“We are far more united and have far more in common with each other than things that divide us.” — Jo Cox (murdered British Member of Parliament; 2015).
In recent years, the number of reported hate crimes and support for extreme right-wing parties have increased in many countries around the globe. One of the main characteristics from supporters of extreme ideological views is their focus on differences between themselves and others (e.g., immigrants, women, Muslims). Ironically, social scientists who predominantly identify themselves as politically left-leaning or liberals may inadvertently support these beliefs by reporting mainly differences between groups of people.
In our recently published paper, we used a novel approach to test whether the attitudes, beliefs, and values from various groups of people would be more different or more similar. Specifically, we compared people from over 60 countries, women and men, people from different age groups, with a different level of formal education, income, and religious denominations. Across all comparisons, we found astonishingly large similarities between all groups, ranging on average between 80% and 96%. In a series of four additional studies, we further found that focusing on similarities between groups of people rather than on differences improves intergroup attitudes and leads to a more accurate understanding of the outcomes of psychological studies. Before we describe our findings in more detail, we first want to walk you through our approach (don’t worry, it is pretty straightforward).
What we did
To illustrate our approach, have a look at the figure below. It shows how more than 3500 people from Poland and the United Kingdom value security. Specifically, participants reported to what degree security (e.g., living in safe and secure surroundings) is a guiding principle in their life, whereas the response option “1” indicates that security is not at all important and “6” indicates that security is very important. If you look at the figure, you probably conclude that security is more important to Polish than to British people. This is also what social scientists typically conclude, some might even say that there are “highly significant differences” between British and Polish people regarding their security values.
However, there are many other ways how we can analyze and graphically display the responses of the 3500 British and Polish respondents. One example is shown below. To create this graph, we superimposed the responses from the British people over those from the Polish people. The impression now is completely different. The overlap between the two groups is now 86%. This indicates that around 86% of the responses that were given by British participants were mirrored by those of Polish participants, suggesting high similarities between both groups. We call this measure that expresses similarities “Percentages of Common Responses” (PCR).
In our paper, we did tens of thousands of such pairwise comparisons for many groups of people (e.g., British and Polish people, younger and older people, or less and highly formally educated people). We compared the groups on a range of psychological variables such as human values (e.g., security, power, freedom, equality), attitudes, and beliefs. The results were pretty clear: The amount of similarities between most groups of people was large. We found the smallest amount of similarities (i.e., the relative largest differences) for countries: The average amount of similarities across 60 countries from around the globe and 22 psychological variables was 80%. In another study, we computed the amount of similarities for 29 European countries and similar psychological variables. As expected, the amount of similarities was somewhat higher, 87%, probably because the countries are politically and economically more homogeneous. The data in both studies came from national representative samples.
The figure below shows some selected results. For example, when asked questions concerning attitude towards (domestic) violence, the amount of similarities between Buddhists and Hindus was 71% (panel A). That is, 71% of the responses given by Buddhists were mirrored by Hindus. Attitudes towards (domestic) violence were measured by questions such as whether it is justifiable for a man to beat his wife or parents to beat their children, with answers given on a scale from 1 (never justifiable) to 10 (always justifiable). The amount of similarity between the poorest 10% in a society and the richest 10% was 87% for political attitude, measured on a scale from 1 (left-wing) to 10 (right-wing; panel B).
It is important to note that this does not mean that two people are the same. It just means that if you know that someone comes from a specific country or is of a certain age, you can tell very little about this person’s attitudes, beliefs, or values. For example, if you meet a person with a very low and a very high income, there is only a 59% chance that the poor person is more left-wing than the rich person. In contrast, there is a 41% chance that the poor person is more left-wing. Or, to put it differently, people within countries or age groups can be quite different from each other (see also this great article by Ronald Fischer and Shalom Schwartz).
So why is it important that we focus more on similarities between groups of people? We believe that there are at least three reasons. First, highlighting similarities between groups of people can improve intergroup attitudes. In a few studies, we showed participants graphs that either highlighted differences between groups (see the first graph) or similarities (see the second and third graph). For example, when British participants saw graphs that highlighted the similarities (vs differences) between British and Polish people, they evaluated Polish people on average more positively and judged them to be closer to British people.
Second, participants who saw graphs that highlighted similarities were better able to estimate the actual amount of similarities/differences between the two groups compared to participants who saw graphs that emphasized differences. Participants who saw graphs that emphasized differences – recall that this is the default way to present findings in social sciences – they drastically overestimated the actual differences. For example, they believed that the similarities in human values between British and Polish were only 28%, when they are in reality 88% (averaged across all values, see graph below). This suggests that the default way how findings are presented distort the understanding of scientific results.
Third, reporting both similarities as well as mean differences results in a more balanced portrayal of research findings. Note that we do believe that analyzing and interpreting mean differences has its merits. However, based on our findings, we recommend that researchers report both mean differences and similarities. If the focus of the research is on polarized groups, it might be better to focus more on similarities.
We found across a range of psychological variables that people from different countries, with different incomes, gender, formal education levels, age, and religious denominations are typically far more similar than different in a range of psychological variables such as attitudes, beliefs, and values. This does not mean that people from different countries are similar. But it means that trying to infer someone’s attitudes or values based on their group membership (e.g., nationality, gender) is quite error-prone.
Highlighting similarities between groups of people has several important ramifications. For example, people from other countries are evaluated more positively when the similarities rather than the differences are highlighted. Further, highlighting similarities (vs differences) leads to more accurate perceptions of research findings.
These findings are described in the article entitled A new way to look at the data: Similarities between groups of people are large and important, recently published in the Journal of Personality and Social Psychology.