Biases that everyone should know

August 19, 2024 - 9 minutes

In research, there are some errors that have to be avoided. I’m not talking about errors in calculations or bugs in the code, but I’m talking about reasoning fallacies. Those errors are called biases. The ability to find biases are fundamental for being a critical thinker, this means that we are owners of our opinions and we have an own understanding of the world.

If we can recognize the biases, we can also avoid them. However, sometimes we ignore some aspects or opinions because we want to be accepted by the people. It is intrinsic in human behavior, we follow a leader that shows the way. But we want to be critical thinker, so, there is a formula that we have to remember:

\text{Critical Thinking} = \text{Claim} + \text{Argumentation}

It is very easy, isn’t it? Unfortunately it is not so easy to apply. If it was, several problems in the world, like pollution or wars, were already solved.

The claim has to satisfy two properties to be a proved:

Arguable: we can discuss this opinion or hypothesis. It has to be possible that someone else could think about it differently, it should present a point of view and it can be true or false, based on different points of view.
Precise: the claim should be as much precise as possible, in this way we can take strong argumentation in favor of it, and make its falsification difficult. This also makes it easier for others to understand and evaluate it.

Let’s do an example.

Bad claim:

My friends say that iOS is better than Android

Imprecise claim:

iOS is better than Android

Better claim:

iOS is safer and more user-friendly than Android

There are some aspects that we need in an argumentation:

Premises: they allow to define the background context of the argumentation. They can show observable facts, or assumptions that in the current time are established truths and universally accepted. For instance, the Earth is a globe (sorry flat earthers).
Connection with the claim: the argumentation should connect well the premises to the claim that we want to prove. Here the fallacies and biases could appear.
Internal or external: based on who we want to convince, if ourself, internal; if someone else, external.
Irrefutable conclusion: ideally at the end of our speech, the arguments should be irrefutable.

Now that we know the basics of critical thinking, let’s enter the topic.

Correlation implies causation

It consists to confuse correlation with causation, so when two things happens at the same time, we consider that one causes other.

Example

Let’s do an example to clarify:

The bottles of water are made of plastic. The plastic is polluting. Therefore, water causes pollution.

In this example, the simultaneity between the bottles of water to be made of plastic and the fact that plastic causes pollution does not imply that water causes pollution. It does not make sense. But both sentences are true at the same time.

Hasty generalization (sample selection bias)

It is a fallacy where the conclusion is a generalization of an event. In this case a small number of cases about a phenomenon are considered as the final conclusion about it. This kind of bias is very common in math, often it is called proof by example.

Example

I make this example to explain:

3 and 5 are prime numbers. 3 and 5 are odd. Therefore, all prime numbers are odd.

In the example, the final conclusion is clearly wrong, since 2 is a prime number and it is even.

But I want to be more clear, this is an example that could happen in reality. We heard on the TV news that a man from another country was stopped by the police because he was a pickpocket. Then we might think that all the people from that country are pickpockets. This is not true, this fallacy is the cause of many prejudices and discriminations against people.

Cherry picking

The cherry picking is a fallacy that consists of looking at the data necessary to validate a thesis. This is very common in arguments because everyone wants to validate their own opinion, but sometimes to hold up the claim the data cited are not complete.

Example

To prove that cigarettes are not so bad for health, a friend told me that his grandma, a 84-years-old woman who smokes a packet in 2 days, is healthy. So smoking is not so dangerous.

In this case, my friend considered only one case, his grandma. But if I check the scientific literature and the statistical data, for sure, they will confirm that smoking is bad.

Straw man fallacy

This fallacy is a technique used especially in discussions, when someone misrepresents an argument in order to attack it easily. Often, there is an oversimplification of the claim in order to make it less defensible. Whoever exploits, consciously or unconsciously, this bias to hold up their own reasons, create a distorted reality to gain advantage in the discussion.

Example

An example is the following discussion. This is an example. There is a person who likes to eat meat that is having a discussion with an environmentalist. The environmentalist says: “We should reduce the amount of meat that we eat in order to mitigate the climate change”, and the other answer: “So you are saying I have to become vegetarian and never enjoy a good steak again? Forget it!“.

These kinds of arguments are very common, both in reality, and in the TV shows or public debates. It’s easy to fall for it and make the same mistake.

Confirmation bias

I will be repetitive, but also this bias is very common. This fallacy consists of taking for granted that a sentence in the argumentation is true, even when the facts are not completely proved. Sometimes we assume that something is true because it is our belief even though there is no evidence about it.

Example

For instance, an argumentation like this:

Since everyone thinks that the best cuisine is Italian, pizza is the best dish.

It is wrong from the beginning. There is no proof about that, so the argumentation falls. (Unfortunately, I met people that don’t think Italian food is the best.)

Begging the question

It consists to answer to the question repeating in another way, with the same meaning, the question. Instead of supporting the answer with an argument, the argument is the question which we are answering to.

Example

For instance, someone in a bookshop asks the seller: “Why is this book so popular?” and the seller answered: “Because everyone likes it!” In this answer we can see that the seller is not actually answering to the question but they are using a circular reasoning. At that point the reply should be: “Why does everyone like this book?“.

HARKing

This kind of bias occurs in research. It consists of formulating a hypothesis just after having seen the results. This is not a correct way to proceed. The hypotheses should be formulated before to produce results, in order that our prior opinions are not biased by the outcome.

Example

A group of researchers collect data on exercise and happiness without a clear hypothesis. After analyzing the data, they notice a pattern and only then, they claim that exercise would increase happiness.

Survivorship bias

Survivorship bias is a type of sample selection bias where someone takes as an example or reference a successful group as the whole group. It occurs when the items that didn’t survive does not appear in the selection. This is commonly used to convince people, especially during motivational speeches.

Example

This is a clear example:

Invest in cryptocurrency too! Look at my friend, he earned a lot of money, with zero effort!

In this sentence there is a survivorship bias. The example that the speaker does is to to take as example one successful case that encourages you to act in investing. The reality is not so simple, there are plenty of people that did not earn but this is not mentioned.

Jumping to conclusion

It happens when someone makes a conclusion directly based on insufficient evidence. Sometimes the conclusion is not wrong but it is wrong the reasoning path used to reach the conclusion. This is similar to correlation implies causation.

Example

For instance, you see a person driving a supercar and you think that that person is very rich and successful. But you are ignoring different factors, for example, that they may have rented the car or borrowed it from someone else. Maybe they are not successful because they aren’t working and their family is rich. Therefore, you need further information to formulate a correct conclusion.

Post hoc fallacy

It is a fallacy where the causation of another event is attributed to the previous event just because it happened before. It consists in the following assumption: since A happened before B, then A caused B.

Example

A silly example could be:

Since the rooster crows before the sunrise, then the rooster is very powerful because it makes the sun rise.

Reduction fallacy

This fallacy is strongly connected to the straw man fallacy. It consists of describing a complex phenomenon using few and simple concepts.

Example

Let’s do a simple example:

Poverty is caused simply by laziness

This sentence ignores a wide range of other factors like lack of education, economic opportunities, and social inequalities.

Sunk cost fallacy

This fallacy is very important to know, because everybody can find themself in this kind of situation. It consists of defining our tendency to follow through on something just because we have already invested a lot of resources, even though giving up takes clearly enormous advantages. The resources invested can be time, money, effort or emotional energy, so we consider that investment important. But in some cases, the best idea is to stop and move on to something else.

Example

An example may be someone who continues to attend a course that they hate, just because they have already paid for it. In some cases it’s better to drop it in order to save money and time, in addition the mental health will benefit of this.

False dichotomy

This fallacy is an oversimplification, it consists of presenting an argument defining two extreme and opposite options as the only possible choices, without taking into account the shapes and the middle way of a scenario.

Example

This sentence is an example:

If you like playing video games, then you must be lazy.

This is a fallacy, because it wrongly suggests that enjoying video games and being active are mutually exclusive.

# To know more about these topics, I strongly recommend to read Lucis Philosophy blog