Data is present everywhere. From the moment you wake up to check your phone to what you ask Alexa or Siri. In the U.S alone, households consumed an average of 268.7 gigabytes (GB) of data a month. During the school day, 78 percent of students use technology devices, according to eSchool News. Universities can harness this data in new and unique ways. It can help campuses develop capabilities and systems to serve students with personalized messages and support. Additionally, with predictive analytics, it can also lead to addressing factors that can help increase the chances of student success. Yet with all this data, campuses need to know how to use this data in a way that promotes effective and efficient practices, while addressing the needs of the students’ experiences and security as their starting point. In this article, we’ll go over best practices that can help universities enable student success and serve a diverse population.
Predictive Analytics
Many higher ed institutions are seeing the benefit of analyzing student data to improve the quality of services they offer. Analyzing past student data to predict what current and prospective students might do has given higher ed institutions more targeted recruiting and use of institutional aid. InAnalytics in Higher Education: Establishing a Common Language, Hawkins, and Watson caution that, “analytics is not a one-size-fits-all endeavor and that one has to consider that analytics is a goal-directed practice.”Ideally, vendors can facilitate the ethical use of data all of the way through the student life cycle. Vendors can help ensure that data is complete and integrated correctly to diminish the chances of misidentifying students. They can be transparent about their algorithms and test them for disparate impact on student populations. They can be flexible with permissions and use reasonable security protocols to help preserve student privacy and security. They can train staff on the correct interpretation of data and on the dangers of implicit bias.
It’s no secret that there is implicit bias in many algorithms and AI. According to a recent MIT article, “Bias can creep in at many stages of the deep-learning process, and the standard practices in computer science aren’t designed to detect it.” For example, some predictive algorithms use factors outside of the student’s control, such as ZIP codes, race/ethnicity, and their high school. If a university uses these factors for admissions or other student support services, an underserved student may find themselves further excluded because of the structural inequalities in those data points. It is important for university officials to be hypersensitive to these algorithms to reduce the amount of bias that is present. This can be done by having an internal analytics team or having technology providers that are constantly refining algorithms to ensure equal treatment and access for all students.
Train Staff
The effective use of data can help campuses make a shift from an institutional mindset to one that is student-serving. University faculty and staff need to be trained to have a deeper understanding of the data and conversations with students based on the data insights. Some conversation topics that might come up with these data insights are financial aid, life situations, and academic issues. For example, if a student was flagged as at risk for drop out due to low grades, an adviser needs to raise the issue and help the student find assistance with tutoring.
Give the Students an Option
With data breaches happening every day, it is important to give the students a choice to opt-in or opt-out of providing data to universities. It is also important to educate students on how campuses are using their data as well. Some students may not want to share information regarding mental health or location tracking.For example, Sacramento State University has a pilot program that collects data from students who opt-in that tracks what services freshman students are using. The data being collected will be used to improve the first-year experience without any personal identifiable information.
Create a Process to Analyze the Data
As with data collection, if there isn’t a set process in place to go over the data, look at insights, and make executable plans, then there’s really no reason to collect data. It is important to bring all the data pieces together to create a truly holistic student view. Whether universities create processes in a business process management platform or institutions create committees, there should be something in place to collectively make decisions on the data. For example, the University of South Florida (USF) created a Persistence Committee:
“USF built its student success initiative on the principle that every student admitted to the university will succeed when given the opportunity to do so…. We had to find ways to use "live" data to identify struggling students—in real-time—so that we could provide them with the support they needed in and out of the classroom. We couldn't wait for mid-term grades.”
Using predictive analytics and home-grown tools, USF was able to raise its retention rate from 88% to 91%.
Conclusion
While there is no standard frameworks or industry-wide structures in place to properly govern student data, these best practices can help universities stay on top of their game. By focusing on being student serving, universities can continue to see success in retention rates, student involvement, and more.
