What if personal attitudes could predict trends in the United States economy? On Wednesday, Dr. James Lepkowski of the University of Michigan visited Kalamazoo College to discuss one area of his research: how regular people can predict vast economic trends simply by participating in a basic questionnaire, as proven through decades of surveys.
“So what do individual consumers and their attitudes have to do with how the economy is going to behave?” The data scientist specifically discussed a survey that began in 1946 called “Measuring Individual Attitudes about the Economy” and aimed “to measure the attitudes for personal and national economic attributes.” The personal attributes, he explained, ask basic questions regarding individual finance: ‘Do you think your finances are going to get better, worse, or stay about the same? In the next month? In the next 12 months?’ The national attributes consist of similar but larger-scale questions: ‘Do you think that the business condition in the next 12 months are going to get better, worse, or stay about the same?’ “The questions are that basic.”
Over time, the survey found that “people’s thoughts about the economy actually are going to translate into what happens in the economy.” Using the example of unemployment, the survey found that people’s expectations would predict financial outcomes three months in advance. “Is there a sound science to connect that kind of aggregate attitude and the outcome in terms of the economy? Not exactly,” claimed Lepkowski. Rather, he suggested that these statistics were the result of an association that has developed over time. “This has been going on month by month for 70 years in the US economy.”
People’s thoughts about the economy actually are going to translate into what happens in the economy.
Why are these predictions important? Dr. Lepkowski explained that large corporations such as Ford, Honda, and Toyota use this kind of data to predict the finances of their consumer base. “When you hear about shifts being closed down or people being laid off, it’s in part because this is part of the prediction process. They want to know in advance what’s going to happen in terms of unemployment, because that will tell them whether people will be buying cars at the current rate.”
K’s guest concluded by advising students planning on entering into the data science sector. “I don’t know anyone who has a degree in statistics or computer science who is unemployed,” said Dr. Lepkowski, prompting chortles from audience members. Additionally, the data scientist shared that all of his PhD students are funded–thus, Lepkowski emphasized, “it’s best to look at institutions that do research because they’re more likely to be able to support their students.”
Math, Physics, and Computer Science triple major Marlisa Pennington ‘17 shared that this event “made [her] think about the intricacies that go into applying math to real world situations.” Next time, she would enjoy also learning “more about the math behind those changes and how the data was actually used.”