Listening to Podcast Episode 4 on “The Numbers Game” had an impact on my interactions with statistics. Currently, I’m enrolled in a Business Statistics course here at UR, where over the course of this semester, my peers and I have to work in groups to present three written and oral presentations on a research topic of our choice. My group decided to research the following question: During the COVID-19 pandemic, have some age groups in the United States been experiencing depression more frequently than other age groups in the U.S.? To effectively answer our research question, we retrieved reported data from the U.S. Census Bureau’s Household Pulse Survey conducted over thirteen days. Participants in this survey were part of the United States population. They self-reported various demographic information and their frequency of feeling down, depressed, or hopeless over the thirteen days that the study was conducted.
My group used relative frequency to test if 18 to 29 year-olds are more likely to experience depression more frequently than 40 to 49 year-olds and 60 to 69. We learned that if a person is between the ages of 18 to 29, there is a 71% chance that they experience depression more frequently than 40 to 49 year-olds and 60 to 69 year-olds. Knowing what I now know about statistics, I am highly skeptical of the people who self-reported not experiencing depression at all, specifically if these persons were older in age. From my viewpoint, older people generally do not admit that they are experiencing health issues even if their concession could save their lives. Whether this is because of their pride, upbringing in history, or other relevant factors, I wonder if, in reality, 40 to 49 year-olds or 60 to 69 year-olds experience depression more frequently than 18 to 29 year-olds.
I think this is an excellent example of thinking about how the numbers may not be what they seem. It has definitely been my experience as well that older people will downplay mental health issues even if it is affecting themselves.
The point you made about self reporting is really important. When combined with a tought, touchy topic like depression too, results can definitely be biased and not accurate. Furthermore, people can present this data that makes these numbers look like either a lot or very little at all (as shown in the 25% vs. 1/4th topic). I was also surprised by how much emphasis Huff places on needing to get a large enough sample size for surveys like this to be applicable and accurate.
Your research is super interesting and you make a great point here about the self-reporting concerns in surveys! Some people may not feel comfortable reporting medical information like depression. But the errors in self-reporting may also be unintentional – some people might just not remember! However, it would be difficult to get data on depression just from observing people and their actions, so I guess it’s a trade-off.