Home | Contact

Classical Test Theory

     

Classical test theory is based on the notion that a test consists of a series of items, with each item being one attempt to measure the psychological characteristic that is being assessed (e.g. extraversion, abstract reasoning ability, etc.). Each attempt at measuring the trait (i.e. each completed test item) is expected to produce a slightly different result, because each test item is not perfectly reliable. (That is, the person’s response to each test item is not only influenced by the psychological characteristic being measured, but is also influenced by random error.). The extent to which all the items are correlated with each other is taken as an estimate of the extent to which the items are reliably measuring the latent trait that is being assessed, and the extent to which they are not correlated with each other is taken as an estimate of the extent to which random error is effecting these items. As the error in each item is considered to be random (i.e. any given item is as likely to over estimate the characteristic being assessed as it is to underestimate it) averaging across the items increases the tests’ reliability by averaging out the random errors.

 
 

 

 
 

 
 

 

   

Copyright Psychometrics Ltd 2006 | Privacy policy
Disclaimer | email: info@practicetests.co.uk