A new working paper with Clem Aeppli on SocArXiv. We look at different measures to capture agreement, consensus, polarization, whatever you want to call it — in ordinal data. Using simulations and an empirical example, we show commonalities and differences between measures. The paper ends with recommendations for researchers wanting to measure consensus, agreement — whatever — in ordinal data.
When asking questions about gender identity in surveys in Switzerland, I often faced the problem that a tiny fraction of respondents did not answer the question seriously. Normally, we can live with this, but it’s a real hindrance when trying to capture relatively small sections of the population.
Here’s a typical case from Switzerland in 2015:
We offered “female”, “male”, and “other” as response categories with the option to specify which “other” identity applies. If we go by estimates elsewhere, we should expect between 0.1% and 2% of the respondents picking “other”. At first sight, we seem to be at the lower end, but there’s likely serious under-reporting because more than half of these “other” responses are not referring to other gender identities. We get responses like “cat”, or “there are only two genders” — definitely not on the useful side of open questions (beyond noting that some people are probably frustrated about the fact that we do talk about non-binary identities, I guess).
Offering more choices for gender identity seems to discourage nonsense and protest answers, leaving us with a better measure of non-binary gender identity
I’ve had this in several surveys, but recently we tried something else: we offered more choice! Yes, rather than “female”, “male”, and “other” we spelled out a few of the “other” category: “female”, “male”, “non-binary”, “transgender female”, “transgender male”, “other”. From a conventional survey design point of view, this was bordering the ridiculous because we only expected some 500 respondents in this survey, which would yield between 1 and 10 respondents in those categories combined (going by existing estimates). We’re still at the lower end of this range, but we had none of these nonsense and protest answers.
Given that we’ve run an almost identical survey just months earlier with the three category format (“female”, “male”, “other”) and had more than half of the “other” answers that did not refer to gender identity, we might be onto a solution…
For a long time, racism has been studied without references to discrimination and was mainly conceived as a specific expression of prejudice. The turn to more subtle and systemic forms of racism has paved the way for studies on ethnic and racial discrimination and inequalities. Research on discrimination against immigrants and their descendants has grown significantly in the last twenty years, paralleling the settlement of immigrant populations. Studies document differential treatment and discrimination in different markets (e.g. labour market, housing) and social spheres regulated by principles of equality (e.g. school, health service, police). Patterns of discrimination are embedded in institutional contexts and a larger societal environment, characterized not only by economic uncertainties and increasing political polarization in public debate around immigrant related issues, but also by increasing diversity and opportunities of contact. Such changes in the context are likely to affect attitudes and ideology diffusion in majority and minority members. However, studies about discrimination do not refer specifically to racism, and the methodological gains in measuring discrimination did not transfer directly to the measurement of racism. How far racism and ethnic and racial discrimination are distinct, and how they relate to each other are key issues we would like to explore in this panel.
The panel will bring together researchers on discrimination, racism, and inequalities, tackling these issues from various disciplines, theoretical backgrounds and methods. We welcome empirical studies of discrimination patterns across a large variety of domains, theoretical perspectives on how the prevalence of ethnic discrimination and racism should be explained and conceptualized, and studies on the consequences of anti-discrimination policies and legislation, including considerations inequalities in health and racial inequalities and how these can be overcome. We also welcome papers which use and discuss theories about cross-country differences, ethnic hierarchies, and evolution over time.
This post might serve as a reminder to myself and others doing research on immigrants and their descendent that nationality is not a mechanism. Put differently, if you discover that people with nationality A differ from people with nationality B in a given characteristic, you have not explained anything at all.
It feels rather obvious when put this way, but it’s usually harder when it comes to multiple regression models. So often we throw in a control variable like “foreign national” or “foreign born” without thinking why we do so, what alternative explanation we think we are capturing. Obviously, a person’s passport or place of birth is used as a shorthand or proxy of something else, but what exactly?
Let’s consider the commonly used variables of migration background or migration origin. Short of calling a particular section of society different in essence (which we probably don’t want to), there are a range of concepts we might be trying to capture, like the experience of (racial) discrimination, having a different skin colour, having a different religion, holding different values, having poor language skills, being of the working class, having additional cultural perspectives and experiences, transnational ties, or a combination of these.
Knowing what we’re after is essential for understanding. Sometimes it is necessary to use proxies like immigrant origin, but we need to specify the mechanism we’re trying to capture. Depending on the mechanism, who should be counted as of immigrant origin, for example, can be quite different, especially when it comes to children of immigrants, individuals of “mixed” background, and naturalized individuals. Having poor language skills, for example, is something most likely to affect (first generation) immigrants; but likely experience of racial discrimination is probably not disappearing just because it was my grandparents rather than me who came to this country.
In a new paper, Graham Brown and Arnim Langer introduce a general class of social distance measures. They follow the general feeling that many measure of diversity and disparity may be closely related by demonstrating how they are all related. By clarifying how these different measures are related, we should find it easier to choose an appropriate measure for the analysis at hand.
The one thing I’m still not convinced is the title of the paper: While they clearly define what they mean by social distance, my sociological training keeps interfering and social distance doesn’t seem fit to express a characteristic of a society. Perhaps it’s easier to talk of the more concrete instances of ethnic diversity, or income disparity.