Simple random sampling

Explanations > Social Research > Sampling > Simple random sampling

Use | Method | Example | Discussion | See also

Use

Use in simple experiments that require a single sample to be taken from a given population or a representative sample frame.

The people in the sample frame must all be accessible and available.

Use when the target group is sufficiently large. Do not use when the target is a relatively small subgroup that might be missed by this method.

Method

Create the sample by selecting randomly from the sample frame or population.

This can be done using a paper list of random numbers, although these days a computer is often used.

Example

A person researching education levels within a company takes the full employee list and applies a random number algorithm to this in order to select people to interview.

Discussion

The basic principle of simple random sampling is like drawing names out of a hat and is based on the mathematical property that a truly random sample (if big enough) will be representative of the target population.

The simple random sample has two key properties:

1. Unbiased: Each unit has the same chance of being selected.
2. Independent: The selection of each unit is not affected by the selection of other units.

Random number generation is easy these days with a computer where, for example the Excel '=RAND()' function (just type it into any cell) generates a random number between 0 and 1. To generate a number between 0 and 5, multiply this by five and take the integer to round it down (eg. '=INT(5*RAND()').

A problem with random selection is that this is not always possible. For example to do a true random sample of the population of the USA, you would start with a list of everyone there, then select a random sample from this (very big) list, then access those people selected, no matter where they lived and whether or not they wanted to partake in the study.

See also