Читать книгу Germany's Freefall - Hermann Dr. Rochholz - Страница 8

Statistics only provide valid results when they don’t include systematic errors or false basic assumptions. This, in turn, refers to “statistics one has faked oneself” (cf. previous chapter).

Statistics thus open the door to manipulation: If you assume a minor thing to be false, then the statistic is false. If you deliberately assume it to be false, you can then “generate” almost any result using statistics.

An article was published about seven years ago on the drastic increase in the proportion of “grave defects” found in motor vehicles after car inspections performed by TÜV (German technical inspection agency). An outcry rippled across the automotive world. You should sit up and take notice to press reports like this because if 1/4 of the inspected cars had exhibited "grave defects” over decades, and 1/3 of these cars exhibited the same in the following year (these are theorized figures), then this would mean a sudden increase of 32%. That alone is statistically unlikely. The “cause” was to be found elsewhere: TÜV had hitherto distinguished between a “defect” and a “grave defect”. A new regulation had categorized all defects detected by TÜV as “grave defects”.

German Railways (Deutsche Bahn) has issued instructions to prioritize its express trains over its commuter trains. Doing this would make its ICE (Intercity Express) trains statistically more on time. Commuter trains are not covered by the press. This is something you need to know. You find out about it from the “petty” railway official. Presumably, the salary of the Deutsche Bahn CEO is linked to the punctuality of its ICE trains: His contract, for example, would then contain the clause that every percent the ICE is not on time would cost ½ million annually in salary or bonuses.

How about a more current event [76]? In mid-April, the “Tagesschau” evening news had reported on an “above-average number of deaths in Germany” in its “Corona Live Blog”. They compared the average from the past five years with the average from the year 2020. To prove this, they evaluated data between March 23rd and April 12th. Correct: The mortality rate during this period is higher than the average of the past five years. Does this necessarily mean that the mortality rate in Germany is higher due to Corona? Conversely: Is manipulation possible using correct statistics? Of course. You just have to choose the right period: The 2020 mortality rate was lower prior to March 23rd. If you examine this figure between January 1st and April 12th (January 1st is, of course, arbitrary as well!), then 8,300 fewer people (!) had died by 12 April than the average of the previous years.⁴ Almost exactly 1 million people die in Germany in any average year, meaning 2,750 per day. Early June saw less than 9,000 deaths caused from or with Corona, thus less than 1%, or less than those who normally die in two days. These figures cannot be used to make a valid statistical statement.

The press is reporting on a high share of electrical consumption covered by renewable energies in June of 2020. Electricity consumption has decreased during Corona. So, it is of no surprise that the share of renewable energies is on the rise.

Reports circulated about “particularly sensitive” medical detection methods with a “high hit rate”. The problem is the “false positives” since these are “particularly sensitive”. These had also recorded hits in many where they should not’ve done so at all. To put it oversubtly: Had the detection method identified all examined people as “positive”, then it would have resulted in a 100% hit rate. But this is of little use, since it would have identified nobody as negative in this case.

Another example are the aforementioned unemployment statistics. How exactly this is being accomplished is shown in the chapter “Manipulated Unemployment Figures”.

No positive proof is possible in principle in individual cases. For example, the television talk show rounds introduced young people who never went to school but still managed to get by in life. We can hence conclude that this kind of lifestyle can work. Nothing more. If one were to invite those 100 people who had never attended school and who are now hooked on the needle or struggling through life as petty criminals, for example, it would immediately become clear that the statistics are being manipulated here. School students like these kinds of TV broadcasts. Their obvious interpretation is: “School is redundant”. But they would need the education of a school to realize that they are being badly manipulated now.

It’s obvious that a “school system” cannot be ideal for every student. It's supposed to help most students receive a good education. A compromise. This, too, is statistics.

Unfortunately, statistical manipulation is increasingly common in scientific publications, where results can be evaluated only partially and which continues to depict what is en vogue or, in particular, contradict political correctness: Nutritional science is to be mentioned here, but publications on environmental toxins and gender studies are affected by this as well. Sometimes the analyses are even “sound”, but the results are formulated in a way that is “officially desired”. Otherwise, funds are cut off from the institutes. Research at universities is no longer free of charge because they are more and more forced to finance themselves. Due to their dependence on companies, they have to provide “appropriate” results. Politically incorrect results are sometimes not published at all because the journal reviewers either don’t approve or otherwise prevent their publication. Various indirect or direct methods are available in such cases.