Using daily interactions with e-learning platforms to predict achievement

In the second part of her interview with BOLD, developmental psychologist Margherita Malanchini talks about leveraging technologies to create personalized learning environments, and how such environments might help all students reach their full potential.

Sabine Gysi: In the first part of our interview, you mentioned that leveraging technology to provide personalized learning is one possible way to improve learning. Can you please elaborate?

Margherita Malanchini: Personalized learning, leveraging recent technological developments, is an interesting avenue for promoting learning experiences and improving learning. For a long time we’ve taught to the middle! But we know that there’s an enormous amount of variation in students’ abilities , even within each classroom, and this is the case even when children are assigned to classrooms on the basis of their ability.

Teaching to the middle and failing to equip students with the tools they need to learn at the appropriate level can have profound negative consequences, including demotivation and anxiety. This happens, for example, when students are always working above their level and cannot quite grasp the concepts they’re working with.

It will take careful planning and a great deal of work to develop appropriate technology, potentially building on existing online learning platforms,  and to create systems that are responsive to students’ performance and abilities and adjust the level of their workload accordingly.

“For a long time we’ve taught to the middle! But we know that there’s an enormous amount of variation in students’ abilities.”

At some point, I hope, we will be able combine our knowledge of new technologies with what psychological and education research tells us about the factors that matter for educational success. This will allow us to create tools to help all students achieve as much as they are capable of achieving – which may be different for every student. There is no single goal for every child; instead, we want to enable all children to learn as successfully as they can.

SG: How far are we from developing sophisticated learning environments that can be adapted to different countries, education systems, and types of institutions?

MM: I think there’s still quite a lot of work to do in terms of understanding the factors that contribute to improving the learning experience at every level – not just the cognitive, but also the emotional level.

One avenue that researchers, including my mentors at the University of Texas and myself, are investigating is how students use online learning platforms. We are interested in whether day-to-day interactions with online learning programs predict differences in academic success beyond more stable characteristics such as intelligence, self-regulation, and personality.

“It will take careful planning and a great deal of work to develop appropriate technology, potentially building on existing online learning platforms.”

We’re looking at students’ behaviors as they interact with an online platform. For example, how much time do students spend interacting with a platform, and when? What do they do on the platform? How do they look for support? This may provide educators with insights into the day-to-day behaviors that matter for educational achievement, and help us identify targets for interventions.

Right now, we’re investigating all of these behaviors and analyzing the data. This project might lead to very interesting findings about how students interact with online learning tools and, in turn, allow us to draw conclusions about how to improve online learning experiences.

So to return to your question: We’re getting closer. But there’s still a lot of work to do.

SG: I understand that one of the tools you’re using to explore these questions is a SMOC (Synchronous Massive Online Course) at the University of Texas? Please tell me more.

MM: My mentor Paige Harden teaches a huge online learning class, an introduction to psychology. About 1,400 students take this SMOC every semester. Each student’s activity on the online portal is recorded. We also gather information about the students’ grades and performance throughout the semester. In addition, we ask them to complete questionnaires that tell us more about their personality, intelligence, and motivation, as all of these factors may contribute to differences in learning.

What we’re really looking at is this: Is there anything in students’ day-to-day interactions with online learning platforms that could help us predict achievement on a finer scale and in a shorter time frame, beyond the personality and intelligence factors that remain stable throughout life?

“There is no single goal for every child; instead, we want to enable all children to learn as successfully as they can.”

Examining students’ activities, we found that some are more closely related than others to how well the students do at the end of the course. For example, we discovered that the amount of time students spend navigating and familiarizing themselves with the online platform, and the amount of time they spend doing assignments, both graded and non-graded, are much more closely correlated with their final grades than, for example, how much time they spend looking for help or doing optional reading. So, from this preliminary work, we see these as potential “target behaviors” that are closely associated with how well students do at the end of the semester.

Importantly, we found that the association of these behaviors with students’ final scores was meaningful beyond such stable traits as motivation, personality, and intelligence.

Of course, we are still in the early stages. We need to do further analyses. We need to go into much greater detail in identifying the relevant factors and how they predict achievement. But once we know more, we will likely have the ability to intervene on a day-to-day basis.

The hope is that one day, we will be able to provide educators with the tools to intervene in everyday classroom experiences. A further hope is to contribute to progress in the e-learning sector, consequently making education more accessible to everyone even beyond geographical boundaries.

In the first part of our interview, Margherita Malanchini explained why curiosity, creativity, and self-confidence are more important for a child’s success in reading and math than commonly assumed. She also discussed recent findings about the role of genes in educational attainment and the implications of this research for future interventions.