Hacker News new | past | comments | ask | show | jobs | submit login

The answer is that criminology is affected by a wide variety of factors, and making statements like the guy in the article that 90% of crime can be attributed to one factor is nonsense.



> making statements like the guy in the article that 90% of crime can be attributed to one factor is nonsense.

And yet, the r^2s when you plug in your data for lead exposure are really high. Odd how that works.


I've mentioned elsewhere a plausible third factor that drives both, and that the data for Australia does not match this pattern. And that I'd like to see this European data that apparently matches it - the linked papers in the article do not mention it.

It's far from the ironclad causative link that you're implying.


> making statements like the guy in the article that 90% of crime can be attributed to one factor is nonsense.

> I've mentioned elsewhere a plausible third factor that drives both

You see the problem here? You claim that a factor being able to drive that much is nonsense, and then you... claim to have identified another factor which drives that much (of both lead and crime).


No, that's you putting words in my mouth. I mean, your first reply to my commentary was specifically against me saying that there are multiple factors in criminology.

Yes, you can argue that because I said 'a factor that drives both' that I meant only one single ironclad factor, but in the context of all that I've written, it's clear that I'm talking about a range of changes - including, again, my comment that elicited your first response, where I explicitly point out multiple factors affect criminology. My first comment in this thread (the heavily downvoted one up there) pointed out two other factors that have been attributed to the decrease in crime as well.

I've also presented graphs of Australian violent crime that don't appear to follow trends in each other, let alone the supposed 'lead-free' crime reduction that this theory presents.

Exactly how many different factors do I have to present to potentially explain variability? To suggest that it's more than one factor that has a significant effect on crime?

Edit: I should probably also point out that the "a" factor I mentioned isn't a single thing that can be measured with a single number like "atmospheric lead", but a suite of many different things dealing with a wide-ranging cultural change (see response to DanBC below)


> I mean, your first reply to my commentary was specifically against me saying that there are multiple factors in criminology.

No, my first reply was that there very clearly was a high fit to the data, and so something is going on that is in need of explanation, be it lead or another confounding factor, and so it was nonsense to claim that it was nonsense that a single factor could matter.

> I've also presented graphs of Australian violent crime that don't appear to follow trends in each other, let alone the supposed 'lead-free' crime reduction that this theory presents.

I don't see any links, so you haven't presented anything. Here, have some graphs of preschool lead vs lagged crime in Australia: http://mpra.ub.uni-muenchen.de/35338/1/MPRA_paper_35338.pdf

> Exactly how many different factors do I have to present to potentially explain variability? To suggest that it's more than one factor that has a significant effect on crime?

You should present something. Otherwise you're engaged in know-nothingism: 'it could be something else! who knows? whoooo knooowwwwssssss...'




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: