Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : The Trump Presidency -- Ignore unavailable to you. Want to Upgrade?

To: combjelly who wrote (92730)	9/5/2018 12:23:26 PM
From: i-node	Read Replies (1) \| Respond to of 365889

"For one, if the relationship holds up over time, that is compelling. Does it prove it? It depends. For one, if the relationship holds up over time, that is compelling. if there is some mechanism that links the two, then yes. If ice cream consumption actually correlated with shark attacks, that should get you attention. If it holds up over time, then you look for a mechanism that links the two. Say ice cream consumption goes up in the summer and people tend to go in the water during the summer and that leads to more shark attacks. In this case, there is a causal relationship, but the cause is it being summer." You are entirely on the right track here. But you left out a truly essential, most time-consuming, part of the process, which is to actively LOOK for alternative explanations -- other variables that may be covariant with an independent and dependent variable contained in a study. In this case, actively seek out explanations. A step which has not always been done in modern science (e.g., global warming). We got our hockey stick, let's go with it. This step is necessary to rule out what Feynman referred to as Cargo Cult science. Back to my original point, "In no case does correlation between two variables imply a causal relationship." It just doesn't. Ever. At most it can lead you to a guess that further research may be able to establish. A near perfect correlation proves nothing more than a relationship between two series of observations. There is no "it depends." Correlation cannot do that, period. Just saying, "Well, that makes sense. Someone hypothesized that increasing CO2 creates the "greenhouse effect" and now we see that CO2 goes up, temperature seems to go up, so increasing CO2 causes temperature to go up." No, it doesn't work like that.