The Stories Behind Data Points
A few years ago, while working within a major Canadian University, I lead the development of a project called Cohort. In essence, we were trying to understand how students were navigating their first year at this university and how we could help them succeed.
As we gathered more and more data, we started conducting interviews and I took a stronger interest in the stories that students were telling, both about their journeys before coming to university and how they were experiencing their first year. It was so diverse and fascinating.
This reminds me of a talk by Jer Thorp (a fascinating leader in how to extract stories from data) where he talks of how data is fundamentally human.
Of course we always need to understand how well our data reflects reality, but in the end, there are fascinating stories that unveil themselves behind those data points.
And that’s kind of why I got interested in customer analytics. The Cohort project made me realize this kind of analytical work was fundamentally human and could reveal a little bit about how we behave. And that’s what I aim to always better understand when working on any sorts of customer analytics projects.
On Student Socialization
This provides a bit of context to the little experiment I ran lately.
As a new HigherEd student, your socialization process changes drastically. Not only does socialization changes, but it also accelerates. It can contribute to either making or breaking how successfully you will manage your integration to this new life.
Students multiply connections through multiple circles (family, friends, co-students and co-workers), which themselves provide opportunities for new encounters. It’s an exponential phenomenon.
Patterns of Friendships
For this experiment, I built a cohort of students that I think are attending the same university and I’ve started gathering a few weeks worth of their tweets. I then extracted the posts which mentioned one other member of that same cohort. And that’s how we’ve established friendships (or sometimes feuds).
I then had fun looking at patterns of friendships, how they evolved and what the tweets were about. As I just mentioned above, some of those discussions are actually not that friendly. Just like any relationships, some of them just start off on the wrong foot and sometimes just end abruptly.
So on August 2017 for example, those were the most prolific friendships (I should call them relationships) we’ve seen with samples of their discussions…
A Signal In Your At-Risk Students’ Monitoring
Why is this of interest? Well as anyone that read their fair share of Tinto’s work on student retention knows, socialization is such an important factor in how a first-year student copes with the transition to HigherEd. And with the multiplication of “at-risk” predictive tools out there, you have to take into account all signals that could inform you into how a student is living that transition.
Of course social media data doesn’t tell the whole story, as many students are just not really present on social media. But a warning signal could just be a sudden drop in social media socialization. With a more sophisticated automated analysis of the social media data that’s coming in, an institution could add a good signal to its monitoring system of at-risk students.