Unbiased estimates from raw data
The masses, in grams, of a random sample of 10 apples from an orchard are shown below.
(a) Find unbiased estimates of the population mean and variance of the masses.
(b) Explain briefly what is meant by saying that an estimate is unbiased.
- 1
Find , and from the raw values. Square each value before adding — is not .
- 2
The unbiased estimate of the mean is simply the sample mean.
- 3
Use the MF19 formula for the unbiased variance, which divides by , not . Keep exact.
- 4
(b) A single estimate may be too high or too low, but the method is correct on average: if many samples were taken, the mean of all the estimates would equal the population value.
Answer
(a) Mean g; variance g. (b) Individual estimates vary, but on average (over many samples) the estimates equal the true population value.
Examiner tip
Dividing by gives the biased value , which scores no method mark. The March 2026 examiner report noted: "A few candidates found the biased variance rather than the unbiased variance." Calculator statistics modes show both and — know which one is .