Science can be WEIRD (Western, Educated, Industrialized, Rich, and Democratic), but researchers are working to change that.
Four years ago three researchers in the Department of Psychology at the University of British Columbia published an article in the journal Behavioral and Brain Sciences entitled “The weirdest people in the world?” The authors, Henrich, Heine, and Norenzayan, reviewed research in behavioral sciences and found that 96% of research subjects were from what they dubbed WEIRD societies—that is, Western, Educated, Industrialized, Rich, and Democratic societies. As well, many of these participants came from a common pool of subjects: undergraduates at research universities. The article suggests some interesting implications when the subject pool is so limited, and there is further discussion about some of these implications by Greg Downey at Neuroantropology.
How might one go about making science less WEIRD? The Making Science Less WEIRD project is one initiative that hopes to find ways to include a more diverse group of people in behavioral research. One way they’ve identified is to make use of the web to connect with populations that researchers might not otherwise have access to in traditional studies.
A diverse participant pool is important, explains Joshua Hartshorne, one of the principal investigators of the Making Science Less WEIRD project, saying, “If we test a wide range of people of different backgrounds, cultures, languages, etc., on some task and they all behave more or less the same, then that’s pretty good evidence that people simply don’t differ along that dimension.” And what if researchers do suspect a difference? “Then we are in a good position to understand those differences,” says Hartshorne. Traditionally this kind of research has been difficult because investigators did not have access to such a large and diverse group of participants as the web affords. Of course, there are also cases where a homogenous population might very well be exactly the population one wants to study, Hartshorne reminds us. But what the web affords, and what the Making Science Less WEIRD project aims to capitalize on, is the ability to undertake those studies of diverse participants that have previously been so difficult.
While there are some 2.4 billion people online, the project investigators are very much aware of the current limitations of that sample. Obtaining a good sample is more complicated than access to more people. “We don’t need a representative sample in order to probe human diversity,” Hartshorne writes, but rather “we need a diverse sample.” The web affords access to more diverse participants than those undergraduate students who are so frequently studied.
However, there are some limitations. Projects currently featured by Making Science Less WEIRD are designed for English speakers. As the initiative grows, and more projects become involved, the principal investigators hope to make the site multilingual. Katharina Reinecke, another principal investigator at Making Science Less WEIRD and a co-lead at Lab in the Wild, said her team is “currently translating Lab in the Wild into five different languages” and hope to include more languages as the project grows. This kind of work is particularly important, Reinecke notes, because there is evidence “showing that people test differently in their native language as opposed to when they are being asked questions in a non-native language.” Even so, Hartshorne reminds us, there is still significant diversity in English speaking populations that has typically not been included in behavioral research studies. An interesting example of this is the vocabulary quiz Hartshorne has been running, which aims to understand differences in the size of vocabularies across English speaking nations. Visit Games with Words to participate in this study and others that help us understand language. As well, Lab in the Wild is working on creating “more usable and intuitive user interfaces for people around the world,” according to Reinecke, which they hope will eventually work to draw in populations that have been less likely to participate. The Visual Preferences Test is one way this kind of work is happening and one way to get involved.
All of the work being put into Making Science Less WEIRD helps develop better approaches to behavioral research. As well, the lessons for citizen science more broadly are many, including the careful consideration in developing a participant pool and considering the linguistic and technical barriers to access.