Obama takes last place in survey of the greatest black presidents of the US.
This statement is true… technically.
This statement, as well as many other examples, were the main focus of Feb. 17’s FSL Discover session; misleading representation of data.
Catherine gave us a presentation all about this topic, providing us with many intriguing examples of data misinterpretation.
One of the ones which really stuck with me was an article which was released by the Toronto Star a couple of days ago regarding the HPV vaccine (protects against cervical cancer and only given to the female gender). The Toronto Star released an article originally titled “A wonder drug’s dark side”, which gave the reader anecdotes about the “dark side” of the HPV vaccine, featuring 5 of the 60+ cases of vaccine injuries since 2008. But, there was no evidence which showed the vaccine caused the injury or death. A rebuttal letter was sent to the Star against this article, signed by 65 experts in the field, leading the Star to issue an apology letter and to change the name of the article. This really showed how presenting information incorrectly can lead to negative consequences; one of the reasons why I found this case interesting.
This led us to talk about correlation vs. causation. Things that are correlated (which happen around the same time) aren’t necessarily caused by each other. For example, stating that countries who consume more chocolate produce more Nobel Prize winners (?!), or the number of people drowning by falling into a swimming pool decreased because the number of films Nicholas Cage appeared in was less that year (??!!!), or that an increase in ice cream consumption leads to murder (??????!!!!!). Correlation doesn’t imply causation.
Next, we looked at how numbers can be presented in different ways, and you can use different methods for different situations. For example, 100-125 could be present as 1.25%, 25% increase, 125%, 125% increase, or 25% increase. To further illustrate this point, we were given an activity. We were given a situation and 2 different numbers, and asked to come up with a “tag line” using the numbers, and apply it to our scenario. Our group was give the numbers 100-200 and a scenario about a stock broker promoting investment in a drug. Our tag line- “Your investment in this drug will increase by two-fold!”. From this, we learnt using different methods to present numbers can give different impressions.
We then ventured into a DNA test which Jenny and Catherine had partaken in on the site 23 & Me. We looked at their ancestry, as well as their risks for certain diseases, and how this information was presented. When we were looking at their risks for diseases, we found that the way the information presented is tailored for different people. There were 3 columns, “Your Risk”, “Average Risk” and the risk “Compared to Average”. Jenny looked at the “Average Risk” and “Compared to Average” columns first, while Catherine looked at “Your Risk” first.
After, we learnt how to make family pedigrees, which was by far my favorite part of the session. Pedigrees were useful ways to provide lots of information, as well as to help us learn about our families. Catherine used her family as an example, and led us through step-by-step on how to draw a pedigree, and how we can distinguish certain members of the family to trace back a certain disease or disorder. I personally have a huge family, so I was excited to illustrate my family in a pedigree, and trace a certain trait.
Finally, we received an informative talk from our guest speaker, Michael Peters, a science fair guru going into medicine at UBC. He has participated in many science fairs, going as far as the international fairs. He came to give us advice on science fairs, as well as answer any questions we had on a wide variety of topics. I found it very useful and intriguing, especially because my future endeavors are related to medicine, so I was able to ask questions which gave me a lot of insight into the field.
Overall, it was a really intriguing session in which we learnt valuable lessons about misinterpreting data. Many of the lessons we learnt today can be applied to many real-life situations which are yet to come, and I look forward to applying the knowledge I gained for the future.