A confidence interval is a range of plausible values for a true population quantity, calculated from your sample, together with a stated level of confidence such as 95 percent. The level describes the long-run reliability of the method: if you repeated your study many times, about ninety-five percent of the intervals built this way would contain the true value. So a confidence interval reports not just your best estimate but the uncertainty around it, which a single number can never do.
Why the interval beats a lone point estimate
A point estimate, a single mean or difference, hides the fact that your sample is only one draw from the population. The confidence interval makes that uncertainty visible by showing how far the truth could reasonably sit from your estimate. A narrow interval signals a precise estimate, usually from a larger sample; a wide interval warns the reader that your data cannot pin the value down tightly. Reporting the interval rather than the bare figure is exactly the discipline that separates an inferential claim from a descriptive one.
Reading the bounds correctly
The subtle part is what the confidence level attaches to. The ninety-five percent describes the procedure, not a single interval; once you have computed one specific interval, the true value is either inside it or not. The honest reading is "I am ninety-five percent confident that this range captures the true value," meaning the method that produced it succeeds ninety-five percent of the time. A 90 percent interval is narrower but less reliable, and a 99 percent interval is wider but more cautious, so the level you choose trades precision against certainty.
What the interval tells you about significance
A confidence interval also answers the question a p-value answers about significance, and it does so with more information. If a 95 percent interval for a difference between two groups excludes zero, the result is significant at the .05 level; if it includes zero, it is not. For an odds ratio or relative risk, the no-effect value is one rather than zero. The advantage over a bare p-value is that the interval shows the plausible range of the effect, so you can see at a glance whether a significant result is also large enough to matter, the same logic behind a reported effect size.
| Interval feature | What it tells you |
|---|---|
| Narrow width | Precise estimate, usually a larger sample |
| Wide width | Imprecise estimate, treat with caution |
| Excludes zero (or one for ratios) | Effect is statistically significant |
| Higher confidence level | Wider interval, more cautious claim |
Writing it up in your dissertation
Report the point estimate first, then the interval in square brackets, for example a mean difference of 4.8 with a 95 percent confidence interval of 1.2 to 8.4. State the units, name the quantity the interval refers to, and interpret the width in words rather than leaving the reader to do it. Pairing the interval with its effect and significance gives a fuller picture than any single number, and the precise formatting, including brackets, decimals, and spacing, follows how to report statistics in APA style. Done this way, your confidence intervals turn raw output into a claim your committee can trust.