«USING DATA MINING TO PREDICT FRESHMEN OUTCOMES Nora Galambos, PhD Senior Data Scientist Office of Institutional Research, Planning & Effectiveness ...»
Conclusion It is clear from studying the decision tree model that weaker students from high schools with lower average SAT scores, who additionally are interacting with the LMS at diminished rates are over-represented in the lower GPA groups. The model can assist in identifying these students before the end of the semester so they can be assigned to interventions that may help to improve their outcomes. Since enrollment in courses with higher failure rates is also a factor appearing in the decision tree, developing a pre-orientation model could assist advisors in steering some students from course loads that may be excessively burdensome. The model results can also be shared with departments to inform their advising and intervention efforts. Automated methods for easily sharing the results are being planned. The goal is to find the students who need assistance in fulfilling their potential, thereby reducing the number who end up leaving due to poor performance.
Bahr, P.R. (2008). Does mathematics remediation work?: a comparative analysis of academic attainment among community college students. Research in Higher Education.
Bean J. (1983). The application of model of turnover in work organizations to the student attrition process. Review of Higher Education, 6(2), 129-148.
Breiman, L., Friedman, J., Olshen, R., Stone, D. (1984): Classification and Regression Trees.
Chen R. (2012). Institutional characteristics and college student dropout risks: a multilevel event history analysis. Research in Higher Education. 53:487–505.
Friedman, J., Hastie, T., Rosset, S., Tibshirani, R., Zhu, J. (2003) Discussion of Boosting Papers.
Retrieved from http://web.stanford.edu/~hastie/Papers/boost_discussion.pdf Herzog S. (2005). Measuring determinants of student return vs. dropout/stopout vs. transfer: a first-to-second year analysis of new freshmen. Research in Higher Education.
Johnson I. (2006). Analysis of stop-out behavior at a public research university: the multi-spell discrete-time approach. Research in Higher Education. 47:905-93.
Johnson I. (2008). Enrollment, persistence and graduation of in-state students at a public research university: does high school matter? Research in Higher Education. 49:776-793.
Parker M. (2005). Placement, retention, and success: a longitudinal study of mathematics and retention. The Journal of General Education. 54:22-40.
Robbins S, Allen J, Casillas A, Akamigbo A, Saltonstall M, Campbell R, Mahoney E, Gore P.
(2009). Associations of resource and service utilization, risk level, and college outcomes.
Research in Higher Education. 50: 101-118.
SAS Institute Inc. (2014). SAS® Enterprise Miner™ 13.2: Reference Help. Cary, NC: SAS
Singell L, Waddell, GR. (2010). Modeling retention at a large public university: can at-risk students be identified early enough to treat? Research in Higher Education. 51:546-572.
Stater M. (2009). The impact of financial aid on college GPA at three flagship public institutions.
American Educational Research Journal. 46:782-815.
Stinebrickner R, Stinebrickner T. (2014). Academic performance and college dropout: using longitudinal expectations data to estimate a learning model. Journal of Labor Economics. 32:601-644.
Thomas EH, Galambos N. (2004). What satisfies students? mining student-opinion data with regression and decision-tree analysis. Research in Higher Education. 45:251-269.
Tinto, V. (1987). Leaving college: rethinking the causes and cures of student attrition. Chicago, IL: The University of Chicago Press.
Zwick R, Sklar JG. (2005). Predicting college grades and degree completion using high school grades and SAT scores: the role of student ethnicity and first language. American Educational Research Journal. 42:439-464.
Variable List Demographics Gender Ethnicity Area of residence at time of admission: Suffolk County, Nassau County, New York City, other NYS, other US, International Pre-college Characteristics High School GPA College Board SAT Averages by High School Average High School Critical Reading Average High School SAT Math Average High School SAT Critical Reading + Math SAT: Math, Critical Reading, Writing, Math+Critical Reading College Characteristics Number of AP STEM courses accepted for credit Number of AP non-STEM courses accepted for credit Total credits accepted at time of admission Total STEM courses Total STEM units Total Non-STEM courses Total No-STEM units Class level Dorm Resident Intermural Sports Participation Fitness Class Participation Honors College Women in Science and Engineering Educational Opportunity Program Stony Brook University Math and Writing Placement Exams College of student’s major or area of interest: Arts and Sciences, Engineering, Health Sciences, Marine Science, Journalism, Business Major Group: business, biological sciences health sciences, humanities and fine arts, physical sciences and math, social behavioral science, engineering and applied sciences, journalism, marine science, undeclared, other Major type: declared major, undeclared major, area of interest High DFW Rate Courses: enrollment = 70, percent DFW =10% Total high DFW STEM units Total high DFW non-STEM units Highest DFW rate among the DFW Courses in which the student is enrolled Highest DFW rate among the DFW Courses in which the student is enrolled Proportion of freshmen in a student’s highest DFW rate STEM course Proportion of freshmen in a student’s highest DFW rate non-STEM course Type of math course: high school level, beginning calculus, sophomore or higher math Financial Aid Measures Aid disbursed in the Fall 2014 – Spring 2015 academic year Total grant funds received Total Loans recorded by the Financial Aid Office Total scholarship funds received Total work study funds received Total athletics aid received Athletic aid, grant, loan, PLIS loan, subsidized/unsubsidized loan, scholarship, work study, TAP, Perkins, Pell indicators Adjusted Gross Income Federal Need Federal Expected Family Contribution Dependent status Services/Learning Management System (LMS) Advising Visits/Tutoring Center Usage Tutoring center appointment no shows Number of STEM Course Center Visits, weeks 1 to 6 Number of non-STEM Course tutoring Center visits, weeks 1 to 6 Advising Visits during week 1 Advising visits during weeks 2 – 6 Course Management System Logins F14_Stem_Login_N F14_NonStem_Login_Week1_N Non-STEM course related logins during weeks 2 - 6 Non-STEM Course related logins during week 1 STEM Course related logins during week 1 STEM Course related logins during weeks 2 to 6 Number of STEM course logins per STEM course using the CMS, weeks 2 – 6.