/sci/ - Science & Math

>>10789231
>not understanding hyperreals

>>10752159
>humans in the 1900s wouldn't have developed nuclear energy for purposes other than war
>therefore no alien race ever would develop nuclear energy for purposes other than war

>>7535110

>>7352613
>>7350510
Your understanding of science is laughably bad dude. You don't understand how burden of proof works in science. I'll explain why you're wrong kid:

By default science has to start with a "null hypothesis" or "default positions". In its most general sense, this just means that we start with assuming that there is no statistical correlation between things we wish to study.

Our job is then to make alternative hypothesis (aka ideas about correlations) and test them. We test to find the correlations. If we do find them, we set a new null hypothesis to encompass the new findings. That is literally how all science is done.

You see, not all hypotheses face the same burden of proof. That is where you are fucking up. The Null hypothesis by definition is as justified as you can be currently be. Saying "I believe in the null hypothesis" doesn't need further fucking justification. You don't need to go around debunking alternative hypothesis to justify the null hypothesis.

The null hypothesis about the relationship between "gays" and babies", IS THAT THERE IS NO FUCKING relationship. And you don't need any data to justify that. That is how the null hypothesis works kiddo. The null hypothesis = no correlation/relations. And the alternative hypothesis would then test for correlations (which it didn't find).

You might ask, "Well why can't the null hypothesis just assume inherit correlations?" Well....THEN YOUR HYPOTHESES BECAUSE UNFALSIFIABLE (so you're not doing science anymore) and you violate the rule of parsimony (aka your over estimating correlations, aka making up shit)

>>6689609