"Linear graphs show the strength. Say that there is a scatter plot of some data that has been collected, and you find that the correlation of the data is close to one. What does this say about the graph? It means that the data elements are very well related. I could give an entire tutorial on this, but that is not the point of the blog. Basically, all the points of the data plot are all very close to the linear graph that they generate if the correlation is "good" or "strong". If the correlation is "weak," the points are very loose about the line, and tend to be more scattered and random than like a line. A strong analogy is like a data plot with a strong correlation: the elements of the analogy are very closely related to each other, not scattered and random. Analogous thoughts are very close in what they share in common, not distant. They will have some separate elements, but what they hold in common, they hold very closely."