Lecture 6: Selection decisions

Question 1

Validity

Answer 1

The degree to which inferences from test scores are justified by the evidence (Does the instrument measure what it is supposed to measure?)

Question 2

Constuct validity

Answer 2

Are you measuring what you’re supposed to measure

Question 3

Content validity

Answer 3

The extent to which the items on a test are fairly representative of the entire domain the test seeks to measure

Question 4

Criterion validity

Answer 4

The extent to which an operationalization of a construct relates to a theoretical representation of the construct

Question 5

Concurrent validity (Part of criterion validity)

Answer 5

Your measure should correlate highly with other related measures which you expect it to correlate with and don’t correlate with constructs that don’t measure the same

Question 6

Predictive validity (Part of criterion validity)

Answer 6

Your measure should be correlated with a future outcome in the way you expect it to be

Question 7

Face validity

Answer 7

The extent to which a test is subjectively viewed as covering the concept it is supposed to measure –> subjective

Question 8

Reliability

Answer 8

The extent to which a score from a test or from an evaluation is consistent and free from error (Does the instrument measure and perform well?)

Question 9

Temporal stability

Answer 9

Test-retest reliability–> testing at different time points should give the same results

Question 10

Form stability

Answer 10

Different forms of tests should give the same results

Question 11

Internal reliability

Answer 11

It all the items in the test measure the same (Cronbach’s alpha)

Question 12

Scorer reliability

Answer 12

Objective evaluation of scores–> In structured interviews scorer reliability is high, panel members often agree

Question 13

Reliable, not valid

Answer 13

Consistent but not measuring what it’s suppost to measure

Question 14

Valid, not reliable

Answer 14

Not consistent but measures what it is supposed to measure

Question 15

Neither reliable or valid

Answer 15

Something that doesn’t measure what it’s supposed to and is different every time you test it

Question 16

Both reliable and valid

Answer 16

Measures what it’s supposed to and is consistent

Question 17

Utility

Answer 17

The degree to which a selection device improves the quality of a personnel system, above what would have occurred had the instrument not been used

Question 18

Taylor-Russel tables

Answer 18

Designed to estimate the percentage of future employees who will be successful on the job if a particular selection method is used

Question 19

Lawshe tables

Answer 19

Tables that use the base rate, test validity and applicant percentile on a test to determine the probability of future success for that applicant

Question 20

Proportion of correct decision

Answer 20

Refers to a utility method that compares the percentage of times a selection decision was accurate with the percentage of succesful employees

Question 21

Utility formulas

Answer 21

Provides an estimate of the amount of money an organization will safe if it adopts a new testing procedure

Question 22

Measurement bias

Answer 22

Technical aspects of the test A is biased if there are group differences in test scores (race, gender) that are unrelated to the construct being measured

Question 23

Predictive bias

Answer 23

A situation in which the predicted level of job success falsely favours one group over another

Question 24

Unadjusted Top-down selection

Answer 24

A “performance first” hiring formula

Question 25

Passing scores

Answer 25

Who will perform at an acceptable level? (passing score is a point in a distribution of scores that distinguishes acceptable from unacceptable)

Question 26

Banding

Answer 26

A compromise between the top-down and passing scores approach. Takes into account that tests are not perfectly reliable because of error variance

Question 27

SEM Banding

Answer 27

Based on the concept of Standard Error of Management –>
- Non sliding band: hire anyone whose scores fall between the top score and the band score
- Sliding band: start with the highest score and subtract from it the band width