Refer to the Academy of Management Journal (August, 2015) study of the charisma of top leaders in business, Exercise 11.28. Recall that data were collected on 24 U.S. presidential elections from 1916 to 2008. The dependent variable of interest was Democratic vote share (y), measured as the percentage of voters who voted for the Democratic candidate in the national election. One of the independent variables of interest was a categorical variable that represented whether or not the presidential election was affected by a World War. A second qualitative variable of interest was whether a Democrat or Republican incumbent was running for president. This incumbent variable was recorded at three levels: Democrat incumbent, Republican incumbent, or no incumbent running.

a. Write a model relating Democratic vote share (y) to the qualitative independent predictor, World War. Use "no World War" as the base level.

b. In terms of the β's of the model, part a, give an expression for the mean Democratic vote share during all years when there is no World War.

c. In terms of the β's of the model, part a, give an expression for the mean Democratic vote share during all years when there is a World War.

d. Fit the model, part a, to the data in the file. Is there sufficient evidence to indicate that the mean Democratic vote share during all years when there is a World War differs from the mean when there is no World War? Test using α = .10.

e. Write a model relating Democratic vote share (y) to the qualitative independent predictor, incumbent running. Use "no incumbent running" as the base level.

f. In terms of the β's of the model, part e, give an expression for the mean Democratic vote share during all years when there is no incumbent running.

g. In terms of the β's of the model, part e, give an expression for the mean Democratic vote share during all years when a Republican incumbent is running.

h. In terms of the β's of the model, part e, give an expression for the difference between the mean Democratic vote share for all years when a Democratic incumbent is running and when there is no incumbent running.

i. Fit the model, part e, to the data in the file. Is there sufficient evidence to indicate that the mean Democratic vote share differs depending on the incumbent running? Test using α = .10.