Identifying At-Risk Students Before It’s Too Late

dice

Colleges and universities expend enormous time and resources trying to figure out how to fix one of their most critical challenges: identifying at-risk students.  Unfortunately, schools are discovering that by the time an at-risk student is detected through traditional means, the student is often beyond help and in jeopardy of dropping out.

Though many universities and colleges recognize the need to improve retention rates by salvaging the at-risk population, most continue to take the same approach as they have in the past.  In addition, many fail to use readily available data.  As a result, schools are now scrambling for new ways to recognize potential at-risk students earlier and at the same time, improve school retention rates.

There may be a solution to this crisis – and it’s one institutions have at the ready, though few are applying it and fewer still are even aware of it.

Identifying At-Risks Students

Many factors help to detect at-risk students. A student who misses class on a regular basis, a student who hasn’t done well on the first assessments in a course, a student who isn’t interacting with peers or activities on campus, or a student who is having difficulty paying for college can all be indicators that the student is at-risk.  Once a flag is raised for a student being at-risk, institutions use different ways of reaching out to that student to help reengage them to the course or institution’s atmosphere.  This could be done by the professor of the course that the student is struggling in, by a resident assistant of the dorm if the student lives on campus, or by the student advising team.   The goal is to reach out to the student and assist them so that they can become successful.

Universities and schools are now realizing that the greatest potential for identifying probable at-risk students is data analytics. This data, gathered during the recruitment and admissions processes, combined with a few other outside processes, enables institutions to isolate at-risk students early enough to make a difference.

Even though most universities have mountains of data, few understand how to use data or even how to employ technological solutions to assist with at-risk identification and analysis.  For example, some institutions have begun using Customer Relationship Management (CRM) programs to log student interactions, such as a student calling the institution to figure out last minute details around financial aid or housing. These logs can be used to flag potential at-risk situations.  However, while most institutions are monitoring this activity few are taking any action with the data results.  What’s more, a lot of schools and universities that do analyze the data are simply reacting to at-risk student scenarios rather than heading off the risk before it is too late.

Some institutions are going a step further and using CRM to gather data from Student Information System (SIS), financial aid departments, and learning management systems (LMS) and then compiling all of the gathered data into one central location.  This approach brings a wider understanding of the student’s circumstances and gives the institution a better chance of reaching out to the student in the early stages of risk.  The key to the success of this data compilation is that it must be gathered and analyzed quickly in order to reach the at-risk students in a timely manner.  Early detection and taking appropriate measures is critical as most at-risk students drop out of college within the first year.

Applying Business Intelligence

The answer to early detection and proactive measures may be the application of business intelligence (BI) during data analysis. BI can assess factors, build predictability models, and help with early detection.

According to a custom research brief from the University Leadership Council, at-risk students can be grouped into three categories:

  1. Students who encounter academic challenges
  2. Students who do not engage socially in the campus community
  3. Students who encounter financial challenges

Through the use of BI, institutions can develop predictive models for these groups. For example, analysis can determine certain factors, such as the percentage of students who are struggling academically. BI can also include social aspects as well as a demographic understanding of the student population, presenting the institution with a broader picture of at-risk factors. Armed with these probability models, the institution can react more quickly and engage a more holistic approach to helping students succeed.

Incorporating socio-economic factors into data gathering offers even greater insight about the student population as a whole while providing a framework for proactive measures to increase retention. If, for instance, the institution is able to identify a certain percentage of commuter students who are not regularly attending class, the institution can use this data to find motivating factors to increase students’ participation. This data can then be used to develop activities both in and out of the classroom. Measures such as these may ward off at-risk factors.

Making Use of Early Detection

Using data analytics on at-risk factors has additional benefits as well, because BI has an even more aggressive step in increasing student retention.  Some institutions are developing early warning systems by going beyond reactive retention practices of gathering and analyzing data. One case study at Indiana State University outlines a solution for increasing retention by not only integrating CRM with LMS and SIS systems, but with social media data as well.  This combination helped them develop a true early warning system for at-risk students.   (For details, check out increased freshman student retention to find out more on how Indiana University’s system works.)

Below is a table of retention rates since Indiana State University put this new system in place.  This table shows retention rates for new fall freshmen that returned to school after their first year was completed.

Retention Rates for Indiana State University

2005

2006

2007

2008

2009

2010

2011

2012

66.44%

67.41%

63.93%

62.84%

61.82%

55.45%

60.53%

63.46%

Student retention will always be a critical factor for colleges and universities.  The more successful their students are, the more attractive the institution becomes.  As time goes by, for institutions to have a real way to identify at-risks students, they must put in place a solid integration of data, early warning systems, and processes to identify and engage the at-risk students before it is too late.