As a student of Economics at Colby, I place a lot of value on data as a tool to support the theories that I learn in the classroom. In fact, without the use of data in Economics, we would not be able to prove any of the theories or show evidence for any of the studies that are conducted. Data is a fundamental tool in the discipline. Thus, I found professor Hanlon’s discussion on revolutions in Data to be particularly insightful, as it made me think about how my own Colby career has been impacted by data becoming the main form of evidence. At the end of his discussion, Hanlon conjectured that the revolution regarding data was when it became the main form of evidence.
As recently as last Friday, I saw professor Hanlon’s conjecture in practice. In my Behavioral Economics class, we were looking at a paper detailing the Anatomy of an Experimental Political Stock Market, economists use data from the Iowa Presidential Stock Market to yield predictions of the expected vote shares of the presidential candidates in presidential elections.
Hanlon cited Francis Bacon and Robert Hooke’s use of images to convey how data functions better than words when it comes to scientific knowledge. For example, in Robert Hooke’s Micrographia, the author publishes images of the things he observes under a microscope so that the reader of the book has a visual on what it is he is describing. The inclusion of images in Robert Hooke’s book is interesting because it shows that data at one point was visual, and not necessarily what we consider it to be today. Today, the word data is synonymous with numerical and quantitative evidence.
Today, when I hear the word data, I think not of the images of Hooke’s Micrographia, but instead of the election polls that detail the number of people who are voting for each candidate in a specific state. The ever-changing numbers that appear on the news each morning are symbolic of how it is numbers, not pictures, that are the data we care about. This idea is interesting however, because it is with these poll percentages, that we almost always create a map showing which regions of the country are dominated by which candidate. It is via this manner that the political data is displayed that draws me back to professor Hanlon’s conjecture about all data being visual.
After reading some of the other student’s responses, I found one idea particularly interesting. Mike described how data can misrepresent somebody or something. I am wondering how can we think about this idea given Professor Hanlon’s discussion. He doesn’t seem to mention how data can misrepresent some things. However, given my discussion of how we display political data on a map, is it possible that when a state is shown as red or blue, that it is a misrepresentation? For example, in a state like Maine (which usually votes democratically), the weight of the population lives in York and Cumberland Counties. Is this truly how most of the state of Maine feels, or is the data a misrepresentation because a large percentage of the population live in Southern Maine?