It is commonly referred to as the normal range or normal values. This term should be discouraged as not everyone outside the range is abnormal, and people who have a particular condition may still fall within this range.
A reference range for a particular test or measurement, is usually defined as the values that 95% (or 2 standard deviations) of the population for into. It relies on the fact that for many biological phenomena, there is a normal distribution of values.
For some types of investigations, there may not be a normal distribution. There could be a bimodal distribution, often caused by difference values in males and females. Or it may change with age, (such as prostate specific antigen[?]). There may also be another distribution such as lognormal distribution, where a mathematical function can be applied to correct it.
A good test (or measurement) in the diagnosis of a disease will have a clear distinction between the ranges for normal and the ranges for diseased individuals.
A graphical illustration of reference ranges

In an ideal situation, the reference range can be obtained by simply measuring a population and taking two standard deviations either side of the mean. 

If a normal population shows a bimodal distribution[?], it is useful to find out why this is the case. Two reference ranges can be established for the two different groups of people. (This pattern is commonly seen in tests that differ between men and women, but could conceivably be seen in a bicultural society and other situations.) 
See also:
Search Encyclopedia

Featured Article
