UL’s Dr Jason Power discusses how self-efficacy and teaching formats influence how students think and behave in STEM education.
Dr Jason Power is an associate professor in the School of Education at the University of Limerick (UL).
While his academic career began with an undergraduate degree in engineering teaching, he says he was drawn to the human side of the discipline early on through a seemingly simple question; why do some students thrive in engineering programmes while others that are equally capable struggle to persist?
A PhD and a research fellowship later, and Power now teaches and researches in the field of engineering education, intent on exploring how people learn, and how that process can be made more effective and inclusive.
‘Knowing how feedback influences motivation can guide how we assess and support students’
Tell us about your current research.
My current research sits at the intersection of cognitive psychology, pedagogy and engineering practice. At its core, it’s about understanding how students think, believe and behave in STEM learning environments. This includes investigating factors like spatial reasoning ability, self-efficacy (a person’s belief in their capability to succeed in specific tasks), and the influence of teaching formats – whether in-person, online or hybrid – on student outcomes.
Much of my work is evidence-based and data-driven. For example, I use surveys, psychometric tests and performance assessments to capture a nuanced picture of student learning. I’m particularly interested in longitudinal data, which allows me to see how a student’s confidence, skills and attitudes evolve across the course of a degree programme. One recent project tracked the sources of self-efficacy among first-year and final-year engineering students, revealing how their self beliefs shift as they gain more experience and tackle more complex projects.
Another strand of my research looks at project-based and problem-based learning. These approaches are widely celebrated in STEM education, but the details matter. Different students respond differently, and the mode of delivery (online versus face-to-face) can change the experience significantly.
Why is your current research important?
At a time when STEM skills are in high demand globally, universities face the dual challenge of broadening participation while maintaining high standards. We can’t address that challenge without understanding what drives student success. My research provides evidence that can inform how programmes are designed, how teaching is delivered and how support structures are built.
For example, if we know that self-efficacy is a major predictor of persistence in engineering, then we can actively design experiences, especially in the early years, that boost self-belief as well as competence. If we know that certain groups of students have lower spatial reasoning skills on entry, we can embed targeted interventions to close those gaps before they impact performance.
In short, this work moves beyond ‘one-size-fits-all’ teaching. It equips educators with actionable strategies grounded in data and theory, helping to ensure that more students not only enter STEM fields but succeed in them.
What impacts can psychology have in educational environments?
Psychology offers powerful insights into how people learn, remember and stay motivated. Concepts like cognitive load, metacognition and self-determination theory might sound abstract, but they translate directly into classroom practice. For example, understanding working memory limitations helps in structuring complex engineering problems so they don’t overwhelm learners. Knowing how feedback influences motivation can guide how we assess and support students.
Motivation, in particular, is central. A student who believes they can succeed is far more likely to persist when facing a challenging assignment. Psychology also helps us recognise the diversity of learners, not everyone arrives with the same prior knowledge, confidence levels or learning preferences. By integrating psychological principles, we can create more adaptive, responsive environments that cater to different needs without compromising the high standards required in the engineering profession.
‘Modern STEM environments excel when they balance authenticity and accessibility’
What are your thoughts on modern STEM learning environments? How can they be improved?
STEM education has evolved rapidly in the past decade, driven by technology, industry needs and the pandemic. We now have a greater mix of online, hybrid and in-person learning than ever before. This flexibility is valuable, but it also raises questions about quality and equity.
I think modern STEM environments excel when they balance authenticity and accessibility. Students should engage with real-world problems and industry-relevant tools, but they also need accessible resources, clear scaffolding and timely feedback.
Improvement comes from intentional design. There are many recent developments that show considerable promise. Using blended models not just for convenience, but to capitalise on the strengths of each format (hands-on labs in person, theory discussions online, pre-recorded lectures etc). Embedding early interventions to develop core cognitive skills like spatial reasoning and problem-solving. Making diversity and inclusion a central design criterion, so that activities and assessments don’t unintentionally disadvantage certain groups.
None of these improvements come easy. Educators need supporting research and evidence to achieve these ambitious goals.
Through your investigation of performance factors in STEM education, have there been any particularly surprising or striking discoveries?
One of the most striking findings has been the persistence of self-efficacy as a predictor of performance, even after controlling for prior grades or aptitude measures. In other words, two students with the same baseline ability can have very different outcomes simply because one believes more strongly in their capacity to succeed.
Another surprising insight came from comparing online and face-to-face project-based learning. While the assumption might be that online formats reduce engagement, the reality is more nuanced. Some students actually flourished in online settings, especially those who valued flexibility or who found in-person group work intimidating. The takeaway is that learning modes shouldn’t be judged in absolute terms, the key is aligning them with the needs and strengths of the learners.
What do you think lies ahead for the education landscape?
I think we’re heading toward a more personalised and data-driven education system. Advances in learning analytics will make it possible to give students real-time feedback on their progress and tailor support to their individual needs. This could be transformative for retention in challenging fields like engineering.
I also hope we’ll see stronger integration between academia and industry, not just through placements but in co-developed curricula and shared problem-solving projects. This keeps learning relevant and prepares graduates for the rapidly evolving demands of the workplace.
Finally, I’d like to see a continued shift toward valuing the process of learning, not just the outcomes. Resilience, adaptability and curiosity are just as vital as technical competence, and they’re qualities we can nurture if we design our courses with them in mind.
Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.
