Quantitative/ Analysis
Hierarchical block regression
Ask what a prespecified predictor block adds
Hierarchical block regression enters predictor sets in a substantively justified sequence and compares nested models, commonly through changes in explained variance. The hierarchy is the order of blocks, not a hierarchy of people inside schools or firms. It differs from automated stepwise selection and from multilevel regression with group-level effects.
WHEN IT FITS
Use to evaluate whether a theoretically motivated block adds information beyond a prespecified baseline model. Explain why the order answers the research question and ensure models use comparable observations.
Strengths
- Makes incremental explanatory questions explicit
- Separates baseline adjustment from focal predictor blocks
Limitations
- Incremental variance depends on entry order
- Data-driven blocks invite selective reporting and unstable conclusions
Know the boundary
Hierarchical block entry is not multilevel modelling and cannot account for clustered observations merely by adding blocks.
USED ACROSS
Banking & financeBusiness & MBAPsychologyComputer scienceEducation