EN
PL
Statistics help us understand and describe the world.
They simplify it, reduce it to a level we can interpret and comprehend with our minds. Statistical models are always a kind of generalization, a simplification of reality. We accept this in order to observe and understand key trends, fundamental patterns and the world around us. We’re interested in what happens in 95%, 99%, 99.99%, 99.99999% of cases. There is always that shadow of risk that we might be wrong. So, can we apply statistics to individual people? Where do we draw the line between a statistical unit and a human being? What if that one-in-a-million error turns out to be us?
In this post, I would like to address a few controversial topics for which I don’t have definitive answers. I suspect such answers might not exist — but each of us should reflect on them personally.
1. Statistical stereotypes
As I mentioned in the introduction, statistics is a simplification of reality.
We observe behaviors, identify frequently recurring patterns, and build generalizations. Even in everyday life, we apply these principles subconsciously. Take, for example, “first impressions.” We meet someone and within seconds form an opinion about them — who they are, what kind of personality they have, what they might be thinking.
Our mind processes numerous variables at once: clothing, posture, speech patterns, gestures, eye contact, and many other traits. It does so automatically, without our awareness, comparing them to previously learned patterns — generalizations based on our past experiences, or what we’ve seen in films, etc. Often, these first impressions prove accurate. But who among us hasn’t been completely wrong before?
Statistical group descriptions work similarly. Some studies suggest that women tend to outperform men in memory tasks, verbal abilities, and emotional intelligence, while men tend to excel in mathematical and visuospatial tasks.
Does this mean every woman would make a better psychologist and every man a better programmer? Of course not. I personally know many women who are more “technically inclined” than I am, and men who are more sensitive than most women. You cannot generalize statistical trends to individuals.
That said, I do know more men in programming and more women in humanities-related fields. If I were designing a marketing campaign targeted at programmers, I would consider that most of them are men. But when speaking to an individual, I would never categorize them based on that.
2.Psychological and medical care
With the rise of technology, the internet, and the digital world, we are facing an increasing number of modern-day ailments and civilizational diseases. Psychological and social problems are also becoming more prevalent. At the same time, public awareness and the trend toward healthy living are growing. Meanwhile, our population is aging.
All of these factors are contributing to a rising demand for psychological and medical support. High demand and a limited number of specialists lead to higher prices or reduced consultation time and quality.
This is where statistics come into play. Doctors may not always run statistical analyses themselves, but they rely on the results — clinical guidelines, scientific studies, decision-making algorithms.
As already mentioned, statistics allow us to build simplified models of reality — for example, models that classify symptoms or predict treatment effectiveness. These models make diagnosis easier and faster, but they also depersonalize the individual.
A doctor often has only a few minutes to make a diagnosis and recommend treatment. It’s similar during a psychologist’s session. There is a risk that the psychologist will, after an initial assessment, place the person into a predefined category.
Due to limited resources, we increasingly risk being treated as objects rather than subjects. I don’t see — and perhaps there isn’t — a good solution. On the one hand, the system doesn’t have enough capacity to treat every case individually; statistical approaches help as many typical patients as possible. On the other hand, this leads to growing mistrust toward healthcare.
As a result, more people turn to dietitians, personal trainers, naturopaths, and so on. Where is the golden mean between the statistical and the individual approach to patient care? Where do we find the resources so doctors can treat patients more personally? I don’t know.
3. Vaccinations
Now we arrive at the most controversial topic: mandatory vaccinations.
Most people likely agree that vaccinations are one of the greatest achievements of medicine. They protect millions from deadly diseases and have eradicated conditions like smallpox. Besides protecting individuals, vaccines safeguard entire communities through so-called herd immunity.
But should vaccinations be mandatory? Take, for example, one of the most studied and safest vaccines — the flu vaccine. The most severe potential side effects include strong allergic reactions and Guillain-Barré Syndrome (GBS). These complications are extremely rare (fewer than 1–2 cases per million).
The Spanish flu, one of the deadliest influenza pandemics in history, devastated populations globally. But is that historical lesson enough to justify mandatory flu vaccination? Imagine this: someone who strongly resists vaccination is forced to take it and develops a severe side effect. That one-in-a-million case. Would their sense of injustice not be justified?
And what about vaccines that — albeit rarely — cause more severe complications, including death? If we force hesitant individuals to be vaccinated, are we not also responsible for their harm?
In voluntary vaccination programs, the patient makes an informed decision — weighing potential benefits and risks. But when vaccinations are mandatory, the choice is imposed. In such cases, informing patients of the risks remains a legal and ethical obligation, but the person cannot refuse.
This raises the question: does responsibility for adverse effects shift entirely to the state? And if so, is the state prepared to bear that responsibility in practice? Does an individual, forced to act against their will, not lose their autonomy — even when the action serves the greater good?
Should personal freedom outweigh the collective good?
It’s a question — a dilemma — for which every answer is both right and wrong. The answer depends entirely on the values one holds.
Summary
Statistics is a powerful tool for understanding the world and facilitating decision-making. It makes decisions more objective and efficient. But it views the world through the lens of numbers and probabilities. It doesn’t necessarily see the people behind them.
It seems impossible for state authorities to make decisions that consider every individual’s needs. Using generalized statistical models to maximize the common good becomes a necessity. But where is the line? When must we stop and say: going further would violate individual freedom and personal space?
This is a very difficult question — and one I will leave to the reader.
Author: Mateusz Żołneczko, senior data analysis consultant at StatSoft