Correlation of STEM Interest and Career Intent in High-School Students

Understanding high school students’ perceptions and dispositions toward STEM, and the role science and math self-efficacy play in establishing STEM career aspirations is imperative to preparing the STEM workforce of the future. Project STEMulate is an industry-aligned and technology-rich Problem-based Learning (PBL) model. The goal of this NSF ITEST grant-funded study (2018-2020) was to improve students’ attitudes towards STEM. Project STEMulate focuses on Upward Bound students in Hawaiˋi and was implemented at three sites: Maui, Hilo, and Oahu. The participants voluntarily selected to participate in this program. The current study reviews year one data collected on the impact of Project STEMulate on low-income and underrepresented and/or native Hawaiian students' STEM career interest, and their science self-efficacy. Students’ reactions to the STEM learning experience were extremely positive. 80% of students expressed a desire to pursue a career in STEM at the post test. High school students who listed their plan to pursue a career in STEM also showed a higher self-efficacy and motivation. Analysis of the results demonstrates this program was effective in empowering students with insights into careers, enhancing knowledge that would serve them in pursuit of a career in STEM. In addition, the project fostered a can-do attitude and increased students’ science selfefficacy.


Introduction
Studies have repeatedly reported the gap observed in student interest in Science, Technology, Engineering, and Mathematics (STEM) and their desire to pursue STEM major or careers (ACT, 2015;Blotnicky et al., 2018;Christensen et al., 2015;Kier et al., 2013). This is despite the current recognition of STEM careers as the most versatile careers (Mokter Hossain & Robinson, 2012). Moreover, the rapid expansion of STEM careers demands an increase in the preparation of the high school graduates who are prepared for post-secondary education, training and careers in STEM (Hayes, 2017). However, many STEM-interested students are not prepared to succeed in the rigorous college math and science coursework required of STEM majors. While many believe proficiency and interest in STEM should be initiated in middle school (Christensen et al., 2015), others selected later elementary school years as the right time (Tai et al., 2006). We agree that earlier engagement of students' interest in STEM is preferable, however, students who are already in high school and will be heading to college soon are also a concern. How can we help this group? Other researchers (Kitchen et al., 2018;Maltese & Tai, 2011) showed majority of students interested in STEM made that choice in high school. This puts us in agreement with Maltese et al. (2014) that many pathways toward STEM study and careers exist, with none being singularly preferred (p. 937). We want to present the results of a study showing that an out-of-school program, i.e., Upward Bound program, can provide opportunities-particularly to low-income and first-generation to college students-to promote their interest in STEM.

Lack of Interest in STEM Workforce
Although research on the relationship between student interest in and pursuit of STEM careers has been on the rise (Christensen & Knezek, 2017), the issue of increasing student interest in STEM is of greater magnitude when it comes to the achievement of underserved and underrepresented students in the STEM fields. The National Academy of Sciences (2011) reported less than 10% of minority students to be college educated in science and technology while they make up close to 30% of the population. It is a critical and growing need to draw minority students into STEM fields and to increase the number of minority graduates from STEM programs (May & Chubin, 2013). Numerous studies at the undergraduate level provide support for increasing minority students' retention in STEM majors (Carpi & Lents, 2013;Junge et al., 2010;Kardash, 2000). Nevertheless, preparing students to sustain study at undergraduate level is also important. Thus, intervention programs such as Project STEMulate, focusing on high school students, are imperative to ensure that minority students are learning the skills needed to be successful in completing college degrees, especially in STEM fields.
Native Hawaiian students are often underprepared, underrepresented, and underserved in STEM fields. Various national and international assessments show that Native Hawaiian students perform far below white students in STEM skills and reading (DeSilver, 2018;NAEP, 2017). Furthermore, in a report on Hawaiian students and STEM education, it was determined that these students are taught by inexperienced science and math teachers (Gay, 2013). Research also has demonstrated that students' lack of exposure to STEM career possibilities is a reason for why they are less likely to pursue STEM careers (Mokter Hossain & Robinson, 2012). The National Science Foundation (NSF) has clearly seen the need to address this problem and funded many studies where their goal has been to support, implement, and assess any program that fosters academic success of minority students majoring in a STEM field.

Informal or Out-of-School Education
Informal or out of school education refers to learning that occurs outside of traditional schooling (Dierking et al., 2003). Common informal learning environments include after-school and weekend and/or summer camp programs. Reports on such programs confirm they have increased students' interest in STEM majors (Bicer et al., 2015(Bicer et al., , 2018Vela et al., 2020), improved students' mathematics and science vocabulary knowledge (Bicer et al., 2015), enhanced students' artistic self-efficacy in STEM (Capraro et al., 2014), enriched students' communication skills (Bicer et al., 2015), advanced students' self-identity (Barroso et al., 2016), and improved students' scientific reasoning (Gerber et al., 2001).

Upward Bound Program
The Upward Bound (UB) program, established in 1965, is designed to provide services to high school students identified as low income and first-generation-tocollege and support their transition to, and enrollment in postsecondary educational institutions (U. S. Department of Education, 2012). The federal fund allocated to UB program is to address existing educational inequities, and to provide required resources and support to prepare students both academically and socially for enrollment and retention in postsecondary education (Strayhorn, 2011;Villalpando & Solorzano, 2005). Given the kind of preparation needed and the support required, a major goal of UB programs at all of their sites MUST be to offer instruction in math, laboratory science, composition, literature, and foreign language.

Perspective(s) or Theoretical Framework
Underrepresented students are a significant school population in the United States, and their educational access is particularly jeopardized and lag behind other students. The full power of ubiquitous learning for educational transformation can be conceptualized through the overcoming of challenges related to infrastructure, human learning and ability, and motivation. This paper focuses on the implementation of a STEM PBL program drawing on constructivism (Dewey, 1933(Dewey, /1998, social cognitive career theory (SCCT) as articulated by, Lent et al. (1994), and culturally relevant education (Dover, 2013). The tenet belief of constructivism is that learners actively participate in interpreting information and creating their own knowledge (Piaget, 1972). PBL provides an active learning environment (Dahlquist & Cutucache, 2013) that engages learners in their learning process by transferring some responsibilities from teachers to students (Nariman & Chrispeels, 2016). According to SCCT, the other theoretical basis of this study, individuals pursue college or career majors that are aligned with their interests and match with their academic and career goals (Lent et al., 1994). Therefore, career choices are influenced by the quality of educational experiences. Consequently, increasing the opportunity to engage students in STEM-related experiences will increase the likelihood of pursuing STEM majors and careers. Although recommendations on when to act in order to have an impact on students' college and career pathways are different, we agree with Beier and Rittmayer (2009), Bicer and Lee (2019), and Hansen (2011) that before and during high school is the most effective time. Seventy-eight percent of college students reported that they decided on their selection of a STEM major in high school (Microsoft Corporation, 2011).

Culturally Relevant Education (CRE)
The CRE emerged from the union of culturally responsive teaching (Gay, 2010) and culturally relevant pedagogy (Ladson-Billings, 1994). The goal for culturally relevant pedagogy is to create equal opportunity for students from diverse cultural backgrounds (Ladson-Billings, 1994) with the view of creating a meaningful connection between students' background knowledge (i.e., culture, language and previous life experiences) and what they learn at school so they can see the relevance of their learning. According to Gay's culturally responsive teaching "the cultural knowledge, prior experiences, frames of reference, and performance styles of ethnically diverse students," (2010, p. 31) have to come together to make meaning and find relevancy in what is learned. This connection only comes through providing all students with equal opportunities to be academically successful (Banks, 2008;Gay, 2010Gay, , 2013Ladson-Billings, 1995). CRE supporters believe that valuing students' cultural backgrounds and cultural identities creates the optimal learning environment for students to thrive (Gay, 2010;Ladson-Billings, 1994;Nieto, 1999) because it demands a student-centered instruction (Irvine & Armento, 2001) where teachers are acting as facilitators with high expectations of students, creating a learning environment within the context of culture (Ladson-Billings, 1994). This requirement matches perfectly with PBL strategy and its tenets. PBL has proven to have the capabilities to help students in this process and to guide and inspire them to relate their previous knowledge to the present, and further connect it to their future studies and career selection. Therefore, the purpose of this project is to support Upward Bound students with a PBL intervention that encourages and motivates them to successfully navigate towards an undergraduate degree in a STEM field.

Problem-based Learning
PBL is an innovative learning and instructional approach that empowers learners to conduct research, integrate theory and practice, and apply knowledge and skills to develop a viable solution to a defined problem, a problem very relevant to the learners (Savery, 2006). Essential tenets of PBL include: 1) real-world focus; 2) collaboration; 3) student-driven and student-centered design; 4) open-ended outcomes; and 5) an interdisciplinary approach (Savery, 2006). In such a PBL setting, students actively participate in learning (individually or in small groups) to address real and relevant problems contributing to their own understanding and achievement of concrete outcomes (Barrows, 1985;Hmelo-Silver, 2004;Marx et al., 2004). Students are consequently better able to apply their learning to new problems in a variety of settings (Barrow, 1985). Furthermore, PBL has proven to be effective for teaching critical thinking, communication, collaboration, and applying knowledge to realworld situations (Walker & Leary, 2009;Darling-Hammond et al., 2008, Strobel & van Barneveld, 2009). The promising results of several high school PBL studies indicate PBL is "as or more effective" than traditional teaching approaches (Boaler, 1998;Mergendoller et al, 2006), especially with low-income students (Lynch et al., 2005;Cueva, 2005;Gallagher & Gallagher, 2013).

Need for New Programs
Many researchers agree that real-world hands-on problem/project-based learning that personally and locally connects to students is of value (Christensen & Knezek, 2015).
This study was guided by the following research questions: 1.
What was the likelihood of students selecting a STEM career? a.
What role did gender play in the likelihood of selecting a STEM career? 2.
What level of STEM career interest existed among high school students? 3.
What was the correlation between student science self-efficacy and STEM interest? 4.
How did Project STEMulate impact students' STEM career interest?

Context of Study
Project STEMulate was organized as a STEM Problem-based Learning (PBL) curriculum and model that operationalizes key PBL tenets while meeting programmatic requirements and academic outcomes to develop motivation and interest in STEM. The primary goal of Project STEMulate was to develop Upward Bound (UB) high school students' interest in STEM content and to elevate their perceptions of STEM careers. The program focused on hands-on activities where students explored and researching solutions to a real-world industry-aligned problem. The context for this program was a five-week UB summer academy on three islands, Maui, Oahu, and Hawaiˋi during 2018-2020. This paper draws on year 1 data.
The integration of the culturally relevant research and Bandura's (1986Bandura's ( , 2001) social cognitive theory and constructivist theoretical frameworks in a PBL setting were used as an analytical lens along with a mixed method approach. The data collection included: pre-and post-surveys, semi-structured focus group interviews, observation of participants' final presentation, and review of their final reports.

Participants
The target population was the low income, underrepresented, first-generation, and/or Native Hawaiian 9 th through 12 th grade students participating in the UB summer Academy. Data were gathered from students who participated in Project STEMulate and a comparison group who had a similar summer experience with traditional courses in math, science, and language arts. In total, 113 high school students participated in this study with 64 in STEMulate group and 49 in the comparison group. The breakdown of UB program participants at each participating site was: University of Hawaiˋi Maui College (UHMC) (n=51), University of Hawaiˋi at Hilo (UHH) (n=37), and the Windward Community College (WCC) -University of Hawaiˋi (n=25). Overall, there were 62% female and 38% male students. The STEMulate group comprised of 58% female and 42% male students and the comparison group had 67% female and 33% male students.

Problem Explored
The problem explored in Year 1 was: "How can the island of meet the statewide goal of 100% energy from renewable sources by 2045 considering different strategies along with pros, cons, and potential hurdles to overcome."

Measures and Instruments
Science Self-Efficacy (SSE). This eight-item scale was used to measure student selfefficacy and ability in science, partially adapted from the science section of the STEM Career Interest Survey (Kier et al., 2013

Findings
Research Question 1: What was the likelihood of students selecting a STEM career?
Students' responses to the question: "I plan to have a career in science, technology, math, or engineering" was calculated at the pre-post survey for comparison. As Figure  1 demonstrates, the percentage of STEMulate group students planning for a career in STEM increased at the end of the program. In particular, there was a 19% gain for the STEMulate students who aspire to have a career in science. However, Figure 1 demonstrates no consistency in the increase or the decrease of the likelihood of selecting a STEM career for the comparison group. Furthermore, when students' responses were dichotomized to create two separate measures: STEM selection (i.e., science, technology, engineering, and mathematics) and non-STEM, a large impact was observed in the STEMulate group as those who planned for a career in STEM gained 39% compared to the comparison group that lost 7% (see Table 1 shows). When the data for the likelihood of selecting STEM career were further analyzed for gender, the program effects on girls were more noticeable. Table 2 shows the changes in students' planning for careers in STEM vs. non-STEM by gender, for both STEMulate and comparison groups. Overall, male students from both groups were equally divided into STEM and non-STEM careers while a greater percentage of females indicated preference for STEM careers at both the pre-post survey for both groups. In fact, both genders in the comparison group lost interest in planning for a STEM career, and the females in the STEMulate group showed more intention for pursuing a STEM career. SCI is also consisted of three subscales, Support: perception of being in a supportive environment for pursuing a career in science, Education: intent to pursue educational opportunities that would lead to a career in science, and Importance: perceived importance of a career in science. Table 3 shows the Mean and Standard Deviation for SCI and its subscale at both Time 1 and Time 2 for both groups. The mean of the subscales ranged from 2.11 to 4.17 across the subscales and groups. For both groups, SCI Part 1 (Support) had the lowest Mean while SCI Part 3 (Importance) had the highest Mean at both Time 1 and Time 2. The result of an independent sample t-test indicated a significant difference in career interest satisfaction between the STEMulate and comparison group, t(108) = .834, p <.001. Nevertheless, no difference in career interest was observed based on gender.

Research Question 3: What was the correlation between student science self-efficacy and their STEM interest?
For the science self-efficacy (SSE) scale, students' responses to the eight statements were dichotomized by assigning a value of "1" to those who were most agreeable with Likert scale ratings of 4 or 5 to the statements, and a value of "0" was assigned to those who disagreed or strongly disagreed (Likert scale ratings of 1 through 3) with the statements. These eight measures were finally summed to create a single SSE scale. The final SSE score ranged from 1 (Low self-efficacy) to 8 (High self-efficacy). The distribution of the Science Self-Efficacy Scale at Time 2 is shown for both groups.

Figure 2: Distribution of Student Science Self-Efficacy (SSE) Scale for STEMulate and Comparison group at T2
For further analysis, the SSE scale were again divided into two subgroups: low SSE (scores of 1 -4) and high SSE (scores of 5 -8). This breakout identified students who were the most comfortable and confident in their science experiences. The results for the SSE scale were validated using confirmatory factor analysis (CFA) and reliability analysis. The factor analysis was statistically significant (KMO = .719, p <.001). These results suggested for students in the STEMulate group to have a higher SSE than the comparison group (see Figure 3). Figure 3: Comparing students' Science Self-Efficacy for STEMulate and Comparison groups at T2

Research Question 4: How did Project STEMulate impact students' STEM career interest?
To respond to this question, students' final reports were evaluated, in addition to their responses to the focus group interviews. All students engaged in a wide variety of activities that would support their competency to pursue STEM careers. Major examples included conducting experiments, performing mathematical calculations, doing research, administering surveys, and conducting interviews which will be further discussed.

Conducting Experiments.
Students conducted experiments such as building a simple alternating current generator, using a voltmeter to measure the number of kilowatts per hour used by an old refrigerator vs. a newer energy-efficient model, and created a prototype of a concrete slab equipped with thermoelectric plates and then tracking and measuring the voltage generated over time. These and other experiments demonstrated students' engagement in skilled explorations relevant to their topics which required carefully executed scientific processes.
Mathematical Calculations. Students also engaged in complex and extensive mathematical calculations such as determining the amount of energy which could be generated by converting human waste into biogas, or calculating the price and amount of power generated by different types of solar cells. The various requirements of students' research made evident that they were learning important mathematical concepts essential to supporting their scientific inquiry.
Research. All of the groups conducted research related to their topics, the problems they were addressing, and their proposed solutions. The cited sources in their final papers included government and industry websites, scholarly journals, books, and a variety of online resources. Their work was well supported by the resources they relied on and it was presented in academically robust ways.
Surveys. Almost all the projects involved administering surveys, most often to assess awareness of and opinions about renewable energy issues. Students developed questions and conducted surveys with community members, experts, and local companies, often executing these online through email and social media. They also presented their quantitative results in graphs and charts. Their presentations and final papers reflected that their efforts in seeking out this kind of real-life data were principled and meaningful to them.
Interviews. Students also engaged with and interviewed a variety of professionals with positions in STEM fields, including government and industry representatives as well as relevant academics. By doing so, they obtained essential background information, got advice on how to conduct their research, and elicited comments relevant to the particular problems their projects focused on, for example by asking what these experts would say to those expressing concerns about the challenges inherent to switching to renewable energy resources (concerns that were raised by the community members they had surveyed). Having contact with those working in STEM fieldspeople who could serve as role models-provided opportunities for students to be exposed to and inspired by the kinds of careers they might one day pursue.

Additional Skills
In the focus groups, in addition to the skills discussed above, students specifically mentioned a number of other competencies they gained or improved upon by participating in the program. Examples include: presenting, critical thinking, time management, writing up papers, data collection, communicating ideas, and teamwork.

RQ 2. How did using technology-rich STEM PBL affect participating student's interest in STEM careers?
In the focus groups, 14 of 23 students explicitly indicated they were thinking about a STEM career, with examples including engineering, computer programming, forensics, aerospace, medicine, animation, and game design. Most responses were simple statements of career plans but many of the Maui students (aided by some probing by the interviewer) specifically indicated the impacts of this program on those decisions: • STEMulate really opened new ideas towards science... like the hands on is really fun. I want to go into a job with lots of hands on. (Maui student) The program also provoked student interest in STEM careers by affecting key steps along such path. The majority of focus group participants (19 of 23) agreed that the STEMulate program would help them be more successful at school, and over twothirds said they were now more likely to take STEM classes in high school. One students said: • I think I would want to take more STEM classes in my upcoming years, because this experience really gave me more insight on the different sides of STEM. (Oahu student)

Problem-based Learning
One of the major ways the program facilitated student interest in STEM and students' possible interest in STEM careers-was by making the learning relevant and focused on real life problems. Without exception, all the student projects addressed issues with both local and cultural relevance. Their problem-based explorations were rooted in things students could relate to, often involving existing controversies within the community. For example, an Oahu participant mentioned: • With this STEM course, we dealt with real-world problems and I'm really interested in that.

Cultural Relevance
The cultural relevance of the work students engaged in was unmistakable throughout the focus group interviews.
• A lot of people think culture is more like hula and chatting and stuff, but there's more to it than that, there's culture to science. (Hilo student) Likewise, in the final papers, students' respect for and investment in the cultural aspects of their research was apparent. For example, in the Maui groups, they acknowledged, valued, and proactively addressed community members' concerns: • As we continue the process of getting to net-zero by 2045, we need to be aware of the deep historical meanings that the land possesses. By recommendation, it would be healthier … to place solar panels on houses/buildings rather than on the land itself, ensuring that we are not damaging the land and its historical roots. (Excerpt from Maui Group 6 Final Paper) The solutions students proposed also drew upon culturally relevant connections. It is well established in educational research that cultural relevance can enhance student interest in their learning, something that is definitely observed in the data from this study. By encouraging and facilitating this kind of scientific research, Project STEMulate is also providing yet another reason why students might consider a STEM career in their future.

Discussion
The focus of the current study was to determine if participation in an industry-aligned technology-rich Problem-based Learning (PBL) model influenced the likelihood of students' selecting STEM careers. Prior research has indicated that the PBL environment can impact student's recognition and selection of STEM Careers (Christensen & Knezek, 2017;LaForce et al., 2017). This study is framed by culturally relevant research and Bandura's (1986Bandura's ( , 2001 social cognitive theory, which suggests that students' behaviors are influenced by their learning environment. Results from the first research question indicated that the likelihood of selecting a career in STEM for the STEMulate group increased at the end (39% gain). In contrast, the likelihood of selecting a STEM career by the comparison group decreased by the end of the summer. These results imply that engagement in Project STEMulate positively exposed students to a variety of STEM career options, something that the comparison group was not exposed. The results also imply that students might have grasped the benefits associated with STEM careers as they explored to find a solution to their problem. In other words, they may not have been aware or exposed to such experiences. Additionally, the results indicated a higher likelihood for the female students in the STEMulate group to select a STEM career at the end of the camp, compared to the male students.
Results from the second research question showed that the level of STEM career interest among high school students was low at the beginning of the program, and it increased by the end of the summer: STEMulate group Time 1 (M = 2.97, SD = .42), Time 2 (M = 3.58, SD = .75), and Comparison group Time 1 (M = 2.68, SD = .43), and Time 2 (M = 3.05, SD = .67). Although the mean increased for both groups after the program, the paired-samples t-test was statistically significant for the career interest score of the STEMulate group only. These results align with previous research (Christensen & Knezek, 2017) stating that engagement in hands-on PBL activities will increase interest in a STEM career. As part of Project STEMulate, students had the support and guidance of a team of three teachers who facilitated their learning daily, they went on many field trips where they listened to STEM partners, and they had access to University of Hawaiˋi math and science instructors.
Results from the third research question displayed a high positive correlation between student science self-efficacy and their STEM interest. Students who were most comfortable and confident in their science experiences showed a higher interest in STEM careers. On various field trips, the STEM partners explored traditional indigenous ways the renewable energy problem has been approached and they connected students' cultural references to mainstream science skills and concepts. Both STEM partners and the program facilitators engaged students in critical reflection, facilitated students' cultural competence to learn about their own and others' cultures, and provided opportunities for students to critique discourses of power and find opportunities to pursue social justice. This concurs with Lemus et al. (2014) in infusing traditional knowledge and ways of knowing into science education.
Also, to be effective, culturally relevant education demands for student-centered instruction where teachers are acting as facilitators with high expectations of students and creating a learning environment within the context of culture (Ladson-Billings, 2014;Lemus et al., 2014;Zaffos, 2013). The PBL setting of this project created the right environment for students' learning and supported them in recognizing, acknowledging, and applying their own cultural identities, strengths, backgrounds, and knowledge. It also acknowledges various ways of knowing and cultural strengths that students and teachers bring by creating space for STEM connection through PBL. This clearly existed in students' final presentations.

Conclusion
Prior research has implied the rising demand for the STEM workforce and the need to prepare students for STEM careers (Christensen & Knezek, 2017;Vela et al., 2020). The overall results from the present study indicated how an industry-aligned technology-rich PBL program can improve student likelihood of selecting a STEM career. This could be the result of hands-on engaging experiences, exposure to many field trips and access to STEM professionals. These experiences provided students with opportunities to learn more about potential STEM career options along with the benefits of those careers. This study is in alignment with Blotnicky et al. (2018), and Vela et al. (2020) that creating opportunities for students to learn about STEM careers directly enhances their interest in those careers. A special contribution of this study is that hands-on STEM PBL science activities, such as those embedded in this study, are particularly effective in enhancing STEM career interests for high school students. The hands-on real-world activities were effective in promoting students self-reported intent and interest in pursuing a career in STEM.