Maybe the student comes from a culture where trains are uncommon and therefore doesn't have some secret knowledge that would help him and/or her get the right answer.
Your sarcastic comment is, despite its disingeniousness, not as far off the mark as you believe.
Culture *does* influence modes of reasoning. The question about the trains carries a huge amount of cultural baggage: it applies the idea of continuous numerical measure to the speed of moving objects indirectly, based on a non-trivial mapping of the number produced by certain culturally advanced mathematical operations whose inputs are, again, continuous numerical measures of space and time, the latter which requires advanced technology (precise clocks) to measure to any significant degree of accuracy in the scales in question.
And even more basic things are culturally dependent, like spatial reasoning. Different cultures have different manners of conceptualizing space, time and geometric shapes. IIRC cultures whose habitat is in jungles are not as good as city dwellers at reasoning about diagonal lines (or something like that, I really can't remember the exact result right now).
The criticism seems mostly "informed" by the Postmodernist school of thought. To get an idea how these people work, imagine a small village with one fisherman, who leaves every day at dawn and returns with fish. A postmodernist analysis might be of anything: the status gained by the ritual of fishing, links to sun-worshipping religions, quasi-Marxist analysis, anything at all. The only thing it would omit would be the fact that the fish the fisherman brings back feed the village.
I will refrain from analysing the fact that this fantasy scenario of yourse tells us hardly anything about the real world, and the converse to this, that it tells us quite a deal about how little-informed you are on these matters.
If you want to know about cultural bias in "ability" testing, the right place to check is work on cultural and anthropological psychology. E.g. Levinson and his associates' work on spatial reasoning, which relates a culture's spatial orientation system (anthropocentric vs. absolute) to their reasoning about space, by means of psychological experiments.
Of course, you and I know that the purpose of word problems is to test the student's ability to see the underlying math in a language-based description. In other words, the skill is to disregard any aspect of the problem that is about baseball or trains other than the math
And this is precisely part of the ideology dominant in the Western world about what "intelligence" is: "reasoning ability in the abstract", supposedly divorced from actual application. This very idea *is* cultural bias. And, I really believe, ultimately wrong. There is no such thing as "pure reasoning ability"; reasoning has as one of its crucial components a set of learned cultural artifacts, e.g. math.
I commend you, though, on your choice of word: "skill".
[ Parent ]