Home
Back
Error Of Difference Formula:
| From: | To: |
The Error of Difference (SED) is a statistical measure that quantifies the standard error of the difference between two measurements or statistics. It's commonly used when comparing two means or proportions to determine if their difference is statistically significant.
The calculator uses the Error of Difference formula:
Where:
Explanation: The formula combines the standard errors of two independent measurements to estimate the standard error of their difference. This is based on the property that variances (squared standard errors) of independent variables add.
Details: The SED is crucial for hypothesis testing when comparing two statistics. It's used to calculate confidence intervals for differences and to perform significance tests (z-tests or t-tests).
Tips: Enter the standard errors for both measurements. Both values must be non-negative. The result will be the standard error of the difference between the two measurements.
Q1: When should I use the Error of Difference?
A: Use it when you need to compare two independent statistics and want to know if their difference is statistically significant.
Q2: Can I use this for dependent measurements?
A: No, this formula assumes independent measurements. For dependent measurements, you need to account for their covariance.
Q3: How is SED different from pooled standard error?
A: SED is for comparing two statistics, while pooled SE combines estimates from multiple samples assuming equal variances.
Q4: What's the relationship between SED and confidence intervals?
A: The 95% CI for the difference is typically: (difference) ± 1.96*SED (for large samples).
Q5: Can I use this for proportions as well as means?
A: Yes, the same formula applies to the standard errors of proportions when comparing two independent proportions.