November 25, 2025 | By Krystal Thomas and Arif Rachmatullah

When we talk about student success, we often focus on outcomes like grades or persistence. But behind those outcomes are the skills, mindsets, and affects that students bring to—and build during—their school experiences. Skills like self-directed learning, problem solving, and adaptability are central to how students navigate their education.

The challenge is that these skills don’t show up the same way for every student. Two students both may be “self-directed,” but one demonstrates it by planning every detail of their semester while another shows it by confidently seeking out new resources when challenges arise. If we focus on individual student characteristics, these important differences get lost.

That’s where person-centered analyses come in. By “grouping” students according to patterns in a set of their characteristics, including skills, affects, and experiences, person-centered analyses give us a nuanced picture of the multitude of ways students approach learning.^1,2

Understanding Students: Person-Centered Analyses vs. Variable-Centered

Education studies have traditionally used variable-centered approaches such as correlation and regression analysis. These approaches examine how different variables—each measuring a single student attribute—are related within a population. These approaches are often used to quantify the role of individual variables, such as student motivation and student academic achievement, in predicting an outcome, such as graduation.How do person-centered analyses differ from this approach?

Let’s consider this example: A new college initiative wants to understand how self-efficacy, mindset, and use of learning strategies relate to student success.

Both person-centered and variable-centered approaches answer the question How do student skills, mindsets, and strategies relate to success? but from two different angles. Take a look below:

In this example, the person-centered analysis shows students’ success depends on the configuration of different skills and mindsets—the particular mix of self-efficacy, mindset, and strategy use—rather than the average effect of any one skill considered in isolation. While traditional regression models can test how individual predictors relate to outcomes (and even whether those relationships vary across groups or levels), person-centered analyses identify patterns of traits within individuals to reveal distinct profiles of students who approach learning in different ways. These profiles offer a holistic way to describe student differences—capturing patterns of different attributes that together shape learning—which open new possibilities for practice and design.

Why person-centered approaches, specifically student profiles, matter

Prior studies show that person-centered analyses can reveal subgroups of learners who use strategies in very different ways. Educators and designers can use these learner profiles to match strategies and interventions to student needs.^3,4

For instance, imagine an analysis that identifies three profiles of online learners:

1. Self-directed navigators: high self-efficacy, proactive help-seeking, and consistent time management
2. Motivated but overwhelmed: high motivation but low confidence and irregular study strategies
3. Disengaged: low motivation and limited self-regulation

These profiles aren’t labels or judgments—they represent patterns in how students approach learning. They make it easier to design targeted supports and interventions, whether in classrooms, online courses, or learning technologies. A faculty member might use them to decide who would benefit from additional scaffolding or resource support. A researcher might use them to gather insights into whether an intervention helps some learner types more than the others. For an edtech designer, they might use them to tailor feedback tools for each profile.

In addition, profiles can be used to understand and highlight the role of cultural and contextual factors, reminding us that learning strategies do not look the same across all students. For example, when testing a new intervention, profiles can reveal its affordances for different learners, showing not just if it works but for whom. By embracing this broader view, profiles become tools for personalization and more responsive learning environments.

Coming up next: Student profiles in action

In our next blog post, we’ll move from the “why” to the “how.” Drawing on new findings from Postsec Collab—a research and development center promoting postsecondary self-directed learning skills in online settings—we’ll show how we create student profiles from data and what they look like in practice.