Ch 6 Flashcards
reliability
the degree to which interviews, tests, and other selection procedures yield comparable
data over time
validity
the degree to which a test or selection procedure measures a person’s attributes and
predicts future behaviour
- Reliability is required for a measure to have validity
Initial screening
cover letters and résumés, application forms, Internet
checks and phone screening
Employment interviews
nondirective, structured
nondirective interview
an interview in which the applicant is allowed the maximum amount of freedom in determining the course of the discussion, while the interviewer carefully refrains from influencing the applicant’s remarks
structured interview
an interview in which a set of
standardized questions with an established set of answers is
used
- Situational interview (applicant is given a hypothetical incident and asked how they would respond to it)
- Behavioural description interview (applicant is asked questions about what they did in a given situation)
Pre-employment tests
job knowledge tests, work sample tests, cognitive ability tests, personality tests, polygraph tests, honesty and integrity tests, physical ability tests, medical examinations, drug testing,
assessment centres
Criterion-related validity
the extent to which a selection
tool predicts, or significantly correlates with, important
elements of work behaviour
content validity
the extent to which a
selection instrument, such as a test,
adequately samples the knowledge and skills
needed to perform particular job tasks
construct validity
the extent to which a
selection tool measures a theoretical construct or trait
Summarizing
Decision-making
(clinical approach, statistical approach)
clinical approach
those making the selection decision review all of
the data on the applicants; then, on the basis of their understanding
of the job and the individuals who have been successful in that job,
they make a decision
statistical approach
involves identifying the most valid predictors and weighting them using statistical methods, such as multiple regression