I confess I was most interested in the information about how you label products you're testing. It never occurred to me that people would gravitate toward A rather than B; or 1 rather than 2. And yet I should know this. As a teacher who made multiple-choice tests for years, I used to tell my students (who were trainee teachers themselves) that when they had no idea what the answer was on a multiple-choice test, they should always choose C. The reasoning was simple, and borne out by the experience of every teacher I ever discussed it with: the teacher doesn't want you to read the first answer and not the other options (because they you feel like you wasted your time writing four choices), so the teacher won't make A the correct answer - certainly not on question 1, anyway. The teacher thinks that B is also 'too soon' for the correct answer, but that if you read C and think it's right, you'll probably take the extra second to read D and make sure it's wrong, and thus the teacher's time spent writing four options is not wasted. So C is the answer that teachers will most often make the correct answer on an exam. Of course, our student-teachers were trained how to write false answers: one that's wildly, obviously wrong; one that's wrong in a detail; one that's tricky-wrong, and then the right answer. And it's easier to write absolutely, definitely wrong answers than right answers (that cannot possibly be questioned, for example, on a grammar test). So again, the teacher is likely to write three wrong answers with the correct answer prepared last - and then swap wrong-answer D to the C position, because having the wrong answer last becomes 'too obvious' after you've done that once or twice already. We were always having to sort of psych ourselves out of choosing C as the default position for correct answers. I'm going to keep thinking about your system of numbering samples with three-digit numbers that begin with at least the number 4. I'd love to know more about the science behind that.

• Reply