Compute the CI of a mean, a proportion (Wilson), a mean difference (Welch) or a correlation (Fisher z), at 90, 95 or 99%, with the APA 7 sentence ready.
A confidence interval gives the range of values of the parameter (the population mean, proportion or correlation) that are compatible with your data. "95%" describes the procedure: if you repeated the study many times and computed the interval each time, 95% of those intervals would contain the true value. With a single study you cannot know whether yours is one of the hits; what you do know is that the narrower it is, the more precise the estimate. I explain it slowly in what a confidence interval actually is.
In 40 patients, the mean anxiety score is 24.3 with a standard deviation of 6.1. The standard error is 6.1 / √40 = 0.96; the critical t with 39 degrees of freedom is 2.023 (it is in the t, z, chi-square and F tables); the margin is 2.023 × 0.96 = 1.95. The interval runs from 22.35 to 26.25. In APA 7: "M = 24.30, 95% CI [22.35, 26.25]".
State the confidence level once and put the interval in square brackets, with the two limits separated by a comma: "95% CI [22.35, 26.25]". If every interval in the paper is 95%, saying so the first time is enough. For statistics that cannot exceed 1 (a correlation, a proportion given as a decimal) drop the leading zero: r = .35, 95% CI [.16, .51].
If your data break the assumptions, the resampling interval is the alternative: see how to report bootstrap confidence intervals in APA 7. And to read the intervals in a paper correctly, interpreting confidence intervals.
If you use it in your thesis or in a paper, cite it so that readers of your Methods section know how the calculation was done.
Ciudad-Fernández, V. (2026). Confidence interval calculator [Web application]. Ciudad Science. Retrieved October 14, 2026, from https://analisisdedatospsicologia.com/en/tools/confidence-interval-calculator
For your Methods section: Confidence intervals were computed with the Ciudad Science confidence interval calculator (Ciudad-Fernández, 2026): from the t distribution for means and mean differences (Welch, 1947), by the Wilson (1927) method for proportions and through Fisher’s z transformation for correlations.
@misc{ciudadfernandez2026intervaloconfianza,
author = {Ciudad-Fern{\'a}ndez, V{\'i}ctor},
title = {{Confidence interval calculator}},
year = {2026},
publisher = {Ciudad Science},
howpublished = {Web application},
url = {https://analisisdedatospsicologia.com/en/tools/confidence-interval-calculator},
urldate = {2026-10-14},
note = {Accessed 2026-10-14},
}