IMED2003 Study Guide - Final Guide: Confidence Interval, Statistical Parameter, Observational Error
10 Feb 2019
School
Department
Course
Professor
Document Summary
It reflects the precision of the estimate. We took a sample to estimate our measure of association. We usually calculate 95% ci as a measure of this. The wider the ci, the less precise is our estimate measure of association. This provides indication of statistical significance (if have null = not significant) We will find that 95% of the time the estimate would be contained within the interval. What the 95% ci represents is: the inaccuracy of the sample in estimating the population parameter of the measure of association for a given health outcome. I. e. represents the inaccuracy of the sample in estimating the population parameter caused by random error. Observational comparative studies will allow us to estimate the association between an exposure and an outcome. But this does not necessarily imply causation! In cohort studies, the relative risk and risk difference and odds ratio can be calculated.
