Compare and rank
Main-effects plot
Shows average or model-adjusted responses at each factor level.
- Plot
- Effect comparison
- Comparison
What it answers
How does response differ across each factor's levels?
- Shape
- Response summaries by factor level
- What it shows
- Main differences across factor levels.
Example
The question
Which factors move delivery time the most?
Average delivery days by level of each factor (main effects plot)
Illustrative data. It belongs to no company.
When to use it
Use it when
Factorial exploration and model summaries.
Avoid it when
Confounding or interactions hidden behind a single average effect.
How to read it
Compare level summaries within each factor and verify raw/adjusted meaning.
Note
Distinguish raw means from adjusted estimates.
The data you need
- Layout
- Long table
- Fields
- Factor: text; Level: category; Response: number
- One row is
- One response assigned to a factor level, or a precomputed level estimate. What is the grain?
- Rules
- Specify raw mean versus adjusted estimate. In multifactor raw data, use one column per factor and a response column.
- Optional
- Precomputed alternative: Factor,Level,Estimate,Lower,Upper.
- Preparation
- Calculate raw or model-adjusted level estimates, distinguishing main effects from interaction-dependent differences.
- Units
- Use the measured variable's stated unit consistently. No denominator is implied unless the value is a rate or proportion.
Worked example
Low responses 10,12 average 11; High 18,20 average 19. Raw mean difference is 8; adjusted effects may differ.
| Factor | Level | Response |
|---|---|---|
| Temperature | Low | 10 |
| Temperature | Low | 12 |
| Temperature | High | 18 |