1. Think about your field of interest within Political Science and locate an interesting data set. Download the data set and the accompanying codebook and/or questionnaire and look for interesting variables (Do not use the exercise data set we have used in class). Present the relationship between two variables of your choice in a cross table and calculate an appropriate measure of association. If one of your variables has many values, group them together in a reasonable way. Interpret and discuss your findings. Is there a relationship between those two variables? Is the relationship strong? In which direction goes the relationship? Also, make sure you cite the source of your data set, that is indicate which data set you are using and the source of the dataset (website).
2. In the ANES exercise dataset, there are 4 variables asking about traditionalism (trad_adjust, trad_lifestyle, trad_tolerant, trad_famval). Respondents were asked to indicate whether they agree or disagree with those 4 statements on traditional values. Two statements (trad_adjust and trad_tolerant) are coded so that higher values (indicating disagreement with the statement) indicate higher traditionalism, while for the other two statements (trad_lifestyle, trad_famval), higher values (indicating disagreement with the statement) indicate lower traditionalism. In order to combine all 4 values to a scale, it is necessary to recode the latter two variables (trad_ lifestyle, trad_famval) so that higher values indicate agreement with the statements and thus higher traditionalism. Use the recode option in SPSS for this (TRANSFORM->RECODE into different variable) and recode the old value 1 to new value 5, old value 2 to new value 4, old value 4 to new value 2 and old value 5 to new value 1. Do a frequency table of the new variable and check whether you did it right (compare with a frequency table of the old variable). Then, add all 4 variables to one new scale using the compute function (TRANSFORM -> COMPUTE VARIABLE) by calculating the average that every respondent has on the 4 traditional variables. Don’t forget to use the two newly created, recoded variables instead of the original trad_tolerant and trad_famval. So, the computation should look like this: (trad_adjust + trad_tolerant + trad_ lifestyle_rec + trad_famval_rec) /4 ]. Make a frequency table of the new variable and check whether you did it right. Present this frequency table in your homework. (Make a proper table, do not copy SPSS output.) Finally, see whether traditionalism hurts
   Clinton’s feeling thermometer scores. That is, compute the correlation coefficient (Pearson’s r) between Clinton’s feeling thermometer and the new traditionalism variable that you have just created. How large is this coefficient (state it in your homework)? How strong is the relationship between feelings for Clinton and traditionalism? Interpret the strength and direction of the coefficient.
3. Give an example for each of the following: additive relationship, spurious relationship, interaction relationship. Don’t use an example we have had in class. Your examples can come from any discipline, they need not come from Political Science. Indicate your three variables for each example and briefly explain why they could form an additive, spurious or interaction relationship. You need not present data or graphs or numbers for this. Your examples should just be somehow plausible.
4. Can I calculate Cramer’s V from ordinal or interval data? Why, or why not? Can I calculate Pearson’s r from nominal data? Why, or why not?
5. Imagine you have asked 10 respondents about their ideological self-position on a liberal-conservative scale that ranges from 1 (very liberal) to 10 (very conservative). Your respondents gave the following answers:
   1 4 10 5 6 7 4 8 5 5
   What is the mean, the variance, and the standard deviation of ideological self positioning in this example? Show how you calculated all three of them.
6. Assume that a variable X is normally distributed with mean 50 and standard deviation of 2. How many respondents will have a value

1. between 50 and 55?
2. between 45 and 55?
3. lower than 45?
4. higher than 56?

What is the value (cut-off point, raw score) above which only 5% of the respondents score? What is the value below which only 5% of the respondents score?

Price: $26.66

Solution: The downloadable solution consists of 13 pages, 1366 words.
Deliverable: Word Document