I recently wrote about a discussion with Ingrid on this subject, and I ended with the question:
"So now I have these two pieces of information dealing with the same issue, one qualitative, on quantitative. Which one is the most memorable, most true, most relevant?"Since I have no background at all in the study of this distinction between quantitative and qualitative evidence, I am slowly re-discovering the literature on this subject, in my own, rather passively, Google Reader driven, way. There is probably some great literature on this, both quantitative and qualitative, but given my personal preferences, this article that recently appeared in PNAS (Motivating voter turnout by invoking the self by Bryan et al.), has a clue to answering one of these three questions. In a real experiment, they asked potential voters how importing voting is for them, but with two experimental conditions:
"In each experiment, participants completed one of two versions of a brief survey. In one version, a short series of questions referred to voting using a self-relevant noun (e.g., “How important is it to you to be a voter in the upcoming election?”); in the other, questions that were otherwise identical referred to voting using a verb (e.g., “How important is it to you to vote in the upcoming election?”)."Based on previous research:
"So we hypothesized that using a predicate noun (e.g., “to be a voter”) as opposed to a verb (e.g., “to vote”) to refer to participation in an upcoming election would create a greater interest in and likelihood of performing that behavior—registering to vote and voting. If this hypothesis were confirmed, it would be evidence for the more general theoretical idea that simply framing a future behavior as a way to claim a desired identity can motivate that behavior."This simple difference in three words ("to be voter" versus "to vote") increased the actual, measured, voting by more than 10%! Now if we combine this amazing result with an assumption that it is easier for somebody to self-identify with a story about somebody else versus a graph or a statistic, I think that the starting hypothesis to the question "Which one is the most memorable?", is the qualitative piece of information. Whether this makes it also more true, and consequently more relevant, is still a completely different matter.
The results of this study also have other implications, though. One of the basic tenets in course design are learning objectives (see for instance this document from the University of Guelph). I advocates SMART (specific, measurable, achievable, relevant, and timely) verbs. The document already stresses the importance of writing them from the student's perspective, which makes it easier for the students to self-identify with them, but how about a study that randomly divides a class in two, and one half of the students get the learning objectives with action verbs, the other half with action nouns. Would this also result in a 10% increase in learner outcomes?