In psychology, there often is a common demographic among research subjects. And among the researchers, themselves. And, in its own way, among research questions, processes and interpretations.
A few years ago, a University of British Columbia research team noticed this trend and came up with an acronym for this demographic: WEIRD, or Western, educated, industrialized, rich and democratic. WEIRD people, WEIRD research, WEIRD-generalized results.
The trouble is, three University of Washington researchers say in a new article, that the field of psychology tends to overlook, or even leave out, people, cultures and issues that could be labeled "non-WEIRD": people of color, of different socioeconomic classes, levels of education and cultural traditions. And that has implications for how the science is interpreted and applied, and how it actually reflects society.
While ideas of diversity aren’t new to the field, the actual process of diversifying has been slow in coming. It’s time, the UW authors say, to change.
"We’re talking about an entire field that has many people and institutions, and how you change longstanding practices and functions," said Laura Brady , a research scientist in the UW Department of Psychology. "We’re saying, take a step back and ask, what is the culture cycle that drives the practices in our field? Let’s look at what our institutions can do, practically speaking, to get more people paying attention to culture, and build a knowledge base with a working understanding of how culture shapes people’s behaviors and thoughts."
Brady wrote the article , along with UW psychology professors Stephanie Fryberg and Yuichi Shoda. The piece was published online Nov. 6 in the Proceedings of the National Academy of Sciences.
The paper represents a call to action, the authors say, to enhance the "interpretive power" of psychology - "the ability to understand individuals’ experiences and behaviors in relation to their cultural contexts." They point to various ways that existing practices overlook the importance of culture, and to how individuals and institutions can be more inclusive.
One frequent practice, the authors say, is generalizing research results, which starts with a lack of diversity among subjects in a given study. College campuses - the populations of which are largely white and middle class - make for a ready pool of participants, and then when results are reported, the language is empirical, as if all people think, behave or act in a way that was determined from studying a much smaller sample.
Demographics are not always reported in the description of a study’s methods, Shoda said, and there is generally little attention to how the results might reflect populations that weren’t represented.
"There has been a tremendous challenge in how you go from findings involving specific individuals to building a science of human beings," he said.
The issue of generalizing research findings can also apply to media coverage of those findings, Shoda said. Take a recent study of delayed gratification among a group of children, a revisiting of questions and methods from what is known as the "marshmallow test" from decades ago. Shoda, who was involved in both studies, points out that the new study clearly identified the limits of the study pool. But one headline implied that "kids today" are better at holding out for a treat than kids from a generation ago. Too general, Shoda said of the headline, despite the fact that he and his co-authors had explicitly cautioned against generalizing from the findings. More research, the authors noted, would need to be conducted among diverse populations.
The PNAS paper also explains how research tends to leave out underrepresented, or non-WEIRD populations, or ignore the cultural factors that could affect results. For example, Native Americans, who, according to the 2010 U.S. Census, number 5.2 million nationwide, were the subject of 0.2 percent of papers about stereotypes, prejudice and relationships, the authors say. And in less than 0.5 percent were Native Americans mentioned at all. Fryberg reported those findings in an earlier paper.
"A lot of times, Natives are left out of a sample, and instead of oversampling, we just leave out the group. Natives experience all kind of prejudice in this country, and yet people don’t think we exist," she said. "We have a lot of responsibility as scientists to make sure science is an accurate reflection of society, and for the most part, we don’t live up to that obligation."
The authors suggest several solutions: developing "culture conscious" research questions and placing more value on ethnographic observations; revising the language that’s used to report research findings and adding details about study samples; and among journals, funders and institutions, supporting and highlighting research that focuses on non-WEIRD populations.
"As a science, we’ve gone forward without engaging with ideas of inclusion, so scientists and those who read our papers too often assume that if the paper is peer-reviewed and published then it reflects basic, fundamental ways of being," Fryberg said. "We’re trying to lead people to a more accurate understanding of human functioning-one that reflects other good and legitimate ways of being, calling into question the universal. This can be threatening to people in powerful positions. But science depends on our ability to be relevant to the global world, and we’re trying to call out the science to meet that challenge."