Monday, August 15, 2011

Are the Rich Really Rude? What Science Actually Says

In a previous post I wrote about how cool it can be for one's research to be picked up by the national media. Last week I was lucky enough to have this happen here, here, and here. The article described in the news is based on a program of research that Professor Dacher Keltner and I started about 7-8 years ago. The goal of our research was to study social class-- that is, the collection of social and material conditions of life that rank us in American society relative to others. More specifically, we wanted to understand how a person's position in society-- based on their income, their educational attainment, or their occupation status-- could shape their behavior in social interactions, their perceptions of others, and their emotions. According to the media, our research concludes that the rich are rude/selfish/and lack empathy. But is this really the case? Let's summarize the evidence!

We started this research examining just university students. We asked these students to estimate their social class rank relative to others who were at the university with them by ranking themselves on a ladder representing social class at the university. We then asked these students to explain increasing economic inequality  in society--giving participants different options as potential explanations. These options included that inequality was caused by contextual factors like "political influences," or "differences in educational opportunities," for example. We also gave participants options to explain inequality in terms of individual factors like "poor money management," "hard work," or "ability and talent." What we found in that first study was a striking preference--among students at the top of the social class hierarchy-- to explain inequality more in terms of individual effort and talent, and less in terms of contextual factors like political influence.

Upper-class students blamed this on hard work and talent
After this initial work, we moved to behavior in the laboratory. We recruited university students again, but this time we recruited these students based on the education and income levels of their parents (30% of participants were first time college students). We then had these participants engage in a getting-to-know-you social interaction with another student who was a stranger. Interestingly, we found that the students who were from more educated and wealthier backgrounds tended to be more disengaged in the interaction. That is, they displayed less head nodding, less eye contact, more checking of cell phones, and more doodling on a paper relative to their less wealthy and educated counterparts.

From there, we moved to studies examining prosocial behavior and empathy. First we used a series of studies to examine prosocial patterns. We found among parallel samples of university students and adults that people with higher income, or higher perceived social class tended to be less helpful to an experiment partner who was suffering, less likely to think that people should give a higher proportion of their salary to charity, and less likely to give 10 points (later converted to money) to an anonymous experiment partner. With regard to empathy, we found that adults and students with higher educational attainment and higher perceived social class rank tended to perform worse at reading others' emotions both in standard emotion recognition tasks using static facial images and in dynamic social interactions where they were asked to read others' emotions.
College educated participants worse at reading emotions

Over nearly a decade of work we have evidence suggesting that people from relatively upper-class backgrounds tend to believe economic inequality is caused more by talent and effort than by external factors, to be more disengaged in social interactions, to be less prosocial in their behavior, and to be less able to accurately read others' emotions. When we look across all these studies, and if I think back to all the work in the laboratory we've done over the years, this feels pretty unequivocal: The rich really do seem to be ruder/more selfish/less empathetic than the poor, and science says so.

However, I wouldn't be a "scientist" if I wasn't at least a little skeptical about such a grand conclusion. After all, though 12 studies does represent the bulk of my early career, in comparison to the wealth of knowledge that psychology research has accumulated over it's century-and-a-half existence our understanding of social class has yet to even scratch the surface. Sure, we have some initially promising research, but the reality is that social class has a much more nuanced influence on the social realm-- one that I plan to understand more fully over the course of my academic career. So, how can we move forward from this initial work?

First, it is important to understand that being self-interested in one's behavior isn't necessarily bad. In the case of academic work, it's great. If I am selfish in school, that means I'm going to focus on my studies, get better grades, and achieve more. The same is true at work. In addition, being other-oriented is not always good. For instance, if I am always perceiving your emotions, and you are experiencing anger or shame chronically in your daily life, I am likely to accurately read those emotions, and in some cases re-experience them. This will lead to more negative emotion in my everyday life, and that's bad for my health. Understanding that this research is not about being bad or good will help move the work forward to new research questions.

Second, upper-class individuals seem to be flexibly self-interested. In one study, when we asked participants to view a video aimed at inducing compassion (the video showed people suffering or in need), people with elevated income tended to be just as prosocial as lower-income participants. The results of this work suggest that upper-class individuals are not more selfish and less altruistic, just that they are in certain contexts. In other contexts that call for compassion, upper-class individuals helping behavior rivals that of the average working class person.

Third, the debate in this research focuses on the rich vs. the poor, but the reality is that most of our research is conducted on people in the middle of the social class range-- we rarely/never have corporate CEOs or homeless people taking part in our studies. To really understand how social class influences behavior requires a broadening out of our research to other communities of very wealthy or very poor individuals. Such broadening will increase our confidence in these initial results.

How else can our understanding of social class be improved? I'd love to read your comments! 

Kraus, M. W., Piff, P. K., & Keltner, D. (2011). Social Class as Culture: The Convergence of Resources and Rank in the Social Realm Current Directions in Psychological Science, 20 (4), 246-250 : 10.1177/0963721411414654


  1. What's really being tested, though, is rearing stratum - what class the subjects were brought up in. An older group, where the current stratum might have been moved into, is what I think folks imagine when hearing about a study like this. That seems an important distinction, maybe not for the research, but for how the research is viewed.

  2. Thanks for the comment Michael! In our research we've basically found consistent patterns whether we are measuring social class in terms of family origin or in terms of an adult's current status. Other work suggests that family social class origin is a stronger influence, but that is a preliminary conclusion. Thanks!