1. According to data from a large-sample survey of countries, the Pearon’s r correlation coefficient between right-wing extremism and political violence is .55 with a p-value of .04. Meanwhile, the correlation coefficient between migration and political violence is -.22 with a p-value of .96.
With respect to the relationship between right-wing extremism and political violence, what is the null hypothesis?
First, interpret the correlation coefficient between right-wing extremism and political violence. Second, can we reject the null hypothesis at the .05 level of significance?
First, interpret the correlation coefficient between migration and political violence. Second, is this coefficient different from 0 with 95% confidence?
2. Professor Billie Joe Armstrong is studying the factors that account for variation across global conflicts in terms of casualty rates. He uses multiple regression analysis to estimate the effects of troop training, financial investment in a state’s military system, and the impact of guerrilla warfare. The variable Training is the mean number of hours per day that new recruits undergo basic training. MilitaryInvest is an ordinal variable ranging from 0 to 10, with higher values indicating greater levels of financial investment in a state’s military. Finally, Insurgency is a dummy variable that equals 1 if a military faces non-state actors who employ guerrilla tactics, and 0 otherwise. The regression equation and a table of regression coefficients are presented below:
Casualty Rate = a + b1 • Training + b2 • MilitaryInvest + b3 • Insurgency
Variable Coefficient Standard Error
Intercept 21.0 16.1
Training 4.0 2.5
MilitaryInvest -1.4 0.2
Insurgency 3.2 0.4
According to these estimates, what is the effect of military training on casualty rates?
Suppose a military trains recruits 6 hours a day, scores 8 on the military investment rating, and does not encounter an adversary who employs guerrilla tactics. What is its predicted average number of casualties in a given conflict?
State A scores 4 on the military investment scale; State B scored 7 on the scale. The states are identical in all other respects. What is the predicted difference in their mean casualty rate?
Which of these coefficients are statistically significant at the .05 level?
View keyboard shortcuts
3. The Secretary of Defense recently declared the military would no longer deploy American troops to African nations to assist with counterterrorism operations. Using survey data, Professor Jello Biafra tests the hypothesis that republicans are more likely to agree with the policy than democrats. After controlling for education, he creates the table below:
High School Diploma College Degree Graduate Degree
Agrees? Republican Democrat Republican Democrat Republican Democrat
Yes 20% 22% 53% 55% 78% 80%
No 80% 78% 47% 45% 22% 20%
State the null hypothesis in the scenario.
Identify the controlled effect(s) of the independent variable on the dependent variable.
What type of relationship does the controlled comparison reveal? Explain your answer.
View keyboard shortcuts
4. Professor Henry Rollins is studying whether Muslim majority countries experience more terrorist attacks than non-Muslim majority countries. Rollins’ data are presented below:
Muslim Majority? Mean Terrorist Attacks Standard Error Frequency
Yes 8 2 1,980
No 6 4 1,287
What is the standard error of the difference in the mean number of terrorist attacks?
Can we be 95% confident that Muslim and non-Muslim countries differ in terms of the number of terrorist attacks? Explain your reasonin
5. Suppose that among European countries the mean number of terrorist attacks is 14 with a standard deviation of 2.
What percentage of European countries experience more than 16 terrorist attacks?
What percentage of European countries experience more than 12 terrorist attacks?
Approximately what percentage of countries experience between 10 and 18 attacks?
A European country is picked at random has a score of 1.5 on the Z scale. How many terrorist attacks does this country experience?
