table 1). Approximately 81% of the sample identified as Black and 9% as Hispanic White. The remainder were either non-Hispanic White or Asian. Most participants were either high school sophomores (35%) or juniors (35%), and the majority identified as female (70%). Among students in treatment schools, 10% received special education services, and 2% received EL services, whereas no students in the control group received special education, and slightly more (4%) were EL. Final Algebra I report card grades were slightly higher for treatment than control students (3.07 vs. 2.89).
Baseline comparability between students in treatment and control schools was tested (refer to Table 1). Statistically significant differences between treatment and control groups were evident only for students’ prior year’s school day attendance. The average daily attendance rate of students in the treatment group was 9 points higher than those in the control group (92.3 vs. 83.4). Similarly, students in the control group had much higher levels of chronic absenteeism (i.e., being absent 10% or more of total days on roll) than those in the treatment group (55.6% vs. 24.6%). As a context, chronic absence levels among all high school students in the district have been troublingly high in recent years. In 2022–2023, approximately 46% missed 20 days of school or more [5].
Regarding baseline comparability on the SCCT constructs (refer to Table 2), the mean for each scale collected during the pre-year survey administration was comparable between students in treatment and control groups. Though students in the treatment group had slightly higher responses for STEM interests (4.12 vs. 3.97), this difference was not statistically significant, nor were any other comparisons on these baseline measures.
Data on the three forms of BOAST participation are presented in Table 3. On average, participants in the treatment group spent 3.77 h working on modules within the LMS. This measure does not capture other ways that students could interact with the BOAST material, which may not have been logged during in-person afterschool meetup sessions where the instructor could have used the whiteboard, material demonstrations, and peers may have worked together in real time. In additional analyses, it was found that the length of total time students spent inside the LMS learning modules during the year varied considerably, with a median time of 0.50 h and the 90% percentile of 9.02 h. Because of the dedicated time in each module to view the videos of STEM professionals, time spent watching them was slightly higher for students in the treatment group (0.46 vs. 0.39 h), though the difference is not statistically significant. Field trip attendance was significantly higher among students in the treatment group (1.6 vs. 0.95 trips).
It is also notable that school day attendance patterns during the program year differed substantially between treatment and control groups; these differences were also statistically significant. Students in treatment schools had higher average daily attendance rates than in control schools (90.3 vs. 81.6) and lower levels of chronic absence (34.4% vs. 66.7%). However, further analyses did not suggest an association between daily attendance and LMS participation for students in the treatment group (r = 0.19, p = 0.08).

3.2. Quantitative Findings

At the beginning and end of the program year, students in both the treatment and control groups were asked to complete a 15 min survey that measured their math self-efficacy, STEM interest, outcome expectations, and goal choices. All survey constructs at each time point were determined to be reliable using Cronbach’s alpha analysis (refer to Appendix A). As a first step, relationships between the pre-survey constructs were tested based on the hypothesized conceptual model, where math self-efficacy is thought to positively impact STEM interest, which, in turn, impacts STEM outcome expectations and choice goals.
Standardized estimates from bivariate regression analyses of pre-survey responses (Table 4) show that math self-efficacy is unassociated with STEM interest, outcome expectations, and choice goals. However, STEM interest and outcome expectations were strongly related (β = 0.50, p = 0.001). STEM choice goals were also strongly associated with both STEM interest (β = 0.65, p < 0.001) and outcome expectations (β = 0.60, p < 0.001).
Next, we examined the extent to which participation in the BOAST program was associated with a pre- to post-change in the SCCT constructs. A summary of regression analyses predicting change scores for each construct is provided in Table 5 (the full set of estimates is provided in Appendix B). Results shown for Model 1 represent the estimate for the number of hours spent on BOAST material in the LMS. Though small, positive effects on math self-efficacy, STEM interest, and STEM outcome expectations were statistically significant.
Models 2 and 3 (Table 5) present estimates for the indicator of experimental group membership; we find that whether controlling for levels of exposure to the BOAST components, demographic characteristics, Algebra I grades, or school attendance rate, there were no significant differences between treatment and control groups in change in math self-efficacy, STEM interest, or STEM outcome expectations. However, students in the treatment group had STEM choice goals that increased, on average, by 0.59 points relative to the control group (p = 0.01), but only when accounting for demographic and academic characteristics. This finding is unexpected, given the null effects of LMS hours for STEM choice goals.

3.3. Qualitative Findings

Interviews with students in the treatment group (pseudonyms protect their confidentiality) were collected to gain insights into how students were thinking about their educational and career plans, their perceptions of what might be required to reach them, and what they believed might help or hinder them during that process.

3.3.1. Developing Postsecondary Interests

Most students interviewed described some level of interest in a STEM-related career. However, participant responses ranged considerably between firm, bourgeoning, and vague plans. For example, Renee, who appeared to be relatively firm in her plans, said, “[I want to] get my bachelor’s degree like my sister has before … I want to major in engineering. It’s engineering. I don’t know what type exactly yet. But it’s engineering”. Another student, Nivea, shared her plan along with a specific college: “Johns Hopkins is actually one of my first choices … I would say software engineering, or computer science is my top two [choice of majors].”
In contrast, some students seemed to have doubts about their emergent plans. A male sophomore, Jean-Paul, shared:
I would like to go to college to get my engineering degree or computer engineering degree because that’s what I wanted to do. Right now I’m just having second thoughts … When I first came [to this country], my plan was just computer engineering … But right now, I’m just having second thoughts about how everything’s working out and how to get to everything. I’m just having doubts about myself being good enough to do the job that I’m supposed to be doing … I may have good grades, but I don’t think rationally like everybody else. I don’t think clear. I mostly learn things from just watching.
Similarly, Vivian expressed her plans in vague terms but expressed confidence about her interests:
I know I want to enroll to a four-year college … I do want to do something in STEM. I don’t really like politics or history. I love science, I love math. I think more, I love engineering, but I don’t really know if I would really want to major in that. I really like astronomy and physics. I could spend hours reading, and from abstract papers, research papers, Stephen Hawking, you know, all of those. I really love learning about that.
Two other students—Marla, who responded to how she was preparing for college, and Darwin, responding to what his career at age 30 might be—appeared to have what might be considered unrealistic goals. Their responses reflect ideas that may have formed without appropriate consideration of timing before graduation or prior experiences that would shed light on what a STEM professional’s workday entails:
Marla: I want to play college sports … I was thinking for my senior year in high school, I would try to do some sports to see which ones I’m really interested in and try to get a scholarship from them.
Darwin: The plan is to join the Navy, and after joining the Navy to finish my degree or start my degree in a business course and also in marine biology. … I see myself maybe seven in the morning getting up, going and boat out to one of our research places, doing a little research, diving to check on some of our sharks that we have in the research lab. And then coming home, stopping by my house, showering up and heading out to prepare for the nightclub that I will own in the future.
While Marla and Darwin’s ambitions are admirable, Marla’s expectation of gaining an athletic scholarship without years of training and Darwin’s plan to be a marine biologist and own a nightclub by age 30 highlights a need for practitioners in the career preparation space to offer students more information about the requirements for both postsecondary access and particular careers.
Indeed, participants’ responses to questions about their plans corresponded with the experiences they shared about how school, family, and peers might be supporting (or not supporting) their educational goals and career interests. The ways that each of these factors potentially influenced students’ goals are illustrated next.

3.3.2. Social Supports

Renee and Nivea, who had very clear plans, attended a selective high school that offered pathways of study in STEM. Describing opportunities she had that are helping her prepare, Renee explained, “I’ve done multiple internships and I have one right now at John [sic] Hopkins where I studied greenhouse gas emissions and I’m comparing them to Baltimore to see what Baltimore could do better to lower their greenhouse gas emissions”. When asked why she applied to the BOAST program, Nivea stated:
Nivea: It was actually one of my teachers recommended me to join. It was my engineering teacher. And also I’m just really interested in engineering. And I heard that BOAST had algebra and engineering combined. So that’s also why I wanted to do it, and I thought that it would help me with my algebra, which it actually does.
Nivea also shared ways her school was connecting her with experiences that were connected to her interests and goals:
Nivea: We’ve actually done some things in school where we actually choose our major or have some ideas for what we want to do when we’re older. … I’ve been trying to just try to join as many programs as I can, especially STEM related programs. … I’m in engineering and this is my third year in engineering. I’m also in BOAST, obviously. I’m trying to do [an engineering internship] in the summer, which is a Johns Hopkins program. So that’s probably what might help me decide what I want to do.
Other participants attended schools where such concerted efforts did not appear to occur. For instance, Marla, who was waiting until senior year to think about sports participation that would lead to a scholarship, was asked if she had any conversations with adults in her school about postsecondary options. She responded, “Nope. Not that I know of”. Darwin, whose intentions to be both a marine biologist and a club owner, described a situation that appeared to reflect haphazard support:
Darwin: I’ve not really heard much about STEM, so I never really thought of entering any field in STEM … I currently don’t have any science classes or engineering classes… I’ve never took engineering classes and right now, they just put me on some courses to get my schedule filled. I haven’t really had much say in it.
Some participants, particularly those with uncertain or nascent ideas about a desired career, also shared frustrating educational experiences. For example, Maliah described an interest in nursing, saying, “My goal is to become a traveling nurse. … But I always wanted to travel, so it’s like me being a traveling nurse, I can go experience stuff and still do what I want”. Yet, Maliah also described recent disappointments in her math class:
Maliah: I used to enjoy math, but once COVID started, I guess, me being at home for three years … I guess that chunk away from school really knocked me off of really being a good student in math … When I went back in person this year, it was a big struggle for me to get adjusted back to a math class, actually any of my classes. And once we got to school, we didn’t really have teachers. We still don’t. And then our math teacher is actually gone … So I’m just in a math class sitting there every day … We have subs there, but they’re different every day and they just keep telling us to get on something called Imagine Math but that really doesn’t help me. And because I don’t have a teacher, I can’t ask questions the way I want to. I’m not really feeling math anymore.
This vignette from Maliah emphasizes how negative school experiences, not to mention COVID-related school closures, can potentially derail interest in particular subjects. To succeed in pursuing a nursing career, this student will need to maintain subject matter self-efficacy, and this example suggests that self-efficacy and interest in math (i.e., “feeling” it) are interdependent.
Concerning how family members informed students’ interests and goals, it is notable that Renee couched her goal of obtaining a bachelor’s degree as being “like my sister has before”. Other participants also relayed how their families and others encouraged them. Jean-Paul, who expressed a goal of becoming a computer engineer but with emerging doubts, said that his brother was supportive. Yet, the lack of specific ways that his brother would help mirrored the hesitancy of his goals:
Jean-Paul: I told my brother about what I wanted to pursue. He just said he would help me get it and all that. He would just be there to help me get it and he would just get me the program that I needed and then get me the things I need to get it.
Jean-Paul also shared that his friends had similar interests and goals, but his emphasis on “trying to do online things” and lack of resources potentially resonate with his prior expression of inadequacies to achieve his goal. He shared, “Most of [my friends] want to be an engineer as well. I talk to them about engineering. We go over some things about it. We try to do online things. We try to do NFTs and stuff, but we just don’t have the resources for that”.
Two male participants, Richie and Jonas, could be categorized as having nascent plans. For instance, Richie’s goal was to be a lawyer or investment banker. When Richie was asked how he and his friends talked about plans after high school, he stated, “I just don’t talk to them about the future … Most them not trying to go to college anyway”. Concerning his mother’s support for his plans, he shared, “My mom’s kind of really religious, and she says all lawyers are heartless, soulless. They sell their soul. And she was freaked out when I said I wanted be lawyer so switched over to investment banking and I learned about it, and it’s okay”. Richie’s explanation of his change in career plans points to the key influence of parents; Richie also described his goals as a contrast to those of his peers, as though they need protection from negative influences.
In describing his postsecondary planning, Jonas, attending the same school as Richie, stated, “I would really love to go to college. Same time, I would love to start doing gigs and promoting what I do after school”. Following up, he explained that outside of school, he did video editing. When asked how he felt about taking math coursework in college, he shared, “Math is just like a brain stimulant. I love it”. But later, when responding to a question about his experiences in high school math class, he said, “Math classes are usually very boring. And it’s as if I’m being stilled with the learning … I usually do really well in math, but when I’m feeling like the class is really slow, my assignments or the grades would go down”. Jonas’ comment indicates a strong connection between competency and interest, and his impressions may reflect varying experiences of engaging in math instruction. Sentiments such as his suggest that self-efficacy and bourgeoning interests in STEM subjects may be precarious for minoritized and low-income students. When the researcher asked Jonas if he’d had the opportunity to talk to anyone at the school about what he might do after high school, he replied:
Jonas: No. They said that we could reach out to them. They’ve talked to students as a whole group … about things and plans we can do after high school and then college. But the message altogether, I don’t feel like it’s hitting 80% of the classes, especially since nobody really cares. They [students] don’t really care for their futures. They’re okay with being on the streets, selling drugs and all that. Or just working at McDonald’s. They don’t really care for future problems or any of that.
Jonas’ response illustrates how a lack of deliberate postsecondary guidance, in combination with negative peer influences and unengaging classes, may be especially detrimental to productive career ideation among this population of high school students.

3.4. Mixed Methods Findings

In combination, the quantitative and qualitative results offer unique insights into the development of self-efficacy, interest, and career goals. Although not all participants expressed interest in STEM necessarily, their data remain relevant to understanding the complex interplay of socio-environmental factors and the other constructs featured in the SCCT model among minoritized, low-income high school students. Figure 2 offers a joint display of findings from both strands in terms of how social supports and barriers relate to the outcomes of interest.
Among students with nascent or uncertain postsecondary goals, whether they concern STEM or non-STEM careers, the data suggest that an accumulation of disappointing or haphazard educational experiences and other negative social influences had left some students with a weak understanding of the concrete steps needed to achieve their individual goals. Their schools did not appear to be organized in deliberate ways around ensuring each student had clear postsecondary interests and plans and that the curriculum or set of experiences on offer was perceived by students as outside of their control and/or less than optimal. Conversely, students with clearer plans had benefited from a range of beneficial enrichment opportunities and more engaging coursework in STEM, and the adults in their school and at home were reportedly helping them achieve their goals. Notably, further quantitative analysis found enrollment in a particular school explained approximately 20% of the variance in STEM interest, 23% in STEM outcome expectations, and 33% of math self-efficacy at baseline.
These results also shed light on some of the surprising quantitative findings regarding the impacts of BOAST. While participation in the BOAST program had no significant effect on math self-efficacy, STEM interest, or STEM outcome expectations, students accessing all components of the program had significantly higher levels of STEM choice goals. This is consistent with SCCT research in demographically similar samples. For example, one study used SCCT to predict the math/science goal intentions of low-income prospective first-generation college students (n = 305); contrary to hypotheses, the relationship between barriers and goals was mediated neither by self-efficacy nor interests [36]. We surmise that BOAST participation, in either the treatment or control condition, helps maintain baseline math self-efficacy and STEM interests. However, concerning the significant boost in STEM choice goals among the BOAST treatment group, we argue that through BOAST, participants gained an increased desire to learn new STEM-related skills and take steps to pursue STEM careers.

4. Discussion

The current research investigated the extent to which a STEM-focused afterschool program was associated with improved math self-efficacy, STEM interests, outcome expectations, and choice goals. Through the lens of the SCCT framework among a low-income, predominantly Black high school student population, we found that socio-environmental factors are particularly salient to how students’ postsecondary plans develop or fail to develop. Our findings also reinforce existing studies employing SCCT, particularly those employing samples of minoritized, low-income populations. While it was surprising that math self-efficacy appears to be unrelated to other SCCT constructs, Garriott et al. [36] also found no effects of self-efficacy on choice goals. However, our findings confirm previously identified associations in more heterogeneous samples regarding relationships between interest and outcome expectations and choice goals, as well as between choice goals and outcome expectations [30,33,36].
Members of our sample who benefited from a greater number of STEM-focused opportunities in school, positive family role models, and concerted coordination of interest-based opportunities had clearer, firmer plans for what they would do after graduation. On the other hand, our sample included many students who did not benefit from well-resourced schools and were subject to haphazard curriculum, understaffing, and questionable peer influences. These students expressed far less certainty about their goals or had stated goals that appeared to be less well-informed.
The results of this study offer promising evidence that BOAST has beneficial impacts on some of the psychological constructs associated with productive career planning, though the effects were weaker than expected. Yet, like other school-embedded programs, student engagement with BOAST is influenced by many external factors. Of special importance was the high degree of variation in students’ completion of the algebra for engineering modules and STEM role model videos. This variation is a reflection of the many competing opportunities that this population of students confronts. It may be that an afterschool program, in comparison to the larger set of academic, social, and environmental influences to which students are subject, is too negligible to produce strong effects.

4.1. Limitations

This study is limited in two primary ways, the first of which is its relatively small sample size. Small samples naturally yield higher variation, which can mask effects that may be evident among larger samples. However, though small, our sample is comparable in demographic characteristics and school representation to the larger Baltimore City Schools district. A larger sample would be amenable to more sophisticated statistical techniques, but the critical need for qualitative insights from high school students themselves requires a smaller, more purposeful sample of participants. Second, the current study is limited by our inability to randomly assign students to experimental conditions. Random assignments within schools would risk contamination effects and potentially constrain our ability to qualitatively account for school-wide factors that impact individual students. Our inability to randomly assign schools to experimental conditions is a related limitation and is evidenced by the pre-existing differences between groups in school day attendance the years before and during BOAST program receipt. It should thus be noted this reflects a strong mediating effect of school engagement on STEM choice goal growth, which may help explain why BOAST’s effects on STEM choice goals were significant only after controlling for academic characteristics.

4.2. Implications

This investigation set out to explore whether BOAST, an afterschool algebra for engineering program, increased students’ self-efficacy in math, interest, outcome expectations, and choice goals in STEM. Through the data collected, this study’s purpose expanded to also identify ways that high school experiences support students’ math self-efficacy and developmental plans for postsecondary education and careers. Of note is that all BOAST participants attended an under-resourced school, yet some benefited from schools that were organized in strategic ways that emphasized postsecondary planning and marshaled resources to encourage students’ specific interests. For high school practitioners, these findings suggest that ensuring students successfully graduate on time is inadequate when it means that students are not equipped with the information needed to navigate life after high school. Graduating “career and college ready” requires not only completing coursework, even if students’ courses are on par with the rigor of college; it also requires that adults in schools talk to students about their interests so that they can share related postsecondary pathway information, including possible careers and what credentials are required for entry. Further, students need to understand what will be necessary for any destination and when, i.e., how to be competitive for particular college admission systems and appropriate timing to reach necessary milestones. In addition, families and caretakers must be engaged by schools as partners along the way so that students have the necessary social support at home to pursue their desired goals.
Many high schools prioritize efforts toward meeting graduation requirements, which is common when serving student populations who disproportionately arrive in high school with many academic needs; however, practitioners can help students meet graduation requirements as well as strategically plan for postsecondary possibilities by using asset-based approaches, which would recognize and leverage each student’s unique interests as a resource to encourage school attendance and engagement [37]. Such interests should also be used to guide the selection of programs offered. Each student’s interest profile should be updated at regular intervals upon meeting again to learn how their interests or values may have evolved.
To meet these goals, adults in school should be organized in ways that facilitate trusting student relationships and purposeful conversations (e.g., small cohort models, leveraging community-based partnerships), which will serve the purpose of ensuring that students feel agency over their education experiences. Although few high schools are resourced to provide every relevant experience, all should at least be able to connect students with enriching opportunities in the local community. Successful planning among high school students in under-resourced contexts is only possible if individual student interests are captured and adults honor interests by helping co-develop a plan.

Author Contributions

Conceptualization, R.E.D., M.L.F., A.G.D., A.S., A.R., M.W. and E.J.Y.; methodology, R.E.D. and A.G.D.; formal analysis, R.E.D.; investigation, R.E.D. and A.G.D.; resources, M.L.F., A.S., A.R., M.W. and E.J.Y.; writing—original draft preparation, R.E.D. and A.G.D.; writing—review and editing, R.E.D., M.L.F., A.G.D., A.S., A.R., M.W. and E.J.Y.; visualization, R.E.D.; project administration, M.L.F., A.G.D. and A.S.; funding acquisition, R.E.D. and M.L.F. All authors have read and agreed to the published version of the manuscript.

Funding

This material is based upon work supported by the National Science Foundation under Grant No. DRL-2005790.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Boards of Baltimore City Schools (2021-045, 6 June 2021), Johns Hopkins University (HIRB00012557, 19 April 2021) and Notre Dame of Maryland University (IRB00010753, 14 February 2021).

Informed Consent Statement

Informed consent and/or child assent were obtained from all subjects involved in the collection of data for this study.

Data Availability Statement

The administrative data used in this study are not readily available due to student privacy regulations per FERPA and the data sharing agreement governing the authors’ use of Baltimore City Schools data. Student interview data are unavailable because of the nature of the project’s parental consent agreement.

Acknowledgments

The authors are grateful for the assistance and of district partners Levi Straight, Liepa Boberiene, Chris Wohn and other staff in Baltimore City Schools Office of Achievement and Accountability who offered comments on earlier drafts, as well as other members of the BOAST project team including Jaracus Copes, Cynthia Glass, Nantambu Kohlbatz, Tyair Manning, and Karen Watkins-Lewis.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A

Table A1. Cronbach alpha reliability tests for SCCT constructs pre- and post-survey.
Table A1. Cronbach alpha reliability tests for SCCT constructs pre- and post-survey.
Construct (N Items)Pre-SurveyPost-Survey
Math self-efficacy (24)0.9180.941
STEM interest (6)0.7110.730
STEM outcome expectations (6)0.6480.757
STEM choice goals (6)0.7860.779

Appendix B

Table A2. Regression of math self-efficacy change on BOAST participation levels and student characteristics.
Table A2. Regression of math self-efficacy change on BOAST participation levels and student characteristics.
Model 1Model 2Model 3
Intercept0.04 (0.24)−0.01 (0.16)−0.48 (1.21)
BOAST participation
 Treatment group (vs. control) −0.19 (0.12)−0.11 (0.16)
 Total LMS hours0.02 (0.01) *
 Role model video hours0.31 (0.21)
 Numb. field trips−0.19 (0.10)
Demographics
 Male −0.65 (0.12) **−0.58 (0.07) **
 Black a 0.02 (0.20)−0.06 (0.29)
 Hispanic a 1.23 (0.26) **1.24 (0.17) **
Academics
 Algebra I final grade −0.16 (0.15)
 School day attendance rate 1.04 (1.83)
R20.160.590.71
Note. Coefficients and robust SEs presented. a Reference group is White or Asian. ** p < 0.01; * p < 0.05.
Table A3. Regression of STEM interest change on BOAST participation levels and student characteristics.
Table A3. Regression of STEM interest change on BOAST participation levels and student characteristics.
Model 1Model 2Model 3
Intercept0.19 (0.18)−0.16 (0.12)−1.23 (1.41)
BOAST participation
 Treatment group (vs. control) 0.08 (0.16)0.29 (0.19)
 Total LMS hours0.03 (0.01) **
 Role model video hours−0.26 (0.07) **
 Numb. field trips−0.10 (0.08)
Demographics
 Male −0.38 (0.39)−0.70 (0.52)
 Black a 0.07 (0.04)0.12 (0.11)
 Hispanic a 0.39 (0.52)0.56 (0.61)
Academics
 Algebra I final grade −0.03 (0.10)
 School day attendance rate −1.11 (1.54)
R20.480.170.41
Note. Coefficients and robust SEs presented. a Reference group is White or Asian. ** p < 0.01.
Table A4. Regression of STEM outcome expectation change on BOAST participation levels and student characteristics.
Table A4. Regression of STEM outcome expectation change on BOAST participation levels and student characteristics.
Model 1Model 2Model 3
Intercept0.20 (0.66)−0.26 (0.61)2.55 (3.07)
BOAST participation
 Treatment group (vs. control) 0.30 (0.51)0.46 (0.74)
 Total LMS hours0.03 (0.01) *
 Role model video hours0.07 (0.52)
 Numb. field trips−0.16 (0.31)
Demographics
 Male 0.49 (0.42)0.19 (0.53)
 Black a −0.01 (0.44)−0.31 (0.87)
 Hispanic a 0.05 (0.29)−0.09 (0.54)
Academics
 Algebra I final grade −0.13 (0.32)
 School day attendance rate −2.54 (3.82)
R20.070.120.19
Note. Coefficients and robust SEs presented. a Reference group is White or Asian. * p < 0.05.
Table A5. Regression of STEM choice goal change on BOAST participation levels and student characteristics.
Table A5. Regression of STEM choice goal change on BOAST participation levels and student characteristics.
Model 1Model 2Model 3
Intercept−0.14 (0.26)−0.28 (0.30)0.52 (1.17)
BOAST participation
 Treatment group (vs. control) 0.18 (.22)0.55 (0.10) *
 Total LMS hours0.01 (0.01)
 Role model video hours−0.20 (0.20)
 Numb. field trips0.16 (0.15)
Demographics
 Male −0.35 (0.20)−0.67 (0.11) **
 Black a 0.25 (0.28)0.49 (0.14) *
 Hispanic a 0.69 (0.32)0.88 (0.15) **
Academics
 Algebra I final grade −0.09 (0.05)
 School day attendance rate −0.77 (1.36)
R20.140.190.71
Note. Coefficients and robust SEs presented. a Reference group is White or Asian. ** p < 0.01; * p < 0.05.

References

  1. Maltese, A.V.; Tai, R.H. Eyeballs in the Fridge: Sources of Early Interest in Science. Int. J. Sci. Educ. 2010, 32, 669–685. [Google Scholar] [CrossRef]
  2. Nicholls, G.M.; Wolfe, H.; Besterfield-Sacre, M.; Shuman, L.J. Predicting STEM Degree Outcomes Based on Eighth Grade Data and Standard Test Scores. J. Eng. Educ. 2010, 99, 209–223. [Google Scholar] [CrossRef]
  3. Tai, R.H.; Sadler, P.M.; Mintzes, J.J. Factors Influencing College Science Success. J. Coll. Sci. Teach. 2006, 36, 52–56. [Google Scholar]
  4. Tyson, W.; Lee, R.; Borman, K.M.; Hanson, M.A. Science, Technology, Engineering, and Mathematics (STEM) Pathways: High School Science and Math Coursework and Postsecondary Degree Attainment. J. Educ. Stud. Placed Risk JESPAR 2007, 12, 243–270. [Google Scholar] [CrossRef]
  5. Maryland State Department of Education Baltimore City Public Schools. State Assessment Results; Maryland State Department of Education Baltimore City Public Schools: Baltimore, MD, USA, 2024. [Google Scholar]
  6. Connolly, F.; Olson, L.S.; Durham, R.E.; Plank, S.B. Indicators of College Readiness: A Comparison of High School and College Measures; Baltimore Education Research Consortium: Baltimore, MD, USA, 2014. [Google Scholar]
  7. Maryland Department of Legislative Services Blueprint for Maryland’s Future. Final Report; Maryland Department of Legislative Services Blueprint for Maryland’s Future: Baltimore, MD, USA, 2020. [Google Scholar]
  8. Baltimore City Public Schools STEM Plan: 2019–2021 [Unpublished Internal Document]; Baltimore City Public Schools: Baltimore, MD, USA, 2019.
  9. Bound, J.; Lovenheim, M.F.; Turner, S. Why Have College Completion Rates Declined? An Analysis of Changing Student Preparation and Collegiate Resources. Am. Econ. J. Appl. Econ. 2010, 2, 129–157. [Google Scholar] [CrossRef] [PubMed]
  10. Jennings, J.L.; Deming, D.; Jencks, C.; Lopuch, M.; Schueler, B.E. Do Differences in School Quality Matter More Than We Thought? New Evidence on Educational Opportunity in the Twenty-First Century. Sociol. Educ. 2015, 88, 56–82. [Google Scholar] [CrossRef]
  11. Lauer, P.A.; Akiba, M.; Wilkerson, S.B.; Apthorp, H.S.; Snow, D.; Martin-Glenn, M.L. Out-of-School-Time Programs: A Meta-Analysis of Effects for At-Risk Students. Rev. Educ. Res. 2006, 76, 275–313. [Google Scholar] [CrossRef]
  12. Dabney, K.P.; Tai, R.H.; Almarode, J.T.; Miller-Friedmann, J.L.; Sonnert, G.; Sadler, P.M.; Hazari, Z. Out-of-School Time Science Activities and Their Association with Career Interest in STEM. Int. J. Sci. Educ. Part B 2012, 2, 63–79. [Google Scholar] [CrossRef]
  13. Young, J.R.; Ortiz, N.; Young, J.L. STEMulating Interest: A Meta-Analysis of the Effects of Out-of-School Time on Student STEM Interest. Int. J. Educ. Math. Sci. Technol. 2016, 5, 62. [Google Scholar] [CrossRef]
  14. Davis, K.; Fullerton, S. Connected Learning in and after School: Exploring Technology’s Role in the Learning Experiences of Diverse High School Students. Inf. Soc. 2016, 32, 98–116. [Google Scholar] [CrossRef]
  15. Nomi, T. The Unintended Consequences of an Algebra-for-All Policy on High-Skill Students: Effects on Instructional Organization and Students’ Academic Outcomes. Educ. Eval. Policy Anal. 2012, 34, 489–505. [Google Scholar] [CrossRef]
  16. Cortes, K.E.; Goodman, J.S. Ability-Tracking, Instructional Time, and Better Pedagogy: The Effect of Double-Dose Algebra on Student Achievement. Am. Econ. Rev. 2014, 104, 400–405. [Google Scholar] [CrossRef]
  17. Cortes, K.E.; Goodman, J.S.; Nomi, T. Intensive Math Instruction and Educational Attainment: Long-Run Impacts of Double-Dose Algebra. J. Hum. Resour. 2015, 50, 108–158. [Google Scholar] [CrossRef]
  18. Nomi, T.; Allensworth, E. “Double-Dose” Algebra as an Alternative Strategy to Remediation: Effects on Students’ Academic Outcomes. J. Res. Educ. Eff. 2009, 2, 111–148. [Google Scholar] [CrossRef]
  19. Stein, M.K.; Kaufman, J.H.; Sherman, M.; Hillen, A.F. Algebra: A Challenge at the Crossroads of Policy and Practice. Rev. Educ. Res. 2011, 81, 453–492. [Google Scholar] [CrossRef]
  20. Mau, W. Factors That Influence Persistence in Science and Engineering Career Aspirations. Career Dev. Q. 2003, 51, 234–243. [Google Scholar] [CrossRef]
  21. Bandura, A. Social Foundations of Thought and Action: A Social Cognitive Theory; Prentiss Hall: Englewood Cliffs, NJ, USA, 1986. [Google Scholar]
  22. Bandura, A. Self-Efficacy: The Exercise of Control; Freeman: New York, NY, USA, 1997. [Google Scholar]
  23. Pajares, F.; Miller, M.D. Mathematics Self-Efficacy and Mathematics Performances: The Need for Specificity of Assessment. J. Couns. Psychol. 1995, 42, 190–198. [Google Scholar] [CrossRef]
  24. Lopez, F.G.; Lent, R.W. Sources of Mathematics Self-Efficacy in High School Students. Career Dev. Q. 1992, 41, 3–12. [Google Scholar] [CrossRef]
  25. Burnham, J.R. A Case Study of Mathematics Self-Efficacy in a Freshman Engineering Mathematics Course; Washington State University: Pullman, WA, USA, 2011. [Google Scholar]
  26. Dorsen, J.; Carlson, B.; Goodyear, L. Connecting Informal STEM Experiences to Career Choices: Identifying the Pathway; ITEST Learning Resource Center: Waltham, MA, USA, 2006. [Google Scholar]
  27. Parker, P.D.; Marsh, H.W.; Ciarrochi, J.; Marshall, S.; Abduljabbar, A.S. Juxtaposing Math Self-Efficacy and Self-Concept as Predictors of Long-Term Achievement Outcomes. Educ. Psychol. 2014, 34, 29–48. [Google Scholar] [CrossRef]
  28. Lent, R.W.; Brown, S.D.; Hackett, G. Career Development Theories Anchored in Learning Theory. In Career Choice and Development; Jossey-Bass: San Francisco, CA, USA, 2002; pp. 255–311. [Google Scholar]
  29. Lent, R.W.; Brown, S.D. Social Cognitive Model of Career Self-Management: Toward a Unifying View of Adaptive Career Behavior across the Life Span. J. Couns. Psychol. 2013, 60, 557–568. [Google Scholar] [CrossRef]
  30. Turner, S.L.; Joeng, J.R.; Sims, M.D.; Dade, S.N.; Reid, M.F. SES, Gender, and STEM Career Interests, Goals, and Actions: A Test of SCCT. J. Career Assess. 2019, 27, 134–150. [Google Scholar] [CrossRef]
  31. Ali, S.R.; McWhirter, E.H.; Chronister, K.M. Self-Efficacy and Vocational Outcome Expectations for Adolescents of Lower Socioeconomic Status: A Pilot Study. J. Career Assess. 2005, 13, 40–58. [Google Scholar] [CrossRef]
  32. Kantamneni, N.; McCain, M.R.C.; Shada, N.; Hellwege, M.A.; Tate, J. Contextual Factors in the Career Development of Prospective First-Generation College Students: An Application of Social Cognitive Career Theory. J. Career Assess. 2018, 26, 183–196. [Google Scholar] [CrossRef]
  33. Lent, R.W.; Lopez, A.M.; Lopez, F.G.; Sheu, H.-B. Social Cognitive Career Theory and the Prediction of Interests and Choice Goals in the Computing Disciplines. J. Vocat. Behav. 2008, 73, 52–62. [Google Scholar] [CrossRef]
  34. Usher, E.L.; Pajares, F. Sources of Self-Efficacy in Mathematics: A Validation Study. Contemp. Educ. Psychol. 2009, 34, 89–101. [Google Scholar] [CrossRef]
  35. Roller, S.A.; Lampley, S.A.; Dillihunt, M.L.; Benfield, M.P.J.; Gholston, S.E.; Turner, M.W.; Davis, A.M. Development and Initial Validation of the Student Interest and Choice in STEM (SIC-STEM) Survey 2.0 Instrument for Assessment of the Social Cognitive Career Theory Constructs. J. Sci. Educ. Technol. 2020, 29, 646–657. [Google Scholar] [CrossRef]
  36. Garriott, P.O.; Flores, L.Y.; Martens, M.P. Predicting the Math/Science Career Goals of Low-Income Prospective First-Generation College Students. J. Couns. Psychol. 2013, 60, 200–209. [Google Scholar] [CrossRef]
  37. Budinoff, H.; Subbian, V. Asset-Based Approaches to Engineering Design Education: A Scoping Review of Theory and Practice. In Proceedings of the American Society for Engineering Education Virtual Annual Conference Proceedings, American Society for Engineering Education Conferences, Virtual Conference, 26–29 July 2021; p. 36727. [Google Scholar]
Figure 1. Conceptual model. Adapted from [33].
Figure 1. Conceptual model. Adapted from [33].
Education 14 00789 g001
Figure 2. Joint display.
Figure 2. Joint display.
Education 14 00789 g002
Table 1. BOAST student characteristics by experimental condition.
Table 1. BOAST student characteristics by experimental condition.
TreatmentControl
Entering grade
 Grade 90.26 (0.44)0.37 (0.49)
 Grade 100.38 (0.49)0.26 (0.45)
 Grade 110.34 (0.48)0.37 (0.49)
Demographic characteristics
  Female0.73 (0.45)0.63 (0.49)
  Male0.27 (0.45)0.37 (0.49)
  Black0.83 (0.38)0.78 (0.42)
  Hispanic, White0.07 (0.25)0.15 (0.36)
  White or Asian, non-Hispanic0.10 (0.30)0.07 (0.27)
  Special education0.10 (0.30)0
  English learner0.02 (0.13)0.04 (0.19)
Academic characteristics
  Algebra I grade (4-point scale)3.07 (0.77)2.89 (1.05)
  School day attendance rate, year before treatment92.3 (0.09) *83.4 (17.32)
  School day chronic absence, year before treatment24.6 (0.43) *55.6 (50.6)
n6027
Note. Means (and standard deviations) presented. * Group difference significant, p < 0.05.
Table 2. Baseline comparison of SCCT constructs.
Table 2. Baseline comparison of SCCT constructs.
SCCT ConstructTreatmentControl
Math self-efficacy3.87 (0.51)3.69 (0.51)
STEM interest4.12 (0.62)3.97 (0.61)
STEM outcome expectations3.72 (0.62)3.76 (0.76)
STEM choice goals3.86 (0.72)3.81 (0.78)
Note. No mean differences were statistically significant (α = 0.05).
Table 3. BOAST participation levels.
Table 3. BOAST participation levels.
TreatmentControl
Total LMS hours 3.77 (7.64)--
Total video viewing hours0.46 (0.74)0.39 (0.46)
Number of field trips attended1.60 (1.21) *0.95 (0.22)
School day attendance
Average daily attendance rate90.3 (8.3) *81.6 (17.1)
% chronically absent34.4 (47.9) *66.7 (48.0)
n6027
Note. Means (and standard deviations) presented. School day attendance is presented to provide context on student participation in BOAST’s out-of-school time components. * Group difference significant, p < 0.05. LMS hours do not capture in-person exposure to algebra for engineering content.
Table 4. Relationships between pre-survey SCCT constructs.
Table 4. Relationships between pre-survey SCCT constructs.
STEM InterestSTEM Outcome ExpectationsSTEM Choice Goals
Math self-efficacyNSNSNS
STEM interest--0.50 **0.65 ***
STEM outcome expectations----0.60 ***
Note. Standardized betas presented. NS: not statistically significant. *** p < 0.001; ** p < 0.01.
Table 5. Treatment effects.
Table 5. Treatment effects.
Outcomes (Pre/Post Delta)Model 1Model 2Model 3
Math self-efficacy0.02 (0.01) * 0.19 (0.12)−0.11 (0.16)
STEM interest0.03 (0.01) **0.08 (0.16)0.29 (0.19)
STEM outcome expectations0.03 (0.01) *0.30 (0.51)0.46 (0.75)
STEM choice goals0.01 (0.01)0.18 (0.22)0.59 (0.16) *
Note. Treatment group coefficients and robust SEs presented. Model 1 includes only treatment effects where ‘treatment’ estimate is for LMS hours. Model 2 controls for demographic characteristics. Model 3 controls for demographic and prior academic characteristics. ** p < 0.01; * p < 0.05.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Durham, R.E.; Falk, M.L.; Daniels, A.G.; Reigel, A.; Sparks, A.; Williams, M.; Yanisko, E.J. Encouraging STEM Careers among Minoritized High School Students: The Interplay between Socio-Environmental Factors and Other Social Cognitive Career Constructs. Educ. Sci. 2024, 14, 789. https://doi.org/10.3390/educsci14070789

AMA Style

Durham RE, Falk ML, Daniels AG, Reigel A, Sparks A, Williams M, Yanisko EJ. Encouraging STEM Careers among Minoritized High School Students: The Interplay between Socio-Environmental Factors and Other Social Cognitive Career Constructs. Education Sciences. 2024; 14(7):789. https://doi.org/10.3390/educsci14070789

Chicago/Turabian Style

Durham, Rachel E., Michael L. Falk, Alexis G. Daniels, Allison Reigel, Alisha Sparks, Margo Williams, and Emily J. Yanisko. 2024. "Encouraging STEM Careers among Minoritized High School Students: The Interplay between Socio-Environmental Factors and Other Social Cognitive Career Constructs" Education Sciences 14, no. 7: 789. https://doi.org/10.3390/educsci14070789

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop