The coefficient of determination

- For our sample of basketball player weights and heights,…we have only five values in each category.…Graphing these on our scatter plot is simple.…Evaluating them by sight is also rather simple,…and the appropriate regression line…was relatively easy to anticipate visually.…For our data, we could see a pattern,…and we could tell the formula for our regression line…was a good fit, but what happens when you have…a huge data set, when the scatter plot is messy,…and the regression line is not necessarily logical?…How can you tell if the regression line…is a good fit for your data?…Well, for this, we have something called r-squared.…

It's called the coefficient of determination.…R-squared is a number between zero and one.…Zero indicates that our regression line…is a very poor fit for our data points.…An r-squared of 1.0 on the other hand…indicates that the line is a perfect fit…for our data.…So, how do we calculate r-squared?…Well, the formula for r-squared…looks pretty simple.…

R-squared is equal to SSR divided by SST.…