Inferential statistics use a random sample to draw conclusions about the population. Typically, it is not practical to obtain data from every member of a population. Instead, we collect a random sample from a small proportion of the population. From the sample, statistical procedures can infer the likely properties of the population.
For example, it is impractical to measure the height of every adult woman, but you can measure the heights of a random sample and use that information to make generalizations about the heights of all women. For example, a confidence interval provides a range that the population mean height is likely to fall in.