How can data be used for more than merely "reporting"?


Jacqueline Bichsel: Data can be used to explain, predict, and provide insight into complex issues that may involve multiple factors, such as student learning. Explanation and prediction elevate the value of data beyond mere reporting. Using data to explain and predict allows for intervention, prevention, change, or some other action that facilitates strategic transformation or decision-making.

Could you please give some concrete examples from American higher education of how learning processes have been positively affected by Analytics?


Jacqueline Bichsel: Examples are now available by the dozens (check EDUCAUSE Review Online for more), but here are a few recent ones illustrating how analytics can be used to increase student success.

• Researchers at the University of Wisconsin-”Milwaukee used student behavior from a course management system (e.g., quiz attempts) along with other information to show that a particular form of instruction (U-Pace) was associated with greater academic success.
• Paul Smith's College used demographic information and analytics to target at-risk students before the semester started for early alerts and academic support. This intervention increased student retention and grades.

How would you explain the significance Analytics has gained in the American educational landscape? Were there any precedents? For example, was Analytics previously used extensively in corporate training?


Jacqueline Bichsel: The importance of analytics for corporate success certainly predated and still mightily outpaces the use of analytics in higher education. Corporations have long been using vast amounts of data collected on consumers to predict interests, tastes, and spending behavior. Since competition has increased among American universities to attract and retain students, analytics has now gained a foothold in predicting student learning behaviors.


The more administrators, faculty, and other higher education professionals are exposed to examples of analytics being used to increase retention, graduation rates, engagement, grades, and other markers of student success, the more they desire and seek to create a culture of data-driven decision making on their own campuses.

Do you know whether the use of Analytics has required any type of attitude change or the acquisition of new skills on the part of university decision-makers?


Jacqueline Bichsel: There have been many examples of analytics "wins" that have changed attitudes among university administrators, faculty, and other staff. The examples mentioned in the previous question constitute such wins. A win helps to change the culture of a university from making gut-driven decisions to making data-driven decisions. To institute this change, one needs to convince decision-makers that the use of data results in higher success rates than the use of intuition or tradition in an area that has been deemed strategically important (e.g., student persistence).

The audience for the results produced by analytics is broad and may include administrators, faculty, and other staff who are not necessarily skilled in data interpretation or statistics. It would not be practical to expect all these individuals to undergo training to understand the sometimes sophisticated analyses that are involved in analytics.


Part of a good analytics strategy is having analysts with the know-how to present results in such a way that it does not require new skill development on the part of university decision-makers. This may involve enabling current analysts with additional training (e.g., in data visualization) or hiring new analysts who know how to present results that are understandable to decision-makers not trained in analytics.

What important factors or actions have contributed to the success of Analytics?


Jacqueline Bichsel: Although most respondents (approximately 85%) in the 2012 ECAR Study of Analytics in Higher Education considered analytics to be important for higher education's success, and although nearly all were collecting data that could be used for analytics, far fewer (around half) were actually using their data proactively or to make predictions.


Moreover, those using analytics were focusing on only a few targeted areas: enrollment management, finance and budgeting, and student progress. Only about 1/3 were using analytics applied to student learning. Analytics has not yet reached the level of successful employment in American higher education that it has enjoyed in corporate America. However, analytics is gaining in popularity in American universities and most of our study's respondents foresee that analytics use will increase in the coming years.

We found that those higher education institutions that are successful in analytics have achieved maturity in the following areas:

1. Culture/Process (having senior leaders who are committed to using data to make decisions and having a process for moving from what the data say to making strategic decisions)
2. Data/Reporting/Tools(having the right kind of clean, quality, standardized data; having reports in the right format to inform decisions; and having the right tools and software for analytics)
3. Investment (having the appropriate amount of funding and number of analysts for analytics)
4. Expertise (having professionals who are trained in and know how to support analytics)
5. Governance/Infrastructure (having information security policies and practices that safeguard the use of and access to data for analytics; and having sufficient capacity to store, manage, and analyze increasingly large volumes of data)

Have there been any widespread obstacles or barriers?


Jacqueline Bichsel: The largest barrier to analytics is affordability. Many institutions have not begun analytics programs out of fear they cannot afford them. However, increased competition among higher education institutions means that they cannot afford not to invest in analytics. Institutions will not make progress in analytics until they view analytics as an investment rather than as an incremental expense. In addition, what is needed most for analytics is not expensive tools, but qualified, trained professionals. There are plenty of examples of institutions with few resources using analytics to make strategic improvements, including the Paul Smith's College example cited previously.

Other barriers in adopting analytics include concerns about the quality of and misuse of data and concern that the university culture will not accept the use of data to make decisions. However, an institution cannot afford to wait for perfect data or for a culture of widespread acceptance. Analytics programs start with important but achievable wins in targeted areas and expand when those initial successes drive a more accepting culture and better quality data and infrastructure.

In Germany, privacy regulations are extensive, well defined, and anchored in law. In the US, is it permissible to collect personal data regarding an individual's learning process without gaining prior consent?


Jacqueline Bichsel: Certain federal laws (e.g., the Family Educational Rights and Privacy Act; FERPA) regulate against making personal student information (e.g., grades,demographics) public. Universities also have policies regulating which individuals have access to such information. Part of a good analytics program is having policies in place that regulate data access and usage. These policies may differ depending on the institution.


It may be safe to say that the deeper dive you take into measurement of the "learning process", the more likely you are to need prior consent. For example, examining individual quiz grades might require prior consent, whereas examining the number of clicks made in a course management system probably would not. The degree of anonymity of the data also plays a part in whether prior consent is needed.

Has anyone ever inquired as to whether students like the approach?


Jacqueline Bichsel: Students certainly have reported they like the degree of personalization that results from analytics use. It is important to point out, however, that students-”much like consumers-”may be unaware of the data that is collected and used in analytics. This fact raises an ethical red flag among many who suspect that analytics may be used for nefarious purposes. This is why it is important that policies safeguarding the use of data are in place and that these policies are publicized.

How do universities benefit? Do they save money because they optimize processes? Are their reputations enhanced because they "produce" better academic results? How would you characterize the advantages they derive?


Jacqueline Bichsel: Analytics can help universities better understand student demographics and behaviors, which can aid with enrollment management, student learning and progress, recruitment, and retention. Analytics can also be used to optimize the use of resources, which can help with instructional and classroom management, finance and budgeting, and the improvement of service delivery, such as IT services. Analytics can help both with lowering costs and enhancing a university's reputation with increased student success. This combination of results is what has made the use of analytics such an important and ubiquitous topic in the higher education arena.




Read more in: Jacqueline Bichsel, 'Analytics in Higher Education: Benefits, Barriers, Progress, and Recommendations,' (Research Report). Louisville, CO: EDUCAUSE Center for Applied Research (August 2012)