The present study examines the profiles of student agency across different courses from six faculties in a Finnish university. It also analyses the impact of learning contexts and student characteristics on profile membership. The aim is to enrich our understanding of student subgroups with respect to their agency, as well as provide insights into factors associated with diverse patterns of student agency. Our findings have implications for the development work of curriculum designs and pedagogical practices in higher education.  

Student agency is linked with learning achievement and satisfaction (Reeve & Tseng, 2011), personal growth (Yang, et al., 2023) as well as a successful transition from education to work-life (Schoon & Heckhausen, 2019), and lifelong learning (Su, 2011). Agency is seen to be critical for enabling purposeful changes in society (Bazzani, 2023), and therefore, it is included in various key educational policy documents (OECD, 2022). Recent conceptualisations of student agency emphasise the multidimensional and dynamic nature of agency, and acknowledge at the same time the interrelations between an individual and environment, and the subject’s experiences and judgments inherent in the construction of student agency (Jääskelä et al., 2023; Mameli et al., 2023). The conceptualisations have been impacted by work in several scientific fields, such as psychologically oriented research (e.g., Bandura, 2018), and social (e.g., Archer, 2003), and educational sciences (e.g., Lund & Vestøl, 2020; Matusov et al., 2016; Su, 2011). Our conceptualisation of student agency involves having access to and being empowered to act through 1) resources linked to an individual’s beliefs of one’s capabilities, and competence; available relations involving issues of power and support; participation such as opportunities to influence and get involved in addressing complex phenomena. These resources allow or support an individual to 2) take stances (willingness) to 3) engage in purposeful, intentional, and meaningful action and learning in study contexts. (Jääskelä et al., 2023; 2024).

The investigation of student profiles in the present study was grounded in three major dimensions of student agency: agentic resources, stances, and action (Heikkilä et al., 2020; Jääskelä et al., 2023, 2024; Klemenčič, 2017). Agentic resources include students’ perceptions of their capabilities to act and learn; relations with teachers and peers involving power balance and support; opportunities for participation in interactive learning, and for development and new ways of thinking (Jääskelä et al., 2023; Lund & Vestøl, 2020). Agentic stances concern students’ willingness or motivation to act and become engaged in learning, such as perseverance in challenging tasks and joint problem-solving (Klemenčič, 2017). Agentic action represents students’ purposeful, intentional, and meaningful action, such as goal-oriented learning, influential participation, and collaborating with and supporting other students (Heikkilä et al., 2020).

Previous research suggests that students’ background characteristics (e.g., gender and prior achievement) predict students’ membership in different agency profiles (Collie et al., 2024). Little is, however, known about the potential differences in experiences of agency among university students across various disciplines and courses (Jääskelä et al., 2017). To address this issue, we examined how factors such as the faculty the student enrolled in, course size, and student characteristics (e.g., age, gender, final grade from the study course, and total amount of grades at the university/study program) predict students’ membership in distinct subgroups of agency.

To sum up, we explored the following two research questions in this study: (1) Are there distinct agency profiles among university students and how are students distributed across these profiles? (2) What is the extent to which the profiles of student agency can be predicted using the information on the faculty the student enrolled in, course size, and student characteristics?

The participants comprised 1249 students (371 males; 876 females, missing data for 2) from six faculties in a Finnish university. Those faculties included Faculty of Humanities and Social Sciences, Faculty of Information Technology, Faculty of Education and Psychology, Faculty of Sport and Health Sciences, Faculty of Mathematics and Science, and School of Business and Economics. The students’ mean age was 24.8 years (SD = 7.02; range = 18–66). The students completed the revised version of the AUS Scale (Jääskelä et al., 2024) at the end part of their courses. Demographic and course information were gained from the university’s student register. The original AUS scale was comprised of 10 factors and 54 items using a five-point Likert scale (1 = fully disagree; 5 = fully agree) (Jääskelä et al., 2017). The construct validity and psychometric properties of the AUS have been supported by previous studies (Jääskelä et al., 2017; 2020). The revised version of the AUS (AUS-R) added more factors, such as transformation. In total, the AUS-R consisted of 15 factors, including nine factors for agentic resources (1. self-efficacy, 2. competence beliefs, 3. trust for teacher support, 4. peer support, 5. power balance among students, 6. opportunities for influence, 7. opportunities for interactive learning, 8. interest and utility value, and 9. opportunities for transformation); three factors for agentic stances (1. perseverance in challenging tasks, 2. willingness for interactive learning, and 3. willingness for joint problem-solving); three factors for agentic action (1. intentional goal-directed learning, 2. collaborating and supporting others, and 3. active impactful participation). Latent profile analysis (LPA) was performed in Mplus (Muthén & Muthén, 2017) to identify distinct profiles of student agency. The best fitting model was selected based on the low values of AIC and BIC, high entropy values, significant LMR test, the smallest class with at least 5% of the sample, and the solution's alignment with existing theory and its interpretability (Jääskelä et al., 2020). Fifteen standardised variables reflecting different agency factors were entered into the model as indicators. The following predictors of profile membership were tested using the three-step approach for adding auxiliary variables to mixture models (Asparouhov & Muthén, 2014): the faculty the student enrolled in, course size, and student characteristics (age, gender, final grade, total grades for university study and the degree program, and years of university enrolment).

Four distinct profile groups of student agency were identified. Profile 1 (n=139) contained students with below-average levels of agency across all latent variables. Profile 2 (n=528) included students with average values of agency. Profile 3 (n=104) comprised students with average levels of intrapersonal (e.g., self-efficacy) and interpersonal (e.g., power balance) agentic resources, and goal-related agentic action (i.e., goal-directed learning). However, those students reported below-average levels of contextual agentic resources (e.g., opportunities for influence), collective agentic stances (e.g., willingness for joint problem-solving), and collective agentic action (e.g., collaborating with and supporting others). Profile 4 (n=478) included students with above-average levels of agency across all latent variables. Covariate analyses revealed that the faculty the student was enrolled in and final grade the student obtained were most strongly associated with the profiles of student agency. Specifically, students representing the above-average level of agency profile were most likely studying in the Faculty of Education and Psychology and the Faculty of Sport and Health Sciences. Moreover, students with a high final grade were most likely in the high agency profile. Age and gender were only slightly predictive of student agency profiles. The impact of the total number of grades, course size, and years of enrolment on the profiles of student agency was minimal. Our findings indicated that most students (around 80%) reported moderate to high levels of agency, aligning with a previous study (Jääskelä et al., 2020). The results paved the way for further investigations of student agency among diverse universities in national and international contexts. Importantly, we found one profile with low levels of collective agentic stances and action. Future research is needed to explore why some students do not favour collaborative learning and how they can be supported to gain successful learning experiences in interactive learning situations (Bearman et al., 2024).

Archer, M. (2003). Structure, agency and the internal conversation. Cambridge University Press. Bazzani, G. (2023). Agency as conversion process. https://doi.org/10.1007/s11186-022-09487-z Asparouhov, T., & Muthén, B. O. (2014). Auxiliary variables in mixture modeling: Three-step approaches using Mplus. https://doi.org/10.1080/10705511.2014.915181 Bandura, A. (2018). Toward a psychology of human agency: Pathways and reflections. https://doi.org/10.1177/1745691617699280 Bearman, M., et al. (2024). Life-on-campus or my-time-and-screen: identity and agency in online postgraduate courses. https://doi.org/10.1080/13562517.2022.2109014 Collie, R. J., et al. (2024). Personal agency among students from low socio-economic backgrounds: An examination of student profiles, perceived teaching support, and achievement. https://doi.org/10.1007/s11218-023-09881-0 Heikkilä, M., et al. (2020). Voices of student teachers' professional agency at the intersection of theory and practice. https://doi.org/10.1016/j.lcsi.2020.100405 Jääskelä, P., et al. 2024. Multidimensional student agency in university courses. Paper presented at the EARLI SIG 4 & SIG 17, 27th September. Utrecht, the Netherlands. Jääskelä, P., et al. (2023). Assessment of students’ agency in Finnish and Spanish university courses: Analysis of measurement invariance. https://doi.org/10.1016/j.ijer.2023.102140 Jääskelä, P., et al. (2020). Students’ agency profiles in relation to student-perceived teaching practices in university courses. https://doi.org/10.1016/j.ijer.2020.101604 Jääskelä, P., et al. (2017). Assessing agency of university students: Validation of the AUS Scale. https://doi.org/10.1080/03075079.2015.1130693 Klemenčič, M. (2017). From student engagement to student agency: Conceptual considerations of European policies on student-centered learning in higher education. https://doi.org/10.1057/s41307-016-0034-4 Lund, A. & Vestøl, J. M. (2020). An analytical unit of transformative agency: Dynamics and dialectics. https://doi.org/10.1016/j.lcsi.2020.100390 Mameli, C., et al. (2023). Student agency: Theoretical elaborations and implications for research and practice. https://doi.org/10.1016/j.ijer.2023.102258 Matusov, E., et al. (2016). Mapping concepts of agency in educational contexts. https://doi.org/10.1007/s12124-015-9336-0 Muthén, L. K., & Muthén, B. O. (2017). Mplus user's guide (8th ed.). Muthén & Muthén. OECD. (2022). The future of education and skills education 2030. OECD Accessed May 10, 2024. https://www.oecd.org/education/2030-project/teaching-and-learning/learning/student-agency/ Reeve, J., & Tseng, C.-M. (2011). Agency as a fourth aspect of students’ engagement during learning activities. https://doi.org/10.1016/j.cedpsych.2011.05.002 Schoon, I & Heckhausen, J. (2019). Conceptualizing individual agency in the transition from school to work: A social-ecological developmental perspective. https://doi.org/10.1007/s40894-019-00111-3 Su, Y-H. (2011). The constitution of agency in developing lifelong learning ability: The ‘being’ mode. https://doi.org/10.1007/s10734-010-9395-6 Yang, L., et al. (2023). Agency and student development in higher education: A cross-cultural and cross-disciplinary exploration. In: Y.I., Oldac, L., Yang, & S., Lee (Eds.), Student Agency and Self-Formation in Higher Education (pp.67-87). Palgrave Studies in Global Higher Education. https://doi.org/10.1007/978-3-031-44885-0_3