Even if an author cites his/her sources, it may be difficult to verify if he/she represented their content correctly, due to:
1. Sources being undigitized
2. Language/palaeography barrier
3. Sources simply being too difficult to understand *correctly*. E.g. much of Rosstat data
1. Sources being undigitized
2. Language/palaeography barrier
3. Sources simply being too difficult to understand *correctly*. E.g. much of Rosstat data
https://twitter.com/gamestales_com/status/1621997525346689025
For example some Russian official statistics may not be necessarily "wrong". It's just that they are represented in a way that a layman is 100% guaranteed to misinterpret them, unless he/she conducts a special research on what does Rosstat mean exactly by this or that figure
Imagine you are trying to estimate Russian import dependency in light swords. The obvious solution would be to look up Rosstat data on
1. Import/Export
2. Domestic production
of light swords and compare them
It would be totally wrong though and will lead to absurd conclusions
1. Import/Export
2. Domestic production
of light swords and compare them
It would be totally wrong though and will lead to absurd conclusions
Why? Well, because of how they count. An agency that is counting import/export of light swords is using one code (= understanding of what constitutes a light sword), while the one that counts domestic production uses another
These figures seem to be comparable, but they are not
These figures seem to be comparable, but they are not
Therefore, anyone who tries to calculate the light swoed import dependency based on import/export vs domestic production data will be fooled, without knowing he is. And once he publishes his wrong estimate in a paper, others will quote it further, mindlessly. It will become legit
Does the Rosstat produce an accurate estimate of light sword import dependency for its own needs? Of course. It is being constantly referred to in the official documents. They just do not publish it. The quantitative estimates based on publicly available data sets are wrong
Making conversion from one code to another based on publicly available data is difficult to impossible. Rosstat makes this conversion, but doesn't show it to anyone except for the government. Meanwhile, estimates based on publicly available unconverted data are just wrong
NB I am not discussing statistics in general. I pointed out to one particular mistake that became widespread in the literature on the light swords market to explain why it is a mistake and why it was made in the first place
• • •
Missing some Tweet in this thread? You can try to
force a refresh
