Final Peer Review #13
Description
This project analyzed demographic data in swing states to predict the results of the 2020 presidential election. Their data sets included a number of different features, such as race, prior political affiliation, and education level.
One thing I didn't understand was how certain features could be continuous... like how can "white" and "no HS diploma" be continuous? I think you could have explained this further, unless I missed it. In which case, ignore that comment! I think it also would've been helpful to know the legitimacy of the Iterative Imputer library. Why did you use it / what's its error or accuracy? I think in general you guys need to emphasize why you're using whatever method you choose and make it very obvious.
This was a hard problem, especially considering all of the other factors that cannot be measured. Like, the public's perception of the pandemic and all the neverending Trump scandals and drama. I think you guys did a great job assessing the problem with what you had. There were a lot of wild cards over the last year so we cannot fault you for not getting the best accuracy. I think you also did a great job with the visuals, which were clear and descriptive. Finally, the sections on mail-in voting and the Weapons of Math Destruction were very well done. Those were very important to address and I'm glad you talked about its impacts. Overall, good job!