LogFAQs > #923854571

ROBANN_88 posted...

i'm talking about this https://www.svd.se/bra-okning-av-sexbrott-gar-inte-att-koppla-till-flyktingar
(it's now under a paywall, but i found a copy https://www.reddit.com/r/svenskpolitik/comments/bu2isw/br%C3%A5_%C3%B6kning_av_sexbrott_g%C3%A5r_inte_att_koppla_till/ep60bca/ )
all it did was:
1, check how many refugees were placed in a county
2, check how manysexual crimes took place in said county
3, divide one with the other and announce proudly that there's no correlation

which has some problems:
1, they assume that everyone would always commit crimes in the city they live in, there is no wiggle room for people crossing city borders.
2. for some reason Stockholm is excluded from the survey
3. it's basically just a few steps above guessing.
4. it was written by Jerzy Sarnecki, who, for example tried to stop Uppdrag Granskning from investigating in the matter before as it would be "problematic"

This work didn't cite conclusions as much as it cited data, so I'm not sure. Like I said, that one essentially did some guesswork (and it admitted as much). The limitations you mention in 1 and 3 were brought up in the report, and Stockholm was excluded precisely because it is so easy in such a region to commit a crime in a different kommun.

There were some other tests done as well, for example calculating the "verrisk" of refugees if it were true that they were driving the increase in rapes and finding that the numbers were incredibly unrealistic. An outlier before would be 5%, but for that hypothesis to hold the verrisk of that group would have to be like 80-90%. But I'm the end, those weren't the main conclusions from the report, and the report literally said that they were guesswork.

(and frankly, UG did a really shitty job so I can somewhat understand Sarnecki's resistance, but like you say it's a tangent that might not be worth exploring)
---

Stop being so aggressively argumentative for no reason. - UnfairRepresent