Statistics play a vital role in today’s society – they quantify our experiences and, in many cases, make these experiences more tangible for others. When one thinks of statistics, they may think of percentages: about 10 percent of people are left-handed, about 85 percent of American children believe in Santa, and about 59 percent of people don’t read the articles they share.

However, statistics also involves whole numbers, decimals, and intervals; it involves life expectancy, gross domestic product, school enrollment, demographics, and more. Statistics are essential to policymaking as well as describing phenomena. As information becomes more easily accessible via the Internet, journalism and other fields have evolved to incorporate this data into their work. This has led to the rise of organizations like GapMinder and FiveThirtyEight, which use numbers and visuals to convey important information to readers, like FiveThirtyEight’s daily Significant Digits column. If Affinity Magazine had its own Significant Digits, they would look like this:

Infographic provided by the author.

Figures are also useful to, and especially important in, discussions pertaining to the state of society: the longest government shutdown in US history was 35 days, and 80,000 Americans died of the flu last flu season. However, if you have ever been involved in a discussion about statistics, you may have heard some variation of the phrase, “statistics lie.” But what exactly does this mean? How could statistics possibly act as independent entities capable of deception? The answer is: they can’t. The reason most people are confused by statistics is because they don’t know how to interpret them; the ability to use critical thinking in relation to numbers and draw conclusions about data is called statistical literacy. You don’t have to be a math genius to be statistically literate: all you have to do is understand what the information is telling you, and how to communicate it.

Here are a few of the steps to being statistically literate, according to StatLit.org:

  1. Be able to differentiate between association and causation (Saying that X causes Y and saying that X is linked to Y are two different things).
  2. Determine when an association is actually causal (Do X and Y seem completely unrelated? They just might be).
  3. Be skeptical about vague assertions (What is considered high or low? What is considered harmful? If it seems like it needs more explanation, it probably does).
  4. Know the difference between frequently and likely (Frequently = high number. Likely = high rate/percentage).
  5. Know how to read statements involving statistics (The percentage of women who smoke and the percentage of smokers who are women are not the same thing!)
  6. Know how to examine different definitions of a group (Wider definitions for a condition of interest, such as sexual assault or an infectious disease, typically include more people under that definition compared to more stringent definitions.)

Why is this important? First, statistics can be easily misinterpreted. Many people understand percentages as being representative of an entire population when it is really only representative of the surveyed population. While statistics can make experiences tangible, they can also make them intangible; context is incredibly important in statistics and presenting information without context could have adverse results. For example, CNN is facing a libel lawsuit because of their investigative report on the pediatric open-heart surgery mortality rate at a Florida hospital. While the network provided statistics to support their claim, the context was missing.

Another example of this loss of context occurs in college admissions. Affinity is written for teenagers by teenagers, and as many of our staff members (including myself) are caught in the whirlwind of college admissions, it is likely that our readership is experiencing the same thing. While prestige is not a top priority for everyone, it is definitely a priority for many students, to the point where college acceptance rates are seen as indicators of “better” schools.

However, lower acceptance rates do not necessarily mean the number of admitted students drops – Frank Bruni, author of the 2015 book Where You Go Is Not Who You’ll Be, asserts that selective colleges try to increase their applicant pool so that their acceptance rates will fall each year. In fact, many colleges accept relatively the same numbers with little statistical significance, and in some cases, the numbers increase. For example, I selected the top five universities in the United States according to Forbes – Harvard, Yale, Princeton, Stanford, and Massachusetts Institute of Technology (MIT) – and looked at how many students they admitted per year beginning with the 2009-2010 admissions cycle. I then found the standard deviation between each school’s numbers.

Graph provided by the author. Dataset and sources can be viewed here.

As you can see, the standard deviation between annual acceptances is not very high (meaning hundreds or thousands) for these universities. Some schools did see a decline in admitted students, but others like Yale are aiming to increase their class size and continue to see an increase in admitted students even as their acceptance rate drops.

The lack of context given with college acceptance rates often confuses students; while their odds of acceptance are low, it’s not necessarily true that less people are being admitted. Knowing where the information comes from and why it is presented a certain way is crucial; even the infographics in this article require context because, without it, the information being presented to you would lose its credibility.

These examples may prove that statistics can be misleading, but in reality, the way we interpret this information depends entirely on the intentions of those that produce it. When researchers perform a study, they are required to declare that they have no conflicts of interest before publication.

However, every study is impacted by the researchers’ individual goals and biases. In many cases, collecting data for studies works the same way you get soundbites; you scope out the Internet, local venues, every nook and cranny where you can find something, or someone, to give you the information that you need. Research and journalism are alike in that the person or persons seeking information spend a lot of time talking to people. You find people, conduct interviews, and then determine which information you want to include in your final product. For researchers, this is empirical data that may fit into a predetermined model or scale. For journalists, these are quotes or audio clips that fit into the story they wish to tell.

While we may be able to interpret others’ intentions based on what we read, it is not always possible to do so. This is why statistical literacy is important; you can decipher the information for yourself and determine your stance. We can criticize fake news and blatant lies all we want, but our work will go to waste if we don’t know how to properly interpret and report the facts presented to us.

Photo: rawpixel via Unsplash

5
HeartHeart
1
YayYay
0
HahaHaha
0
LoveLove
0
WowWow
0
SadSad
0
PoopPoop
0
AngryAngry
Voted Thanks!