Asking the students wouldn't help -- very few of us are aware of our unconscious beliefs, especially since they are often in conflict with our conscious beliefs. Asking the students might be an interesting side project on awareness among 9-year-olds, but it won't shed any light on stereotype threat, which encompasses confidence anyway (that's one of the mechanisms).
After all, if you can get Asian women in the US to score significantly better and then significantly worse on a math test by first reminding them "Asians are good at math" and then "girls are bad at math", do you think they're going to be able to report on that "rationale"? And if they're aware of it, won't they subvert the whole experiment by using other sophisticated techniques like
(1) externalizing difficulty ("that researcher is just a jerk who doesn't know anything!"),
(2) reminding themselves of their complex self-identities and membership in several groups ("I'm a complex person; gender stereotypes are societal constructs; and remember, I'm Asian and I'm good at math -- let's use this stereotype for personal good!"), and
(3) using previous teachers' communication of high standards and assurance of capability, along with some role models,
to blow that test out of the water and confound all the study results?
(All of the above are techniques to combat stereotype threat that have been studied and found somewhat to very effective.)