Assessing the Impact of Human Security Issue Frames on Public Support for North Korean Refugee Women

Introduction

Issue framing involves presenting an issue in a specific way to influence support for that issue (Benford & Snow, 2000; Snow, 2013). Issue framing works by highlighting specific elements of the given issue, emphasizing certain parts of the issue that are in frame, and de-emphasizing other aspects that are out of frame (Snow, 2013). In short, issue framing shapes how audiences process information.

The existing research on framing has examined many important policy issues including public opinion on government spending (Jacoby, 2000), climate change policy (Singh & Swanson, 2017), and support for vaccine cooperation during Covid-19 (Avdagic & Sedelmeier, 2024). An important line of framing research examines framing in relation to refugees and migrants. This research explores how the media depicts the issue of migration and refugee protection (Iberi and Saddam, 2022; Hillenbrand et al., 2024). Frames identified in academic research usually center around human rights and humanitarianism, emphasizing the responsibility to protect refugees, while other frames portray refugees as a cultural or security threat to nation-states. Beyond examining the content of frames themselves, other studies have examined the impact of framing on the public’s attitude towards migrants and refugees.

However, one type of frame that remains unexplored in academic research is human security. Since the publication of the 1994 Human Development Report by the United Nations Development Programme (UNDP), practitioners involved in assisting migrants and refugees have increasingly presented the issue as a matter of human security. Unlike traditional notions of security that emphasize threats to the nation-state, human security emphasizes threats faced in everyday life by individuals and communities and is therefore a people-centric notion of security, rather than a state-centric one (UNDP, 1994). For example, the United Nations High Commissioner for Refugees (UNHCR) categorizes the issue of refugee protection as a human security issue, as “[p]rotection of refugees is now primarily defined as security of refugees and refugee operations rather than in terms of the legal asylum process” (Adelman, 2001, p. 7). Other non-governmental organizations (NGOs) have followed suit (e.g., Amnesty International, 2015; European Council on Refugees and Exiles, 2016). Although NGOs utilize human security framing to communicate the plight of refugees, research has yet to test how levels of public support for refugees change based on different human security frames.

To explore this gap, this research uses survey experiments to test the impact of three different human security frames – political security, food security, and economic security – on public support for the protection of North Korean refugee women. This study utilizes comparative analysis of these three frames across multiple metrics that measure support for refugee protection, which is one of the most critical humanitarian crises in modern times (Betts, 2015). According to the United Nations High Commissioner for Refugees (UNHCR), at the end of 2024, the number of refugees globally reached 42.7 million, and the number has been steadily increasing for over a decade (UNHCR, 2025). Yet, liberal democracies have found it difficult to garner public support for refugee protection policies. Nationalism and populism have been prevalent trends in the United States and elsewhere, making it difficult for the general public, advocates, and policy makers to reach an agreement on how to best address what has become the heavily politicized issue of refugee protection. In light of these trends, the findings have implications for the creation of policy given calls for new solutions to address the predicament of refugees and contribute to the broader academic understanding of how framing impacts public attitudes towards support for refugees.

Literature Review

There is much research on issue framing, and how different ways of framing relate to migrants and refugees (Bleiker et al., 2013; Greenbank, 2014; Greussing & Boomgaarden, 2017; Hillenbrand et al., 2024; Lawlor & Tolley, 2017; Liu, 2022; Urbániková & Tkaczyk, 2020). However, most of these studies tend to focus on threat frames and society’s responsibility towards refugees from a humanitarian perspective; less studied are frames that relate to human security and how they may shape public opinion on refugee protection, even though these frames are widely used by practitioners involved in serving and protecting refugees, and often underlie refugees’ reasons for leaving (Adelman, 2001; Amnesty International, 2015; Aubin, 2011). This literature review begins by reviewing the broader literature on issue framing, before turning to studies with a migrant and refugee focus. It then addresses the framework of human security, and finally discusses North Korean refugees specifically.

The creation of issue frames is a process that involves generating ideas and meanings (Benford & Snow, 2000). In other words, meaning is constructed by an agent through processes of interpretation rather than being an inherent part of a specific social condition. Frames are a part of this process as they are constructed to highlight and call attention to specific aspects of the issue in question, and communicate a narrative with the aim of affecting how a given issue is perceived and understood (Snow, 2013). This involves “meaning work” in which agents define a unified interpretation of a problem, pinpoint its cause, offer solutions, and mobilize people to take action (Benford & Snow, 2000). Thus, the goal of framing is twofold: produce consensus about the nature of a problem and what is needed to remedy it, and to mobilize people to take action in ways that are consistent with the frame (Benford & Snow, 2000; Snow, 2013).

In terms of framing and refugees, the literature shows that media framing has a disproportionately negative bent. For example, one of the most prominent frames is the threat frame, which portrays refugees as a threat to national security, sovereignty, and the economy (Bleiker et al., 2013; Greussing & Boomgaarden, 2017; Lawlor & Tolley, 2017; Liu, 2022; Urbániková & Tkaczyk, 2020). The victim frame and policy issue frame are also common in the literature. Through these frames, refugees are portrayed as othered victims of a crisis (Liu, 2022; Urbániková & Tkaczyk, 2020) and framed as a policy issue involving a burden on the host society (Greenbank, 2014). A few studies have found evidence of a humanitarian frame and a benefit frame (Greussing & Boomgaarden, 2017; Liu, 2022). However, these media frames are rare in comparison to frames that portray refugees in a more negative light.

A stream of this literature examines framing and public support for refugees. Some research has found that the humanitarian frame increases support for refugee assistance (Hillenbrand et al., 2024), as it emphasizes society’s responsibilities towards refugees and highlights various humanitarian concerns. In addition, there is evidence that supports the idea that public support for refugees varies by refugees’ national origin. For example, Iberi and Saddam (2023) found that African refugees fleeing to Europe were more likely to be portrayed as a security threat compared to Ukrainian refugees, who were framed in a more sympathetic light, which likely influences the level of sympathy they receive from the public. Another prominent theme is that support varies by the sex of the refugee. More specifically, male refugees have been found to receive a lower level of support compared to women (Hillenbrand et al., 2024); similarly, support for refugee girls receives more support among European publics (Bešić et al., 2018). Additionally, experimental research has shown that American conservatives are more likely to support refugees when it is portrayed as an issue of protecting women and children from violence (Sullivan & Rich, 2017). The violence against women frame was also tested in the European context and was found to garner high levels of support in France and Germany (Schnyder, 2020).

One frame that has received far less attention in the literature is the human security frame, even though this frame is used quite often by practitioners who work in support of refugees, such as UNHCR (see Adelman, 2001). The human security framework was introduced by the UNDP in the 1994 Human Development Report. As a concept, human security focuses on the security of the individual or community, and is concerned with human life and dignity, as opposed to a traditional focus on threats faced by the nation-state (UNDP, 1994). Unlike territorial or military security, which is a defensive concept, human security is an integrative process (UNDP, 1994). The Human Development Report outlines 7 dimensions of human security: economic security, food security, health security, environmental security, personal security, community security, and political security. The literature on refugee protection has considered how the framework of human security can offer a source of protection in light of eroding refugee rights (Edwards, 2009; Odutayo, 2018; Katsuma, 2018; Khairiah et al., 2021). However, it has not yet been tested in experimental studies to see how it affects support for refugee protection.

Surveys of North Korean refugees over time have shown that several of these dimensions of human security play a central role in their escape (Korea Hana Foundation, 2022, 2023, 2024). Among North Korean women, the top three reasons for defecting have to do with food security, political security, and economic security. More specifically, among those recently surveyed, 38.5% of women reported leaving due to food shortages, 36.1% reported leaving in order to seek freedom because they “hated being monitored and controlled by the North Korean regime,” and 34.3% reported leaving in order to earn more money (Korea Hana Foundation, 2024, p. 110).^[1] As these three aspects of human security are so central to women’s experiences of escape, they are used as the basis of the frames for the survey experiments in this study, discussed in more detail in the methodology section. North Korean women are the subject of the survey experiments because they are the vast majority of escapees. According to the South Korean government, 72% of those who have defected from North Korea are women (Ministry of Unification, n.d.).

Based on this examination of the literature and the gaps that have been noted, the following research question guided this study: How do human security frames impact public support for North Korean refugee women? The hypothesis is that framing the protection of North Korean refugee women in terms of different dimensions of human security will increase public support for their protection.

Materials and Methods

This study uses a survey experiment to test how framing impacts attitudes towards North Korean refugee women. This research method involves the random assignment of participants to either a treatment or control group, utilizing different versions of the survey, and compares outcomes across groups to determine if measurable differences exist. This procedure is described further below.

Experimental Protocol and Dependent Variables

The experiment consisted of three treatment groups and a control group. Each of the three treatment groups were exposed to a unique frame and then asked to answer a series of questions. The first three questions correspond to three dependent variables that measure attitudes towards refugee protection, and are further described in section 3.2. The remaining questions were demographic questions asking respondents to report their gender, age, and education level. Respondents also had to answer a quality control question to ensure that they were paying attention. All respondents in the working data set answered the quality control question correctly. The control group was asked to answer the same series of questions without seeing a frame. Participants were either randomly assigned to one of the frames or the control group. The three frames, which are presented in Table 1, consist of a food security frame, a political security frame, and an economic security frame, which correspond to the top three reasons why North Korean refugee women self-report escaping North Korea, presented in the literature review.

2.2 Data Collection and Analysis Process

Individuals over the age of eighteen in the United States were recruited to the survey experiments during the month of August 2025 using the Prolific online data collection platform, noted for its high data quality, transparency, and consistency (Peer et al., 2021; Douglas et al., 2023; Kothe & Ling, 2019). This study included quality control measures, including a quality control question to ensure respondents were paying attention, and ensuring that the Internet Protocol (IP) address used to take the survey was used only once. The total sample size is 622 participants, which when compared against the population aged 18 and over in the United States, represents a confidence level of 98% and a margin of error of 5%. Before taking part in the survey experiments, participants were shown a short description of the survey topic, the estimated completion time, and compensation details.

A comparative analysis was conducted comparing the three frames and the control group to identify the impact of the different frames on attitudes towards North Korean refugee women. The effects of the different treatments are analyzed using difference-in-means estimation, which assesses differences in mean scores on the dependent variables relative to the control group. Sample weights were included to ensure that the data more accurately reflects the characteristics of the U.S. population.

Results

This section discusses the sample and findings for the three aforementioned dependent variables for each of the issue frames compared to the control group.

Sample Characteristics

The overall sample is relatively balanced between male (50%) and female (48%) respondents.^[2] The modal overall sample member is in the 35-44 age group (31%) and has either completed some university (24%) or holds a 4-year university degree (40%). Compared to U.S. Census data, the sample skews younger and more educated than the general population (U.S. Census Bureau, n.d.-b, n.d.-a). Therefore, results were run with sample weights to reduce over-representation of certain educational and age groups and to align the sample with the known population distribution, which improves generalizability and allows for more accurate inferences about the broader population.

Dependent Variables

After seeing a frame or being assigned to the control group, respondents answered three questions to measure their level of support for refugee protection. The first dependent variable asked respondents to report whether the number of North Korean refugee women allowed to enter the United States should decrease, stay about the same, or increase, on a three-point scale with one representing decrease, two representing stay about the same, and three representing increase. Next, participants were asked to rate the importance of refugee protection policy as it relates to North Korean women, on a scale from one to five, with one representing not at all important and five representing extremely important. Finally, participants were asked to assess the relative importance of refugee protection compared to six other important policy issues: terrorism, international economic policy, the spread of nuclear weapons, democratization, global public health, and climate change. Since it is possible that one could view refugee protection as an issue of high absolute importance but rank it low in comparison to other important issues, this dependent variable measures the relative importance of refugee protection versus absolute support. Table 2 shows the summary statistics for the three dependent variables for the combined sample.

reatment Effects

The average treatment effects of the different issue frame treatments are assessed using difference-in-means estimation. The first dependent variable represents the most specific measure of support by asking respondents whether the number of North Korean refugee women allowed to enter the United States should decrease, stay about the same, or increase. Table 3 presents differences in mean ratings for this variable across the three issue frame treatments, which are considered relative to the control group, whose mean level of support is 2.47.

The results show that the food security frame produces a statistically significant treatment effect. More specifically, respondents who are exposed to the food security frame show increased support for the protection of North Korean refugee women relative to the control group. Expressed as a proportion, among the respondents exposed to the food security frame, 68.4% favor increasing the number admitted and 24.1% favor maintaining current levels, compared to 51.3% and 44.7%, respectively, in the control group.

The political security frame shows similar effects, although the results for this frame are just outside of conventional levels of statistical significance. Here, respondents who are exposed to the political security frame also show increased support for protecting North Korean refugee women relative to the control group. Among the respondents exposed to the political security frame, 62.8% favor increasing the number admitted and 31.4% favor maintaining current levels. Overall, these results provide support for the hypothesis that how the refugee protection issue is framed influences levels of support.

The second dependent variable measures individuals’ opinions about the absolute importance of protecting North Korean women refugees as a policy issue. Again, Table 3 shows differences in mean ratings of this variable across the three issue frame treatments. Differences are considered relative to the control group, for which the mean absolute importance rating is 3.48 on a 1-5 scale. For this measure of support for refugee protection, none of the issue frame treatments produce a statistically significant framing effect, which does not support the hypothesis.

Finally, the relative measure of refugee protection policy importance is examined in which the importance of refugee protection is ranked relative to six other important policy issues. The scale ranges from 1 (most important) to 7 (least important). The mean ranking of refugee protection importance for the control group is 4.37. To reiterate, Table 3 shows differences in the mean rankings of the importance of refugee protection policy across the three issue frame treatments relative to the control group. Like the absolute importance variable described above, the results do not show a statistically significant framing effect for any of the framing treatments, which again does not support the hypothesis.

For these last two dependent variables, these statistically insignificant findings may be explained by the fact that the means show little variation across treatment groups. Asking respondents about the general importance of this policy issue shows that opinions are similar across the three frames, and between the frames and the control group. One possible explanation is that phrasing the question in terms of general importance is not a specific enough measure of support. Another possible explanation is that the economic and political security frames resonate less with the general public. These norms may not be as emotionally salient for a few reasons. If the public holds the pre-existing belief that refugees are an economic threat because they increase labor market competition, then the economic security frame will be less salient. As a result, individuals may be unlikely to increase their support for refugees in response to economic security framing. The insignificant results from the political security frame may reflect weaker norm internalization. Political security threats may seem vague or distant compared to local, everyday problems, and therefore difficult for people to imagine. Because political security threats can seem far removed, this frame may resonate less among the general public.

Discussion

Overall, these findings only partially supported the hypothesis. The findings suggest that certain dimensions of human security can influence public perceptions of refugees, and food security is a particularly salient dimension. While the food security frame increased public support for the protection of North Korean refugee women, neither the economic security frame nor the political security frame had a statistically significant effect. In other words, the food security frame is the only treatment to produce a clear framing effect.

Food security is a factor commonly featured in the literature on the plight of refugees (Alkharouf et al., 2023; Hammond, 2018; Laborde et al., 2017). Aspects of food insecurity such as hunger and food shortages are widely written about as motivators for fleeing; however, the literature has not yet examined food security in the context of the receiving societies in terms of how messages about food insecurity experienced by those in other countries can contribute to greater public acceptance of refugees. That the food security frame produces a treatment effect among those residing in a host society is notable in that its underlying norms are widely considered to be influential (Raymond et al., 2014; Weldon, 2006). The right to food, for instance, is a universal human right that stipulates everyone should have “physical and economic access at all times to adequate food or means for its procurement” (Office of the United Nations High Commissioner for Human Rights & Food and Agriculture Organization, 2010, p.2). The right to adequate food and the right to be free from hunger feature in the Universal Declaration of Human Rights in connection with the right to an adequate standard of living, and are protected in the International Covenant on Economic, Social and Cultural Rights (Office of the United Nations High Commissioner for Human Rights & Food and Agriculture Organization, 2010). Women are especially vulnerable to violations of the right to food, due to traditional gender roles, gender inequality, and disproportionate power relations in many societies (Food and Agriculture Organization, 2023).

These findings complement other research that finds that the strength of the norm underlying a given policy makes a difference in predicting the overall stability of that policy (Raymond & Delshad, 2016). Moreover, the findings are consistent with research that has found many human rights norms to be compelling across a range of individuals and societies (Keck & Sikkink, 1998; Raymond et al., 2014). In addition, research examining other policy areas shows that local level issue frames may produce particularly strong effects on opinion (Hornsey et al., 2016; Spence et al., 2012; Wiest et al., 2015). Given this research context, future research could create and test localized food security issue frames in the context of a specific city, state, or province, for example. Given that research has found that support for women refugees exists even among those who identify as politically conservative (Sullivan & Rich, 2017), future research could continue to highlight the plight of women fleeing food insecure environments. In addition, future research should include more cross-case comparisons examining how different frames impact support for different refugee groups such as those from Syria, Afghanistan, Sudan, Venezuela, and Ukraine. Support could be measured among publics in different receiving country contexts. This type of comparative research could highlight additional factors that could increase public support, which could increase generalizability beyond the U.S. context, and potentially hold relevance for policy makers.

In conclusion, experimental public opinion research such as this can be used to determine how different ways of framing a policy issue affect public opinion, allowing policymakers to potentially create policies that are more stable and resistant to change (Raymond & Delshad, 2016). Individuals’ opinions about refugee protection may not be as fixed as is often assumed when the issue is portrayed as a matter of food security and highlights the plight of women. Non-governmental organizations and policymakers concerned with refuge protection may find it useful to draw on issue frames supported by strong and widely accepted human rights norms.

Acknowledgements

I would like to thank my research mentor, Dr. Melissa Schnyder, for providing me with guidance for this study.

Ethical Approval Statement

This study received IRB approval from American Public University 2025-004.