Even before I began my studies in psychology back in 2006, I have always been intrigued by the scientific research findings in the behavioral sciences. Back then I did not pay much attention to the sample characteristics on which those findings were based. Things changed when I had to write my bachelor and master’s theses, but especially when I began my PhD in Developmental Psychology in the Netherlands. During my PhD I began to read research studies more critically, particularly the Methods section.
On the one hand, it became salient to me that behavioral science research studies are very diverse in the methodology used. I noticed this perhaps due to my interest in methodology. On the other hand, the homogeneity of the samples used became salient to me also. This growing awareness partially reflected my inherent interest in culture, but perhaps in part it also arose due to me being a native of the Caribbean and being an ethnic minority in The Netherlands. For example, in the Netherlands the vast majority of studies in the behavioral sciences are based primarily on ethnically Dutch (i.e., of western/primarily white) persons. Globally, the majority of behavioral science studies are based on samples from the United States followed by the rest of the Western world (the vast majority of those samples consist primarily of ethnic majority, i.e., western/primarily white persons, too).
This homogeneity of the samples used in the behavioral sciences has been a long-debated issue. At least one empirical study has shown that the vast majority of research studies in the behavioral sciences are based on Western, Educated, Industrialized, Rich and Democratic (WEIRD) samples. Below is a conclusion from that well-known article by Joseph Henrich, Steven Heine, and Ara Norenzayan:
“A recent analysis of the top journals in six subdisciplines of psychology from 2003 to 2007 revealed that 68% of subjects came from the United States, and a full 96% of subjects were from Western industrialized countries, specifically those in North America and Europe, as well as Australia and Israel (Arnett, 2008). The make-up of these samples appears to largely reflect the country of residence of the authors, as 73% of first authors were at American universities, and 99% were at universities in Western countries. This means that 96% of psychological samples come from countries with only 12% of the world’s population.” (p. 63).
Hence, there is empirical evidence of a cultural (“Western”) bias in behavioral science research. To be sure, researchers often study those in their subject pools (e.g., undergraduate psychology majors) and those pools are not representative of the population at large. Thus, researchers do not necessarily set out to be biased. Researchers may also believe that culture is not relevant to the basic science questions they want to study.
While it was disappointing to read the empirical findings of this cultural bias, I also felt inspired to tackle this issue for my own research. Thus, for my PhD, I collected research data from a substantial number of ethnically-Dutch minority youth (in addition to ethnically-Dutch majority youth) living in the Netherlands, and this sample was also socio-economically diverse. Additionally, my PhD project included a cross-national aspect, as it was also based on research data collected in the Caribbean.
Taking these initiatives to make my PhD project as less WEIRD as possible, was definitely not easy. But I was motivated by the belief that we as behavioral scientists need to do more and take these challenging steps in order to produce more generalizable and representative research. I think that this will ultimately help us better understand the human mind and behavior. Importantly, I do not believe this should solely be the task of cross-cultural researchers (e.g., cross-cultural psychologists). Instead, behavioral scientists in general could consider contributing to this cause, as most of us would like to establish broad generalizability of our findings or we would at least agree that comparative data across diverse populations would be of added value. For example, I’ve learned that collaborating on cross-national studies is not only a worthwhile way to contribute to this cause, but such research is also exciting. Hence, I would like to encourage other researchers to also aim to make their research less WEIRD.
Some cultural differences that were highlighted in my dissertation (affecting decision making) include differences in depressive symptoms between ethnic majority Dutch youth and ethnic minority Dutch youth (consisting primarily of Moroccan, Turkish, Surinamese, and Dutch-Caribbean youth). In particular, ethnic minority Dutch youth reported more depressive symptoms. As for cross-national differences, whereas cannabis use predicted alcohol use one year later in Dutch youth from the Netherlands, the reverse was true for Caribbean youth from St. Maarten (i.e., an island in the Dutch Caribbean). Namely, on St. Maarten, it was alcohol use that predicted cannabis use one year later. The effects of such substances on adolescent cognition have been intensely debated.
The general conclusion of my dissertation pertaining to cross-cultural differences is that overall there were more differences between countries/geographical regions (i.e., St. Maarten versus The Netherlands) compared to within-country ethnicity differences (i.e., ethnic majority Dutch youth versus ethnic minority Dutch youth).
Below follows an interview I did in 2016 that highlights (1) the main findings of my PhD dissertation, (2) the cross-cultural aspects of it (3) my experiences with making my PhD project less WEIRD as possible, and why I thought this was important. The article below based on this interview was written by Pieter Hofmann, the journalist who interviewed me a few weeks after I defended my PhD dissertation.
During her 5-year PhD, Ivy Defoe collected research data for 3 years on over 950 adolescents living in The Netherlands and on St. Maarten (Dutch-Caribbean), in order to investigate what leads adolescents to engage in risk behaviors. She encountered a large number of factors and a cultural bias in the scientific world. Her PhD dissertation is entitled: “The Puzzle of Adolescent Risk Taking: An Experimental-Longitudinal Investigation of Individual, Social and Cultural Influences”. She received her PhD (cum laude) in November 2016.
Smoking, drinking and other so-called risk behavior is typical for adolescents. “That is what society believes,” Defoe says. “Adolescents and children take equal amounts of risks during experiments. But when adolescents could avoid taking risks, they did that more than children. In the real-world, such age differences in risk-taking are confounded with the availability of a risk-taking situation (i.e., risk opportunity). Adolescents go to parties where people drink and smoke, children do not. The opportunity makes the thief.”
Family relationships
A good relationship with parents (particularly with mothers), protects children from engaging in risk behavior, such as aggression and crime. Defoe: “It is risky if your child has delinquent friends, of course you can say to your child: Don’t hang around with ‘bad’ friends, but you can also make them resistant to such influences by having a good relationship with them.”
Many family therapies for juvenile delinquents are therefore aimed at parents and children. Defoe additionally investigated the role of siblings, which is also important. “The role of siblings can be the same as that of friends, especially if the sibling is older. Therapies often focus on the parent-child relationship. But if, for example, an older sibling also engages in problem behavior, he/she can undermine the therapy. Thus, for therapy purposes, it is very important to include each member of the household. ”
Multicultural research
In the Netherlands, Defoe went through many obstacles to recruit an ethnically-diverse sample for her research. “I have recruited primarily so-called ‘black’ and ‘mixed’ schools for my research. It took three months for a school to say ‘yes’.” Defoe succeeded in her goal. Her research sample in The Netherlands (602 participants) consists of 39% of adolescents with a non-Dutch ethnic background. That is much more than the average Dutch behavioral science research study. “Every research study should reflect the ethnic makeup of its society when relevant, which means including participants from ethnic-minority groups also.” Defoe emphasizes the importance of a well-balanced reflection. “Ethnic minorities are overrepresented in juvenile detention centers in the Netherlands, for example. But most research on psychological issues such as the development of behavioral problems is based on ethnically Dutch adolescents. We need more research studies that include ethnic minorities. When that is the case, I suspect that perhaps then therapies would be better tailored to adolescents from ethnic-minority groups also. Culture eventually plays a role.”
Cultural bias
Defoe feels supported by her colleagues at Utrecht University. “I want to make the cultural bias in scientific research a discussion point and I think my Dutch colleagues appreciate that. Whenever they give presentations about their research, I sometimes point out cultural aspects or emphasize why it is important to additionally have certain ethnic-minority groups in one’s research study. My department head has also asked me to give a presentation about these cultural biases in research.”