Seriously. I will need it for work. I haven’t been able to understand that shit from any online resource until now. Please ELI5
In: 380
Essentially you have to take two measurements, one of an independent variable and one of a dependent. For example, sleep and GPA. Researchers plot the results they get on a graph. From this plot, the researchers use complex formulas to create a line that minimizes the *residuals*, or the difference between plotted data points and the line that they draw. They then use other complex formulas to determine how well the data fits the line; the closer it is clustered to the line, the better the fit. This is known as the *correlation coefficient*, or r.
R is good, but its sign always matches the slope of the graph (if y decreases as x increases, slope is negative). Researchers need a positive value, so they square r to get r-squared, the *coefficient of determination*. If your r-squared value is .64, you’d say that 64% of the variation in y is a result of the change in x.
Latest Answers