Kind of a silly question, but it came up in our Sequences reading group yesterday: in EY’s An Intuitive Explanation of Bayes’ Theorem, we found the following statement:
It’s like the experiment in which you ask a second-grader: “If eighteen people get on a bus, and then seven more people get on the bus, how old is the bus driver?” Many second-graders will respond: “Twenty-five.”
Anybody have any idea where this finding comes from initially? We found several people who referenced EY’s post, including one second-grade teacher who claimed that he’d been largely able to replicate it (a case of guessing the teacher’s password, presumably), plus a bunch of “jokes” where the reader is the driver (so the correct answer is the reader’s age), but I didn’t see an original source for the experimental result. Maybe my Google-fu is weak but I’m curious if anybody knows...
The original source is Reusser 1986 p25 (26), who reports that 3⁄4 of first and second graders give a numerical answer. I learned that from Kaplinsky’s 2013 replication, not with eight-year-olds, but with eighth graders (video). He credits Merseth 1993 with popularizing it. Kaplinsky via Gwern.
A late descendent of the joke appears in Science Made Stupid (1986), which I highly recommend.
If you feel silly about the particular example or doubt its provenance or application, there is also the New Cuyama sign adding up numbers in an even worse way, the classic riddle where the answer is “you’re adding the wrong things,” or Scott’s latest post with similarly bad math being applied to public policy.
I’m particularly fond of the New Cuyama sign as an illustration of the principle because so much online discourse seems to involve using funny pictures to illustrate points. Your quote works well for a Less Wrong audience, as we seem to be mostly text-based here.
In the “remarkable coincidence” department, I just saw this morning an advert for a (UK) PhD studentship on this very topic: “Enabling Success on Science and Maths Problems; The Role of Local and Global Processing”.
… The student will investigate the relationship between success on maths and science misconceptions (e.g. naive conceptions, scientific misconceptions, perceptual bias, overgeneralisations) and: 1) the ability to inhibit local vs. global misleading information; 2) the ability to adaptively switch between local and global processing.
Full info is at this link (eligibility: UK/EU, application deadline 13 March). I am not affiliated with this research group/institution but thought there was enough potential overlap with LW readership to be worth posting.
Kind of a silly question, but it came up in our Sequences reading group yesterday: in EY’s An Intuitive Explanation of Bayes’ Theorem, we found the following statement:
Anybody have any idea where this finding comes from initially? We found several people who referenced EY’s post, including one second-grade teacher who claimed that he’d been largely able to replicate it (a case of guessing the teacher’s password, presumably), plus a bunch of “jokes” where the reader is the driver (so the correct answer is the reader’s age), but I didn’t see an original source for the experimental result. Maybe my Google-fu is weak but I’m curious if anybody knows...
The original source is Reusser 1986 p25 (26), who reports that 3⁄4 of first and second graders give a numerical answer. I learned that from Kaplinsky’s 2013 replication, not with eight-year-olds, but with eighth graders (video). He credits Merseth 1993 with popularizing it. Kaplinsky via Gwern.
A late descendent of the joke appears in Science Made Stupid (1986), which I highly recommend.
If you feel silly about the particular example or doubt its provenance or application, there is also the New Cuyama sign adding up numbers in an even worse way, the classic riddle where the answer is “you’re adding the wrong things,” or Scott’s latest post with similarly bad math being applied to public policy.
I’m particularly fond of the New Cuyama sign as an illustration of the principle because so much online discourse seems to involve using funny pictures to illustrate points. Your quote works well for a Less Wrong audience, as we seem to be mostly text-based here.
Are you sure the sign wasn’t a joke?
In the “remarkable coincidence” department, I just saw this morning an advert for a (UK) PhD studentship on this very topic: “Enabling Success on Science and Maths Problems; The Role of Local and Global Processing”.
Full info is at this link (eligibility: UK/EU, application deadline 13 March). I am not affiliated with this research group/institution but thought there was enough potential overlap with LW readership to be worth posting.