Confidence Interval

A Confidence Interval is a range of values that is expected to contain the true value of what is being measured.

For example, we wanted to find out the average age at which babies learn how to use the potty. Since we cannot survey the entire population of babies, we just take a sample of let’s say 50 babies, and find out the average age at which they were successfully potty-trained. Let’s say that from that sample, we learned that the average age of potty-training is 25 months. Since we expect those 50 babies to be sufficiently representative of the entire population of babies, we can say that the average toilet training age is 25 months.

If we were to get another sample of 50 babies from the same population, we would get a similar result, like maybe 26 months, but probably not the exact same value. If we keep getting other samples, we could get other values like 24 months, or 25.5 months, and so on. This is because every sample is slightly different from another sample even if they all represent the same population.

Now, the actual average age, or the true value from the population, could fall anywhere between those values. In this case, our Confidence Interval may be 24 – 26 months. That is why it is useful to report our findings as a range of values, or as a Confidence Interval.