Populations vs. Samples

A population is all conceivable observations of a particular type. A sample is a limited number of observations from a population. Using statistics, it is possible to make statements about what a population is probably like on the basis of information from a sample. It is also possible to make statements about what the sample will probably be like on the basis of information about the population.

Because of restrictions on research time and finances, it is not feasible to look at the entire population, instead a sample or samples are taken from the population, which are only parts of that population. For example, a researcher is interested in the learning ability of rats that receive a particular diet (he or she wants to compare the effect of this diet with that of another diet). There is no clear theoretical limit to the number of rats he or she could raise on the special diet, but for practical reasons, 50 are settled on for measuring the learning ability of rates. In this example the 50 rats are the sample.

The most important type of sample in research is a random sample. A random sample of a given population of a given size n is defined as a collection of n objects from a population selected in such a manner that each item in the population had an equal chance of being selected for that sample. This concept of drawing samples from a population is extremely important in research for the validity of the experiment and numerous methods have been devised for doing this. Bias can occur when samples are drawn so that all observations, animals or humans do not have an equal chance of getting selected. Randomization techniques include; assigning numbers to all members of the population and randomly selecting by drawing numbers from a hat; or using tables of random numbers. It is important to check books on selecting random samples for the method best suited for a student's research.