Date of Award

Spring 2023

Degree Type

Open Access Dissertation

Degree Name

Psychology, PhD


School of Social Science, Politics, and Evaluation

Advisor/Supervisor/Committee Chair

Kendall Cotton Bronk

Dissertation or Thesis Committee Member

Tiffany Berry

Dissertation or Thesis Committee Member

Saeideh (Saida) Heshmati

Terms of Use & License Information

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Rights Information

© 2023 Elyse L Postlewaite


Adolescence, COVID-19, Developmental period, Montessori, Self-directed learning, Self-regulated learning

Subject Categories



Students who develop and apply their self-directed learning skills have advantages in school over those who do not (Betts & Knapp, 1981; Candy, 1991; Guglielmino, 1977; Schunk & Zimmerman, 2012). This is because self-directed learning (SDL) skills enable autonomous learning where students self-initiate, solve problems, develop new ideas, and monitor themselves with minimal external guidance (Knowles, 1976; Zimmerman, 2000). Despite the importance of these skills, research shows that few students consistently engage in SDL (Dent & Koenka, 2016; Zimmerman & Schunk, 2001). Because of SDL's multi-faceted and complex nature, it is difficult to discern why this is the case (Dent & Koenka, 2016; Zimmerman & Schunk, 2001). Multiple dynamic, interacting factors, including maturation, environmental supports, and historical events, likely contribute to students’ SDL development (Hoyle & Dent, 2017). As such, recent literature suggests that a relational dynamic systems (RDS) approach can help elucidate the dynamic, context-dependent patterns by which SDL skills unfold (Hoyle & Dent, 2017). According to RDS theories, development occurs within multi-level, interacting, relational systems; therefore, the bi-directional relationship between the person and their environment should be the unit of analysis (Lerner et al., 2011). This study aimed to provide a systematic investigation of the development of SDL skills, accounting for important contextual and developmental influences as well as individual pathways. Adolescence appears to be an optimal time for students to gain SDL skills (Brown, 1978; Brown et al., 1983); hence, this study focused on that developmental period. Also, because the Montessori educational approach is conducive to SDL skill development (Lillard, 2017; Rathunde, 2009, 2014), it was used as the school backdrop for this study. Furthermore, during data collection, a global pandemic caused by COVID-19 impacted school environments and was also included as a developmental context in this study. Specifically, this study utilized a longitudinal convergent mixed methods design to (1) identify patterns of SDL skill development across adolescence, (2) illustrate the reasons for those changes, and (3) illuminate the indirect effect of COVID-19 on students’ SDL. Emergent themes from student interviews conducted over four years augmented growth curve analysis results from an accelerated longitudinal design utilizing student surveys to address the research questions. Descriptive, correlational, multi-level model (MLM), and repeated-measures ANOVA analyses of student survey responses across four years (4 waves) of data collection with students grades 7 through 12 ( n = 284) were applied to address the quantitative research questions. Emergent themes, derived through thematic analysis of 29 interviews, or 11 cases of students with a range of SDL skills (average, above average, and below average), addressed the qualitative research questions. Finally, quantitative results and qualitative findings were combined and compared to investigate convergence, divergence, and expansion areas that addressed integrative research questions. Findings shed important light on the development of adolescent students’ SDL skills across adolescence. Quantitative results and qualitative analyses were combined to address the research question: Do adolescent students’ SDL skills increase, decrease, or remain stable throughout middle school and high school? Findings resulted in areas of convergence and divergence across methods. Despite some diverging quantitative results, namely a non-significant growth model, other quantitative results, a non-significant no-growth model and descriptive plots, converged with qualitative findings from student interviews to suggest that within and between students, SDL skill development can include a combination of growth, decline, or stability over time. Findings from this study also suggest that each SDL skill can develop on its own timetable. Also, findings suggest a developmental pattern whereby SDL skills vary more in middle school than in high school. These findings have begun to disentangle contradictory results of earlier SDL research (e.g., Heater, 2005; Pajares & Valiante, 2002; Reio & Ward, 2005). From a practice perspective, the findings imply that it may benefit students to have tailored interventions that meet them where they are developmentally, considering each SDL skill individually and all together. The second integrative research question that was addressed in this study was: What roles do factors like grade level and the Montessori learning environment play in SDL development? Quantitative results and qualitative findings converged to suggest that students' SDL skills develop, at least in part, as a factor of the length of time a student has been immersed in the Montessori program. The findings also show that a student’s maturation may play a role in SDL skill development, especially when environmental contexts are supportive. In addition, qualitative interviews with students identified features of the Montessori program, such as open work time, scaffolded opportunities to be self-directed, autonomy support, and supportive teachers that aided students in their SDL development, which also aligns with the literature (Zumbrunn et al., 2011). This finding strengthens prior research, which found cursory evidence for how Montessori schools support the development of SDL skills (Ervin et al., 2010). In addition, quantitative results and qualitative findings diverged for the third overarching research question: Have changes in the learning environment associated with COVID-19 shaped the development of students’ SDL skills? If so, how? Although the quantitative results from this study failed to detect any indirect effects of the impact of COVID-19 on students’ SDL, qualitative findings found that changes in their learning environment as a result of COVID-19 impacted students’ SDL both negatively and positively. Research has also found that the global pandemic drastically impacted the school environment, so it is most likely that the quantitative measure failed to detect an effect (Huck & Zhang, 2021; Tarkar, 2020). Furthermore, in the interviews, most students reported a combination of negative experiences (e.g., more distractions at home, lowered motivation, fewer social interactions, higher stress, and missing in-school learning) as well as positive experiences (e.g., increased time management, access to resources, multi-tasking, organization, ability to shut out distractions, and time to sleep) that impacted their SDL abilities. Qualitative findings from this study extend prior research by providing student accounts of their experiences, including silver linings (Wilson et al., 2020). Despite its limitations, this study revealed important exploratory findings about how students’ SDL skills can develop across adolescence. Areas of convergence across qualitative and quantitative methodologies underscore the reliability of the study findings. There were also unique quantitative and qualitative findings that extend prior research and provide important implications for future research and practice.



Included in

Psychology Commons