Detailed Explanation
The mean of \( n \) observations depends on the sum of all observations, so if one value changes, the sum changes and hence the mean changes. The standard deviation depends on the sum of squares of the observations. A convenient formula for variance is \( \sigma^2 = \frac{\sum x_i^2}{n} - \bar{x}^2 \). From the given mean we can recover \( \sum x_i \), and from the given standard deviation we can recover \( \sum x_i^2 \). Then we remove the wrong observation's contribution and add the correct observation's contribution to both the sum and the sum of squares. Finally we recompute the mean and the standard deviation.
Formula Used
\[ \bar{x} = \frac{\sum x_i}{n}, \quad \sigma^2 = \frac{\sum x_i^2}{n} - \bar{x}^2 \]
Solution
- Step 1: Write down the given data. Number of observations \( n = 20 \), incorrect mean \( \bar{x} = 10 \), incorrect standard deviation \( \sigma = 2 \). The wrong value \( 8 \) is to be replaced by \( 12 \).
- Step 2: Find the incorrect sum of observations using the mean formula. \[ \bar{x} = \frac{\sum x_i}{n} \] So \[ \sum x_i = 20 \times 10 = 200 \]
- Step 3: Correct the sum by removing \( 8 \) and adding \( 12 \). \[ \text{Correct sum} = 200 - 8 + 12 = 204 \]
- Step 4: Compute the correct mean. \[ \text{Correct mean} = \frac{204}{20} = 10.2 \]
- Step 5: Find the incorrect sum of squares using the variance formula. \[ \sigma^2 = \frac{\sum x_i^2}{n} - \bar{x}^2 \] With \( \sigma = 2 \), \( \sigma^2 = 4 \), so \[ 4 = \frac{\sum x_i^2}{20} - 100 \] This gives \[ \sum x_i^2 = 20 \times 104 = 2080 \]
- Step 6: Correct the sum of squares by removing \( 8^2 \) and adding \( 12^2 \). \[ \text{Correct } \sum x_i^2 = 2080 - 8^2 + 12^2 = 2080 - 64 + 144 = 2160 \]
- Step 7: Compute the correct variance using the corrected sum of squares and corrected mean. \[ \sigma^2 = \frac{2160}{20} - (10.2)^2 \] \[ \sigma^2 = 108 - 104.04 = 3.96 \]
- Step 8: Take the square root to get the correct standard deviation. \[ \sigma = \sqrt{3.96} \approx 1.99 \]
Final Answer
The correct mean is \( 10.2 \) and the correct standard deviation is \( \sqrt{3.96} \approx 1.99 \).
Common Mistake
A very common mistake is to correct only the sum (and hence the mean) but forget to correct the sum of squares \( \sum x_i^2 \) before recomputing the standard deviation. You must remove \( 8^2 \) and add \( 12^2 \), and also use the new corrected mean \( 10.2 \), not the old mean \( 10 \), in the variance formula.