Predicting Student Dropout in Higher Education

Aulck, Lovenoor; Velagapudi, Nishant; West, Jevin D.

doi:10.48550/arxiv.1606.06364

Cited by 33 publications

(47 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Interestingly, boosting seemed to have no appreciable increase in predictive performance, as can be seen with the small performance difference between Ada-boosted logistic regression and logistic regression. Seeing logistic regression perform better than other models is in line with what our group has seen in the past when predicting dropout across all students [3]. As was the case with our previous work, we believe these results are strong given the limited amount of information fed into the models, as data was extracted from registrar records alone.…”

Section: Model Constructionsupporting

confidence: 87%

“…Ram et al used data on about 6,500 freshmen at a large, public US university to predict whether students would drop out after their first semester, and for those that remained, whether they will drop out after their second semester [27]. Our group has also shown some early success in predicting dropout in more heterogeneous data while using a much larger dataset (>32,000 students) [3].…”

Section: Related Workmentioning

confidence: 95%

“…In addition, US census data was used to add additional information based on students' application ZIP code including: average income in ZIP, percentage of high school graduates in ZIP, and percentage of college graduates in ZIP. In addition, we also calculated the distance from the ZIP code to UW using Python's geopy package 3 .…”

Section: Predicting Stem Attritionmentioning

confidence: 99%

“…To impute missing SAT and ACT scores, we used a linear regression model with other demographic and pre-college entry data. In our previous work, we had tried using mean imputation for these missing values and did not see an appreciable difference in performance [3]. Students' preentry and demographic information accounted for 52 total features.…”

Section: Predicting Stem Attritionmentioning

confidence: 99%

See 3 more Smart Citations

Stem-ming the Tide: Predicting STEM attrition using student transcript data

Aulck¹,

Aras²,

Lysia³

et al. 2017

Preprint

Self Cite

View full text Add to dashboard Cite

Science, technology, engineering, and math (STEM) fields play growing roles in national and international economies by driving innovation and generating high salary jobs. Yet, the US is lagging behind other highly industrialized nations in terms of STEM education and training. Furthermore, many economic forecasts predict a rising shortage of domestic STEM-trained professions in the US for years to come. One potential solution to this deficit is to decrease the rates at which students leave STEM-related fields in higher education, as currently over half of all students intending to graduate with a STEM degree eventually attrite. However, little quantitative research at scale has looked at causes of STEM attrition, let alone the use of machine learning to examine how well this phenomenon can be predicted. In this paper, we detail our efforts to model and predict dropout from STEM fields using one of the largest known datasets used for research on students at a traditional campus setting. Our results suggest that attrition from STEM fields can be accurately predicted with data that is routinely collected at universities using only information on students' first academic year. We also propose a method to model student STEM intentions for each academic term to better understand the timing of STEM attrition events. We believe these results show great promise in using machine learning to improve STEM retention in traditional and non-traditional campus settings.

show abstract

Section: Model Constructionsupporting

confidence: 87%

Section: Related Workmentioning

confidence: 95%

Section: Predicting Stem Attritionmentioning

confidence: 99%

Section: Predicting Stem Attritionmentioning

confidence: 99%

See 2 more Smart Citations

Stem-ming the Tide: Predicting STEM attrition using student transcript data

Aulck¹,

Aras²,

Lysia³

et al. 2017

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Previous research has centered on physical or psychological outcomes of experiences of violence with little attention to environment or setting. GPA is the foremost predictor of dropout (Aulck et al, 2016). Existing literature has found associations between interpersonal violence and academic performance such as declines in attendance, decreased GPA, and a decreased likelihood to complete one's education altogether (Loya, 2015).…”

Section: Introductionmentioning

confidence: 99%

Substance Use and Interpersonal Violence: Exploring Potential Threats to Underrepresented Minority Students’ Academic Success

BlackDeer

Wolf

Beeler‐Stinn

et al. 2020

Journal of College Student Retention: Research, Theory & Pr

View full text Add to dashboard Cite

Background College dropout has been described as an epidemic, with underrepresented minority (URM) students having the highest dropout rates at colleges and universities. This study examines interpersonal violence and substance use as potential threats to the academic success of URM students. Methods This study is a secondary data analysis of the National College Health Assessment. Results Significant decreases for grade point average (GPA) in African American students were associated with physical violence and marijuana use. For Hispanic/Latinx students, physical violence, marijuana use, and methamphetamine use were significantly associated with decreases in GPA. American Indian, Alaska Native, or Native Hawaiian students’ decreases in GPA were significantly associated with marijuana use. Conclusions Results of the study indicated that substance use and interpersonal violence are associated with decreases in GPA for various groups of URM. These findings are disconcerting and relay the importance for colleges and universities to undertake strategies to increase the retention of URM students.

show abstract

Contextualizing the current state of research on the use of machine learning for student performance prediction: A systematic literature review

Alalawi

Athauda

Chiong

2023

Engineering Reports

View full text Add to dashboard Cite

Today, educational institutions produce large amounts of data with the deployment of learning management systems. These large datasets provide an untapped potential to support and enhance decision‐making and operations. In recent times, machine learning (ML) has been applied to develop models utilizing this “big” data to assist in decision‐making. This study presents a systematic literature review into the application of ML to predict student performance. A total of 162 research articles from January 2010 to October 2022 were critically reviewed and analyzed by applying Kitchenham's systematic literature review approach. Our analysis categorized the literature predicting students' academic performance into two categories: (i) predicting student performance in assessments, courses or programs, and identifying students at‐risk of failing their course/program (129 studies); and (ii) predicting student dropout or retention in a course or program (33 studies). Classification is the most commonly used approach for predicting student performance (138 studies), followed by regression (25 studies) and clustering (9 studies). Supervised learning methods are used more often than semi‐supervised learning. Five most popular ML algorithms include the Decision Tree, Random Forest, Naïve Bayes, Artificial Neural Network, and Support Vector Machine. Historical records of students' grades and class performance, academic related data from learning management systems, and students' demographics are the most common features used for predicting students' performance. The most common methods used for feature selection are Information Gain‐based selection algorithms, Correlation‐based feature selection, and Gain Ratio. The general platforms/tools/libraries used in the studies include WEKA, Python, R, Rapid Miner, and MATLAB. We also investigated possible actions considered in the literature to help at‐risk students. We only found very few studies that deployed remedial actions and evaluated their impact on students' performance. In conclusion, ML has shown great potential in the prediction of student performance, but also has many areas of further research.

show abstract

Predicting Student Dropout in Higher Education

Cited by 33 publications

References 0 publications

Stem-ming the Tide: Predicting STEM attrition using student transcript data

Stem-ming the Tide: Predicting STEM attrition using student transcript data

Substance Use and Interpersonal Violence: Exploring Potential Threats to Underrepresented Minority Students’ Academic Success

Contextualizing the current state of research on the use of machine learning for student performance prediction: A systematic literature review

Contact Info

Product

Resources

About