Education Evidence

supporting the art of teaching with the science of learning.
Home
Sources of evidence
Classroom evidence
Neuroscience
Workshops
Video
Evidence based books
Links
Contact
About
This section gives a brief outline of the reliable evidence available to educators.
 
Use the navigation to select different sources.  Use the "Links" button to find other sources of reliable advice
Reliable evidence
 
Evidence from the classroom.  To be reliable,  educational research needs to be designed in a similar way to the way medicines are tested.  It must use control groups and combine the results of several studies as a 'meta-study'.
 
To see an introduction to this material look at 'Classroom evidence' in the navigation.
 
Evidence from the brain. Cognitive neuroscience is the study of how the brain thinks and learns.  This is 'hard' science which already follows the rules of reliable evidence.
 
Although this is a new science which does not claim to 'know everything', there is a considerable body of knowledge which has implications for teaching.
 
To see an introduction to this material, choose 'Neuroscience' in the navigation.
 
 
 Unreliable evidence
 
  • "It's in the government strategy...."
  • "I used this last year and my class did well...."
  • "St Xavier's does this and they get good results..."
  • "There's a very good book which shows...."
  • "I read it in the education press...?"
  • "It's in the exam boards guidelines..."
  • ""A study at Branfield University shows..."
 
Almost any intervention improves learning - we need to know, not "does it work?", but "which ones work best?"
 
Because education is not at present 'evidence-based', much of the advice from experts, consultants or government directives cannot be relied on.  They tend to be opinion, backed only by selected evidence.
 
Individual research findings need to be compared with other projects.
 
Directives need to be tested and compared with alternatives.
 
Individual classroom experiments do not usually include sufficient learners nor control enough variables to give results which can be relied on.
 
Many studies use too few pupils or test them over too short a time-scale to be reliable.