This is the 14th in our series of posts by job market candidates.
“There are those that run away from their countries for safety reasons. But there are those that are not good, they come here as refugees but just want to do business here and some are al-shabaab. The terrorists are hiding among them.”
– Research participant, Nairobi, Kenya
How can people take the perspective of others, particularly in situations where they are divided by distrust, fear, and scarce resources? Following a number of divisive elections, people around the world have found it challenging to relate to those they perceive as different from themselves, bringing the question of perspective-taking to the forefront. In my job market paper, my colleagues and I conduct mixed-methods research to test different strategies that people use to understand the perspective of others in the context of citizen-refugee dynamics in Kenya.
Increasingly, governments in the Global South are shifting away from historic encampment policies that required refugees to reside in designated areas and restricted their right to work and movement. More recent legislation in Kenya allows refugees greater access to work opportunities and freedom of movement. These inclusive policies bring increased opportunities for citizens to encounter refugees, who they may perceive as dissimilar to themselves on many dimensions. It is important to understand how we can influence citizens’ personal and policy-oriented attitudes toward greater tolerance and support for these populations.
Perspective-taking interventions in real-world settings: Listening to outgroup narratives or using one’s own personal narratives to understand the experience of outgroups
A psychological intervention that has gained a lot of traction in the last decade among social scientists and political organizers is perspective-taking. Perspective-taking refers to the active consideration of the psychological point of view–thoughts, emotions, motivations–of others. Although researchers have explored several approaches to perspective-taking, the current consensus favors prompting perspective-taking by listening to the narrative of an outgroup member. For example, a canvasser might share a transgender person’s story about enduring discrimination and abuse to improve attitudes toward transgender people. However, in everyday life, people often rely on their own personal experiences and emotions to understand others’ perspectives. For example, to understand the experience of a transgender person, one might recall a time when they personally felt judged for being different. Research suggests that the self is a default reference that people use to understand others, and this can increase positive regard for those whose perspective is being considered (Epley and Caruso, 2008; Todd and Galinsky, 2014). However, this process is conditional: the self may not be the most relevant or accessible referent for those considered too dissimilar. In such cases, people are more likely to rely on stereotypes or assume that the other’s perspective is the opposite of their own (Ames, 2004).
In my job market paper, we conducted a field experiment to test three perspective-taking approaches. Here, I focus specifically on the “self-narrative” based strategy since this has not previously been tested by itself in a field experiment. We hypothesized that if people recalled their own experiences of hardship that related to those of refugees, their attitudes toward refugees would improve.
The setting for the current study and the self-narrative perspective-taking intervention
We conducted our study in Nairobi, Kenya, which is home to roughly 600,000 refugees. We sampled 1466 Kenyan citizens for the self-narrative and control groups. These individuals are a subset of participants recruited by a non-governmental organization for a different study on microenterprise support and mentorship for refugees and Kenyans. Microentrepreneurs in Kenya are highly representative of the working population. We took the design of the larger study into account for our random assignment to ensure any relevant features (timings of the delivery of intervention, treatment assignment, etc.) were balanced across our research arms.
We randomly assigned participants to a self-narrative treatment (N=775) or a control condition (N = 691). Enumerators visited all participants as a part of a routine data collection activity. For participants in the self-narrative treatment condition, enumerators prompted participants to recall and share their experience: “Do you have a personal story about a time when you felt unsafe or felt like you were not in control of your life?” After listening to and engaging with the story of the participant, the enumerator connected their experience to the experience of refugees. For participants in the control condition, the enumerators simply collected the outcome measures, without any experience sharing.
We measured feelings toward, attitudes about, and policy support for two groups of refugees, Congolese and Somali refugees. We asked Kenyan citizens about both groups to avoid basing our results on just one refugee outgroup and because Congolese and Somali refugees are the two largest refugee groups in Nairobi. However, Congolese refugees are welcomed to a much greater extent than Somali refugees, who are often portrayed in the media and policy debates as threatening or dangerous because of past terrorist attacks carried out by al-Shabaab in Kenya. In addition, Congolese share a language in common with Kenyans, while Somalis stand out both because of different language and style of dress, and typically hold different religious beliefs.
Self-narratives influence attitudes and policy support for Congolese refugees immediately after the perspective-taking exercise…
Reflecting on a personal narrative about an unsafe experience led Kenyan citizens to adopt more positive attitudes and policy positions toward Congolese refugees, compared to the control group (Figure 1, Panel A). Although the effects were small–0.11 standard deviations or about a 2% change in each outcome, on average–they were comparable to those of bundled perspective-taking interventions in the U.S., where Americans considered the perspective of immigrants.
… but not for Somali refugees and not in the long term.
We observe no effect of the self-narrative perspective-taking exercise in improving feelings or attitudes toward Somali refugees (Figure 1, Panel B). When we look at the longer-term effects of this one-time perspective-taking exercise, we find no durable effects at either 3 or 6 months for Congolese or Somali refugees.
Perceived similarity is one possible reason for the differential results for Congolese vs. Somali refugees
We use exploratory qualitative, observational, and survey methods to demonstrate that perceived similarity may be one reason why self-narratives were effective at changing short-term attitudes for Congolese but not Somali refugees. Our findings indicate that Kenyans perceive Congolese as more similar, and Somalis as more dissimilar to themselves.
So, should we use self-referential prompts for perspective-taking interventions?
Our results suggest that using the self to imagine others is an under-appreciated path toward better regard for others. This approach builds on a capacity that people already have and utilize in their everyday lives. It is also a more readily deployable strategy when other-narratives or intergroup contact are not possible. In the current excitement for other-narratives, an approach that uses self-referential prompts may be mistakenly overlooked.
However, there is a clear limitation to self-narratives: using the self may not be possible if the self is not a relevant referent. This might be the case when others are perceived as too dissimilar or when more easily accessible stereotypes are available.
The short duration of our effects suggests that these one-off brief interactions may be insufficient to meaningfully shift peoples’ attitudes, but they may still be a useful approach for political organizers. For example, this approach can be utilized to move public opinion and impact voting right in time for a referendum.
Sana Adnan Khan is a PhD student in the Department of Psychology at Princeton University.
Join the Conversation