A true or complete census is a complete dataset about a field of interest. If a dataset is a true census then it is not a sample and no statistical inference or estimation applies to it. For example if we do have the ages of all attenders at a therapy service say in 2019 and the gender of all those people then looking that the difference between the mean ages of employed and unemployed people is accurate: the mean difference is the mean difference, in principle a statistical test of the difference is not needed nor is the confidence interval around the observed difference. However, we often apply these methods to such data as we want sensible ways to extrapolate from the census, to generalise, e.g. to what we might expect in other years.

Try also … #


Chapters #

Nothing here yet!

Powered by BetterDocs