1. Learn
    2. /
  2. Courses
    3. /
  3. Causal Inference with R - Regression

Exercise

Running a Regression Model: Improving Our Model

As we mentioned in the previous question, Dr. Funn predicts that income is positively associated with happiness, he is worried that the relationship between income and happiness might be confounded by their joint relationship with work hours. Specifically, Dr. Funn believes that people who work more hours make more money (which indirectly increases a person's happiness), but that people who work more hours at the same income tend to be less happy (i.e. work hours has a direct negative effect on happiness). This would mean that we can't trust our first model to give us accurate results.

So now using the dataset GSS, help Dr. Max Funn measure the effect of income on happiness, controlling for work hours.

Instructions

100 XP
  • 1) Examine the results of our simple first model from the previous question.
  • 2) Plot the relationship between happiness and income, including a regression line.
  • 3) Look at correlation statistics for confounders between work hours, income, and happiness.
  • 4) Construct a second regression model that measures the effect of income on happiness that controls for work hours.