MGT 315 Lecture Notes - Lecture 3: Happy Hour, Femininity, Extraversion And Introversion
Document Summary
Needs to be consistent and free from error. Repetition in times: someone takes a test and then retakes it and does nothing to change then their results should be the same unless they do something different. Repetition in raters: have multiple people interview someone and see what their scores are to see if there is a human error or if they get the score they deserve. How accurately the measure predicts an aspect of job performance, basically it measures what it"s supposed to. A measure must be reliable in order to be valid. Ex: if we had a test on nuclear physics it could be valid able but not measure what it"s suppose to. Concurrent validation: its easy to implement but doesn"t give us the best result. Predictive validation: you test someone and see if it can predict something useful or if it doesn"t. Are we measuring what we want to measure.
