How salad became a major source of food poisoning in the US

Re:How salad became a major source of food poisoning in the US by AstralStorm on Friday May 04, 2018 @04:13AM (#16992983)

Correlation cannot demonstrate lack of causation, only lack of linear dose response. (Incl. Multivariate)

Any nonlinear enough process will show with wrong low correlations.

Suppose the process is stepwise - silly sample hypothesis, eating over x grams meat a day is harmful. It is binary. Correlation is meaningless for such a process. You have to use significantly more powerful statistic tools. Logistic regression will somewhat work for such a process for instance if the boundary is somewhat fuzzy. (You can put exact mathematical bound on how fuzzy.)

To extend it for multivariate analysis you get to use such complex tools as nonlinear ICA. (Instead of CCA and PCA for linear.)

[ Parent | on HN ]

Re:How salad became a major source of food poisoning in the US by toasterlovin on Friday May 04, 2018 @11:17AM (#16995204)

I’m using the wrong terminology, I’m certain, but what I’m getting at is that observational (which is probably what I mean instead of correlational) studies can negate a hypothesis, but they can never prove one. So, one hypothesis is that red meat consumption leads to heart disease. If the data shows the trend lines of heart disease and red meat consumption going in opposite directions, then you know your hypothesis is wrong. But if the trend lines are going in the same direction, the hypothesis hasn’t been proven; all you know is that it is a potential explanation.

Now, it may be that there is a different relationship, like the one that you mention: that eating more than X grams per day is harmful. The data from an observational study can negate this hypothesis (if there is no threshold pattern evident in the data), but it cannot prove the hypothesis. All it can do, at best, is let you know that the hypothesis might be true.

[ Parent | on HN ]

Re:How salad became a major source of food poisoning in the US by AstralStorm on Friday May 04, 2018 @01:51PM (#16996618)

Observational studies can be post- interventional and in this case can prove or not a hypothesis that a certain widespread intervention was successful.

This is very rare and hard to run, so more direct interventional short term studies are used more often.

Suppose a certain country believed an observational study pointing that eating more than certain amount of sugar is harmful (suppose backed by other evidence). They instated hard tax and limit on sugar in foods as an intervention. After some time a big observational study is ran and it is assessed if the intervention brought results.

This is precisely what observational studies (follow up) should do. Not really try to find hypotheses - these need more direct evidence.

[ Parent | on HN ]

Re:How salad became a major source of food poisoning in the US by toasterlovin on Friday May 04, 2018 @03:39PM (#16997560)

Yeah, but if there is an intervention, then it is not purely an observational study. The big problem with nutrition science is that most of the long term studies are purely observational. There is no intervention at all. All they are measuring is how certain observed measurements correlate (or not) with other observed measurements. These are the types of studies I'm talking about.

[ Parent | on HN ]