MKT 317 Lecture Notes - Lecture 3: Observational Error, Predictive Validity, Job Performance
Document Summary
Work analysis and work design: understand what work needs to be done, design the work to be efficient and motivating, what k ao"s are (cid:374)eeded to get it do(cid:374)e. Now we need to hire people to do the work. The process by which companies decide who will or will not be allowed into the organization. Good measures should meet 5 criteria: reliable, valid, generalizable, utility, legal. Measured score = true score +/- random error. Interviews: unusual circumstances, distractions, biases, tests, poorly worded questions, scoring mistakes, surveys. Item ambiguity, judgment calls: references, mood of reference giver. Are the scores similar: repetition in raters, example: interview applicants with multiple interviewers. How accurately the measure predicts an aspect of job performance; it measures what its supposed to. A measure must be reliable in order to be valid. I design a test that i think predicts future job performance. I want to find out if it a(cid:272)tually does .