It is almost a cliché in Administration to say that management and decision making must be supported by data and information. This obviousness, however, has subtleties that often go unnoticed and can cause serious consequences for a company, sometimes irreversible.
After all, there is only one thing worse than not having information: having bad information or “misinformation”. A study published in the Harvard Business Review estimated that, in the United States alone, the damage caused to the economy by low quality data is $ 3 trillion, equivalent to about 16% of that country's annual GDP or 4% of world production wealth in one year.
The fact is that the abundance of computational resources brought both benefits and pitfalls. Countless are the data visualization tools, commonly called BI (Business Intelligence), which are dedicated to group information and supposedly facilitate its understanding through the presentation of graphs and summary tables.
But is a simple graph of the average monthly sales sufficient to assess the behavior of the commercial sector? Is the work of salespeople the only variable involved? These and other questions refer to some of the dangers of inappropriate use of these tools, often (or almost always) without adequate analysis and reasoning. To remedy such deficiencies, a science, sometimes overlooked and left in the background, is the most appropriate remedy: Statistics. After all, it deals with the provision of theoretical and methodological framework capable of validating conclusions and supporting the adequate interpretation of different types of data and visualizations.
To keep a very simple example, it determines that a measure of central tendency, such as the average, must always be accompanied by its dispersion (how spread these data are), such as the mean deviation or the standard deviation. We can understand this by imagining twenty patients in a hospital, with ten receiving 1.5 kg of daily meals and the rest absolutely nothing (0 kg). On average, they are 750g per individual, which could possibly seem satisfactory, but in a few days, half of these people will be dead.
But if Statistics has such great potential, why is it so often left behind in decision making? Firstly, there is a popularly generalized view that it “does not work”, originating mainly from electoral polls that are commonly poorly conducted, especially when the final results are significantly different from those presented in the surveys. In addition, it has long lived with outdated academic curricula in undergraduate courses, which placed little emphasis on the practical application of theory in organizational contexts as well as the use of more recent technological tools. The good news is that in recent years, albeit late, this update has taken place more intensely and rapidly in educational institutions across the country that
In companies, the thirst for information is so great that sometimes everything that lies ahead in terms of data is blindly drank. But the prudent manager must keep in mind that, first of all, it is necessary to check if what is in front of him is water and not poison, information and not misinformation. For that, nothing better than Statistics.
Jeanfrank Sartori, Master in Information Management, Specialist in Business Intelligence and Bachelor in Administration. He works in the Academic Controllership and Quality sector at Grupo Positivo, in Curitiba (PR).