What are the strengths and limitations of using laboratory and field experiments to research education?
As a student you probably would have been the subject of an experiment within your school at some point during your 13 years of formal education.
Experiments conducted by schools themselves are a lot more common than educational researchers conducting their own field experiments within schools, and so such experiments are a rich source of data for students studying Methods in Context for A-Level sociology.
It is quite usual for schools to conduct small scale field experiments to try out new teaching techniques or to evaluate the effectiveness of banding of streaming.
If there are several classes of students of similar abilities doing the same subject, it is relatively easy to keep some classes being taught in the same way as usual but to change one aspect of teaching of other classes, and to measure the effect this has on student behaviour or learning.
Some years ago we conducted the following experiment in my institution, designed to measure the effectiveness of splitting students into ability bands:
In five subjects we deliberately created one class of all higher ability students (those predicted to get Bs or above at A-level based on their GCSE grades), but kept some high ability students in all of the regular mixed ability classes.
So in A-level sociology we had 7 classes at AS, and we ended up with one ‘high ability class’ and then 6 mixed ability classes, each with 2-3 A/B students rather than 3/5 A/B students.
The hypothesis of the Senior Management was that grouping all the higher ability students into one group would lead to improved results.
The control students are the high ability students still in the regular mixed classes.
We let this run for a year and compared the AS exam results at the end (it was a long time ago, when we still had AS exams.
The results in the end were inconclusive when we looked at the results of the ‘top band’ classes across all subjects – there was no signficant evidence that putting them in once class led to them getting better results.
As far as I am aware the students involved to this day have no idea they were subject to this experiment!
This kind of experiment within education is probably more common than you think!
A lot of schools put students into ability bands, and it makes sense that they review their results from time to time and ‘experiment’ to see if mixing up the bands give better results – for example, if you’ve got 6 maths groups in one school year, you could either have 6 discrete bands, or 2 more general bands, so you end up with a wider range of abilities split across 3 classes.
It kind of makes sense for schools to play around with how the split groups up to see if they can improve behaviour our outcome.
If you think about all of the things that schools can do differently, there are a lot of potential variables schools can change in one class, say, but not in others, just to see if changing that one variable makes a difference after a year in one class, rather than risking rolling out a change across the whole year group.
Variables you might change (all possible in-school experiments):
- The gender mix of classes.
- Seating lay-outs within classes
- The length of lessons/ number of lessons in a week
- The timing of ‘support lessons’ – before or after school, weekends, holidays.
- How support staff are used in classrooms
- Wider school policies on uniform, discipline and punishment
- The teaching techniques used in lessons.
- The ratio of face to face and online learning.
The practical, theoretical and ethical strengths of schools conducting experiments
Don’t kid yourselves, this kind of micro-experiment goes on all the time, but there some good reasons:
- Ethically teachers and schools have to provide students with the best education they can – educational theory about what the most effective teaching techniques changes, technology changes, so teachers and schools have to adapt. Doing an experiment for a year with one class can be a useful way of finding out how to implement changes more effectively across the whole school the following year. Or if some experiments don’t work out, at least it’s not all students who suffer.
- Practically, the students are there, the school is there, it’s relatively easy for teachers and schools to do experiments, rather than having them done externally.
- Theoretically – validity should be very high because one typically doesn’t inform students they are part of an experiment. Reliability should also be good because the conditions are relatively stable over time in most schools.
The practical, theoretical and ethical weaknesses of schools conducting experiments
- There is the ethical problem of deception and some students getting treated differently for the period of the experiment, which goes against equal opportunities.
- Teaching one class differently to the rest can be stressful and demanding for the teacher.
- In very small classes and schools with few classes, it’s hard to get a large enough sample for good representativeness.
For a great example of a really extreme experiment in a school, see ‘Chinese School‘.
For some of the general strengths and limitations of field experiments please see this post.