Robust
Robust statistics represent an alternative approach to parameter estimation, differing from nonrobust statistics (sometimes called classical statistics) in the degree to which they are affected by violations of model assumptions. Whereas nonrobust statistics are greatly affected by small violations of their underlying assumptions, robust statistics are only slightly affected by such violations. Statisticians have focused primarily on designing statistics that are robust to violations of normality, due to both the frequency of nonnormality (e.g., via outliers) and its unwanted impact on commonly used statistics that assume normality (e.g., standard error of the mean). Nevertheless, robust statistics also exist that minimize the impact of violations other than nonnormality (e.g., heteroscedasticity).
In evaluating the robustness of any inferential statistic, one should consider both ...
Looks like you do not have access to this content.
Reader's Guide
Descriptive Statistics
Distributions
Graphical Displays of Data
Hypothesis Testing
Important Publications
Inferential Statistics
Item Response Theory
Mathematical Concepts
Measurement Concepts
Organizations
Publishing
Qualitative Research
Reliability of Scores
Research Design Concepts
Research Designs
Research Ethics
Research Process
Research Validity Issues
Sampling
Scaling
Software Applications
Statistical Assumptions
Statistical Concepts
Statistical Procedures
Statistical Tests
Theories, Laws, and Principles
Types of Variables
Validity of Scores
- All
- A
- B
- C
- D
- E
- F
- G
- H
- I
- J
- K
- L
- M
- N
- O
- P
- Q
- R
- S
- T
- U
- V
- W
- X
- Y
- Z