Work mostly by CE other than CORE (for which, see : www.coresystemtrust.org.uk)
Means what it says but be careful …
It means a subgroup or subset of a population not a sample. So unless you have a complete census of data for a finite population it’s something that will be being extrapolated/generalised to from data from a subgroup of a sample: you will have the statistics for your subsample but are interested in the subpopulation.
For example if you are interested to know whether full time employed clients coming to your service DNA more than part-time or unemployed clients you can know this from your data but you are probably really interested in whether any difference there will generalise to the subpopulation: i.e. all fully employed people who could come to the service ever (yes it’s hypothetical). The maths of that are the best guide you can have to whether the differences you may or may not have seen in the data you have so far will predict a systematic difference in DNA rate in future clients according to their employment status. Here this hypothetical “any fully employed clients ever …” is the subpopulation of the population of “all possible clients ever”.
Mostly chapter 5 but these distinctions of subsamples from whole samples and of sample data from population data, and likewise of subsample data from subpopulation data, run through the chapters 7 and 8.
Powered by BetterDocs