Testing theories

This post first appeared on the Alberta Teachers' Association website here.

By Dennis Theobald

My daughter Siobhan has advised me that she intends to write her January diploma examinations while wearing her pyjamas. Pointing out that she did phenomenally well on the two unit tests she wrote on “pyjama day” at her school, she asserts that this strategy will ensure she achieves the best possible results on her Grade 12 tests. (My response was to tell her to Google the Latin phrase Post hoc ergo propter hoc.)

Siobhan, however, is not alone in her efforts to identify strategies to maximize test performance that do not rely on old-fashioned learning, study, practice and review. The media and scholarly journals are full of accounts of innovative ideas for improving test scores.

For example, the Times Education Supplement (October 8, 2010) reports that Sydenham High School, in London, England, is encouraging students to sniff scented “memory oils” while studying and immediately before they write their standardized tests. To ensure peak results in each subject, students will use grapefruit scent for math, lavender for French and spearmint for history.

A somewhat less odoriferous approach is to have students spend just one quarter-hour writing about personally important values, such as friends and family. Female college students majoring in sciences who completed this value-affirmation exercise achieved a measurable boost in both in-class multiple choice examinations and standardized tests of physics concepts (Science, November 26, 2010). Unfortunately, male students did not experience similar improvement—they were probably confused about these things called “feelings” and “values” and needed a good six hours playing Call of Duty on their Xboxes to recover their equilibrium.

All is not lost for the boys, however. In an article in the journal Emotion (October, 2010), Michael Kraus et al reported that NBA teams whose members touched each other more frequently during play performed better than their less tactile opponents, even after the researchers controlled for player status, preseason projections and early season results. Perhaps adopting a touchy-feely approach to exam preparation would generate similarly positive results.

It has also been noted in the Journal of Neuroscience, Psychology and Economics (November 8, 2010) that there is a strong correlation between the implementation of daylight saving time and lower scores on the SATs, exams written by US students seeking entry to college. Extrapolating from this, I suggest that if we want to blow those damned Finns out of the water on the next round of Programme for International Student Assessment (PISA) tests, we should stand firm and refuse to fall forward or spring ahead. (Then again, if daylight saving time made that great a difference, Saskatchewan would lead the world in student performance.)

So, what does all this mean? I suppose that in pursuit of every ­possible marginal advantage, we could encourage students to sleep in late and show up for examinations in their jammies. Then we could have them engage in a round of pre-examination touching and exploration of their feelings (what could possibly go wrong?) before presenting them with exams printed with delicately scented inks.

Alternatively, we could remember just how artificial our evaluation and testing process is. After all, can you recall any real world situation when you were required to make dozens of critical decisions in a restricted timeframe, having deliberately been deprived of access to ­assistance from colleagues or reference materials? With that in mind, we could then care a whole lot more about learning and a lot less about testing. (Hat tip to David Brooks, whose December 6, 2010, New York Times column “Social Science Palooza” highlighted several of the ­studies referenced here.)

I welcome your comments—contact me at dennis.theobald@ata.ab.ca.