Reliability and validity

The accuracy of psychometric tests is measured by reliability and validity.

Reliability - no measure can be of much value unless it measures in a reliable or consistent manner. Reliability refers to the consistency of test items as selected by test takers as well as scores obtained when re-tested with the same assessment on different occasions.

Test reliability correlation coefficients should be in the .70's, .80's and .90's (U.S. Dept. of Labor, 1999).

The ProfileXT scores an average of .82 coefficient alpha across the nine personality scales it measures.

Validity - this is the extent to which an assessment measures what it is supposed to evaluate. If it is not valid then it is a waste of money and time.

Various approaches to gathering validity evidence have been analysed including Construct Validity, the Guilford-Zimmerman Temperament Survey, Gordon Personal Profile - Inventory, California Psychological Inventory and the Criterion-Related Validity Approach.

These have all reinforced the predictive nature of the ProfileXT.

Statistical analysis of the ProfileXT clearly shows that it consists of nine reliable scales that provide an accurate measure of nine core dimensions of normal adult personality that are useful for predicting a number of important business-related criteria.

Click here to contact us for an Executive summary of the ProfileXT