To determine the appropriate sample size in a road safety study, we first determine the size of the statistical ‘population’. Next, we need to determine the confidence level required and the acceptable margin of error. 

If the statistical population is taken as, say, 1 million road users and a confidence of 95% is required the sample size at different margins of error is presented below. 

The graphs above show that to get reasonable confidence in your outcomes your sample size may not need to be as big as you think.