204 / 762
204
Chapter 5: Bivariate Data
5-6 Regression Analysis and the Least Squares Regression Line
In investigating the relationship between two variables, the first thing one
should do is to prepare a scatter plot after the data are collected. From the
plot, one can observe any pattern of association. If the pattern or association
is linear and the linear correlation coefficient is reasonably large (positive or
negative), then one may want to fit a straight line model which best fits or
models the linear pattern or relationship. One of the purposes for the model
would be to help in making predictions. We will use the broad area of
regression analysis to help determine this model.
The scatter plot in
Figure 5-16
shows three possible straight lines which
may be used to model the data.
Question:
Which of these lines best represents the association between the
two variables?
Figure 5-16:
Scatter Plot with Three Possible
Straight Line Models