Research design and methods in IR Assignment 3

Project description
For questions 1 and 2 below, please use the QoG dataset on the E-learning site (either the reduced one that we used in class or the full standard one) or another

dataset related to your research interests (in which case you should also submit the data in .Rdata format or any format that is easily read by Deducer, such as csv or

sav). You are welcome to work on getting the Deducer output in groups, but when you need to choose a variable for the question, each person should choose different

variables and write his or her own answers to the questions.

1. Using R deducer, produce a contingency table (also called a crosstabulation) for two categorical (nominal or ordinal) variables and perform a Chi square test. Add

either row or column percentages (by clicking the cells tab and ticking one or the other).
You may use categorical variables in the data (see list below) or create them from any continuous variables using the Data-Recode variables command. Some examples of

categorical variables in the standard dataset are: Freedom House status (fh_status), Human rights issues (most of the variables starting ciri_), Political terror

(starting gd_), The Regulation of Participation (most of the variables starting p_), Political regime institutions (chga_hinst), Electoral system (most of the

variables starting idea_), …
See the descriptions of the variables in the codebook.

(a) Create a Word table with the contigency table output that R deducer gives you. You should include the observations and either the row or column percentages in the

table (please do not simply copy/paste the R output from the console into the document).

(b) Interpret one of the percentage values in one cell of the contingency table.

(c) Is there a statistically significant relationship between the two variables? How do you know? What can you conclude about the nature of the relationship?

2. Using R deducer, calculate the correlation coefficient between two ratio-scale variables.

(a) Is there a statistically significant relationship between the two variables? How do you know?
(b) What is the correlation coefficent?
(c) What can you conclude about the direction and strength of the relationship?

3. Answer the following question about the regression analysis presented in the table below. The table is from page 200 of the article, which you can find on the E-

learning site (if needed):

Pippa Norris, Stefaan Walgrave, and Peter Van Aelst, “Who Demonstrates? Antistate Rebels, Conventional Participants, or Everyone?,” Comparative Politics 37, no. 2

(January 1, 2005): 189–205.

You do not need to read the whole article to be able to answer the questions, but it may be helpful to refer to it if anything is unclear. Each column is a model that

explains the effect of social background and media use variables on a different form of activism (voting, campaign work, etc).

(a) In the regression table, how many regression models are presented?
(b) Choose one regression model from the table above for answering the remaining questions. What is the dependent variable?
(c) List one independent variable that is statistically significant. What is the (unstandardized) coefficient for that variable? What can you conclude about the

direction of the effect on the dependent variable?
(d) Using the standardized coefficients (St. Beta), compare the coefficients of TWO independent variables that both have a statistically significant effect and state,

which variable has a stronger effect on the dependent variable.
(e) List one independent variable that is not statistically significant. What do you conclude about that variable?

4. Based on the R deducer output below, please answer the following questions. The dependent variable is the percentage of internet users in the population (0-100).

The independent variables are income inequality (Gini), GDP per capita, and a democracy dummy variable. The coefficients are unstandardized.
(a) Which variables have a statistically significant effect on the percentage of internet users? How do you precisely interpret the coefficients of each of those

variables?
(b) What is the R-squared for the model? How do you precisely interpret it?

Deducer output

PLACE THIS ORDER OR A SIMILAR ORDER WITH US TODAY AND GET AN AMAZING DISCOUNT 🙂