Define and describe types of bias in psychological testing. Discuss different types of predictive bias. Give examples of each type.

Question

What will be an ideal response?

Answer 1

• When the scores on a test taken by different subgroups in the population (e.g., men, women) need to be interpreted differently because of some characteristic of the test not related to the construct being measured, measurement bias is present in the test.

• Predictive bias occurs when the predictions made about a criterion score based on a test score are different for subsets of test takers (e.g., minority vs. majority, men vs. women).

• Predictive bias is indicated when a test has different validity coefficients for different groups.

• When a test yields significantly different validity coefficients for subgroups, we say it has differential validity.

• When a test is valid for one group but not for another group (e.g., valid for Whites but not for Blacks), it has single-group validity.

• Two common types of predictive biases are intercept bias and slope bias.
o When a common regression line for two groups is used to predict performance, but the individual regressions lines for the groups differ where they cross the Y axis, intercept bias is present.
o When intercept bias is present, one group’s performance would be over predicted and another group would be under predicted if the predictions were based on the common regression line.
o When the regression lines that predict performance for two groups have different slopes, slope bias is present.
o Slope bias suggests that there that the validity coefficient for the test would not be the same for each group (i.e., differential validity would be present).