This is a very basic concept that gets used often in mathematics and especially statistics. Graphing has been regarded as a way to visualize the data. How far are the points from each other? Where does the summary statistics fall in terms of your data points? Do I have a normally distributed sample? There are a few questions that come to mind when you graph the data. Why do I need this graph? What relevance does this ‘picture’ have to my data , let alone my life? What is the point? The point is this…Your Y axis typically represents an outcome. Your X axis represents a stimulus. So when you graph, your mainly focused on the way in which the Y axis behaves in terms of your X axis. So whether you are looking at a Kaplan Meier, scatter plot, or residuals and predictors – the theme is the same. You typically set-up the graph so that descriptive are on the X axis (demographics) and the dependent factor is on your Y axis.
This is a very small point I am trying to make here. But when think about statistics and how models are developed and illustrated – the objective becomes very clear. Your slope or Beta estimates in any given regression model translate into coordinates on a graph. For one unit change in your predictors or independent variables, you outcome (dependent variable) will yield a particular result.
If you enjoyed this blog there are more to follow-Amy