Using low-stakes quizzing and interleaving to promote learning and retention of information in primary mathematics
Description of the innovation
This innovation project investigated the impact of using low-stakes quizzes and spaced practice and/or interleaving on mathematics attainment in Year 5 pupils in a large three-form entry primary school.
Summary of the evaluation
Sixty-nine Year 5 pupils in three parallel mixed-ability classes participated in the evaluation. All pupils attended a large primary school based in Sandwell. The percentage of pupils eligible for pupil premium at the school is 39%, which is well above the national average.
The three classes were randomly assigned with a coin toss to each of the conditions described below:
Control class – business-as-usual. Paper-based quizzes were presented after a topic was taught in lessons in a blocked fashion corresponding to the topic which had just been taught in class.
Intervention class 1 – LbQ quizzes with spaced practice. For the purposes of this project we define spaced practice as children only revisiting the material on one occasion after the initial teaching but with a delay of two weeks after it was taught. Post-topic quizzes were therefore presented to children in lessons with a delay of two weeks between classroom teaching and quizzing.
Intervention class 2 – LbQ with spaced practice and interleaving. As for intervention class 1, there was a delay of at least two weeks between classroom learning and testing, but in addition maths topics were tested in an interleaved fashion rather than blocked practice. In intervention class 2, teachers interleaved practice of different types of mathematical content so that newer tasks were mixed together with revision of earlier lessons and maths topics. This meant that key topics were presented multiple times to pupils before the final post-test in this group.
Summary of findings
The results suggest that there is promising evidence that interleaving had a positive effect on mathematics progress compared to both the control and spacing groups (Effect size for intervention class 2 (spacing + interleaving) vs intervention class 1 (just spacing) = +0.80; Effect size for intervention class 2 (spacing + interleaving) vs control class = +0.78). The control group performed slightly better than the intervention class 1 (just spacing) but there was a small effect size for this comparison (-0.08).