How to recognize when science becomes selective

We live in an increasingly noisy world in which news and information bombard us night and day. Trying to make sense of it all is challenging. There are some who want the average person to accept what is being said to them at face value. How can a person decide what is true, what is misleading, and what is downright false?

I am a scientist, so I will approach the topic from a scientific perspective. I will share tools and strategies to help you effectively evaluate how important the messaging is to you.

I do this because I have seen a plenty of questionable, if not misleading information presented to the public as being factual. False information can and has influenced social behaviour, and has become the basis of bad public policy.

Let’s start with some general definitions and build from there.

Science

I understand science to be a systematic and logical approach to understanding how things work in the universe. It is the coherent body of knowledge based on demonstrable, reproducible data accumulated through discoveries about these. Scientific data are facts, numbers, or measurements gathered from research, experiments, or observations using the scientific method.

The scientific method

“Yes, we applied the scientific method. You know, trial and error.”

Humans have been using versions of the scientific method since prehistoric times. An early example is the development and refinement of stone tools and weapons.

The scientific method summarized:

1. 1. Ask a question, conduct basic research about it, and make a hypothesis
   2. Design an experiment and test the hypothesis
   3. Make observations, record the results, and analyze them
   4. Draw a conclusion (findings)
   5. Share the findings

Findings may confirm the hypothesis, or they may not. The experiment must be repeatable, consistently yielding the same results time after time. A hypothesis can be useful in improving understanding about a subject, but until it is proven, a hypothesis is not a fact.

Thinking – a critical endeavour 

Thinking critically involves questioning and being skeptical of the information we consume, examining underlying assumptions, challenging reasoning, and uncovering biases. It can help us recognize what are and are not facts, and aid in forming reliable judgments.[1]

Many aspects of critical thinking are used across all societies, such as asking questions, seeking answers, conducting research, recognizing patterns, and identifying biases. Critical thinking can be used to make daily decisions that affect our lives and those around us.

What is a fact?

Many things are touted as being facts, but this is not necessarily true.

Fact

• A fact is a thing that is known to be true, especially when it is proved to be true.[2]

Not a fact

• Models, prophecies, simulations, hypotheses, predictions, forecasts, paradigms, reconstructions, interpretations, projections, scenarios, ideas, perceptions, thoughts, concepts, notions, opinions, positions, beliefs, suggestions, inferences, assumptions, presumptions, preferences, suppositions, negotiations, consensuses, superstitions, arguments, estimates, guesses, hunches, feelings, rumours, dogma, and narratives are NOT facts.

• Adding strongly worded adjectives or fear-inspiring imagery to non-facts does not elevate them to the status of facts.

• The passionate promotion, approval, or support of non-facts by charismatic individuals, popular celebrities, or powerful people does not turn non-facts into facts.

Should we fix the data?

Data include facts, observations, and measurements that may be assembled for reference or analysis. The following are ways that data can be manipulated:

Interpreted data: When an interpretation is brought into play, the data can become interpreted data that reflects the interpreter’s opinion, introducing bias into the data. Fellow BIG Media contributor epidemiologist Dr. David Vickers showcased how this occurs in the medical research community.

Adjusted data: Data can be modified for many reasons, such as adjusting for differences between older and modern data collected by different recording instruments or methods. Other motivations for modifying data may also exist. For instance, in plotting-up data on a graph, a curve can be fit to the data by hand or by using algorithms, and by rejecting (or largely ignoring) outlier data that does not fit the curve. If this is done, results should be treated with caution and transparency.

Affected data: Some data, if recorded at a particular locale over a long period of time – e.g., at weather stations – can be affected by changes in the recording instruments or the changing of the surrounding environment over time. For example, urban heat island [UHI] effects in urban and nearby rural areas.

Selected data: Where the data “user” carefully selects certain data for various reasons, which can include selecting data that best supports their hypothesis.

Sampling: When data is sourced from the sampling of items, the sampler should take representative samples. Disappointingly, this does not always happen, and bias is regularly introduced in the process.

Labels: The use of such terms as “on record” can be very misleading when the person using these does not define the start and end of the time period for such a record. For example, “this is the hottest or coldest or biggest or smallest on record”. Full context matters.

In the above examples, the precise nature of each should be clearly disclosed, though this is not always done. It is entirely reasonable to expect that scientists disclose such information in their published works. If this disclosure is not provided, it is reasonable to question the data and conclusions formed from them. Furthermore, such lack of disclosure could lead to potentially erroneous data and findings permeating the “science”, if not public policy and society in general.

It is also important to check the data sources, why it is being collected, who is collecting it and using it, and who is funding the process.

These questions and clarifications help identify what are truly valid data and what are not.

Statistics and averages

“Figures often beguile me, particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: ‘There are three kinds of lies: lies, damned lies, and statistics.’ “(Mark Twain, 1907).[3]

Many people, especially scientists, build models, and they may rely on statistics to better understand certain concepts. People building models recognize that models can be useful but can also be wrong. The latter is rarely mentioned.

We have all seen the word “average” used. One must be cautious. For instance,

• • The average of all whole numbers from 1 to 99 is 50.
  • 50 is also the average of all whole numbers from 49 to 51.

The two number ranges are very different, but they share the same average.

A strategy:

• Question how large the dataset is to see if the “average” makes any sense.
• If someone is comparing averages from year to year (or other periods),

• • question if the same number of data points are in each year’s dataset, and,

• • question if the data were recorded at the same locations under the same conditions.

• Confirm that valid comparisons are being made.

Bias in science

Science changes from objective to subjective when bias is introduced. The more subjective, the less valid science becomes.

Ideally, scientists would look at things objectively 100% of the time and consider all available information about a field of study. But this does not happen often. There can be many reasons for this, such as the individual’s beliefs, their level of expertise on the subject matter, the volume of available information and what was considered, the nature of their funding, peer pressure, organizational pressure, and political pressure.

In the scientific method previously listed, the final item is: “Share the findings”. This is what most scientists want, as by doing so, exchanges and discussions can happen, new ideas can be developed, and science can advance.

Publishing articles in scientific periodicals is one way to share findings. Articles typically list the references of books, articles, and other publications that were consulted, deemed pertinent, and cited by the author(s).

The hiccup: Is it reasonable to expect that references to other pertinent work that contradicts any of the author’s conclusions be included in the references? This is done very rarely.

It is important to be vigilant regarding the suppression or obfuscation of data and findings that go counter to popular “accepted” narratives.

House of cards – watch out for windy days

If the foundation of a scientific argument or a scientific study is not based on fact, the entire structure of the argument is weakened.

For example, someone might base the entirety of their research on a model and then build more models based on the initial model. Then they might make conclusions from these as if the entirety is a valid scientific experiment, which it is not. If the initial model is wrong, then the entire structure is like a house of cards that could easily collapse.

Even if the initial model is correct, if any of the derivative models are wrong, the structure can still collapse. Sometimes such houses of cards are incorrectly called facts.

People often use words to qualify that some statements are possibilities and not firm outcomes, such as “may” or “can” or “could”. When one sees these words, it is valid to insert the word “not”, such as “may not” or “cannot” or “could not”.

About questioning

Science is based on demonstrable, reproducible data. Scientists should be receptive if someone asks questions about the nature of the data they used or the conclusions (findings) they reached. This applies even if articles went through peer review and were accepted by a publisher, as there are human beings involved at all steps of the process.

Final thoughts

False information – routinely presented as factual – can and has influenced social behaviour, and has become the basis of bad public policy.

In this article, I have shared some strategies and tools to help sort through the deluge of information that hits us each day. I focused on science, shared definitions of science and critical thinking, and explained the difference between facts. I progressed to what data is, how it is collected, how bias is introduced, and the risks of using non-facts in forming conclusions.

While this article is far from exhaustive, I hope that you find it useful in evaluating what is presented to you in order to make sound, reliable judgments that help in your daily lives.

We must all be vigilant. For if we are not, we risk being sold a bill of goods with potentially far-reaching consequences.

References: 

[1] Inspired from Critical Thinking Tutorial: Definitions of Critical Thinking

[2] Oxford Learner’s Dictionary

[3] Lies, damned lies, and statistics

(Peter Dorrins – BIG Media Ltd., 2026)