What do I use

Public health and epidemiology

Prevalence surveys, outbreak work, health systems research, programme evaluation.

The usual order: Define the population - sample it - describe with weights - compare groups - map the pattern - report the denominator.

Your question, and what answers it

A proportion with a Wilson interval, which behaves near 0 and 100% where the ordinary one does not.

In ZStat: open your study, go to Analysis, and choose survey proportion.

What goes wrong in public health and epidemiology

Comparing rates across sites without checking the denominator means the same thing in both. Most disputes about a rate are disputes about the denominator.
Inferring individual behaviour from district averages. The relationship can reverse.

ZStat runs these checks whether or not you ask, and stops the analysis when one of them would make the result misleading. You do not have to remember this list.

Other fields

Medicine and clinical research Psychology and behavioural science Education Agriculture and food science Urban planning, estate management and architecture Economics, finance and accounting Engineering, manufacturing and quality Sociology, development and social work

Start free All procedures Questions