First you’ll have to understand the two other main types of data, cross sectional and time series. Cross sectional data looks at a snapshot in time, and compares a lot of different attributes of the given subjects in a sample as they exist at that particular time. It generally compares differences among the subjects. Think of the test scores of a particular class of kindergartners, and how they vary according to the differences among the kids, their parents, etc.

Time series data tracks a given variable/measure over time. Think of a child’s test scores as he ages.

Panel data is a mix of both. It starts as a cross sectional data set, by taking a sample of people and measuring some number of their attributes at some time. But then it regularly “checks back in” on the *same* sample, by re-taking all those measurements at regular intervals, say annually. So in our example, we sample the kindergartners, then re-do the sampling each year, all the way through high school.

It adds a time element to cross-sectional data.

## Latest Answers