Great article, thanks for sharing. I think you touch on a lot of exciting areas in prediction modeling and I think the ones around causal inference are particularly important. I feel like we are reaching a stage where there’s a lot of regression analysis being done in determining the association between two or more variables, but due to limitations in hypothesis testing often we cannot infer whether there is a causal link between the two or more variables we are testing. This is where we get into trouble by making false causation assumptions (like chicken consumption causes changes in oil consumption). I think it’ll be interesting to see how much weight Uber and companies like them place on this type of analysis. Theoretically, if you can prove that something has a causal link, it would be only fair to therefore invest a significant amount of resources into the causal agent. However, as you point out, there are some assumptions that are being made and if Uber starts gambling on these analyses where the limitations of the assumptions are not being fully appreciated, there could be some costly mistakes!
Thank you so much for your article. You touch on a lot of important points with regards to the ethics in ‘big data’ usage and highlight some of the key areas that it can be used for in DEI. Trust I think is a crucial aspect in approaching any sort of data analysis that may involve making decisions upon other people. As we know, biased data will only lead to biased outcomes if it is not managed appropriately. Even if the data is unbiased, if it is sourced unethically, that is also a problem. It made me think of the Cambridge Analytica saga whereby a whole range of data points were used to sway voters, which I think goes completely against ‘trust’. Even when some participants requested the data Cambridge Analytica had on them, they refused. To build ‘trust’ is also to reduce the temptation to cross ethical boundaries and harness the power of data for certain causes. This is very tricky and involves a lot of complex power dynamics.
Thank you for your post. This is something that I am all too familiar with. We don’t have USMLE in Australia but use other surrogates to try and differentiate candidates into competitive training programs. There has been a reliance on obtaining extra master’s programs (MPH, MSc, MMed, MSurg, etc.) to add points to a candidate’s CV. What has resulted is that there has been degree inflation amongst candidates and a lot of people are getting ‘too many’ degrees. I agree that the Pass/Fail system doesn’t really solve the issue at hand, because institutions will look for or make their own alternative tests/exams to differentiate people. On the surface, candidates may think that it will help them, but in reality, the institutions may be making very biased decisions or ones without good evidence-based rigor. Would love to chat about this more to see what other suggestions you think may work!