I really enjoyed this assignment. It was my first time using MTurk and I was surprised how quickly you could get a large amount of responses. Conducting the statistical tests in R required some dusting off of cobwebs, but was easy to reproduce many tests once I re-familiarized myself with it. I really enjoyed creating graphics from real data that I collected myself. I feel more comfortable with ggplot now, and tried to make my charts look less basic than they did in Assignment 1. However, I tried to keep it simple.
Interesting research question, and I think your experimental design is simple but directly tackles this question
For the shortest bar question, for those answering incorrectly, which bar did they pick? Because looking at the graph, 2seater does blend in with the background because of the precise gradient used. It wouldn't surprise me that more people went for minivan instead (if that is what occurred). If there was more contrast between the background and the gradient for the shortest bar, I'm not sure you would replicate these results
Interesting catch about bias vs. variance on Q3
Did the multi-channel graph take longer to interpret on average?
I really enjoyed this assignment. It was my first time using MTurk and I was surprised how quickly you could get a large amount of responses. Conducting the statistical tests in R required some dusting off of cobwebs, but was easy to reproduce many tests once I re-familiarized myself with it. I really enjoyed creating graphics from real data that I collected myself. I feel more comfortable with ggplot now, and tried to make my charts look less basic than they did in Assignment 1. However, I tried to keep it simple.