4 ms·
> Without a natural feedback loop, interviewing mostly runs on myth and survivor bias. It is a law of human nature that a chaotic environment breeds magical th
by pron 12y ago
> Without a natural feedback loop, interviewing mostly runs on myth and survivor bias.
It is a law of human nature that a chaotic environment breeds magical thinking and ingroup allegiance, and there are few business environments more chaotic than Silicon Valley, where startups succeed or fail seemingly at random. The noise calls for reason, and because there are no true known causes for success our mind demands some explanation. How many times have we read the line -- often written by some very smart people -- "correlation isn't causation, but we've found a strong correlation between....". What follows the but in that sentence is what is known as superstition. Correlation doesn't equal causation, but no buts follow. Actually, correlation plus confirmation bias equals magical thinking, and "data driven" correlation (with no confirmation bias) equals bigotry. Failing to realize that correlation without causation provides us with exactly zero predictive abilities (it might be "predictive" on unknown present data, but not on future data) is the root of a lot of evil. In fact, "data driven" correlation -- because it disguises itself as knowledge -- creates (or, usually, reinforces) a reverse causation.
> Whatever else one can say about the Mirrortocracy, it has the virtue of actually working, in the sense that the lucky few who break in have a decent rate of success.
This is the worst of all fallacies belying the "SV logic". Even supposing it were true, a large-number statistical observation says little or nothing about the behavior of a single random variable, or a single startup in this instance. Yes, the startup system "works" (for whom is another question) on the whole, but the vast majority of individual startups still fail. Learning back from correlative observations on the large system and implying the so called lessons to an individual company has little grounds in any rigorous reasoning.
> You can protest your logic and impartiality all day long, but the only honest statement is that we're all biased.
Understanding this is one of the keys to progress. How many times have we seen posts discussing sexism in tech deemed "controversial" here on HN? But the truth is that the chance a member of a society imbued with biases for millennia is not sexist (or otherwise biased) is extremely slim. The only way to fight this bias is to seek it out and see it (because it's there).
- sharpneli 12y ago"Correlation doesn't equal causation" I've noticed that people tend to misinterpret this sentence. In a lot of discussions it shows that people think that there can be correlation while the things don't have _ANY_ relationship between eachother, as in the correlation would arise purely trough random chance. That is blatantly false. Correlation implies at very minimum a causal link trough a third effect. The traditional example of drownings and ice cream consumption has that third link. Nice summer days. So those two are linked. I think one should say "Correlation doesn't imply direct causation". That's closer to the truth.
- dragonwriter 12y ago> That is blatantly false. Correlation implies at very minimum a causal link trough a third effect. No, it doesn't. Correlation can be coincidence; while a particular correlation may be unlikely on its own to be coincidence, there are lots of combinations of things in the universe that might be compared side-by-side, and so unlikely coincidental correlations are everywhere, and so if you troll through the universe looking for correlations, you are going to find a bunch that don't mean anything causally.
- sharpneli 12y agoSure. But with enough datapoints the probability of the correlation being a mere accident is going to be vanishingly small. And by enough I am talking about hundreds and more of datapoints. If one looks at things like yearly values of something for within last 10 years like the page spurious correlations is doing one is bound to find some weird ones.
- supine 12y agoPointing out weird correlations that have absolutely no causal link whatsoever is a global sport. https://www.google.com/search?q=weird+correlations https://www.google.com/search?q=weird+correlations The first result is a beauty http://twentytwowords.com/funny-graphs-show-correlation-between-completely-unrelated-stats-9-pictures/ http://twentytwowords.com/funny-graphs-show-correlation-betw...
- pron 12y agoCorrelation is a statistical term that means that two random variables exhibit "similar behavior" across the probability space. Causation is a philosophical concept taken as a scientific axiom. Correlation does not imply causation at all -- neither direct or indirect -- though it may increase the chances of logical consequence that is sometimes equated (albeit occasionally through confusion) with causality. Now, it is true that perfect correlation (which is rarely shown; more often than not in the business world, only "half" of the correlation is shown -- e.g. all successful startups have ping pong tables -- which is as good as nothing at all) in the real world usually implies some hidden causal link, but that link is not what the investigator has in mind. Often, the causation is actually quite direct but flows in the opposite direction from the common interpretation.