4. You are interested in investigating whether people, when buying phones, are influenced by the choices of others. You have access to a sample of individuals who are all going to buy a phone. You sort them into two groups, Control (C) and Treatment (T). In the C group, you observe whether someone buys an iPhone or not. In the T group, you inform individuals about the proportion of people who have an iPhone, and then tell them that information about whatever phone brand they buy will be shared with others. You collect the data, and then run the following regression: iPhone =0 +1 +, where phone is an indicator that equals 1 if i buys an iPhone, and is otherwise zero. is an indicator that equals 1 if i belongs to the Treatment group, and is zero otherwise. Let's assume that you find positive estimates for and 1.

a. What is the interpretation of the and 1 estimates?

b. You are now interested in figuring out whether the treatment group effect is due to social learning or social influence. Is it possible to tell those apart with this design? If not, outline an experimental design that will allow you to do so; write down the estimation equation and discuss it in detail.