Why it is Important to Recognize that Statistical Evidence is not necessarily the same as Truth
Introduction
Numbers do not lie. This statement confers the highest level of trust in statistics to the extent that whoever cites data in an argument has the best chances of winning. That is why newspaper reports often throw numbers in their reporting to crystallize their messages. To be sure, numbers do not lie, but people can use them to lie. This is to say that, while numbers represent truthful information, bad actors can manipulate the numbers to spin the Truth.
The problem of statistical evidence not being equivalent to Truth goes back to the beginning of the field of statistics itself. However, it came to global consciousness after Darrell Huff came out with the book titled “How to Lie with Statistics” in 1954. In the book, Huff joked that the book “may seem altogether too much like a manual for swindlers” but then justified it by saying, “The crooks already know these tricks; honest men must learn them in self-defense.” (Huff, 1993). In the same breath, this essay is an attempt at explaining why it is important to recognize that statistical evidence is not necessarily the same as Truth.
Why are people so quick to accept statistics as true?
Statistics entails the collection and analysis of data with the aim of answering hard questions. Since the process is highly scientific in nature, the conclusions that are drawn at the end of the process are oftentimes authoritative. According to Munafò, Nosek, Bishop, Button, Chambers, Du Sert… and Ioannidis (2017), statistics is readily accepted because of its scientific nature. Usually, scientific studies rely on empirical data that one can easily measure. As such, people assume that whatever information results from such a process is objective.
Nevertheless, this is not always true. According to the article, “Can People Lie with Statistics?” it is clear that people can use objective data to serve subject ends. According to the article, data tells a story but just a part of it. Take the example of data that says many millennials are unemployed and that they stay with their parents for longer than older generations. While this may be true, it is not true to argue that millennials are lazier than older generations. This is because the data is one-dimensional and it does not explore reasons why millennials are out of jobs. It could be that the job market has shrunk due to bad economic conditions or the nature of jobs is different from what it was in the past.
From a scientific point of view, is spinning the Truth acceptable?
The ultimate goal of scientific research is to break new grounds and to solve the world’s difficult questions. To do that, scientific research must be efficient and reliable (Munafò et al, 2017). Besides, the hallmark of good scientific research is the ability to be reproduced. This way, subsequent studies can build on the literature in a preceding study to accelerate discovery. Nonetheless, this is impossible if the preceding scientific study lacks credibility. One of the issues that undermine the credibility of scientific research is using statistics to spin the Truth.
Consider a scientific study aimed at explaining global warming and its climate changing effects. After collecting and analyzing the data, the researcher discovers a strong link between global warming and climate change. However, the researcher goes ahead and spins this information such that the cause and effect relationship between global warming and climate change is unidentifiable. If this study is published and shared to the world, the consequences will be unimaginable. For instance, more companies will continue to emit greenhouse gasses wantonly because, after all, scientific research shows no cause and effect relationship between global warming and climate change.
Is the practice okay from a critical approach?
Social research is critical in the effort to understand social behavior and to explain the happenings in human society. Oftentimes, the research shapes (as well as interpreting) social perspectives, which, subsequently, shape data and facts (Hammersley, 2015). What happens when the social perspectives and the subsequent sociological conclusions are wrong? In this case, taking a critical approach helps to disprove the faulty conclusions. Accordingly, it is essential that a critical approach avoid spinning the Truth. This way, a researcher will be able to correct faulty sociological conclusions.
Biases in a news story
News stories are an essential part of the society because they shape opinions and narratives. Accordingly, any kind of bias in news content, whether positive or negative, does a huge disservice to the society (Soroka and McAdams, 2015). One example of bias in news content is a news report by The Washington Post on November 14, 2018 where the paper insinuated that the US President Donald Trump was unhinged during a phone conversation with British PM Theresa May. The headline of the story read, “Five days of fury: Inside Trump’s Paris temper, election woes and staff upheaval” (Dawsey & Rucker, 2018, November 14).
Present in the news story was spin media bias. Notably, the Post used sensationalist words that painted a dramatic scene involving the president. For example, in the last line of the opening paragraph of the article, the Post wrote, “British Prime Minister Theresa May was calling to celebrate the Republican Party’s wins in the midterm elections — never mind that Democrats seized control of the House — but her appeal to the American president’s vanity was met with an ornery outburst.” In the first place, the Post did not provide any evidence to support the reference to the phrase “American president’s vanity”. Secondly, the Post used the phrase “ornery outburst” which paints a dramatic scene. Again, there is no evidence in the paper to corroborate this claim.
References
Dawsey, J., & Rucker, P. (2018, November 14). Five days of fury: Inside Trump’s Paris temper, election woes and staff upheaval. The Washington Post. Retrieved from https://www.washingtonpost.com/politics/five-days-of-fury-inside-trumps-paris-temper-election-woes-and-staff-upheaval/2018/11/13/e90b7cba-e69e-11e8-a939-9469f1166f9d_story.html
Hammersley, M. (2015). On ethical principles for social research. International Journal of Social Research Methodology, 18(4), 433-449.
Huff, D. (1993). How to lie with statistics. WW Norton & Company.
Munafò, M. R., Nosek, B. A., Bishop, D. V., Button, K. S., Chambers, C. D., Du Sert, N. P., … & Ioannidis, J. P. (2017). A manifesto for reproducible science. Nature human behaviour, 1(1), 1-9.
Soroka, S., & McAdams, S. (2015). News, politics, and negativity. Political Communication, 32(1), 1-22.