Teacher Choices trials explore some of the most common questions teachers ask about their practice, testing the everyday choices teachers make when planning their lessons and supporting their students.
A Winning Start was the first classroom-based ‘Teacher Choices’ trial, and was led by a team at NFER. The key aims of this trial were to ascertain the feasibility of running such trials in schools and to provide recommendations for future similar projects.
In this trial, two approaches to 10-minute lesson starters were explored: a retrieval quiz, focused on content from previous lessons; and discussions, focused on foregrounding new learning. Teachers delivered one type of lesson starter for a half term, with a holiday ‘wash out’ period before delivering the other lesson starter type with a different science topic.
Quiz starters were chosen building on evidence around retrieval practice, while discussions were identified as common practice starter activities and used as an active control group (a comparison group for retrieval quizzes).
As this was a feasibility trial, the main research questions focused on how feasible this new methodology was for the EEF to deploy to answer questions about common teaching practices. The results were fairly positive: recruitment was easy and teachers mostly implemented the correct starters, implying the simple guidance provided to them was relatively effective. Teachers mostly adhered to their lesson starter but only 51% delivered the allotted lessons starter as intended (for under 10 minutes) in all lessons, indicating pupils experienced modest, variable exposure to the lesson starters.
The experiment to measure the impact of the choice by comparing teachers’ different end-of-topic tests was not as successful in this trial, partly due to how different the tests were from one another, as the teacher-developed tests related to different topics and lacked validity and reliability; and partly due to the fact that different attainment groupings and some mixed ability grouping were being compared with one another.
We did not find a difference between the groups in terms of attainment based on the different starter activities, based on either the attempt to compare different end of topic test results or on a standardised score that combined results from all end of topic tests. We have not assigned a security rating to this finding.