The final assignment asks you to use the computational thinking and programming skills you learned in the course to answer an empirical social science question. You are expected to apply the best practices and theoretical concepts we covered in the course to produce a program that not only returns the correct output but is also legible, modular, and reasonably optimized. The assignment assumes mastery of loops, conditionals, and functions, as well as awareness of issues related to runtime performance.
In the assignment, we will study the homophily and social contagion of cheating in the massive multiplayer online game PlayerUnknown's Battlegrounds (PUBG). Cheating in this context means the adoption of unapproved software that gives the player an unfair advantage in the game (e.g. being able to see through walls).
Our hypotheses are that cheaters tend to associate with other cheaters but also, players who interact with cheaters become likely to adopt cheating themselves. To provide preliminary evidence for these hypotheses, we will:
- Observe whether cheaters tend to team up with other cheaters more than chance.
- Observe whether players who observe cheaters are likely to become cheaters more than chance.
- Observe whether players who are killed by cheaters are likely to become cheaters more than chance.
To test the "more than chance" part, we will simulate alternative universes in which the players played the same game but joined a different team or happened to be killed by someone else at a different time. We will then compare how what we observe in the actual data compares to what we would expect in a "randomized" world.