A leader in higher education suggests a clearer view of all that data.
GUEST COLUMN | by Terry Mills
In the wake of the recent college admissions scandals which stunned the country, people are wondering how this fraud went undetected. Prosecuted by the FBI and currently under investigation by the U.S. Department of Education, the multiple scams focused on securing student admissions to elite schools, regardless of their academic or athletic abilities. The methods of fraudulent admission, stewarded by a third-party organizer, included bribing exam administrators to facilitate cheating on entrance exams and the fabrication of sports credentials.
Choosing who gets into college is no small task
Millions of students apply to college each year and entrance has become an increasingly competitive and stressful process for students and admissions staff alike. Admissions administrators have a multi-faceted role – marketing their institutions to attract the best students and maintain academic excellence, evaluating new enrollment markets, and supporting students through the admissions process.
‘The color, shape, size and position of the various geometric shapes provide insights at a glance and allow administrators to compare students, identify patterns, and find the proverbial needle in the haystack by combining data from multiple sources in a way that brings hidden insights to the forefront.’
A critical asset in this process is student data. Data-driven insights give admissions staff vital information about a prospect, applicant, or student, such as whether they’d have the attributes of a great candidate or how they perform compared to others in the candidate pool. This raises two important questions. Armed with massive amounts of student data, how did potential anomalies in numerous college applications go unnoticed and how can it be prevented from happening again?
Data challenges plague admissions teams
Most admissions administrators are not data analysts. Unless they’re looking very specifically for anomalies, even the most diligent staffer can miss that lone fraudulent application among the thousands that cross their desks. Another challenge for administrators is that the data they depend on is often siloed in emails, spreadsheets, databases, CRM systems, and PDFs.
The struggle to collate and analyze that data makes it hard to catch anomalies like major jumps in test scores or applicants with no sports background who mysteriously receive a tennis scholarship.
With “modern” day bar and pie charts dating back to the 1700s, a new approach to data analysis clearly is needed. One approach involves pulling all known data about a student, the good and bad, into one holistic view. Based on two decades of government and academic research, this approach is allowing enrollment teams at many top universities to quickly understand their candidate pool and see, not only obvious candidates that deserve a spot, but diamonds in the rough.
Additionally, visualizations that use this approach can make it much easier to spot anomalies in student data that traditional methods would likely miss.
How do holistic visualizations work?
Instead of pouring over charts and graphs, a holistic visualization, sometimes called a glyph, after the multiple dimensions of hieroglyphics, are known to tap into the human mind’s inherent ability to recognize patterns, shapes, and colors better than it does numbers. Imagine a three-dimensional pictograph of geometric objects and data elements (e.g. academic, extracurricular, and personal information) that represents each student. So instead of looking at spreadsheets or scattergrams, data is transformed into easily understood visual chunks that enable teams to quickly analyze and act on vast amounts of data.
The color, shape, size and position of the various geometric shapes provide insights at a glance and allow administrators to compare students, identify patterns, and find the proverbial needle in the haystack by combining data from multiple sources in a way that brings hidden insights to the forefront.
For example, admissions officers can start off by merging straightforward data into the platform such as gender, ethnicity, test scores, extracurricular information, and other binary indicators. These are represented by different colors. A blue ring on a glyph may indicate that an applicant is an athlete, green indicates that he/she is a straight-A student, and so on.
From here, it becomes easy to quickly assess that individual in the context of the entire student population and spot potential inconsistencies in student credentials. Within moments of scanning admissions data, data-savvy and novice users alike can easily see where performance or athletic achievements fall outside the norm.
For instance, if a student has a SAT score of 1,500 yet their academic record shows a history of mid-level grades and low attendance, this may be something to explore further. Is there an error in the data? Or is something else afoot?
Quickly sift through a sea of applicants
A perennial challenge for admissions teams is that they must review a huge field of applicants. This can translate to a sea of glyphs. A key goal of glyph-based visualization is to democratize data so that everyone can understand data like a data scientist – no matter how big the data pool is. Legends can be built into the glyph software to support ease of analysis. Individual elements can also be turned on and off and users can easily drill down to a single glyph for more granular insights into potential red flags.
Glyphs can also be enriched with third-party data sources such as information from college admissions consultants. Testing companies can also use this approach to validate scores before they are sent to schools.
By filtering and isolating data it becomes easier to ask questions of that data, such as “why is one student excelling at basketball yet their attendance records are poor?” or “how come a student’s application includes a photograph of them on the crew team, yet there’s no official record of them participating in the sport?”.
Using the power of a glyph, admissions teams can quickly determine whether to probe further to identify the real student buried in the application material or refer the candidate to the faculty or admission committee for further review.
It’s not about what you see, it’s about what you can’t see
With the ability to view all their student data in a single view, college admissions teams can quickly uncover atypical applicants, even fraudulent ones, that would normally fly under the radar.
Armed with these insights, they can conduct deeper background checks and possibly avoid falling victim to future cheating schemes, focusing instead on what they do best – steering students through their higher educational experience.
Terry Mills, Ph.D., has spent the last 18 years in higher education, serving in leadership roles within many collegiate institutions to help them to implement and improve diversity and inclusion. He serves as Assistant Provost for Diversity and Inclusion/Chief Diversity Officer at John Carroll University and manages the First in the World Grant program for JCU. Contact Terry through LinkedIn.