Well, university, actually. I had lots of ideas about how to teach spelling using visual mnemonics, but no hard evidence (beyond my own experience teaching adults) that they were actually effective. I particularly wanted to find out whether this approach worked for primary school children, as they seem to spend a huge amount of time learning and being tested on spellings.
I took myself off to do an MSc in Educational Neuroscience at Birkbeck / UCL Institute of Education in London. Every time you read a sentence that starts “Neuroscience shows that …”, alarm bells should start ringing. You can be fairly certain that what follows is a selective interpretation of a small amount of published research which may only be tangentially related to the topic.
However. We should not abandon the drive for a rigorous evidence base for our teaching, but we should interrogate all claims that are justified by neuroscience quite carefully. My dissertation research was a controlled trial of teaching spelling using visual mnemonics to Y4 (8-9 year olds) in a classroom setting. My methodology drew on the work of Partz, M. P. D., Seron, X., & Linden, M. V. D. (1992), who used a similar visual imagery approach for the cognitive rehabilitation of a patient following a stroke.
I recruited 54 primary school children and split them into two evenly matched groups. I taught both groups 20 words using either the Button Spelling method, or a comparable conventional method (definitions and contextual sentences) over a period of four weeks.
When the children were tested on the words one week after teaching, the children taught using visual mnemonics remembered 50% more correct spellings than children taught the same words using the conventional method. If you like statistics, here they are: The experimental group had a greater proportion of correct spellings (20.4%) than the control group (13.3%): χ2 (3) = 13.41, p = .003. This corresponds to a medium effect size: Cramer’s V (df=3) = .15.
(If you are really keen, email me and I will send you my thesis. If you are really, really keen, you might even read it).