ProjectImplicit

Writing for theconversation.com, Tom Stafford produced a similar map of European countries

These are interesting maps to look at, but there are few things to keep in mind. First, the blue-red scale is done to illustrate relative differences across states or countries. While there is some variation across areas, it’s worth noting that even the bluest states showed robust pro-White biases in implicit attitudes, so these maps illustrate which states or countries have reltively higher or lower pro-White implicit bias, not which do or do not have bias at all.

Second, the data underlying these graphs do not come from a “representative sample” of people from each area. That Project Implicit participants are volunteers who are perhaps already interested in the notion of implicit bias and almost certainly skews the sample in some ways (for example, PI participants tend to be more politically liberal and educated than other people living in the same area). Results might look different if these were truly random samples of each area. Moreover, this “sampling bias” may be more or less strong in certain states or countries. For instance, perhaps states or countries where Project Implicit has received more media coverage have a more diverse sample than states or countries where Project Implicit is less well-known.

Finally, these figures represent crude averages of very large areas. There is sure to be meaningful variation within any state or country as well (fortunately, the available datasets also list metropolitan areas for US participants, allowing for a more fine-grained map to be made in the future).

Even with those caveats, these maps lead to some interesting questions about what factors may be responsible for geographical variations in implicit attitudes. Our next post will dive into one possible influence on differences in area’s average levels of implicit racial attitudes.