Tool 2: Mini randomised control trials: six weeks to change your classroom:


As a teacher you will be trying out different ways of teaching to improve the learning of your students. THe MESHGuides are there to provide research evidence to help you.

This toolkit sets out methods for a six to eight week intervention.The approach follows that of the successful Andrews, Rivers and Akhurst (2008) intervention programme (reference below).

We think it is important for the education sector to accumulate evidence of the impact on learning on teachers and students and our particular interest is in testing out the advice in the MESHGuides in different contexts. We ask educators to do this through a series of replicated studies in different schools, countries and contexts. If you are using this approach to test out MESH guidance, please provide a summary of your approach and outcomes for consideration by the authors of the MESHGuide.


  1. First identify an area where you need to improve student learning and colleagues who are interested in the same area. These might be in your school or from other schools or countries.
  2. Find two groups of students in your school  in which relevant characteristics are roughly the same. Rivers et al 2008 used two separate schools which had been matched (on bullying prevalence rates) but for most interventions it should be possible to use two matched groups within a single school. This would reduce some variability (eg socioeconomic status, postcodes*, school ethos etc) and might also be more convenient! You need to decide which characteristics are relevant for matching; these might include:
    • prior achievement profile
    • ethnic mix
    • gender balance
    • age
  3. Make a decision about how the study is to be done. Drawing on evidence for example from a MESHGuide  linked with the teacher’s professional judgement use an existing intervention (or design a new intervention) which would normally take the form a short series of lessons delivered to the test groups by the normal classroom teacher or different teachers with the control group being taught using the previously accepted approach.
  4. To help accumulate a strong evidence base for educational practice, please let the MESH steering group know what you are doing so that we can put you in touch with others with similar interests.
  5. Both intervention and control group get a pre-test to establish the baseline against which the post-test results will be analysed.
    • It might be useful to use the pre-test to determine which group receives the intervention because the group that starts off with lower scores may appear to make progress just through regression to the mean (conversely the group that starts off with higher scores may appear to make less progress for the same reason). We recommend using the higher scoring group for the intervention because it is potentially more difficult for them to make considerable progress.
    • Individual students might be matched using the pre-test scores.
    • The teachers are blind to the pre-test scores
  6. Deliver the intervention to one group, using the matched group as control.
  7. Measure the effect of the intervention on both teacher and students using agreed pre and post-test strategies. This might be a questionnaire or an assessment of current attainment. The items on the post-test would have to be matched such that the control group would have a possibility of answering them (eg the two groups would have to cover the same content taught using different pedagogies).
  8. Analyse the results.
  9. If the intervention is shown to have a significant positive effect then a second intervention might be implemented so that the control group receives the intervention. This removes the ethical objection that the original control group may have suffered disadvantage by not receiving the intervention. It also enables the experimenter to make a further test on the group to see whether they ‘catch up’ when they receive the intervention. Be prepared to abandon the experiment if negative outcomes are apparent.
  10. Submit the report to the main MESH database via .

* IDACI index, based on postcodes, is used in the UK to provide information about relative deprivation.

Appendix: Description of the Positive Psychology intervention programme 2008

See Richards A.; Rivers, I. and Ackhurst, J. (2008) “A positive psychology approach to tackling bullying in secondary schools: A comparative evaluation”. Educational & Child Psychology Vol 25 No 2, pp 72-90.