Key points are not available for this paper at this time.
Knowing the estimate (mean or proportion) for the whole population is not feasible due to a lack of resources. Therefore, we rely on a sample estimate to draw conclusions about the population. The range of values in which the true population estimate lies is called a confidence interval (CI) which is more reliable than single estimate. For descriptive studies, CIs should be reported along with the point estimate (mean or proportion). For effect sizes in inferential statistics (mean difference, odds ratio, risk ratio, etc.), the p-value should be supplemented with a CI for strength and direction of the association.
Hussain et al. (Fri,) studied this question.