I was honored to have a colleague laud my Myths book (she was kind enough to also promote the newer learning science book), but it was something she said that I found intriguing. She suggested that one of the things in it includes “discussing how to read research”. And it occurs to me that it’s worth unpacking the situation a wee bit more. So here’s a discussion about how we (properly) develop learning science that informs us in reading research.
Caveat: I haven’t been an active researcher for decades, serving instead to interpret and apply the research, but it’s easier to say ‘we’ than “scientists”, etc.
Generally, theory drives research. You’ve created an explanation that accounts for observed phenomena better than previous approaches. What you do then is extend it to other predictions, and test them. Occasionally, we do purely exploratory studies just to see what emerges, but mostly we generate hypotheses and test them.
We do this with some rigor. We try to ensure that the method we devise removes confounding variables, and then we use statistical analysis to remove the effects of other factors. For instance, I created a convoluted balancing approach to remove order effects in my Ph.D. research. (So complicated that I had to analyze a factor or two first, to ensure it wasn’t a factor, so I could remove it from the resulting analysis!). We also try to select relevant subjects, design uncontaminated materials, and carefully control our analysis. Understanding the ways in which we do this requires an ability to know about experiment design, which isn’t common knowledge.
Moreover, we then need to share this with our colleagues so that they can review what we’ve done. We need to do it in unambiguous language, using the specific vocabulary of our field. And we need to make it scrutable. Thus, we publish in peer-reviewed journals which mean others have looked at our work and deemed it acceptable. However, the language is deliberately passive, unemotional, and precise, as well as focused on a very narrow topic. Thus, it’s not a lot of fun to read unless you really care about the topic!
There are problems with this. Increasingly, we’re finding that trying to isolate independent variables doesn’t reflect the inherent interactions. Our brains actually have a lot of complexity that hinder simple explanations. We’ve also found that it’s difficult to get representative subjects, when what’s easy to get are higher education students in the developed world. There are also politics involved, sad to say, so that it can be hard for new ideas to emerge if they challenge the entrenched views. Yet, it’s still the best approach we have. The scientific method has led to more advances in understanding than anything else!
There are things to worry about as a consumer of science. For one, there are people who fake results. They’re few, of course. There’s also research that’s kept proprietary, for financial reasons. Or is commissioned. As soon as there’s money involved, there’s the opportunity for corruption (think: tobacco, and sugar). Companies may have something that they tout as valid, but the research base isn’t publically available. Caveat emptor!
Thus, being able to successfully read research isn’t for everyone. You need to be able to comprehend the studies, and know when to be wary. The easy thing to do is to look for translations, and translators, who have demonstrated a trustworthy ability to help sort out the wheat from the chaff. They exist.
I hope this illustrates what reading research requires. You can take some preliminary steps: give it the ‘sniff’ test, see if it applies to you, and see who’s telling you this (and if anyone else is agreeing or saying to the contrary) and what their stake in the game is. If these steps don’t answer a question, however, maybe you want to look for good guidance. Make sense?