Biology, images, analysis, design...
Use/Abuse Principles How To Related
"It has long been an axiom of mine that the little things are infinitely the most important" (Sherlock Holmes)




Using range to estimate SD (σ)

This provides a quick and simple estimate of standard deviation. Providing your data approximate to a normal distribution, their population standard deviation (σ) can be estimated from the range.

Algebraically speaking -

σ =   Range multiplier(n)

where :

  • σ is the estimated standard deviation,
  • the range is the maximum - minimum
  • the multiplier(n) is looked up in the appropriate table for that number of observations.

Table 1 gives the values of the multiplier for n = 1 to 1000.

Let's work this out for our lamb PCVs

Worked example

Lamb No.PCV
Range33-26 = 7

We will assume the data follow a normal distribution.

The range estimate of the standard deviation (σ) is therefore:

Range Estimate of σ    =    Multipliern=7 Range
     =    2.59

Τhis compares with a value of 2.95 using the corrected average of the sum of the squared deviations from the mean.

The range estimate provides a quick and easy way to check the standard deviation when calculated in the conventional way.

For small sample sizes (n<10) the range estimator is surprisingly efficient (>85%). Efficiency in this sense means that to get the same accuracy for s calculated when n=10, we would have to use a sample size of 12 for the range estimator. Hence the efficiency is approximately 10/12 or 85%.