IOP4863
Assessment 05
Unique No: 148704
Due Date 6 October 2025
, IOP4863
Assessment 05
Unique Number: 148704
Due Date: 6 October 2025
PART A: PERSONNEL PSYCHOLOGY
Question 1.1: What are differential validity and differential prediction, and how do
they apply in personnel selection?
Differential validity refers to whether a selection tool has the same level of validity for
different groups. A test may predict managerial success well for one group but poorly for
another, which raises concerns about fairness and bias (Gatewood, Feild & Barrick,
2019).
Differential prediction, on the other hand, looks at whether the tool predicts performance
outcomes equally across groups. A test could show equal validity coefficients but still
systematically over- or under-predict for certain groups. For example, if a cognitive
ability test consistently under-predicts women’s performance, even though they perform
equally well in the role, it reflects differential prediction (Sackett, Borneman & Connelly,
2008).
In a selection committee, it is crucial to analyze test results for both differential validity
and prediction. If either is present, the fairness and defensibility of the selection process
are compromised. Adjusting tools, supplementing them with structured interviews, or
using multiple predictors helps address this issue.
Assessment 05
Unique No: 148704
Due Date 6 October 2025
, IOP4863
Assessment 05
Unique Number: 148704
Due Date: 6 October 2025
PART A: PERSONNEL PSYCHOLOGY
Question 1.1: What are differential validity and differential prediction, and how do
they apply in personnel selection?
Differential validity refers to whether a selection tool has the same level of validity for
different groups. A test may predict managerial success well for one group but poorly for
another, which raises concerns about fairness and bias (Gatewood, Feild & Barrick,
2019).
Differential prediction, on the other hand, looks at whether the tool predicts performance
outcomes equally across groups. A test could show equal validity coefficients but still
systematically over- or under-predict for certain groups. For example, if a cognitive
ability test consistently under-predicts women’s performance, even though they perform
equally well in the role, it reflects differential prediction (Sackett, Borneman & Connelly,
2008).
In a selection committee, it is crucial to analyze test results for both differential validity
and prediction. If either is present, the fairness and defensibility of the selection process
are compromised. Adjusting tools, supplementing them with structured interviews, or
using multiple predictors helps address this issue.
