Related Experiment Videos
Maintaining standards: differences between the standard deviation and standard error, and when to use each
1Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario. streiner@fhs.csu.mcmaster.ca
Canadian Journal of Psychiatry. Revue Canadienne De Psychiatrie
|October 1, 1996
Summary
Standard deviation (SD) measures data variability and should always be reported. Standard error of the mean (SE) estimates population variability and is used for confidence intervals, aiding data interpretation.
Area of Science:
- Statistics
- Data Presentation
- Scientific Communication
Background:
- Many researchers confuse standard deviation (SD) and standard error of the mean (SE).
- Uncertainty exists regarding the appropriate use of SD and SE in data presentation.
- Understanding these measures is crucial for accurate data interpretation.
Purpose of the Study:
- To clarify the distinct roles of standard deviation (SD) and standard error of the mean (SE).
- To provide guidance on selecting the appropriate measure for data presentation.
- To explain the utility of SE in constructing confidence intervals.
Main Methods:
- The abstract explains the statistical definitions of SD and SE.
- It describes the application of SE in calculating confidence intervals (CIs).
- It recommends using error bars representing 95% CIs (±2 SEs) in graphical data presentation.
Main Results:
- Standard deviation (SD) quantifies the variability of individual data points.
- Standard error of the mean (SE) quantifies the variability of sample means.
- SE is essential for calculating confidence intervals (CIs), which estimate the true population value.
Conclusions:
- SD should be reported in all studies to show data variability.
- SE is valuable for constructing confidence intervals (CIs), indicating estimate precision.
- Using CIs (e.g., ±2 SEs) in graphs enhances the reader's understanding of findings' accuracy.