We are concerned that the relationship between the experiment and Duolingo usage could be different for mobile and web users. We want to run a regression where the effect of the experiment on Duolingo usage is different for mobile and web users.
Specify a regression equation that you could use to do this. Use dummy variable D5 to measure how a person uses Duolingo (mobile or web), instead of the dummy variables D2, D3, and D4.
Estimate the regression in R.
