Hierarchical multiple regression, also known as setwise regression, is a method where predictor variables are entered into the regression model in a pre-determined order. This approach assesses the incremental contribution of different predictor sets in explaining the variance in the dependent variable.
We aim to predict job performance (\(Y\)) using:
Step 1: Control Variables – Age, Gender \[ Y = B_0 + B_1 (\text{Age}) + B_2 (\text{Gender}) + e \]
Step 2: Cognitive Ability – IQ, Problem-Solving Skills \[ Y = B_0 + B_1 (\text{Age}) + B_2 (\text{Gender}) + B_3 (\text{IQ}) + B_4 (\text{Problem-solving}) + e \]
Step 3: Work Experience – Years in the Industry, Previous Job Roles \[ Y = B_0 + B_1 (\text{Age}) + B_2 (\text{Gender}) + B_3 (\text{IQ}) + B_4 (\text{Problem-solving}) + B_5 (\text{Experience}) + e \]
| Step | R2 | Delta_R2 | p_value |
|---|---|---|---|
| Step 1: Control Variables | 0.10 | - | < 0.05 |
| Step 2: + Cognitive Ability | 0.40 | 0.3 | < 0.001 |
| Step 3: + Work Experience | 0.45 | 0.05 | 0.06 (NS) |
Step 1: Demographics explain 10% of the variance in job performance.
Step 2: Adding cognitive ability increases Rsquared by 30%, meaning IQ and problem-solving skills are strong predictors.
Step 3: Work experience adds only 5% more, and the change is not statistically significant (𝑝=0.06).