4 ms·
> There is something useful to me in posts that show that I'm not alone, because it suggests a certain momentum towards alternatives, reducing my risk. No, the
by aioprisan 10y ago
> There is something useful to me in posts that show that I'm not alone, because it suggests a certain momentum towards alternatives, reducing my risk.
No, there really isn't anything useful in statistically insignificant samples, except to further your confirmation bias. Stats 101.
- vegabook 10y agoHow are 1800 comments in one HN thread post-Macbook-pro launch, an "insignficant sample"? Stats 101. And should we ignore the qualitative evidence that this issue has been picked up by multiple independent sources, including the ultra-Apple-friendly Macrumors itself?
- aioprisan 10y agoYou analyzed 1800 comments and can say which way each leans? 1800 comments from people who come here and want to be heard is not a good sample and is statistically useless https://en.wikipedia.org/wiki/Self-selection_bias https://en.wikipedia.org/wiki/Self-selection_bias
- vegabook 10y agooh please. Not everything in life has an rsquared and a p-value. It's perfectly obvious, when perusing the HN comments and the rest of the internet, that the Macbook Pro has been generally poorly received by high end users, for credible reasons. You don't need a degree in statistics to figure that out.