The margin of error of a sample must not be within acceptable limits.

The sampling frame for any probability sample is a small list of all the cases in the population from which your sample will be drown.

If 60 per cent of your sample were small service sector companies then, provided the sample was representative, you would expect 30 per cent of the target population to be small service sector companies.

Generalisations about target populations from data collected using any probability samples are based on statistical probability.

Each non-response will necessitate an extra respondent being found to reach the required sample size, increasing the cost of your data collection.

As part of your research report, you will not need to include your response rate.

By defining the sampling frame you are defining the target population about which you want to generalise.

With non-probability sampling, the probability of each case being selected from the target population is not known.

Statisticians have proved that the larger the absolute size of a sample, the closer its distribution will be to the normal distribution and thus the more robust it will be.

For all research questions where it would be impracticable for you to collect data from the entire population by observation, you need to select a sample.