If you've ever gathered data from a sample survey, you might have found that your sample didn't represent the target population as well as you'd like. Perhaps there were too many survey respondents from one particular age, or one particular gender, or so forth. If that demographic category actually affects people's responses, then your non-representative sample might be seriously biased!
We can use statistical weighting to adjust survey data and make up for those imbalances in the sample. This tutorial assumes that you've learned the skills for working with categorical variables & frequency data from earlier videos, as well as at least a little bit of familiarity with the "raking" method of weighting data, but we will still go through the simplest version of that technique from the start. Then, we'll look at how to apply that weighting method quickly and (efficiently for one demographic variable) using formulas in Google Sheets.
Note that, in most real-life surveys, we would use more than one demographic variable to weight our data - usually, many more than one! And the raking method using more than one variable is actually much more complicated: instead of just applying the same process just once for each demographic variable (gender, age, race, etc.), you actually need to apply the process over and over again for each variable until the results stop changing. That requires both calculus and either specifically-designed statistical software or advanced programming skills... so we're not going to bother with it here. This video is just designed to give a quick, high-school-level peek at how weighting data and "correcting" for a variable can work!