Cross Validated
2026-11-13 14:15 UTC
By Emily
AI-113-20261113-social-media-e30293dd
When should confidence intervals be reported for descriptive statistics?
I have noticed that observational studies often report proportions without confidence intervals in Table 1 but include 95% CIs for descriptive statistics elsewhere in the analysis. What determines whether confidence intervals should accompany descriptive statistics? Is the distinction based on whether the aim is simply to summarise the observed sample vs estimate characteristics of a wider population? What is the statistical reasoning for this?
I have noticed that observational studies often report proportions without confidence intervals in Table 1 but include 95% CIs for descriptive statistics elsewhere in the analysis. What determines whether confidence intervals should accompany descriptive statistics? Is the distinction based on whether the aim is simply to summarise the observed sample vs estimate characteristics of a wider population? What is the statistical reasoning for this?
Full article content could not be extracted automatically. Read the original below.
Source:
Cross Validated
· stats.stackexchange.com