Point Estimate vs Parameter

A point estimate is your one-number summary of a parameter before you attach any margin of uncertainty. Common ones are the sample mean $\bar{x}$ (read x-bar) for the population mean $\mu$ (the Greek letter mu), and the sample proportion $\hat{p}$ (read p-hat) for the population proportion $p$. For example, if 63 of 200 sampled voters approve, the point estimate of the population approval rate is $\hat{p} = 63/200 = 0.315$. A point estimate alone hides sampling variability, so it is usually paired with a margin of error to form a confidence interval.

A parameter is a constant for a given population, even though you usually do not know its value. For example, the true proportion of all adults who own a car is a population proportion, written $p$ (the letter p). Parameters are typically labeled with Greek letters, such as the population mean $\mu$ (mu) and population standard deviation $\sigma$ (sigma). You estimate a parameter using a statistic computed from a sample.