Everyone knows that “there is a lie, a big lie, and there is statistics.” And how many times have I heard these words addressed to myself from those with whom I first spoke about statistical research and its importance. Let’s figure out where lies and where manipulation.
Unfortunately, it should be admitted that the quoted judgment has a foundation. And this image was not formed without the participation of the research companies themselves. But first, an important statement: statistics is science! With a huge number of laws, rules and formulas, which are based on the laws of mathematics. Therefore, statistics is not just a science, but an exact science. However, often small and not-so-great companies, determined to receive immediate benefits, are ready to “draw” the desired result. Of course, when intent is revealed, their image suffers. But such companies operate according to the principle “as long as fools live in the world, we should be able to deceive”. And such unscrupulous, if I may say “researchers”, I met regularly in many countries. And I had to work not only in Russia, but also in India, and in China, and in Brazil. And by the way,
This is the first reason why people often don’t believe statistics.
The second reason for mistrust can be illustrated by such expressions: “the average temperature in the hospital” or “the rich eat meat, the poor eat cabbage, but on average everyone eats cabbage rolls.” And therein lies the basic data manipulation tool. This is not a trivial deception, which was discussed above, but precisely manipulation: when, in order to enter the desired result, researchers use limited samples or poor representativeness.
I will give you a concrete example, without naming names and characters for obvious reasons. I saw an ad once that said that X was the number one tire brand. This surprised me, because this tire brand was hardly known even to a narrow circle of specialists. It turned out that the company’s marketers figured out how to rotate the research so as to fit their results into the desired formulation. It turns out that brand X has become “number one” in the ranking for the number of “likes” on one particular social network on one particular day. Bingo! Brand “number 1”, and the mosquito will not undermine the nose! And this is exactly the wording that marketers used to promote the brand. For creativity, marketers – five. But their contribution to statistics is clearly negative.
In the skillful hands of a professional, statistics are a powerful weapon for manipulation, which, by the way, was fully used during the last presidential elections in the United States. The results of numerous surveys were aimed not at providing objective and accurate statistical information about the state of affairs, but at tipping the scales in the right direction. This manipulation is from a series, if you tell a lie a hundred times at every corner, it will become true.
But if we abstract from the “dark side” of statistics, it is worth remembering that this is exact knowledge based on mathematics. It operates with formulas and laws of natural sciences. And in order not to be fooled, you need to know what nuances you need to pay attention to when reading statistics.
Any research should have something like this wording: “This research is based on …”. Because there can be no serious research without author (s) and source. In the media, you can often see research publications without reference to the source. This is a totally wrong approach and gives rise to doubts about the professionalism of such publications. It is imperative to put a link to the source so that the one who reads the statistics has the opportunity to clarify all the information that is important for a correct understanding of the study.
So, friends, statistics can be a tool for manipulation. Don’t be fooled, pay attention to the data source and only trust reliable sources.