You are designing a research project, where you predict that college sports fans

Question

You are designing a research project, where you predict that college sports fans will be more likely to be partisans, since their behavior of supporting sports teams shows a “participatory” personality that will make them more likely to take sides in politics as well. To test this hypothesis, you have a survey of 1200 Idaho residents. Assume this survey is representative of the population of the state of Idaho. Any missing data in your survey is already coded as missing, so there is no need to recode for missing data. The codebook for this dataset is on the last page of the assignment.

(YOU CAN SAEV THIS PICTURE SO YOU WILL HAVE THE FULL RESOLUTION FOR IT PLEASE )

Questions :

2a. Which variable would you select for the independent variable in this analysis? Explain why you would choose it and how it would be coded (6 points).

2b. Which variable would you select for the dependent variable in this analysis? Explain why you would choose it and how it would be coded (6 points).

2c. Explain what statistical model you would use to test this relationship, and why it is appropriate (6 points)

2d. Consider 2 control variables to use in this analysis (hint: there is one variable you MUST control for). Explain why you would control for these (8 points).

2e. Explain, in terms of direction and statistical significance, what beta coefficient on your independent variable would support your hypothesis (4 points)

2f. Imagine you think that the effect of being a college sports fan on being a partisan is higher for Boise State graduates than anyone else, since Boise State sports are obviously much better than any other university. Explain how you would test this relationship, including variable coding and what variables you would include in your model (10 points).

2g. Do you think your results are generalizable to the population of Idaho as a whole? To the population of the United States as a whole? Explain why or why not (8 points).

Ivariabe Descriation 2 A 4 -children Has Children? LNochildren Haschaldren college fan College sports fan Not a college sports fan LCollege sports fan Did donate education Level of education 5 Less than HS degree Some college Ludegree degree 6 Employed Full Time Employed Part Time LUnemployed globaluanm LIs global warming happening No LYes but natural Lhumans Very liberal Liberal LSlight liberal Moderate LSlight conservative Cons cons Continuous A BSU of Idaho Idaho State Lother volunteer Did you volunteer for a campaien in 2016 Did not voluntee Did volunteer

Explanation / Answer

1) Here we are dealing with the analysis where we want to predict that whether college sports fans shows a “participatory” personality that will make them more likely to take sides in politics as well or not. So in this analysis we can consider various independent variables but most important one is college fan coded as 0 or 1 (like binary).

Other independent variables that we can consider are education, partisanship, race, employ etc.

2) Dependent Variable should be vote2016. Because it helps us to understand whether there is any relationship with college fan and his voting preference which is the main motto behind this analysis. We can use Logistic regression to find out the probability of events that can possibly happen.

3) There are various statistical model you can use like linear regression( generally used for qualitative analysis) , logistic regression( used in binary output case generally), linear discriminant analysis (popular for multiple class classification). I will chose LDA because:

When classes are well separated the parameter estimates for logistic regression model are surprisingly unstable. LDA doesnot suffere this problem. If n is small and the distribution of predictors is aprrox normal in each class, lda is again more stable than logistic regression. And lastly lda is popular when we have more than two response classes.

4) Control variables are variables which are remain unchanged to asses the relationship of independent variable with other dependent variables. So once you choose independent variables in your final model then if you want to study the relationship of vote2016 with for example race then you should first makes other variables constant then you can understand its behavior.

Related Questions

Navigate

• Browse (All)
• Browse Y
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.