When discussing confidence intervals, it is important to be aware of the tension between confidence and precision.

As previously stated, confidence is reflected by the confidence level we choose; logically, a higher confidence level means more confidence. The higher the confidence level we choose, the wider the interval gets – and the wider the interval is, the less the precision we get.  

Lower confidence level	Slimmer confidence interval	More precision
Higher confidence level	Wider confidence interval	Less precision

Example Let us return to the example with the age of sociology students at Stockholm University. To start with, we found that their mean age is 26 years. The 95% confidence interval was 25-27. Thus, we could conclude, with 95% confidence that the mean age of sociology students at Stockholm University is 25-27 years of age. But what if we wanted to be more certain about the mean age, let us say, with 99% confidence? Maybe then the confidence interval would be 23-29. We are more confident about this but, at the same time, less precise. And if we instead wanted to be more precise? Maybe then we would find that the mean age is exactly 26. But then we would perhaps have to settle for stating this only with 90% confidence.

Note However, it is important to know that the width of the confidence interval is also affected by the sample size: the larger the sample size, the slimmer the interval (which means better precision).