It's not one or the other. I look harder for possible ambiguities in a
question if the difficulty & discrimination indices are out of whack. FWIW,
I try to emphasize to the class that my adjustments are not a "curve."
Instead, I am introducing a "correction factor" for bad questions. I
suspect
Hi
On Sun, 8 Apr 2001, Chuck Huff wrote:
> I am still not sure how to interpret the item-to-whole correlation,
> though. Is it bad only if it is negative?
Even for a low or zero value, one interpretation would be similar
to a negative or zero part-whole correlation from the reliability
command
Thanks to both Ed Pollack and Michael Renner for suggesting the
following criteria:
1) A cutoff of .25 or .33 of those who got it correct
2) The item to whole correlation as a discrimination index
3) Asking for what students were thinking of when they got it wrong.
I like 3 the most, since it h
Chuck Huff asked
> What are the lousy statistics that folks look for? I have done
> histograms of the number of folks who choose each choice in each
> item, but when I looked at these, I was not convinced to toss a
> question even though most folks got it wrong. Are there particular
> question
Hello,
I have just shown the Sci American tape "Animal Einsteins" to my intro psych
classes. It has Alex the Parrot, chimps using model rooms to find hidden
objects, problem-solving ravens and a bunch more. A student was asking me
about the reports of dolphins saving the lives of drowning huma