Front | Back |
What does a test need to have in order for it to be useful?
|
- tells you the practical payoffs of the test
- supports (or disconfirms) use of test with numerical values - helps in decision making regarding the implementation of a certain test (and can consider the relative to its cost). |
The Utility of Using a Test Tends to be Greatest When...
|
- you have many job openings
- you have many more applicants than openings - you have high test validity - the current selection system is poor (base rate) - there are high dollar values associated with individual differences in job performance - tenure on the job is high |
5 potential components of utility
|
- selection ratio
- validity coefficient (r) - Base rate of current performance - SDy - Cost of testing |
The ratio between the number of openings to the number of applicants
|
Selection ratio
|
The percentage of employees currently on the
job who are considered successful. |
- Base rate of current performance
|
The difference in performance (measured in dollars)
between a high performing and average performing worker. i.e., how many more dollars is a high performer worth as compared to an average performer? A research-supported estimate is 40% of the position’s salary. |
SDy
|
(#applicants)*(cost per application)
|
Cost of testing
|
4 common utility methods
|
- proportion of correct decisions (decision accuracy)
- Taylor russell tables - Lawshe tables - The Brogden-Cronbach-Gleser model |
Can look at varous proportions to determine how much accuracy the test provides in deciding who to hire and who not to hire
|
Proportions of corret decisions
|
You can only do the proportion of correct decisions if you know...
|
- employee or applicant test scores
- criterion scores (eg. job performance scores) |
What is in Q1, Q2, Q3, Q4
|
Q1 = false negatives (incorrect rejections)
Q2 = true positives (correct accpetances) Q3 = false positives (incorrect acceptances) Q4 = true negatives (correct rejections) |
Provides some information about the effects of using the test on hiring decisions.
|
Proportion of correct dcisions
|
Proportion of Correct Decisions provides some comparison information on __ of hiring decisions made with and without a test.
|
Accuracy
|
Estimates the percentage of future employees that will b successful if we use a certain test
|
Taylor russell tables
|
Taylor-Russell Tables uses three sources of information
|
- test validity coefficient (r)
- base rate - selection ratio |