Admissions: How a US university is tackling Big Data

News

  • Author: Carlos Alberto Gómez Grajales
  • Date: 30 October 2013
  • Copyright: Image appears courtesy of iStock Photo

Welcome to the 21st century, the age of Big Data. We all know (or should know) how popular this trend has become, the idea of gathering and constantly updating and analyzing data, usually with predictive or classification goals in mind. In recent months all sorts of companies, from Wal-Mart to VISA, are now embracing this particular use of statistics and, well, in order to keep the party going, another success story was recently revealed. This time, it is about the implementation of models in a company that, using predictive statistics has not only increased profits but it has also improved its customer service. This time, statistics are being used by the Wichita State University.

thumbnail image: Admissions: How a US university is tackling Big Data

Actually, this story started a few years back. It was in 2011 when the University started using predictive analysis for improving their recruitment process. At that time, they partnered with a software company to develop a model that would allow the admissions department to determine in advance, who would succeed in college, who would fail, and who would need an extra push to achieve his/her goals. All this based on information gathered before the courses started.

David Wright, the University's Associate Vice President for Academic Affairs is the man responsible for overseeing these vast amounts of data. This information, that includes not only academic history but also economic and family topics, is being routinely used to track and predict the student's performance. Wright mentions that, at first, the motivation was mainly economical: “The admissions and student services teams were seeking to move to evidence-based decision-making when it came to analyzing student success and course costs.” Apparently, in the old way of analyzing admissions, several variables were being left out, particularly drop-out rates, which are generally hard to predict. Being a business, the University does not only need to attract new enrollments, but also to retain its current level of students. Since the former tools were not capable of managing this information, Wright and his team considered an investment on a more complex statistical model. “We wanted to know the revenue and costs analyzed by department, student and course, and then forecast the figures for the next semester. If we could add student retention rates to the model, it would help us assess the likely budget for new programs or investment projects”.

The new model uses information regarding a student’s academic and family history to predict his future achievement in the University. Some of the variables included in the model are the students' paper grades, the number of hours enrolled, whether they work full-time, part-time or not at all and the amount of financial assistance they receive from their family. Some of the history details of the students were registered at the moment of admission, with some extra details added later to each university profile. Based on this updated information, the University is now able to detect possible drop-outs or students with a high risk of lower performance. With these considerations, students and courses would be matched, based on personalized suggestions that now consider an expected future performance. This way, an adviser may suggest that you skip certain courses or else postpone them until you have a lighter burden at your work, considering your previous academic history. This would, in time, lead to fuller courses and greater student success.

A few days back, the Wichita State University presented a summary of some of the results from this innovative approach. According to the University's data, this recruitment model had a 96% accuracy identifying “high-yield” application prospects. This is a marked improved compared to the 82% accuracy the University used to have before implementing the model, when they relied on external consultants that helped the admissions department to classify the expected results of the students.

Finally, I thought it would be worth to mention some interesting details and insights that David Wright shared on the process of involving every department with the statistical team. It is a testament to both the importance of Statistics and of the barriers that appear when an analytical approach is integrated in the daily work: “Implementing analytics requires getting in people's business. Many units on campus that are overburdened or have low resources could see it as a burden. It means more requests for work, strangers showing up at their meetings, etc. But in terms of admissions, I saw they needed some information that they couldn't get their hands on through external consultants, so I told them that if I worked with them I'd help them find benefits.” Considering the early success of this recent enterprise, I bet every member of the University is now glad Wright got into their business.

Related Topics

Related Publications

Related Content

Site Footer

Address:

This website is provided by John Wiley & Sons Limited, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ (Company No: 00641132, VAT No: 376766987)

Published features on StatisticsViews.com are checked for statistical accuracy by a panel from the European Network for Business and Industrial Statistics (ENBIS)   to whom Wiley and StatisticsViews.com express their gratitude. This panel are: Ron Kenett, David Steinberg, Shirley Coleman, Irena Ograjenšek, Fabrizio Ruggeri, Rainer Göb, Philippe Castagliola, Xavier Tort-Martorell, Bart De Ketelaere, Antonio Pievatolo, Martina Vandebroek, Lance Mitchell, Gilbert Saporta, Helmut Waldl and Stelios Psarakis.