What a shepherd is is completely irrelevant [1] to the problem.
> Perhaps this is also a failure of education, but it isn't a failure of math education.
The inability to identify which things in a word problem are relevant to the question being asked is clearly a failure in education in the area of logic and reasoning, if not specifically in "math" per se.
[1] Well, unless "shepherd" is some thing that has ownership of a number of dogs and/or sheep that has a fixed mathematical relationship to its age, but I doubt any child's lack of experience with the term would leave them with a prior expectation that that was the case.
...says the person privileged to know what a shepherd is. I agree that math education could be improved, but this study may not be the best basis for that improvement.
Even if the children aren't familiar with the term "shepherd", it still comes down to the fact that they do not have enough information to answer the question and still attempt to perform some sort of calculation with no reasoning of why they are doing it.
Yes, the superior abstract reasoning that the computer imparts would definitely solve this!
(I'm also fairly skeptical of this finding).
From the article:
The “How Old Is the Shepherd?” question was popularized by an essay written by Professor Katherine K. Merseth in 1993 and was based on research by Professor Kurt Reusser in a paper presented at the 1986 American Educational Research Association annual meeting. It is noteworthy that Professor Merseth wrote that “researchers report that three out of four schoolchildren will produce a numerical answer to this problem.” Twenty years later that statement still held true.