4 ms·
(153,057 active HN users * 12.6 events per million)/1,000,000 = 1.4*10^-8 This leads to a P value of approximately 0.0014, which is well below the 5% threshold
by neoncontrails 5y ago
(153,057 active HN users * 12.6 events per million)/1,000,000 = 1.4*10^-8
This leads to a P value of approximately 0.0014, which is well below the 5% threshold.
This is a very crude calculation, but it does give an idea of what we can expect. Given the small P value, there’s a good chance that the HN community has experienced a similar event to the one described in the article.