A Practical Application of Business Intelligence and Analytics in College Admissions

Request More Information

To learn more about an online program at Saint Joseph’s University, complete this information request form and a Program Manager will contact you to answer your questions.

Yes, I do want to communicate with Saint Joseph's University via SMS. I am providing my consent by leaving the opt-in checked. Message and data rates may apply. Privacy and Terms.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

How do colleges and universities use the abundance of data available to them in order to achieve institutional objectives and organizational mission? For example, data from a variety of applications and sources, such as geo-demographic and other data, is gathered on prospective students and can be put to use to determine, for example, the future success of students. As the following demonstrates, there are many practical applications for a Master of Science in Business Intelligence & Analytics as it relates to college admissions.

Function of the Analytics
It’s vital for universities to identifying which students will ultimately succeed and which applicants are inappropriate for admission. With an ever-increasing number of applicants, institutions must sift through large amounts of disparate applications that are obviously in neither the “accept” nor “reject” categories. By analyzing existing applicant data, it is possible to develop models that will be able to predict the success potentials of these potential students.

Overview of Enrollment Management
Over the last forty years, enrollment management has been developed to meet the changing needs of the higher education industry. An essential function of enrollment management is to ensure that tuition goals are met (by enrolling a sufficient number of qualified students and retaining those students). Business intelligence is an integral tool used in analyzing the data needed to create the appropriate marketing, recruitment, and admissions strategies to meet those goals.

Use of Predictive Analytics
In the past, data has been analyzed for both students entering undergraduate education (in the form of SAT or ACT scores) and students entering graduate education (in the form of GMAT or GRE scores, undergraduate majors, etc.) in order to predict student outcomes.

Additionally, business analytical models have been used to determine how many students will enroll in a university, based on a number of factors and data that the university has on hand. This information can then be used to make positive changes in areas such as marketing and financial aid.

Data Analysis
Data analysis is a focus of business intelligence & analytics courses. Learning to use the tools can allow educational professionals to perform powerful analyses. For example, student data at a private university was gathered from 2006 to 2008 for entering first-year students. Data sources included various applications, including the FAFSA (Free Application for Federal Student Aid). Types of data included high school quality index, SAT scores, secondary school grade point average (GPA), and others. Unique identifying numbers ensured anonymity. The goal was to discover the End-of-First Year GPA.

Decision Tree Analysis
The first analysis method reduces the number of data points through a process of applying decision rules. In addition to predicting success, this method ranked the importance of data types (with secondary school GPA and high school quality index being the two most important predictors). Reducing the number of data types streamlines the process for enrollment managers, but there is always the option to fine-tune with more variables.

Neural Network Analysis
This type of model works through the construction of structures meant to mimic the brain. The model is composed of three structures: input, “hidden” (which may contain multiple levels), and output. In this model, fewer inputs can have a positive impact on performance.

Multiple Regression Analysis
This more traditional model can both describe the predicted results, and give a level of error that points to how accurate those results will be.

Comparison of Models
The neural network was the best at predicting success, followed by the decision tree analysis (both of which are business intelligence style models). Still, the results from all three models were similar enough that a case could be argued for using any one to predict End-of-First-Year GPA.

Implications for Enrollment Management Leadership
These models are useful to the success of enrollment management in many ways. They provide a more consistent approach to evaluating applicants than the traditional method of review by committee. Models like these can even be employed before the application process during the marketing phase; prospects predicted to be less academically successful can receive resources detailing additional academic assistance, while prospects predicted to be more academically successful can receive information about honor societies. The application of this method could reduce reliance on other, more arbitrary techniques.

See this concept come to life with published research by a Saint Joseph’s University Master of Science in Business Intelligence & Analytics student graduate William Amburgey, “Using Business Intelligence in College Admissions: A Strategic Approach”.

Earning a Master of Science in Business Intelligence & Analytics can help you learn to perform analyses such as these, and provide valuable services to organizations, such as universities, that can help to balance budgets and return more consistent results than some traditional methods.

You may also be interested in

Master of Science in Business Intelligence and Analytics Curriculum
Learn More About

MS in Business Intelligence & Analytics Career Outlook

learn more
Master of Science in Business Intelligence and Analytics Curriculum
Learn More About

MS in Business Intelligence & Analytics Curriculum

learn more
Data Science versus Business Intelligence: What's the Difference?
Featured Article

Data Science vs. Business Intelligence: What's the Difference?