Eric, there were studies done a few decades ago using factual questions. 
Here's a critical round-up of some of the studies: 
http://www.jstor.org/stable/25828215

Basically, 40-60% correct, but possibly the questions were not 
representative -- so possibly the results are really worse :(

kc

On 4/5/16 1:11 PM, Eric Lease Morgan wrote:
>   I sincerely wonder to what extent librarians give the reader
> (patrons) the right -- correct -- answer to a (reference) question.
> Such is a hypothesis that can be tested and measured. Please show me
> non-antidotal evidence one way or the other. --ELM

-- 
Karen Coyle
[log in to unmask] http://kcoyle.net
m: +1-510-435-8234
skype: kcoylenet/+1-510-984-3600