[the_old_crowd] W.O.M.B.A.T.S

[rebecca]

>Snow:

>I was curious as well on this subject point but more so on how you would
>grade a question that began with " in your opinion". This was the most
>puzzling aspect of the test for me. If a question can't actually be graded
>because everyone is entitled to their own opinion, then why ask the
>question?

<snip>

>This test reminds me of the one I had taken in the shopping mall when you
>are asked all types of questions about what brand of cigarette you smoke
>when the survey was actually for beer (they did slide one single question
>about alcohol into the survey but the main drift was about your cigarette
>brand). Made little sense to me at the time but I suppose they aimed their
>advertising at the smokers if the majority of the surveyed also said that
>they drank.

rebecca:

Therein lies the rub, and you hit the nails on the head IMO, my dear. :)

Usually in a test with questions like that, numerical values are assigned 
for each optional answer for a given question - if there are 4 answers in a 
multiple choice, one of them is the one test administrator is seeking and 
will be awarded the most points,  and the others will be ranked with lower 
values. The least likely answer (or in effect, the wrong one) will get 
either very little points attributed to it or none at all. Marketing 
surveys, as you point out for the one you've been approached with in the 
mall, are all collectively analyzed this way - it's mathematically how you 
measure behaviors and market trends or interest on subjective answers. 
Hence why I'd give my right arm to have have access to that database.

rebecca, who has been amusing herself reading all the outcries of dismay 
about the new "LeakyMug.com" :)