An official website of the United States government
In this paper, we describe our solution to predict student STEM career choices during the 2017 ASSISTments Datamining Competition. We built a machine learning system that automatically reformats the data set, generates new features and prunes redundant ones, and performs model and feature selection. We designed the system to automatically find a model that optimizes prediction performance, yet the final model is a simple logistic regression that allows researchers to discover important features and study their effects on STEM career choices. We also compared our method to other methods, which revealed that the key to good prediction is proper feature enrichment in the beginning stage of the data analysis, while feature selection in a later stage allows a simpler final model.
more »
« less
ASSISTments Longitudinal Data Mining Competition Special Issue: A Preface.
This special issue includes papers from some of the leading competitors in the ASSISTments Longitudinal Data Mining Competition 2017, as well as some research from non-competitors, using the same data set. In this competition, participants attempted to predict whether students would choose a career in a STEM field or not, making this prediction using a click-stream dataset from middle school students working on math assignments inside ASSISTments, an online tutoring platform. At the conclusion of the competition on December 3rd, 2017, there were 202 participants, 74 of whom submitted predictions at least once. In this special issue, some of the leading competitors present their results and what they have learned about the link between behavior in online learning and future STEM career development.
more »
« less
- PAR ID:
- 10190344
- Date Published:
- Journal Name:
- Journal of educational data mining
- Volume:
- 12
- Issue:
- 2
- ISSN:
- 2157-2100
- Page Range / eLocation ID:
- I-Xi
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
There is an urgent need for young people to prepare for and pursue engineering careers. Engineering occupations comprise 20% of the science, technology, engineering, and math (STEM) jobs in the U.S. (Bureau of Labor Statistics, 2017). The average wage for STEM occupations is nearly double that of non-STEM occupations, with engineers commanding some of the highest salaries in STEM (Bureau of Labor Statistics, 2017). Moreover, engineering occupations are expected to be some of the fastest growing occupations in the U.S. over the next 10 years (Occupational Outlook Handbook, 2018); yet, there are current and projected shortages of workers in the engineering workforce so that many engineering jobs will go unfilled (Bureau of Labor Statistics, 2015) Native Americans are highly underrepresented in engineering (NSF, 2017). They comprise approximately 2% of the U.S. population (U.S. Census Bureau, 2013), but only 0.3% of engineers (Sandia National Laboratories, 2016). Thus, they are not positioned to attain a high-demand, high-growth, highly rewarding engineering job, nor to provide engineering expertise to meet the needs of their own communities or society at large. The purpose of this study was to examine factors that encourage or discourage Native American college students’ entry into engineering. Using Social Cognitive Career Theory (SCCT; Lent, Brown, & Hackett, 1994; 2000), we examined the correlates of these students’ interests and efficacy in engineering to accomplish this goal. Participants were N = 30 Native American engineering college students from the Midwest; 65% men, 30% women, and 4% other. The mean age was 25.87 (SD = 6.98). Data were collected over the period of one year on college campuses and at professional development conferences via an online survey hosted on Qualtrics. Three scales were used in the study: Mapping Vocational Challenges – Engineering (Lapan & Turner, 2000, 2016), the Perceptions of Barriers Scale (POB; McWhirter, 1998), and the Structured Career Development Inventory (Lapan & Turner, 2004). An a priori Power Analysis (f2 = .50; α = .05, 1 – β = .90) indicated our sample size was adequate. For all scales, full-scale Cronbach’s α reliabilities ranged from .82 to .86. Results of correlation analyses indicated that engineering efficacy was negatively related to lack of academic preparation (r = -.50, p = .016), and perceived lack of ability (r = -.53, p = .009), and positively related to academic achievement (r = .43, p = .043), career exploration (r = .47, p = .022), and approaching engineering studies proactively (r = .53, p = .009). Engineering interests were negatively related to perceived lack of ability (r = -.55, p = .007), and positively to proactivity (r = .42, p = .044), and academic achievement (r = .45, p = .033). Engineering interests were also related to support from parents, teachers, and friends to study engineering and pursue an engineering career. There was no significant relationship between engineering interests and engineering efficacy among these students. The relevance of these results will be discussed in light of SCCT, and recommendations for practice will be included.more » « less
-
null (Ed.)In early 2020, colleges shifted abruptly from traditional in-person to remote distant instruction due to COVID-19 potentially exacerbating science, technology, engineering, and mathematics (STEM) students’ recruitment and retention. This preliminary study using survey methodology was conducted with STEM students at a small (700 students) private college to examine questions related to students’ perceptions of natural science careers, career decision-making factors, barriers influencing students’ career path, including effects of COVID-19 on career goals, mental health, and perceived quality of instruction. A Qualtrics® survey was sent to 180 STEM students, from which we received 53 responses (29.4% response rate). Consistent with other studies, family was one of the most important factors supporting their career path. Students had a relatively upbeat career outlook despite being in the middle of a global pandemic and were only moderately worried about the impact of COVID-19 on their future career. Despite these relatively positive outcomes, the abrupt switch to online instruction was viewed unfavorably by most respondents, who valued the hands-on learning experiences obtained with traditional in-person instruction. It is possible that respondents’ views of online instruction may improve over time as instructors become more adept at using new instructional tools. Future research should evaluate this aspect and whether students’ career goals change across time as the pandemic unfolds.more » « less
-
null (Ed.)There is a strong need in the United States to increase the size and diversity of the domestic workforce trained in science, technology, engineering, and math (STEM). With almost half of all students that earn a baccalaureate degree enrolling in a 2-year public college at some point, the nation’s 2-year colleges provide great promise for improving the capacity of the STEM workforce for innovation and global competition while addressing the nation’s need for more equity between groups that have been historically included and those that have been economically and politically disenfranchized. Almost half of underrepresented minoritized (URM) students begin their post-secondary education at 2-year colleges yet their transfer rates within 5 years are only 16%. This study describes interventions put in place at a 2-year college to support increased transfer rates and STEM transfer readiness for URM STEM-interested students. The program studied, in place from 2017 through 2020, had an overall transfer rate of 45%. Analysis of administrative, transcript, and student survey data connects the program interventions to the existing research on STEM momentum and other research on URM STEM transfer success. Ultimately, this study identifies potential leading indicators of transfer readiness, providing much needed documentation and guidance on the efficacy and limitations of interventions to improve upward STEM transfer.more » « less
-
POSTER. Presented at the Symposium (9/12/2019) Abstract: The Academy of Engineering Success (AcES) employs literature-based, best practices to support and retain underrepresented students in engineering through graduation with the ultimate goal of diversifying the engineering workforce. AcES was established in 2012 and has been supported via NSF S-STEM award number 1644119 since 2016. The 2016, 2017, and 2018 cohorts consist of 12, 20, and 22 students, respectively. Five S-STEM supported scholarships were awarded to the 2016 cohort, seven scholarships were awarded to students from the 2017 cohort, and six scholarships were awarded to students from the 2018 cohort. AcES students participate in a one-week summer bridge experience, a common fall semester course focused on professional development, and a common spring semester course emphasizing the role of engineers in societal development. Starting with the summer bridge experience, and continuing until graduation, students are immersed in curricular and co-curricular activities with the goals of fostering feelings of institutional inclusion and belonging in engineering, providing academic support and student success skills, and professional development. The aforementioned goals are achieved by providing (1) opportunities for faculty-student, student-student, and industry mentor-student interaction, (2) academic support, and student success education in areas such as time management and study skills, and (3) facilitated career and major exploration. Four research questions are being examined, (1) What is the relationship between participation in the AcES program and participants’ academic success?, (2) What aspects of the AcES program most significantly impact participants’ success in engineering, (3) How do AcES students seek to overcome challenges in studying engineering, and (4) What is the longitudinal impact of the AcES program in terms of motivation, perceptions, feelings of inclusion, outcome expectations of the participants and retention? Students enrolled in the AcES program participate in the GRIT, LAESE, and MSLQ surveys, focus groups, and one-on-one interviews at the start and end of each fall semester and at the end of the spring semester. The surveys provide a measure of students’ GRIT, general self-efficacy, engineering self-efficacy, test anxiety, math outcome efficacy, intrinsic value of learning, inclusion, career expectations, and coping efficacy. Focus group and interview responses are analyzed in order to answer research questions 2, 3, and 4. Survey responses are analyzed to answer research question 4, and institutional data such as GPA is used to answer research question 1. An analysis of the 2017 AcES cohort survey responses produced a surprising result. When the responses of AcES students who retained were compared to the responses of AcES students who left engineering, those who left engineering had higher baseline values of GRIT, career expectations, engineering self-efficacy, and math outcome efficacy than those students who retained. A preliminary analysis of the 2016, 2017, and 2018 focus group and one-on-one interview responses indicates that the Engineering Learning Center, tutors, organized out of class experiences, first-year seminar, the AcES cohort, the AcES summer bridge, the AcES program, AcES Faculty/Staff, AcES guest lecturers, and FEP faculty/Staff are viewed as valuable by students and cited with contributing to their success in engineering. It is also evident that AcES students seek help from peers, seek help from tutors, use online resources, and attend office hours to overcome their challenges in studying engineering.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.5281/zenodo.4008048
