In an era of dashboards and performance metrics, we have more data at our fingertips than ever before. Used wisely, data can help us identify students who need early intervention, track progress over time, and evaluate the effectiveness of our programming.
But data can also go wrong. Take a personal example – like many of you, I spend a good part of my week worrying about whether our graduates are going to pass the bar. In fact, right before writing this post, I was staring at the predictive analytics for my bar takers. However, I’ve found that when I’m looking at the studiers’ likelihoods of passing the bar based on the predictive metrics, I often forget to consider what I’ve known about each person since they were 1Ls. By reducing a student to their numbers, there’s a risk that I’ve overlooked their grit, context, dreams, and even the structural barriers that shape their experiences.
It's difficult to walk the line between useful insights and dehumanizing labels when faced with so much data. There are a few principles that help guide me when I find myself lost in the data weeds:
- Remember that Data is a Signal, Not a Sentence
Data can help alert us to patterns, but it cannot tell us the full story. A low GPA might suggest that a student is struggling in their classes, but it doesn’t reveal that they’re caring for a sick parent, working two jobs, or struggling with undiagnosed anxiety. Data can be the conversation starter, but it’s not the conclusion. When looking at data, ask, What might this number be telling me, and what might it be missing? We can provide our best support for a student when we understand the larger picture.
- Avoid Labels
Language matters. Referring to a student as “at risk” or a “bar concern” can shape how we see them. It can also impact how a student sees themselves. When data becomes a label rather than a tool, we risk reinforcing the very challenges we hope to address. Instead, talk in terms of support needs. Keep the focus on what the student needs, not what they lack.
- Share the Power of Data with Students
Students should never feel that decisions about them are being made without them. When appropriate, share the data you’re using and invite students to interpret it with you. This can foster metacognition, trust, and ownership. It also gives you the chance to frame the data in a more positive light. For example, instead of saying, “Your score predicts you’ll struggle,” try: “Here’s how your performance compares to past students. What do you think is working for you right now? What feels hard?”
- Use Disaggregated Data to Address Equity Gaps
Data can reveal troubling disparities in performance, bar passage, and retention. But when we treat these gaps as individual failings rather than structural ones, we miss the chance to make real change. Look for patterns across race, gender, first-generation students, and more – then ask: What systems need to be improved? Don’t limit data usage to triaging students; use it to hold institutions accountable.
- Humanize the Process at Every Turn
Behind every data point is a person. Academic support educators are uniquely positioned to bridge the quantitative and qualitative and combine insight with empathy. This means listening before diagnosing. It means resisting quick fixes in favor of sustained relationships. And it means always asking: Does this intervention honor the student's individuality and humanity?
Used well, data can help us reach students who might otherwise fall through the cracks. But used poorly, it can widen those very cracks we aim to address. As academic support educators, our challenge is to harness the power of data without losing sight of the person behind the numbers. Let’s commit to data-informed – not data-driven – support. Let’s use number to amplify human stories, not erase them.
(Dayna Smith)
