Pragmatic clustered randomised control trial to evaluate a self-regulated learning intervention to be implemented in South Australian primary schools-study protocol.

Abstract

Background: Self-regulated learning (SRL) is described as a process whereby learners actively take control of their learning by setting goals, planning, monitoring, evaluating, and adjusting their learning strategies to improve performance and achieve desired outcomes Panadero (Front Psychol 8:422, 2017). SRL proficiency has been shown to predict educational success and lifelong outcomes, such as income and health. While SRL is recognised as a key lifelong competency, there remain questions regarding how educators can best develop and promote SRL in school settings. A scalable, low-cost intervention targeted at grade 1 students (6-7 years old) in Germany was found to have substantial effects on impulse control and self-regulated learning, with sustained impacts on long-term academic success Schunk (Nat Hum Behav 6(12):1680-90, 2022). This study protocol seeks to adapt the Schunk et al. (2022) randomised trial to the Australian content and extend it to grades levels 2, 4 and 6.

Methods: We will use a standard pragmatic clustered (by school) randomised controlled superiority trial with an additional population-wide matched parallel control arm. Effectively, we will conduct three trials-one for each age/grade level. Each trial will be powered to assess the impact of the intervention on the age/grade groups independently: grade 2 (early primary, 7-8 years), grade 4 (mid primary, 9-10 years), and grade 6 (late primary, 11-12 years). Schools assigned to the treatment group will have all three grade levels (grades 2, 4, and 6) receiving the treatment (at least one class per grade); no classes in the schools assigned to the control group will receive the intervention. A minimum of 56 schools with an average cluster size of 19 children/class will be required to detect a minimum impact of 0.25 SD effect size at 80% power taking into account the clustered design with an intraclass correlation coefficient (ICC) of 0.05. This results in a total sample of 1064 per grade and thus 3192 students in total (56 schools per arm × 19 students in an average-sized class × three grade levels). Randomisation will occur on a 1:1 ratio, such that half of the schools (n = 28), and effectively about half of the students (n = 1596) will receive the intervention. The primary outcome will be improved self-regulation assessed at 6 weeks, 6 months and 12 months post the intervention. Longer-term secondary outcomes will include academic and wellbeing measures obtained through administrative data linkage to the National Assessment Program in Literacy and Numeracy (NAPLAN) outcomes and the Wellbeing and Engagement Collection (WEC) outcomes measures in the year following implementation (grades 3, 5, and 7). Follow-up via the South Australia Data Linkage Systems will allow for longer-term academic outcomes, mental health, school completion, criminal justice, and tax data.

Discussion: This protocol paper provides a detailed record of the trial design. We also discuss our analytical plan, especially highlighting the opportunities associated with the linkage of the trial participants to South Australia's population-wide administrative data linkage systems.

Trial registration: ACTRN12623001331628. Registered on 9th of November 2023.