Statnote 18: comparison of regression lines
Anthony Craig Hilton, Richard A. Armstrong
Abstract
Anthony Craig Hilton, Richard A. Armstrong
Abstract
In many circumstances, it may be of interest to discover whether two or more regression lines are the same. Regression lines may differ in three properties, viz., in residual variance, in slope, and in elevation; all of which can be tested using analysis of covariance. If there are no significant differences between regression lines, an investigator may which to combine the data from different studies and fit a single regression line to the whole of the data.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In many circumstances, it may be of interest to discover whether two or more regression lines are the same. Regression lines may differ in three properties, viz., in residual variance, in slope, and in elevation; all of which can be tested using analysis of covariance. If there are no significant differences between regression lines, an investigator may which to combine the data from different studies and fit a single regression line to the whole of the data.
Key concepts: Regression, Regression diagnostic, Regression dilution, Statistics, Regression analysis, Cross-sectional regression, Segmented regression, Residual