Machine Learning

I mean..ain't most of articles like that, lately? We just go on about our business...

There are several ways to approach this, however, you should keep in mind that particularly in matters of opinion, you may not get any resolution.

In all cases, being respectful in tone rather than disrespectful is much more likely to get some response.

1. The first is the classic “letter” to the editor which today could also be via email. An actual letter may be treated with more respect because hand written or typed it shows a clear amount of human effort to produce.

2. Next is contacting the author. If it’s a mistake of a fact, they may actually appreciate the correction. Again, being kind or professional in tone is most likely to get a response.

3. If the problem you have is with a matter of fact, clearly documented elsewhere and not disputed, you should provide references to the source of your finding.

4. If you have a problem with them methods used to reach a conclusion, small sample size, method of analysis, bias in data collection, you should indicate this very clearly.

For example, to be truly representative, every person in the population to be sampled should have an equal chance of being included in the study.

I’m not going to go into more detail here, but there’s an essential difference between being certain something is wrong, and not understanding whether there is an error in how a conclusion is reached.

5. In matters of argumentative process, where you believe that the author has made an error in constructing the argument they are making, for example, they’ve violated the laws of logic by using faulty foundations for their argument, committed a logical fallacy, or are being deliberately misleading you should be aware of the potential difficulties in communicating this error to the author.

This is one reason that I wish principles/rules of argumentation were taught at an early age and educational level. Taking a class in formal logic at any time is the equivalent of arming yourself with the most effective mental armor you could ever have.

6. In matters of opinion you are least likely to get a “satisfactory” response. Often, people do not understand the reasons behind why they believe certain things are right or wrong. They may be relying on the opinion of an expert, an authority, the majority of people they know, their own gut feelings of right and wrong or their daily horoscope or lucky number predictions to formulate their opinions.

Understanding how an author has formulated their opinion is very helpful in figuring out whether someone else can be brought around to your point of view.

There is also the question of why someone holds an opinion. Is the opinion based on personal or economic self-interest? On political or religious motivations? On a difference in observational point of view? It is sometimes very difficult to impossible to know this for sure. But understanding this is essential for changing someone else’s opinion.

6. Bringing there error to the attention of a public competitor to the publisher with more ability and resources available to make your case for you.

7. Appealing to a political authority. In this case being able to say you are an active voter is helpful. So vote as often as you can.

I’ve attempted to write this without committing errors of my own. I have experience in evaluating research for factual and methodological errors and have attempted to describe some options based on that experience. However, that may be subject to some biases of my own.

Humans are subjective creatures who are heavily impacted by the culture they are in and the opinions of people they know. We exist in bubbles socially, economically, morally and geographically.

Those who live in a desert will automatically have different perceptions and opinions about water rights than those living near a major water source.

Pretty simple.

Email the corresponding or lead author. Be polite. It's possible the authors have considered angles that you haven't.