3 ms·
81k is kind of a shitty sample size to make a sweeping worldwide judgement for 7+billion. Next part, since the focus of 2020 was Covid testing, I'm pretty sure
by NoOneNew 6y ago
81k is kind of a shitty sample size to make a sweeping worldwide judgement for 7+billion.
Next part, since the focus of 2020 was Covid testing, I'm pretty sure flu testing resources were lessened in the process, if not borderline ignored.
At what point was it okay to judge a population on 0.0000001157142% sample size?
While I'm not a fan of the WHO anymore, to be fair to them, where the hell do they say the flu disappeared? This title is disingenuous.
- anoncake 6y agoProperly selected, 1k is a sufficient sample size for any population.
- drivebycomment 6y agoThis is a common misunderstanding of the statistics. The sample size is more a function of the target variance and the confidence level, not as much on the overall population. 81k, if picked properly, would be sufficient for their purpose.
- NoOneNew 6y agoOkay, what are the rules to determining the proper sample? How can you objectively determine if one sample is better or worse than another?
- drivebycomment 6y agoIf you are asking about sample size selection, you can use: http://www.tools4dev.org/resources/how-to-choose-a-sample-size http://www.tools4dev.org/resources/how-to-choose-a-sample-si... If you want to evaluate quality of a sample, that's an entirely different issue. You may want to look up different sampling methods.