With 160,000 students at 45 campuses and sites throughout Indiana, Ivy Tech Community College generates a lot of data—about 100 million new records per day. This forward-thinking learning institution wanted to harness the power of its data to buck an intractable nationwide trend: high college dropout rates.
Ivy Tech wanted to use predictive analytics to spot the warning signs of failure, to intervene early, and to get students back on track. With too few IT resources and no self-service analytics, reports took too long to run, and the data quality was questionable. The college needed a data warehouse and a new approach to big data management. The goal was to create a data democracy, a unique concept to empower data-informed decision-making, accelerate institutional processes and speed innovation.
Single View of Data with Self-Service Analytics Helps Small Team Do Big Things
The college selected Pentaho’s data integration and business analytics platform as a solution. As an open source-based platform with a web-based user interface that easily embeds into existing applications, Pentaho provided both the flexibility and self-service analytics functionality, without costly desktop licenses.
Pentaho powers “NewT” (for the “New Thing”), a system of curated data sets across student records, financial aid, and accounts payable data. Seven analytics experts trained nearly 3,000 employees how to interact with and visually analyze data in only 6 months.
To ensure data quality, data governance practices were established and data preparation and cleansing were shifted away from the warehouse and into production. Functional areas can keep their own data sets clean, and basic reporting is easy. Users can drag and drop data elements into and out of existing templates, reports, graphs, and dashboards and share new models between users and departments without IT’s help.
Additional Funding Claimed, and Lowest Mid-Term Failure Rates in College’s History
Data democracy paid off almost immediately. One report discovered 2,800 unclaimed credentials that could add up to $5.6 million in funding, which more than paid for the entire analytics project.
In 2014, the data science team applied predictive analytics to reduce student failure rates. The team created an innovative data model and identified a data set of 16,247 at-risk students in the first two weeks of a sixteen week term, 5,118 of whom were interviewed by over 800 faculty, staff, and administrators over the next 2-weeks. The team learned that failure was sometimes associated with healthcare, transportation and childcare challenges, and these students were steered toward resources that could help.
The number of failing students at midterm fell 3.3 percent from the previous fall term. More than 3,120 at-risk students continued their studies instead of failing. It was the largest single-year drop in mid-term failure rates the college had ever recorded. Even better, student success rates for those students jumped 4 percent when successfully contacted. Higher graduation rates not only mean increased funding for the college, but benefits Indiana’s economy as a whole, since college graduates earn an average of more than 60 percent more per year than students with only a high school diploma.
As the program is refined, failure rates should continue to fall. The college recently added a new data strategist position to the chief data officer office to work on proactive applications of analytics. Pentaho is helping Ivy Tech create an innovation, data-driven culture that is outcomes-driven.
Bigdata and data center