Prometheus

We don't need to restrict attention to whites or jews, just define the variable of interest which seems to be ethnicity. We don't need to restrict attention to whites or jews, just define the variable of interest which seems to be ethnicity.

Yes, i grew up with Irish Republican terrorism (IRA). But Itoero informs me he doesn't like anecdotes to support an opinion on this thread (fair enough), unless he is giving them of course. So here's the data. What do you reckon with the IRA - should that be classed as religiously inspired violence or as political separatist violence?

Isn't this the exact same data as in the white privilege in academia? If so wouldn't it make sense to merge the two threads and consider ethnic (or whatever variable you want to explore) privilege in academia?

Then we are agreed: if there is a relationship it more subtle than A causes B. It's more like A increases B, but only when a certain quantity of C is present, unless a certain level of D is also present... and so on. Are we also agreed on this? Also did you want to retract your assertion that most terrorists are Muslim in light of the data dimreepr and i provided?

As you might say: most Muslim terrorists are Muslim . So if Islam isn't a sufficient nor necessary condition for terrorism then if Islam does contribute to terrorism the relationship isn't a simple A causes B type relationship. Agree or not? 90% of US terrorists aren't Muslim. This data from Europe shows a similar trend. And this data.

Great, so you'd be happy to say that being Muslim is not a sufficient condition for being a terrorist (as not all Muslims are terrorists)? Nor is it a necessary condition (as not all terrorists are Muslim). Yes?

We want to know whether Islam causes terrorism. We have instances where Islam and terrorism occur together. We also have instances where there is Islam and no terrorism, and instances where we have terrorism but no Islam. Before i proceed further can i just check we are agreed on this?

I'm just amazed at how black and white some people think this all is, like the world can be split neatly between goodies and baddies. Maybe it's the Star Wars generation, there's the light and the dark side and nothing in between. A hefty dose Star Trek should help. In reply to my question 'why is A increasing', you have responded 'because A is increasing.' Can you really not see this? Do you really think that is a legitimate mode of reasoning (i'm presuming you value reason and evidence)? Is it a language thing - is my terminology or posting style ambiguous? There are many things wrong with this statement, but i want to focus on the cherry picking aspect since the theme is already well established. Let's consider China - easily the equal of any European power throughout much of world history, superior to any civilisation for a fair bit of it - until the cultural revolution and the attempted to abolish religion. As soon as religion goes, BOOM! Progress stops.

To get onto any PhD. a student would have already done a few years study. In that time the student would have been exposed to the consensus understanding of the subject: in physics the Standard Model is first taught, not one of the alternative models. So why would climate science be any different? Has science not been making progress over the last few centuries such that we need to abandon its methods for something new?

You keep saying you understand it but then demonstrate you clearly don't. Yes but the confounding variable might be having an effect on two variables which are related. The doctrines of Islam might be having an effect on levels of terrorism - i'm not saying they are or aren't yet. But we need to consider whether there are other factors that are exacerbate or ameliorate this effect this. We have lived with Islam for over a thousand years and in that time it has been no worse than Christianity. There have been significant Muslim populations in Europe for decades, why has terrorism become a problem all of a sudden - the Islamic doctrines certainly haven't changed in that time. To explore this properly we need to first determine what variable we think might cause terrorism in general (unless you think even drug lord terrorism in South America is somehow Islamic too). Poverty, displaced populations, education levels, unemployment, etc... The we collect a data set of all these variables. Then we choose a statistical technique beforehand and analyse them ALL IN ONE GO. No cherry picking, and you are cherry picking whether you realise it or not, which variables best support our biases. Look at what the TheBeardedDude did in post 224. That's the best attempt anyone has made on this thread to actually provide evidence, even though PCA is a poor method in this context, especially to look for confounders, and the data itself is of questionable quality. I imagine the reason why he hasn't explored it further, and certainly the reason i haven't yet, is that even quick and rough attempts like this take time. I'm planning on looking at the data soon, why don't you take the time and join us? We can begin by thinking about what variables may contribute to terrorism in general.

John Milton in Paradise Lost:

What does that even mean? And what do you imagine i mean when i say forget about religion for a second to consider the general case?

If true they are following some of the principles of the scientific method, so all is well. However, i fear they are simply stopping following one person blindly and starting to blindly follow another: we are pack animals after all. Regarding scientific fraud, things like p-hacking are well known problems in the scientific community and measures are taken to mitigate against them. It is a mistake to imagine this equivocates the findings of scientists with the opinions of someone who has a 'theory of everything (but i need someone to do the math for me)'. Sometimes feel like everyone who rails so hard against science should forfeit all the products of science and go back to live in the pre-industrial age.

Apparently not.

I said forget about religion. Your misunderstanding is far too fundamental to begin with a real example, you just keep repeating the same mistakes over and over again. Lets's just deal with Xs, Ys and Zs so no one can accuse any one else of biases. Yes?

This is a new problem for the generations brought up with the internet. People use youtube for all sorts of things in life, and we can't reasonably expect people not to do so when they have a question regarding a scientific subject. With the extra information out there it is more important to teach from a young age how to sift through the rubbish to find the nuggets that are undoubtedly out there - i.e less teaching of facts and more teaching what a fact looks like and how to identify good sources. This isn't off topic because this is post 22 and the Universities Act was passed in 1922. Coincidence?

OK, so you object to the scientific method itself, at least i know where you stand. Forget about religion for a minute, there is a far more fundamental point about confounding variables you are failing to grasp. You have basically just said 'yeah confounding variables might exist, but even if they do they will make no difference to my statistical inference'. Which means you don't know, at a very basic level, what a confounding variable actually is. Worse, you seem to making absolutely no attempt to learn. Even worse you are quite happy to nurture a hatred for a subset of the human population on the back of it.

The point is a far more general one about the scientific method. The point with the margarine and divorce rates show that it is possible that variables are correlated without them being causative. And it's not a small problem. There was a time ecstasy was thought to cause HIV because they were strongly correlated. Turns out the sample they were looking at (gay men in some American city) both took ecstasy and engaged in risky sex. Also a lesson in trying to generalise to a population from an unrepresentative sample. Now all the points you have raised regarding subsets of Muslim populations is certainly a concern and warrants further attention. But it is the start of a scientific investigation, far from the end of it. That means thinking hard about the variables of interest to measure, collecting the data and performing suitable statistical analysis. What part of this process do you object to?

All those divorcees have one thing in common: margarine consumption. Margarine obviously makes divorce rates increase. Yes? Good question. You might start with confounding variables. Better, i think, to learn this concept in the general before looking at specific cases to help avoid conformation bias. There are plenty of other sources of bias but that should be a good start. Try this. All X that cause crime to go up have one thing in common: they are all X. This is the very definition of a tautology, to which 'no shit Sherlock' is an appropriate response.

By your reasoning margarine consumption increases divorce rates in the US. It matters because you are learning to lie to yourself with statistics.

I just meant generalising to any p, but yeah looks like you've got the hang of it generalising to any N too.

The easy and dirty way to generalise this would be to consider [latex]\frac{p}{p+1}[/latex], where p is the probability of getting a B.

As long as the independence assumption still holds. Maybe some people aren't happy if picked together.

If i understand the first problem we can see that we have a 0.5 probability of drawing a W and a 0.5 probability of drawing a B. But as we know a B cannot be followed by a B then this latter is saying there is a 0.5 probability of drawing a BW. So for every B there is also an additional W plus the other (random) Ws in the series. Intuitively we can see that this converges to 1/3 in the long run (there should be twice the number of Ws as Bs). To do it more thoroughly we have to be careful what we mean about the convergence of a series of random variables, there are several types. I assumed you meant almost surely?

Try this one. Been trying to figure out a succinct way to give my take on the theme, but the song covers it much better than i can. What it means in the particular context you're after i'm not sure.