Use the chi-square test to determine whether there is any interaction between location and the tendency of residuals to be positive or negative. If you reject the null hypothesis of no pattern, describe how a researcher might continue the regression analysis.
The following table represents the number of residuals observed in a regression of voter
registration rates on income and age over eight districts:
First, we calculate the expected frequencies of the residuals, given the totals
We use these expected values to calculate the chi-square statistic. For each cell, we divide the
squared difference between the observed and expected values by the expected value: e.g. (20 -
17.7)2
/ 17.7. These values are summed, and their sum comes to 2.885.
We compare this value to the chi-square distribution, for which the critical value is associated with
the chosen alpha (0.05) and the degrees of freedom (the number of rows minus one times the
number of columns minus one, or 1 * 7 in this case for d.f. 7). 14.067 is larger than our chi-square
value of 2.885, so that we fail to reject the null hypothesis of no spatial pattern.
Given the lack of relationship between location and residuals, we need not take any further steps.
You might also like to view...
Detail the major points either for or against the view that there can be a science of politics, providing examples for each view.
What will be an ideal response?
What is a resolution?
What will be an ideal response?
What is the primary weakness of a correlational study?
a. an inability to determine if two variables are associated b. we cannot determine the strength of a relationship c. correlational studies tend to be artificial d. we cannot establish cause and effect with a correlational study
In the context of national grants, Democrats "have consistently favored fewer strings, less national supervision, and the delegation of spending discretion to the state and local governments."
Indicate whether the statement is true or false