#### Comment Preferences

• ##### However, when one via multiple tests(26+ / 0-)

repeatedly finds correlations that point to a common causal factor, it increases the probability of that factor being the cause. I think it is pretty clear that the correlations depicted herein do have a commonality with regard to the implementation of conservative ideology.

• ##### BINGO(8+ / 0-)

If it was just one case or one issue then you could argue that correlation is not necessarily causation.

But when you have SEVERAL cases or issues then there is a DIRECT link between the two.

This is your world These are your people You can live for yourself today Or help build tomorrow for everyone -8.75, -8.00

[ Parent ]

• ##### as a matter of logic, not really(1+ / 0-)
Recommended by:
Geoffrey Williamson

If all these bad outcomes have to do with poverty, then adding lots of examples doesn't necessarily make the inference any stronger.

• ##### Only if the inference is deductive(1+ / 0-)
Recommended by:
neroden

And it isn't.  It's ampliative.  So adding or removing information does change the strength of the inference.

The world does not need billionaires.

[ Parent ]

• ##### but what information is being added?(0+ / 0-)

If the correlation might be spurious, what good does it do to point to other correlations that might be spurious for the same reason?

I'm not saying that the correlation is spurious, only that piling on more dependent variables isn't a good approach.

• ##### Spurious correlation?(1+ / 0-)
Recommended by:
happymisanthropy

Not sure what you're driving at here. When outcomes A and B are correlated, either A causes B, B causes A, or third factor C causes both A and B. By adding more examples, we establish a stronger correlation, meaning that one of those three possibilities, whichever is causing the correlation, is more strongly established.

The problem is that no evidence is explicitly explaining why we choose to explain the correlation as being A causes B instead of B causes A or C causes A and B.

Best response in IMHO was the comment about cutting healthcare, etc. because it directly addresses this question.

• ##### let me try to be clearer(1+ / 0-)
Recommended by:
Geoffrey Williamson

What we have here, very crudely, is (1) the diarist arguing that Bad Outcomes are correlated with Conservatism and therefore Conservatism causes Bad Outcomes; (2) some commenters retorting that Poverty is correlated with and causes Bad Outcomes, and that while Poverty is correlated with Conservatism*, it is far from obvious or proven that Conservatism causes Poverty.

*At the state level, anecdotally. The diary doesn't actually provide a measure of conservatism, so we don't exactly know what we're talking about. That adds to the fun. :)

I'm saying that giving more examples of correlations between Conservatism and Bad Outcomes doesn't in any way address the retort. It may "establish a stronger correlation" (sort of like Robert Putnam tried to do between social capital and Good Outcomes in Bowling Alone), but that doesn't really help much.

I agree that it's not hard to think of ways in which conservatism could contribute to bad outcomes.

Subscribe or Donate to support Daily Kos.

• Recommended (160)
• Community (66)
• Baltimore (48)
• Bernie Sanders (37)
• Civil Rights (36)
• Culture (31)
• Law (25)
• Elections (24)
• Freddie Gray (23)
• Hillary Clinton (21)
• Economy (21)
• Education (21)
• Rescued (20)
• Racism (20)
• Texas (20)
• Labor (19)
• 2016 (18)
• Environment (18)
• Media (17)
• Politics (17)