Experiments in IS research

Bullet Points

CUHK-DSME6622: Advanced MIS Research Seminar

1. Proposal each course (10 min each);
2. There is reading list;
3. About Econometrics of IS;

Basics

1. How to design experiments:

• Experiment
• EC 9A3 Advanced Econometric Theory - Lecture Slides
• CrossValidation:Difference-in-difference-in-differences estimator

1. COS Analysis? Baseline?
2. Randomization ~ statistical test
3. Regression discontinuity design
4. Theoretical Modeling vs behavior vs experiment vs empirical
5. How do we identify the way we want to ~ social psychology ~ mental mapping
6. impact of technology ~ topic is moving on ~ some of the topics are already understood ~ do good literature review
7. treatment? ~ control? ~ outcome⁰
8. Difference in differences (DID): before & after treatment, well controlled

Presentation

Firefox extensions update classification

1. Firefox web browser (open source platform)~ extensions or add-ons
2. Classification : open or close source
3. Mechanism? ~ size or role of users
4. Output: influence : positive or negative
5. Rapid release strategy vs traditional way
6. cross? experimental design
7. whether the add-on supplier update their products in response to the rapid release of Firefox platform?
8. 1 observation: add-on version number?

Review/online experiment⁰

1. Treatment: hit of newest review / high-quality
2. Control : no review
3. Social psychology ~ ab test? ~ Online experiment, see Web-based experiments

Economic analysis of social interactions¹

1. Charles F.Manski, Journal of Economic
2. Empirical analysis
3. mutually exclusive but collectively exhaustive¹¹
4. 3 effects

Paying not to go to the gym²

1. consumer behavior -> maximize utility (rational expectation vs irrational)
2. theoretical prediction & empirical testing & survey
3. behavior IO?
4. Personal view:

• How to quantify the utility function?
• not like financial product: stock, bond or derivatives, the utility of going to gym should also include non-economic factor (price), such as the potential chance of keeping the right of going to gym

Observational Learning: Evidence from a Randomized Natural Field Experiment³

1. One restaurant: whether the randomization is not good enough?
2. Control group: pre-experiment data, one week
3. Raking treatment table vs salience treatment table
4. One week data
5. Difference-in-difference-in-differences (DDD)³¹
6. Coincidence, observational learning (learn the quality, more about yourself)
7. conformity (more from others), compliance, salience³²

The Spread of Behavior in an Online Social Network Experiment⁴

1. Small world network model
2. Network vs photonics??
3. Social reinforcement

⁰. Treatment and control groups ↩

¹. Manski, C. (2000). Economic analysis of social interactions. Journal of Economic Perspectives, 115–136. ↩

¹¹. Quora|What’s the difference between mutually exclusive events and collectively exhaustive events? ↩

². DellaVigna, S., & Malmendier, U. (2006). Paying not to go to the gym. American Economic Review, 96(3), 694–719. ↩

³. Cai, H., Chen, Y., & Fang, H. (2009). Observational Learning: Evidence from a Randomized Natural Field Experiment. American Economic Review. ↩

³¹. CrossValidated|Difference-in-difference-in-differences estimator ↩

³². Chapter 7: Conformity,Compliance, and Obedience ↩

⁴. Centola, D. (2010). The Spread of Behavior in an Online Social Network Experiment. Science, 329(5996), 1194–1197. ↩