Rita Karlsen at Human Rights Service responds to accusations about bad statistics. Read on.

My post in Dagbladet's blog on the report on immigrant marriage trends from Human Rights Service got some replies about methodological errors in their earlier work, which supposedly is the reason why the media ignores their research. Here's a reply from Rita Karlsen, director of HRS:

For our 2003 book Human Visas we bought the same kind of data from Statistics Norway (SSB) [as for this report], (the data must be bought, they are not publicly available.) These are detailed data about the marriage patterns of various groups in a Norway with an immigrant background. But what's interesting is the reason we looked into these data: It has to do with the way SSB presents their statistics. Demographer Lars Østby is the person at SSB who has spoken most about immigration statistics, or, to be precise, about immigration politics. Østby's claim is that integration improves over time, and that immigrants become more and more like us in all variables, including marriage patterns (which Østby has confirmed to me personally). In other words: The minority population is becoming more and more like the majority population.

This is a claim we believe there are many reasons to investigate, not least because of information and research done in other countries. So we used mariage patterns as a starting point, and tried to verify Østby's claim. Our findings, (presented in Human Visas, 2003), did not in our view support Østby's hypothesis, because most people by far marry among their own, and the majority imported their spouse from their country of origin. And the trend was increasing. Ie.: More and more first generation immigrants fetched their spouse from the country of origin in the period 1996 - 2002, (which we then had data for), and the percentage is even larger in the second generation (descendants) than in the first.

SSB apparently didn't like this. They went public and said that we had taken our conclusions too far. Their main point was this: There are many who haven't married yet, and when they do many of them will marry here in Norway! In other words: A new hypothesis is put forward, and it builds on the premise that the majority is becoming more and more like the majority. Perhaps they do in many ways, but when we look at marriage it is our claim that SSB is wrong. (And they've never disproven our claim with facts.)

In other words: this isn't about methodological errors, (as our "opponents" like to refer to it), but about two different hypothesises. Which one is right remains to be seen.

And that's why we looked into this again: We bought new data, extended them to include more countries, and now have data for an eight year period, (not much in a statistical perspective, but we believe it can serve as an indicator). The trend we pointed to earlier is reinforced by these new figures.

Unfortunately, representatives of SSB appear more like politicians than statisticians.

(And to be clear: I'm a former researcher, and have also been a methodological adviser at Riksrevisjonen. I wish a methodological debate welcome.)

The only issue I can think about that would introduce bias is that counting those who married young in itself selects those who are not integrated, so it's hardly a surprise the vast majority of them would marry within the foreign nationality. Norwegian women tend to marry between 25 and 30, if I'm not mistaken. So those immigrants who are better integrated, will not at all marry young, neither a Norwegian nor another immigrant, so this statistic will not catch them. How large this group is, we don't know.

If I've read this wrong, I welcome corrections.

Bjørn: I suspect the media ignores their research because they don't like the results, not because there is anything wrong with it.

A couple of years ago HRS was accused of putting undue pressure on, or manipulating, some of the girls that have served as their witnesses or sources of information in certain cases. To my knowledge, nothing much has come of this.

Also, it seems that Rita Karlsen of HRS may have been proven wrong by Marius Emberland in a certain legal issue. So what?

The issue at hand is however not whether HRS has made mistakes or been proven wrong here and there, but whether there is reason to believe that the latest report from HRS presents a biased interpretation of the statistical numbers that were purchased by HRS from SSB. I have yet to see anything that even resembles a convincing argument that this is the case.

Fjordman probably hits the nail on the head when he says the following: “I suspect the media ignores their research because they don't like the results, not because there is anything wrong with it.”

I find it quite disturbing that large portions of our media behave like this. These are questions of utmost importance, and a thorough rethinking of the entire area of immigration and integration is overdue. The issues at stake include human rights and democracy.

Trackback URL: http://bearstrong.net/cgi-bin/mt/mt-tb.cgi/1587