You will continue to be analyzing data from the class survey

Assignment:

Copy your regression output into the word file (TIP: paste as image to preserve formatting) and after each regression write a comment response to the questions: Which variables are statistically significant (and how can you tell?). What is the interpretation of the result? (eg. females have higher/lower life satisfaction). Is this result expected or not? (what are the reasons you would expect that result, compare with the research that we studied in class).

In each regression, life satisfaction is the dependent variable (i.e. Y variable), and each regression will have different controls (i.e. X variables).

Run a regression with two control variables: female (i.e. the numerical “female” variable which equals to 1 for female and 0 for male) and stress.
Add to the above regression a variable called senior (equals to 1 for seniors). (i.e. now you have 3 variables in your regressions: female, stress and senior).
Add to the above regression a variable called lowincome (equals to 1 for people in bottom and second quintiles of income distribution). (i.e. now you have 4 variables in your regression: female, stress, senior and low income)
Remove variable senior (you should move that column to the end of the sheet – cut and paste, remove blank column) and add a variable called Social media over 6 hours (equals to 1 for people who use over 6 hours of social media per week). (i.e. now you have 4 variables in your regression; female, stress, low income and social media).
Remove Social media and copy year 2020 next to low income column. Run regression with the following 4 variables: female, stress, low income, year2020. Are people in 2020 more or less happy than previous years?
Which of the 5 regressions you run above is the best in terms of predicting the outcome, and how can you tell? (i.e. which has the strongest predictive power?) – review week 6 PPT if you need to.
: 1 point: Pick one of the remaining categorical (i.e. text) variables (i.e. those that have not been used yet). Create a numerical variable (use IF command – see how the numerical variables are created in columns that we used above. For examples, you can create a numerical variable for different work categories, or one for debt, or economic outlook, etc. (Note – if you use Friends and social life you will have less data points since it is not available for two earlier years). Add this numerical variable to the regression in item 1 above (so you will have 3 variables in your regression: stress, female and the new one you created).
Note that you will need to put the column with your new variable next to the Female column since Excel has to have all dependent variables columns next to each other.