As lifelong learning (LLL) is a significant condition for employability, social inclusion and active citizenship, the European Council has been emphasizing the importance of adult learning for the last two decades (European Commission, 2001; 2016; 2019). While good-quality motivation and good-quality learning strategies are important determinants of continued learning in adulthood (European Commission, 2016; Lüftenegger et al., 2012), learning processes in adult education are an understudied terrain, especially among low-educated adults, who we can expect to be insufficiently developed in these learning skills.

Educational psychological research agrees that learning is a complex interplay of motivation, regulation and cognitive processing and, regardless of the age of the studied population, individual differences in learning quality exist (Vermunt & Donche, 2017). It can therefore be expected that also low-educated adults should not be considered a homogeneous group of learners, but learner profiles, differing in the quality of learning motivation and use of learning strategies, may be present. In person-oriented research, motivation and learning strategies have so far mainly been studied separately, rather than as an integrated whole. Yet literature points to the strong reciprocal relation between the two components, in which neither motivation nor learning strategies are the protagonist (e.g., Alexander, 2017).  For this reason, the current study seeks to answer the question of which qualitatively different learner profiles exist among low-educated adults, based on learning motivation and learning strategies used, investigating both components, relative to each other.

Learning motivation is conceptualized in this study according to Deci and Ryan's (2000) Self-Determination Theory (SDT), in which quality is understood as the degree to which behaviour is self-determined. Amotivation is situated at the lower end of the SDT-continuum, which is the same as a lack of motivation. Next on the continuum are various forms of extrinsic motivation. The least self-determined form of extrinsic motivation is external regulation. This behaviour is initiated by external pressure, such as rewards or power. Introjected regulation refers to behaviour that is self-imposed, such as behaviour to avoid guilt or boost the ego. The third and most self-determined form of extrinsic motivation, is identified regulation. It refers to behaviour that is posed because the learner finds it valuable. At the very top of the continuum is intrinsic motivation which refers to behaviour that stems from inherent interest or pleasure.

The distinction in quality for the component of learning strategies is conceptualized according to the Learning Patterns Model (LPM)(Vermunt & Donche, 2017). Students tending toward a meaning-oriented learning pattern process learning content in a deep way, combined with a high degree of self-regulation strategies. Students with an application-oriented learning pattern prefer to make connections to concrete situations and prefer both self- and external regulation strategies. Students with a reproduction-oriented learning pattern process in a surface manner and prefer strong external regulation by the learning environment. Students can be identified lacking any regulation strategies and using few to none processing strategies and whom the model labels as the undirected learning pattern. The former two patterns are considered good-quality patterns, while the latter patterns are considered poor-quality learning.  

Although both theories have a tradition of variable-oriented research, person-oriented studies have increasingly appeared to distinguish between individual quality. For each component of learning (motivation, regulation and processing strategies), typically, four profiles are found, differentiating between a high- versus low-quantity and a good- versus poor-quality profile (e.g., Cents-Boonstra et al., 2019; Shum et al., 2023). Based on the insights of earlier person-oriented research, we hypothesize learning profiles among low-educated adults to be distinct not only in terms of quality but also in terms of quantity.

1. Context and participants The present study was conducted in six institutions for adult education in Flanders (northern part of Belgium). To reach the target population of low-educated adults, we compiled a convenience sample of 512 adults participating in a second-chance education program, allowing every participant to complete the survey during class hours. 2. Instrument and measurement Motivation, regulation and processing strategies were measured by means of a paper and pencil version of the LEarning and MOtivation questionnaire (LEMO, Donche et al., 2010), a 49-item self-report inventory including 15 items measuring learning motivation based on SDT (Deci, & Ryan, 2000) and 34 items measuring regulation and processing strategies, as conceptualized in the LPM (Vermunt & Donche, 2017). All items were measured on a seven-point Likert scale to reduce ceiling effects and ranged for motivation from one (totally disagree) to seven (totally agree) and for regulation and processing strategies from one (never) to seven (always). Inspection of the psychometric properties showed acceptable construct validity and reliability of the different scales (motivation (CFI = .92, RMSEA = .08, SRMR = .06)(.70<α<.89); regulation strategies (CFI = .87, RMSEA = .08, SRMR = .07)(.69<α<.78); processing strategies (CFI = .91, RMSEA = .06, SRMR = .06)(.66<α<.73). 3. Data analysis To distinguish learning profiles, a latent profile analysis was conducted. To evaluate how many groups best describe the data, typically, LPA uses several information criteria. As multiple information criteria can point to different conclusions, we mathematically combined different model fit criteria (AIC, AWE, BIC, CLC, and KIC) into a composite relative importance vector (C-RIV), with the highest value representing the model with the most optimal number of profiles (Akogul & Erisoglu, 2017). For LPA, inspection of missing data, outliers and normality of the distributions is recommended (Spurk et al., 2020). This resulted in the use of multiple imputation of missing values, removal of multivariate outliers using the Mahalanobis distance indicator and log-transformation of highly skewed scales. Key variables were standardized by rescaling to z-scores. All analyses were carried out in the statistical software R.

Analyses revealed for the motivational component the four expected profiles. A distinction was made between a high-quantity profile (25.14%), a low-quantity profile (18.08% ), a good-quality profile (40.11%) and a poor-quality profile (16.67%). For the variables measuring regulation strategies, a two-profile solution proved most optimal. Both profiles are particularly distinct in their scores on self-regulation strategies. The profiles were labelled self-regulated profile (62.15%) versus unregulated profile (37.85%). For the processing scales, the five-profile solution yielded the most optimal results. Of the profiles found, 4 of 5 are quantitatively distinct, scoring either relatively high or low on all processing strategies. We labelled these profiles active (21.47%), moderately-active (49.15%), moderately-inactive (19.21%) and inactive profile (4.80%). The fifth, but underrepresented profile was labelled deep profile (5.37%), because of its relatively low levels of surface processing strategies and relatively high levels of deep processing strategies. When integrating the three components of learning, five motivational-learning profiles could be retrieved. For the learning strategies component in these profiles there is little variation in quality: the mean scores are either relatively high, moderate or low. In other words, homogeneous subgroups of learners can only be discerned in the quantity of learning strategies used. A distinction in quality however, was made for the motivational component in these integrated profiles. Results showed that patterns found in this study are very similar to motivational-learning profiles identified among primary school students (Heirweg et al., 2019). Previous longitudinal person-oriented studies suggested that the high-quantity learning profiles have the potential to further evolve into good-quality profiles by gaining more learning experiences (e.g., Vanthournout et al., 2009). This developmental hypothesis may hold true for low-educated adults who often did not have had a trouble-free prior educational trajectory and where further development in good-quality learning strategies and motivation is possible.

Alexander, P.A. (2017). Issues of Constructs, Contexts, and Continuity: Commentary on Learning in Higher Education. Educational Psychology Review, 29(2), 345–351. Akogul, S., & Erisoglu, M. (2017). An Approach for Determining the Number of clusters in a Model-Based Cluster Analysis. Entropy, 19(9), 452. Cents-Boonstra, M., Lichtwarck-Aschoff, A., Denessen, E., Haerens, L., & Aelterman, N. (2019). Identifying motivational profiles among VET students: differences in self-efficacy, test anxiety and perceived motivating teaching. Journal of Vocational Education and Training, 71(4), 600–622. Deci, E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry, 11(4), 227–268. Donche, V., Van Petegem, P., Van de Mosselaer, H., & Vermunt, J. (2010). LEMO: een instrument voor feedback over leren en motivatie. Plantyn: Mechelen. European Commission (2001) Making a European Area of Lifelong Learning a Reality. European Commission COM 678 final. Available at: http://aei.pitt.edu/42878/1/com2001_0678.pdf (accessed January 30, 2024). European Commission (2016) on Upskilling Pathways: New Opportunities for Adults (2016/C 484/01). Available at: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=OJ:JOC_2016_484_R_0001 (accessed January 30, 2024) European Commission (2019) Directorate-General for Education, Youth, Sport and Culture, Key competences for lifelong learning, Publications Office (2019) https://data.europa.eu/doi/10.2766/569540 (accessed January 30, 2024) Heirweg, S., De Smul, M., Devos, G., & Van Keer, H. (2019). Profiling upper primary school students’ self-regulated learning through self-report questionnaires and think-aloud protocol analysis. Learning and Individual Differences, 70, 1555-168. Lüftenegger, M., Schober, B., Van de Schoot, R., Wagner, P., Finsterwald, M., & Spiel, C. (2012). Lifelong Learning as a goal - do autonomy and self-regulation in school result in well prepared pupils? Learning and Instruction, 22, 27-36. Shum, A., Fryer, L.K., Vermunt, J.D., Ajisuksmo, C., Cano, F., Donche, V., Law, D.C.S., Martínez-Fernández, J.R., Van Petegem, P., & Yu, J. (2023). Variable- and Person-centred meta-re-analyses of university students' learning strategies from a cross-cultural perspective. Higher Education. Spurk, D., Hirschi, A., Wang, M., Valero, D., & Kauffeld, S. (2020). Latent profile analysis: A review and “how to” guide of its application within vocational behavior research. Journal of Vocational behavior, 120, Article 103445. Vanthournout, G., Donche, V., Gijbels, D., & Van Petegem, P. (2009). Alternative data-analysis techniques in research on student learning: Illustrations of a person-oriented and developmental perspectives. Reflecting education, 5(2), 35-51. Vermunt, J. D., & Donche, V. (2017). A Learning Patterns Perspective on Student Learning in Higher Education: State of the Art and Moving Forward. Educational Psychology Review, 29(2), 269–299.