Related Experiment Videos
Correct and incorrect estimation of within-day and between-day variation
Clinical Chemistry
|September 1, 1986
Summary
Estimating between-day variance is often done incorrectly, leading to significant errors. This study highlights the biases in common variance estimation methods and recommends standardized approaches for accurate precision estimates.
Area of Science:
- Statistics
- Biostatistics
- Experimental Design
Background:
- Between-day variance is frequently misinterpreted, leading to inaccurate calculations.
- Existing statistical methods like analysis of variance (ANOVA) are underutilized for precise estimation.
- Intuitive but incorrect methods can introduce substantial systematic bias.
Purpose of the Study:
- To predict the error magnitude from common intuitive variance estimation approaches.
- To evaluate the impact of estimating total variance versus pure between-day variance.
- To assess the effects of biased estimators on variance component accuracy.
Main Methods:
- Application of statistical theory to model estimation errors.
- Comparative analysis of different variance estimation techniques.
- Evaluation of bias in variance component estimation.
Main Results:
- Intuitive methods for estimating variance components can yield errors of several hundred percent.
- Estimating total population variance instead of pure between-day variance introduces bias.
- Biased estimators significantly skew the accuracy of variance component estimates.
Conclusions:
- Commonly used methods for estimating between-day variance are prone to significant systematic bias.
- Recommendations are provided to mitigate bias and standardize precision estimation.
- Accurate variance component estimation is crucial for reliable scientific conclusions.