# Residual plot (method comparison)

A residual plot shows the difference between the measured values and the predicted values against the true values.

The residual plot shows disagreement between the data and the fitted model. The ideal residual plot (called the null residual plot) shows a random scatter of points forming an approximately constant width band around the identity line.

It is important to check the fit of the model and the assumptions:

Assumption | How to check |
---|---|

Model function is linear | The points will form a pattern when the model function is not linear. |

Constant variance | If the points tend to form an increasing, decreasing, or non-constant width band, the variance is not constant and you should consider using weighted regression. |

Normality | A histogram of the residuals should form a normal distribution. This is an assumption of linear regression. Deming regression with Jacknife standard errors is robust to this assumption. Passing-Bablok regression is non-parametric and this assumption does not apply. |

- What is Analyse-it?
- What's new?
- Administrator's Guide
- User's Guide
- Statistical Reference Guide
- Distribution
- Compare groups
- Compare pairs
- Contingency tables
- Correlation and association
- Principal component analysis (PCA)
- Factor analysis (FA)
- Item reliability
- Fit model
- Method comparison
- Correlation coefficient
- Scatter plot
- Fit Y on X
- Fitting ordinary linear regression
- Fitting Deming regression
- Fitting Passing-Bablok regression
- Linearity
- Residual plot
- Checking the assumptions of the fit
- Average bias
- Estimating the bias between methods at a decision level
- Testing commutability of other materials
- Difference plot (Bland-Altman plot)
- Fit differences
- Plotting a difference plot and estimating the average bias
- Limits of agreement (LoA)
- Plotting the Bland-Altman limits of agreement
- Mountain plot (folded CDF plot)
- Plotting a mountain plot
- Partitioning and reducing the measuring interval
- Agreement measures for binary and semi-quantitative data
- Chance corrected agreement measures for binary and semi-quantitative data
- Agreement plot
- Estimating agreement between two binary or semi-quantitative methods
- Study design
- Study design for qualitative methods
- Measurement systems analysis (MSA)
- Reference interval
- Diagnostic performance
- Control charts
- Process capability
- Pareto analysis
- Study Designs
- Bibliography

Version 5.60

Published 27-Apr-2020