A high PA Relevancy means that this test can indicate whether or not someone have prosopagnosia.
(This is all you really need to know. If you want the details, keep reading)

The PA Relevancy for a test ranges from 0% to 100%. If the PA Relevancy is 0%, then there is no correlation at all between the testresult and whether or not the person taking the test has PA (prosopagnosia). If the PA Relevancy is 100%, then all people who have PA have got a higher score than all the people who do not have PA (or the other way around). Both 100% and 0% PA Relevancy is of course highly unlikely in a test. Pure chance will always let someone get some points who did not know the right answers, and also make someone make a mistake on something they would usually do without problem.


A word of warning





How is the PA Relevancy calculated?

The PA Relevancy is based on the percent of people with PA who score like NT:s, and the percent of NT:s who score like people with PA . The sum of these percents I call X. (Is there a mathematical term for this already?)
"scoring like NT:s" and "scoring like PA:s" is defined as having a score above or below a the average score. Eg: If the average score of PA:s is higher than the average score NT:s, then "scoring like PA:s" is defined as scoring higher than the total average and "scoring like NT:s" is defined as scoring lower than the total average.


The PA Relevancy is calculated like this:
PA Relevancy = ( 100 - X )




The PA Relevancy will not be calculated for a test until at least three people who have PA, and three people who don't have PA, have taken the test and registered their scores.




If you wish to quote this index, please call it "Burman's PA Relevancy". It is of course possible to use this way of calculating a test's relevancy for other things than prosopagnosia. In that case, please call it "Burman's Relevancy".

I would be interested to hear when and where you use it. My mail address is cb@prosopagnosia.com.






Home / Face-Blindness