After all of my work this semester, I have finally finished my experiment. While it took five days of running the experiment, I was able to get a pretty decent yield. Overall, I had 83 good results, which is over a fifth of the entire school. I ended up having to throw out a handful because something or other went wrong during the experiment. Most often this was because they choose to reject but never proposed a counter offer.
Conducting the experiment ended up being a lot more enjoyable than I thought it would be. It was curious to see what people would choose as their answers. It was also intriguing to figure out the different play styles people would fall into. There were the good samaritans, the rare few who accepted every answer put in front of them. Another rare group was the punishers, who said no to everything that was not above fifty. Much more common were the splitters, which rejected everything under a certain threshold and then countered with an even split. The most common, however, were the people who were more flexible and would counter differently based on the original offer. The majority of the participants fell into this category, which is pretty much what I was expecting.
After the experiment came data entry, which was about as annoying as I expected it to be. I was a bit worried at first since it took me an hour to enter the first six participants into the excel sheet. However, I ended up developing a rhythm in the data entry, and my pace quickly increased. The crunching of numbers ended up being pretty simple once I got the sum method working in Excel (For a while I was just multiplying all the data points together instead of adding). The deliberate layout of my data entry helped immensely, making it easy to find the average accept and counter offer for each person.
Unfortunately, the data isn’t showing much of a significant difference in any direction. The percent of accepts between different groups never differs by more than 3.5 percent. The same goes for the counter offers, which never vary more than 3 percent. There is one more way I want to try sorting the data to see if that comes up with any correlations. Hopefully, that will turn up something interesting.
Thanks for reading.
For those who want to read some of the studies I have been using: