Learning Resources

As analysts, we work to create clear, quickly understood visualizations and explanations of the data in our care.  We share them through emails, dashboards, infographics, and presentations so colleagues can access the information they need.  But that, in itself, is not really a data story and may not have the impact we’d like.   … Read More

If you’ve ever refreshed a report showing student headcount “by state” and thought, “Wait… didn’t we have more students from Tennessee when I looked last month?”, you’ve run into a surprisingly common issue.  As real-world data changes, sometimes our data warehouse needs to remember those changes. Other times, our data warehouse needs to forget and simply reflect the… Read More

What Do We Mean by Data Storytelling?  During World War II, the U.S. military faced a serious problem: bombers were being lost at an unsustainable rate. To address this, analysts examined patterns of damage on aircraft that returned from missions and found a heavy impact on the… Read More

Small Sample Size Suppression: Reasons and Techniques  Imagine a campus-wide survey asking Vanderbilt students about their study habits, course satisfaction, and time management. The survey is heavily marketed as “anonymous.” However, when the final dashboard goes live, a table shows that “100% of transfer students in a… Read More

Career, Program, Plan – The Structure Beneath the Student Journey  Most people think of a student record as a collection of courses and grades. In reality, it rests on a deliberate and carefully layered structure. At its core is a hierarchy of career, program, and plan, organized… Read More

What does DSA do? The primary focus of the Office of Data & Strategic Analytics (DSA) is to ensure that Vanderbilt’s decision-makers can access data they trust to do their jobs better. We want to make data the oil that keeps the Vanderbilt engine humming, and to keep it… Read More

Requests to add additional data to a list are common and often straightforward. The information exists, the desired output is clear, and the goal is simple: connect one dataset to another and return a more complete picture.  Whether that connection happens smoothly depends on the structure of the file… Read More

Why Understanding Effect Sizes Are Important  Imagine you’re conducting a survey and want to understand how meaningful your results really are. Is this difference small or large compared to previous surveys conducted at my institution or with other populations? Enter Cohen’s d – one of several effect size measures.  What exactly is Cohen’s d? Cohen’s d is a statistical calculation that helps you compare the magnitude of differences between two groups, regardless… Read More

  Granularity and Cardinality in Higher Ed Data  “Just give me a big table and I’ll figure it out in Excel” is a sentiment often expressed to the Data and Strategic Analytics team. If only it were that easy.  Sometimes it is. You have a… Read More

  If you’re new to SQL, one of the first concepts you’ll encounter is joining data.  At Vanderbilt we have thousands of tables of data stored separately. One table might hold core student information, another demographic attributes, another course data, and another faculty assignments. The question is how those pieces connect.  If you’re fortunate, a data model… Read More