Simple Linear Regression Analysis

Graphics to include

Scatter plot with the response on the vertical axis and independent variable on the horizontal axis. Overlay the fitted regression line. Make sure the two axes are labeled

Scatter plot of the Residuals Vs Fitted values. I highly recommend that you include a horizontal line beginning at zero on the vertical axis. This will help you identify lack of linearity and/or constant variance.

qqplot of residuals. Use this to assess Normality of the residuals. Make sure you include a reference line that can be used to assess how close the residuals are matching to the Normality assumption.

Histogram of residuals. Use this to assess Normality of the residuals. Make sure you include a density overlay as well as the Normal curve. In the example solution, the Normal curve is in red and the density estimate is in blue. Numerical measures to include.

Estimates of the intercept and slope parameters (these are often referred to as 0 and 1). Interpret the slope parameter in context.

Compute the correlation coefficient and R-square value. Interpret R-square in context.

It is okay to use counts here. For example, in Baseball, it would acceptable to look at the number of Runs scored as a function of At-Bats. This should be correlated, right?! It is inappropriate to use the regression techniques used in this class to model 0 and 1 type responses (i.e. Surival or Presence/Absence of an attribute).