You may think that scrolling endlessly through social media is a harmless way to decompress after a long day of work and let your mind relax. And the latest research on the mental and emotional effects of sinking hours into social media suggests that it has a relatively limited effect on your well-being.
While social media may not be the cause for the increasing youth depression rates, it does have a perhaps, more insidious effect on our critical thinking skills. The average adult spends 2 hours and 24 minutes every day on social media. It’s impossible to spend that much time doing a single activity in your day without the repetitive behaviors associated with that activity carrying over into how you do other activities.
How Social Media is Undermining Your Critical Thinking Skills
Said another way, the way you engage with social media is, likely without you knowing it, training you how to think when at work, when interacting with friends and family, and when running into strangers on the street. Patricia Greenfield, UCLA distinguished professor of psychology and director of the Children’s Digital Media Center in Los Angeles, puts it this way: “the mid-21st century mind might almost be infantilized, characterized by short attention spans, sensationalism, inability to empathize and a shaky sense of identity.”
When you thumb through Instagram or Twitter posts, you’re building cognitive habits around how you process and make sense of information. And research suggests that our habits for processing information on social media are far from exemplary. We know this because many people fail to identify fake or false information from true information. In one study, 44% of millennial participants failed to correctly identify whether information was true or false in at least four of nine questions.
The more time people spend on social media, the more likely they are to fall prey to false information. A study sponsored by the Reboot Foundation found that 36% of people who check social media hourly or more frequently held at least one wrong belief about COVID-19, while only 22% of people who checked social media once a week held at least one wrong belief.
Our use of social media limits the development of robust critical thinking skills. Professor Greenfield explains that the visual media we consume on screens “do not allow time for reflection, analysis or imagination — those do not get developed by real-time media such as television or video games.”
While identifying fake news is a key critical thinking-related challenge when it comes to social media, there is another challenge that goes beyond deciphering fact from fake. This is the challenge of determining whether the reasoning that underlies a post or article is rigorous and rationale. Unfortunately, social media is littered with posts that contain critical thinking fallacies. We must learn to identify them or we will fall prey to them not just on social media, but in every area of life.
Here are some examples of common critical thinking fallacies.
Examples of Critical Thinking Fallacies on Social Media
Our goal is not to convince you of any particular point of view found in the examples. Critical thinking doesn’t care about the answer. It only cares about the rigor behind the support for the answer. As we’ve explained, critical thinking is providing a robust answer to a question.
Undermining the Messenger Fallacy:
Many times, people on social media immediately dismiss an idea because of the person sharing it. This is a cognitive shortcut that leads to lazy thinking. There is no law of logic or nature that dictates that if people made statements that are wrong or false in the past, they can no longer make any statements that are right or true.
When you discredit ideas because of their source, you operate out of “stereotype thinking.” Stereotype thinking says that because a certain condition has been statistically probable in the past, it is true in the present. While stereotypes can help people make snap decisions when absolutely necessary, they create significant problems as we can clearly see from the stories of racial inequity that are becoming more visible.
Because most people fall prey to this fallacy, those arguing on social media often resort to a cheap and often irrelevant strategy for dismissing the ideas of those with whom they disagree. Rather than engaging in a debate around the idea shared by their opponent, they simply hurl personal attacks at the opponent. The goal is to discredit the messenger so that we will automatically dismiss the idea.
Here’s a simple, but common example:
The other consequence of this fallacy is that we are much more likely to reject an idea posed by someone we dislike even if we would have supported the idea had it been presented by someone we like – and the opposite is true, we are quick to support ideas shared by our friends even if they aren’t rigorous enough to warrant our support.
Correlation vs. Causation Fallacy
This is a more well-known fallacy that is beat into the head of every statistics student: correlation doesn’t mean causation. Just because two events trend together doesn’t mean that one caused the other. For example, let’s imagine hypothetically that you found data asserting that people drive slower in urban areas when it rains. The conclusion that most people would jump to is that the presence of rain causes people to drive more slowly. If your job is to eliminate the slow-downs, you might try to solve this problem by requiring drivers to go through rain driving training or increasing regulations on tire conditions during vehicle inspections.
However, it’s easy to see that a third factor may be the cause of reduced driving speeds. When it rains, more people in cities are likely to drive (rather than walk, bike, or take public transport and get wet), creating more traffic, which, in turn, could cause people to drive more slowly.
You can see how the tendency to believe that correlation equals causation can cause you to arrive at very different conclusions. @DanaSawan1’s tweet below contains an example of equating correlation with causation below:
Why don’t you just save time by saying:— Bad Panda 428 (@DanaSawan1) July 1, 2020
ALL Cities run by @TheDemocrats are GONE!
George Floyd died in Minneapolis, with:
DEMOCRAT City Council
But it’s @realDonaldTrump’s fault and let’s defund the Police!
The logic used in this Twitter thread is that Democrats are the cause for riots and racist police brutality because the leadership in those cities and states are largely Democratic. While this could be the case, the data shared in this tweet only establishes correlation, not causation. There are many other possible explanations for how both of these facts can be true without one causing the other.
Wrong Denominator Fallacy
Dividing the incidence of an event by a denominator helps achieve what statisticians call normalization of the data. For example, imagine you take a test that has 200 questions and you get 20 wrong and your friend takes a test with 100 questions and gets 11 wrong. If you simply compare the number of wrong answers, you would think you did worse. But you answered more questions than your friend, so you have to divide the number wrong by the total number of questions:
- 20/200 = 10% wrong
- 11/100 = 11% wrong
When you normalize the data by dividing by the right denominator, you can see the that conclusion is reversed: you did better, not worse.
Sometimes people run into critical thinking fallacies because they don’t normalize the data (see @PatriotCowboy2’s tweet below); that is, they don’t divide by a denominator. But a more subtle fallacy is dividing by the wrong denominator.
For example, many have been debating police brutality rates against Black vs. White individuals. To normalize the number of killings by police, many have used the number of Black people vs. White people in the United States as the denominator (see @lilmochababy’s reply below).
we know this and the rates are disproportionate. African Americans only make up approximately 13% of the US population. black Americans are twice more likely to fall victim to police brutality than white Americans. pic.twitter.com/1RdhiHtUEi— j duh (@lilmochababy) July 1, 2020
However, this is a misleading denominator if you’re trying to figure out whether police kill more Black or White people because police don’t interact with and have the opportunity to kill all people in the country. Instead, the denominator of police-civilian interactions is much more robust because interactions represent the number of opportunities police have to kill people.
Our point is not that police don’t exhibit racist tendencies, that police are justified in their killing of roughly 1,100 civilians a year in the United States, or that either of the aforementioned Tweeters’ data is correct. Our point is that if you’re trying to prove the police aren’t more likely to kill Black people (as @PatriotCowboy2 was doing), you have to normalize the data by dividing the number of killings by a denominator. And if you’re trying to rebut @PatriotCowboy2 (as @lilmochababy was doing), you have to be sure you’re using the right denominator.
False Comparison Fallacy
When trying to make sense of information, particularly data points, we commonly compare data to other numbers that we better understand. In business, this is called benchmarking. However, if you compare the data point of interest to other data points that don’t possess analogous traits, then you may misinterpret the original data point.
For example, this spring, many news sources and others on social media shared that the number of people in the United States that had died from COVID-19 had surpassed the number of deaths of Americans in the Vietnam, Korean, and Desert Storm wars combined. This comparison seemed to be made to demonstrate the large number of deaths caused by COVID-19, but is it a fair comparison logically?
Vietnam War 58,220; Korean War 36,914; Desert Storm 292: combined =95,449— Clifton Gerring (@clifton3052) May 24, 2020
2401 more deaths from COVID-19
than in Vietnam, Korea and Desert Storm!
Many could have been avoided with competent leadership. SAD
From a logical standpoint, this seems to be a poor comparison because the number of people at risk for dying in these two scenarios was vastly different and the actions that led to or averted deaths in these scenarios were vastly different. For example, only the soldiers sent to Vietnam, Korea, and the Middle East could have died in those wars, while the whole US population was at risk of succumbing to the coronavirus (see the Wrong Denominator Fallacy above).
The point of benchmarks is to learn from them. But you can only learn from benchmarks if they possess enough similar traits to the scenario of interest that the lessons are transferrable. The way people die in wars and the way you prevent deaths in war is very different from the way you prevent deaths from infectious diseases, making the lessons unlikely to be transferrable.
A much better comparison would be to look at deaths to other diseases that could afflict the whole population, like the flu (which caused 61,000 deaths in 2017-2018 season and 34,000 in 2018-109) and cancer (which is estimated to cause 607,000 deaths this year). Using these comparisons which possess many more similarities to COVID-19, we can quickly deduce that the coronavirus at the time was 50% worse than a bad flu season, but still not nearly as fatal as cancer.
The point here is not how many COVID-19 deaths have been preventable. The point of benchmarking is to identify appropriate analogs from which to learn. We can learn little from the deaths in recent wars that will help us stop the pandemic, but maybe, we could learn from what we have done to thwart other diseases if we understood which diseases affected us at the same magnitude.
You might argue that learning from benchmarks wasn’t the point of @clifton3052’s tweet. This is fair. In fact, it’s more likely that his and others’ point in sharing this information was to make the number of COVID-19 deaths appear large. But even should this be the intent, the reasoning above still holds. You need to compare COVID deaths to deaths by similar causes.
For example, we wouldn’t tell a 12-year old little league baseball player that he didn’t hit many home runs during his 20-game season because he only hit 22 home runs and Mark McGwire hit 71 in a 162-game Major League Baseball season.
Firsthand Experience Fallacy
When people experience something firsthand, we tend to give their opinion more credence than those who lack firsthand experiences. This is why there is occasional outrage on social media after celebrities share their political views. People think that celebrities’ lack of political experience automatically makes their idea wrong. While this makes sense to an extent, carelessly rejecting ideas that come from those without firsthand experience or blindly accepting ideas that come from those with firsthand experience is irrational.
Consider @goldengal_22’s reaction to Taylor Swift’s foray into politics:
Well, it was primarily Trump fans and their families who started with her and brought her to fame…not you.— CrazyTimes!🇺🇸 (@goldengal_22) June 19, 2020
She should leave politics out of her area…she’s stepping out of her lane and it isn’t becoming of her.
The point is not that celebrities always or even often have good political ideas, but only that being a celebrity, and not a politician, doesn’t make it impossible to generate compelling political ideas. This is a nuanced version of the “Undermining the Messenger” fallacy.
We do the same when we veto rich people’s ideas for helping lower-income people simply because they are rich. The criticism @supitsshekinah has received for sharing her opinion on matters related to Black people is another example of this fallacy:
If my black voice isn’t legitimate because my parents are white then neither is Colin Kaepernick’s.— Shekinah (@supitsshekinah) June 29, 2020
Generalizability is the extent to which something that works in one place works in other places too. When something works once or in one place, we can be quick to assume that it will work everywhere, but this over-simplifies the role context/environment play in determining the success of an idea or intervention. Every context or environment consists of thousands of variables that influence the applicability and success of ideas. When a celebrity shares her experience using a particular product and the amazing benefits she derived from it, we are quick to assume that it will work for us, not recognizing the fact that our lives are far different from hers. Just because it worked for her doesn’t mean that it will work for us.
There are countless examples of the Generalizability Fallacy. Here’s one:
The thought of Monday morning stressing you out? Discover how some of the world’s most successful people start their days. https://t.co/9Wm4umIh9x— District Ventures Capital (@districtvcap) August 12, 2019
The point isn’t that these are bad or counterproductive rituals or that we shouldn’t learn from successful people. Of course, we should. However, you need to consider the differences between your life and context and the life and context of the noted celebrities before assuming their suggestions will automatically work for you.
This is the value of academic research studies. They intentionally recruit a representative group to participate in studies so that we can know that if an intervention works for them, it will likely work for us.
Assuming Motivations Fallacy
We can know the actions of others but we can’t know their intentions or motivations unless they share them. Motivations are “the reason or reasons one has for acting in a particular way.” We can gather evidence from which we can deduce others’ motivations, but at the end of the day, statements about others’ motivation and intent will always only be assumptions.
Many of the criticisms that we lob at each other on social media and at our leaders and others in authority are based on assumptions about their intentions. We think that because they acted in a certain way, they must be motivated by negative or deceitful intentions. For example, many of those opposing government lock-downs during COVID-19 have jumped to the conclusion that liberal governors and mayors have employed expansive restrictions in order to move toward their supposed vision of greater government control:
Because for our Democratic Governor’s like Newsom, it’s all about making sure President @realDonaldTrump DOES NOT GET RE-ELECTED ‼️Dems are now all about Socialism and Dictatorship 😡‼️ We can’t let them win! #Trump2020LandslideVictory 🇺🇸 https://t.co/dyo7vJt1cM— Sungodess (@SDSUgrad1983) June 30, 2020
The problem is that we can’t actually know that Governor Newsom is imposing lockdown and other restrictions to hurt President Trump’s chance of re-election and/or to advance more socialist policies unless Newsom says that is true.
In this particular case, some may argue that Newsom has admitted to these intentions when he said on April 1, “There is opportunity for reimagining a [more] progressive era as it [relates] to capitalism. So yes, absolutely we see this as an opportunity to reshape the way we do business and how we govern.”
While this statement makes it more likely that @SDSUgrad1983’s tweet is accurate, Newsom’s statements can be interpreted in multiple ways. Newsome may have imposed lockdowns and asserted greater government control with the primary motivation of limiting the harm of COVID-19, and at the same time, he may appreciate and welcome the added byproduct of moving California into a “more progressive era.”
Our point is not to defend Newsom or other Democratic governors against accusations of socialism, but only to demonstrate how difficult it is to determine others’ true motivations. Because it is so difficult, critical thinkers avoid assuming intentions or appropriately humble and caveat their statements when making assumptions about others’ motivations.
Many are quick to believe an idea is not good if it comes with downsides. However, those who fall prey to this fallacy possess a naive view of ideas or actions. Almost every idea contains some downsides. You don’t reject ideas because they contain downsides. You reject them when the downsides outweigh the upsides.
One recent example of this is the implication that protesting police brutality is not a good idea because following such protests, police are less likely to engage civilians and crime tends to go up.
The Ferguson Effect is real, and it is spectacular.— Honey Badger Antebellum (@SalsaPrice) July 2, 2020
"would lead to an additional 2,000 black homicide victims in 2015 and 2016"https://t.co/GzqWIhVSkk
From a logical standpoint, you could decide the extra homicide victims indirectly caused by protests are “worth” the reforms that protests could bring about. At the same time, the Ferguson Effect may not be worth the potential benefits of protesting. Our point is not whether it’s worth it or not, but only that when evaluating a decision, you must pit the upsides against the downsides and see which are greater. Police protests are not, by default, bad ideas just because they may lead to the Ferguson Effect.
For example, it would be silly to claim that exercising is a bad idea because it can leave you sore and tired. These downsides do not outweigh the upsides. When confronted by an idea’s downside, avoid immediately rejecting it.
How to Deal with Critical Thinking Fallacies on Social Media
While long already, this is not an exhaustive list of critical thinking fallacies that abound on social media. We will continue to add to this list and share more examples of the above fallacies.
Why bother calling out critical thinking fallacies on social media when we are focused on helping ambitious professionals accomplish their goals faster?
There are 2 main reasons:
- The 2.5 hours a day you spend on social media is training your mind how to think. Every time you fall prey to a fallacy on social media, you make yourself more likely to fall prey to that very fallacy in the future, when you may be at work rather than liking a tweet.
- The opinions people form about what is happening in the world are no longer isolated from their experiences at work. The virtual protest by Facebook employees concerning Facebook’s handling of President Trump’s posts show that people are increasingly bringing their perspectives on world issues to the workplace.
Simply put, it’s important to your work performance that you practice robust critical thinking practices when scanning social media.
To that end, our hope is that this article leaves you feeling better equipped to do just that. Though this article may leave you feeling more confused or less hopeful about your ability to make sense of the world around you. And, to some extent, we hope that many will become more sober-minded because of the ease with which we can all draw fallacious conclusions.
However, there is hope. One study found that simply exposing participants to a series of guidelines for evaluating news online led to a reduced likelihood to trust, like, and share fake news. While the effect was small and the challenge (i.e., identifying fake news) was a bit different, this study shows that we can learn to become better critical thinkers through simple interventions. That is the intent of this article – not to pick sides or tell you what is true – but to help, even just a little, to train you to become a better critical thinker.
If reading this has made you realize that you need some training on critical thinking, consider these three resources: