The following questions (7 & 8) refer to the Uzonyi article and data.
7. The troops variable indicates the number of troops a third party committed to a peacekeeping mission. The polity2 variable indicates the third party’s regime type, where values between -10 and -7 represent autocracies, -6 and 6 represent anocracies (mixed regimes), and 7 and 10 represent democracies. For all parts of this question, use only data from the year 2000.
a) Create a new variable named regime that takes on a value of 0 for autocracies, 1 for anocracies, and 2 for democracies. Report the average troop commitments for each of these groups. Also provide a table with that information.
b) Estimate an Analysis of Variance (ANOVA) F test using troops committed as the dependent variable and the regime type as the independent variable. Provide the F statistic and the p-value.
c) Briefly discuss your findings statistically and substantively for a general audience.
8. Bivariate Linear Regression. The troops variable indicates the number of troops a third party committed to the peacekeeping mission. The refinflow variable indicates the number of refugees entering that particular third party from the conflict. Use all data.
a) Report the first, second, and third quartiles (the 25th, 50th, and 75th percentile), as well as the mean and standard deviation, for both variables.
b) Report the covariance and correlation between these variables. Interpret the latter.
c) Estimate a bivariate linear regression model using refugee inflows to predict troop commitments as the dependent variable. Provide a regression table showing the estimated coefficients and standard errors for ???? and ???? parameters, as well as indicating the p-value.
d) Write down the equation for the model predicted.
e) Calculate the substantive impact of refugees on the number of additional troops the model predicts will be committed with:
i. an increase of 10,000 in refugee inflows
ii. an increase from the first quartile to the third quartile of refugee inflows
iii. an increase of one standard deviation in refugee inflows
f) Create a graph showing a scatterplot of these data (with troops on the vertical axis and refugee inflows on the horizontal axis) overlaid with the line of best fit estimated by this model.
g) Briefly discuss your findings statistically and substantively for a general audience.
h) In a few sentences, consider the utility of estimating the same model on only those country-year observations that did experience an inflow of refugees.
i) Now do so. Provide a regression table showing the estimated coefficients and standard errors for ???? and ???? parameters, as well as indicating the p-value.