On Wednesday 05 March 2008, Otheus (aka Timothy J. Shelling) wrote:
> To everyone:
>
> I apologize that my words came across as "combative". 

<snip>

Sir, I salute you. A post like this takes courage, and more often than 
not a good deal of self-reflection.

> Moving on, however, I'd like to get to the meat of the matter. First,
> does anyone have disagreement with my "view" on the 7 questions. You
> have these in your mailbox, and you can refer to them by number. 
> Second, assuming we agree on those 7 questions as being problematic,
> can anyone calculate the probability of error rate for that one
> exam?  Let's say there's a pool of 500 questions, and these are the
> only 7 that are defective, what are the chances I got all 7
> questions?  Third, is it possible (financially feasible) to strike
> these questions from future exams? You think with the exam being
> computerized and all....

I think you have valid concerns with the questions. It is true that some 
errors can creep into the vetting process, no matter how diligent the 
review is. Sometimes the question is just faulty, other times it's due 
to changing circumstance.

I think that only the LPI statisticians are truly in a position to 
evaluate the percentage errors, so no comment from me on that aspect.

What we can do as a community instead is to help make the next review a 
thorough one. People like yourself who have passed the certification 
and want to see the exam be better should sign up to be reviewers and 
examine each question multiple times. That way we can weed out the bad 
ones. Think of it this way - someone reviewed the questions before you 
wrote them so you could write a better exam that they did. You can 
return the favour to the next one in line.

-- 
Alan McKinnon
alan dot mckinnon at gmail dot com

_______________________________________________
lpi-examdev mailing list
[email protected]
http://list.lpi.org/cgi-bin/mailman/listinfo/lpi-examdev

Reply via email to