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

Exercise

Increasing the Auction Prices on eGulf

To increase their commissions, the popular online auctioneer, eGulf, wants to help its sellers increase the final sales prices of their merchandise. For each auction on eGulf, sellers are allowed to post up to 10 pictures of an item that they are selling. Past studies suggest that customers' appraisals of merchandise (and subsequently, final bids of merchandise) improve when sellers include more pictures of their merchandise; therefore, eGulf considers raising the limit on the number of pictures that a seller can post on an auction's webpage.

To test whether eGulf should allow sellers to post more than 10 pictures of their items, eGulf conducts an experiment: eGulf finds a random sample of dedicated used WePhone sellers who typically post 10 pictures of their used WePhones, and temporarily allows them to post up to 15 pictures for each auction of their WePhones. Using the results of this experiment, listed in dataset eGulf, test whether there is an added benefit to posting more than 10 pictures on WePhone auctions. Specifically:

Instructions

100 XP
  • 1) Use the structure function on eGulf to get a sense of the dataframe
  • 2) Test whether sellers who posted more than 10 pictures (Seller_Opted_Into_Treatment==1) on their WePhone auctions (Final_Sales_Price) sold their WePhones at higher prices than sellers who posted 10 or fewer pictures on their WePhone auctions (Seller_Opted_Into_Treatment==0)