The current #Immigration debate is fascinating.
Proponents of it usually argue that modern Western states should be defined through (liberal) values. But detractors point out that most immigrants, particularly Muslims, aren't really liberal.
So I decided to have a closer look.
Proponents of it usually argue that modern Western states should be defined through (liberal) values. But detractors point out that most immigrants, particularly Muslims, aren't really liberal.
So I decided to have a closer look.
I started with Women's rights since this topic is hotly debated.
To that end, I took the Global Gender Gap index from the (2018) Global Gender Gap report and used the percentage of #Muslims per country as a predictor. (Note the confusing fact that a high index means a small gap)
To that end, I took the Global Gender Gap index from the (2018) Global Gender Gap report and used the percentage of #Muslims per country as a predictor. (Note the confusing fact that a high index means a small gap)
The regression revealed a strong relationship with R2=0.395. This means that almost 40% of variation in the country-level gender gap is explained by the % of Muslims in a given country.
That's way higher than I expected.
That's way higher than I expected.
Next I looked at #antisemitism, since it's also a hotly debated topic, particularly in my home country Germany.
In addition, Antisemitism is a half-way decent proxy for out-group hostility in general (e.g., #racism).
To that end, I took the 2014 ADL data.
In addition, Antisemitism is a half-way decent proxy for out-group hostility in general (e.g., #racism).
To that end, I took the 2014 ADL data.
The linear regression revealed an even stronger correlation of R2=0.569. This means that almost 57% of the variation in country-level rates of Antisemitism are explained by the percentage of Muslims in that country.
That's extremely high, to say the least.
That's extremely high, to say the least.
Finally, I looked a #homophobic opinions, since it's a good proxy for hostility towards misfits and outcasts in one's own in-group.
I couldn't get decent polling data, so I took the Wikipedia data on country-level support for #LGBT
rights at the United Nations.
I couldn't get decent polling data, so I took the Wikipedia data on country-level support for #LGBT
rights at the United Nations.
Since these data are binary, I did a logistical regression. There is no R2 for this type of regression but I computed McFadden's pseudo-R2. This came out as R2=0.637.
Take this with a grain of salt, but again the relationship is striking.
Take this with a grain of salt, but again the relationship is striking.
These results provide surprisingly strong support for the notion that liberal support for (Muslim) immigration is an conspicuous opinion, or luxury belief.
This means, it's main aim is not to reflect reality but to provide a (costly) signal of support for one's own peer group.
This means, it's main aim is not to reflect reality but to provide a (costly) signal of support for one's own peer group.
A lot of things could be improved like better, more and more varied data; different models; different predictors; etc.
That's it so far. Good-faith feedback and criticism is highly welcomed.
That's it so far. Good-faith feedback and criticism is highly welcomed.
