Populistic rhetoric is characterized by discourses and frames dividing society into "us" and "them" (Araújo & Prior, 2020). Immigrants and refugees, as the ultimate other, are a frequent target of populist discourses, portrayed as threats to the economy, culture, and security (Levy et al., 2016; Wirz et al., 2018). Negative representation is associated with the increase of xenophobia and hate speech, weakening existing anti-discriminatory regulations and strengthening avoidance and discrimination towards immigrants and refugees (Bilewicz & Soral, 2020; Kroon et al., 2020). Media discourses may help shape and construct negative stereotypical frames that present immigrants and refugees as bad, threat or victim, in a process that divides in-groups from outgroups (Seate & Mastro, 2017; Matthes & Schmuck, 2017). Since the media influences the public agenda (McCombs & Shaw, 1972; McCombs & Valenzuela, 2007), it is essential to understand media representation and frames over time.
There are numerous studies focusing on media framing and representations of immigrants and refugees in Europe (Amores et al., 2019; Eberl et al., 2018; Lams, 2018). A robust literature review from 2018 examines the diversity of findings from those studies; however, it is noteworthy that none of these studies compares European and Latin American countries (Eberl et al., 2018). A vast majority of the studies have focused on European territories, with a minority also comparing European countries with the United States or African countries (Benson, 2013; Fengler et al., 2020). The prevalence of studies that focus on Western Educated, Industrialized, Rich and Democratic (WEIRD) societies was highlighted in communication and media studies (Chakravartty et al., 2018).
This systematic review seeks to contribute in filling this gap by focusing on studies about media and migration in Brazil and the United Kingdom (UK). Drawing on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines for systematizing literature, this paper aims to provide a better understanding of the differences and similarities in research related to media representations of immigrants and refugees in Brazil and the UK. It focuses on the following two questions: (RQ1) which research methods were used; and (RQ2) how the representation varied between Brazil and the UK in the findings of previous scholarly literature.
The study focuses on these two countries because of their recent interactions with the populist and far-right end of the political spectrum. Scholars have shown that immigration is an important factor in the success of right-wing political parties (Shehaj et al., 2021). In the UK, two events represent the political bond with populism: the UK Independence Party's ascension and the Brexit Referendum in 2016 (Hughes, 2019). In Brazil, the most prominent event is the election of President Jair Bolsonaro in 2018 (Lugo-Ocando, 2020; Novoselova, 2020). In both cases, politicians portrayed immigrants as enemies (Hughes, 2019; Silva, 2020).
Although populism rose in different countries worldwide, this study aims to compare one country from the global south and one country from the global north with different political and media systems with similar experiences on populism in the past years. The UK has a parliamentary form of democracy, while Brazil has a presidential system. Moreover, the UK belongs to the liberal model of media, while Brazil has characteristics from different models. Some aspects of Brazilian media are: private media enterprises' control of the press; the politicization of public broadcasting media; low level of newspapers circulation; and tradition of non-objective journalism style (Albuquerque, 2012; Hallin & Mancini, 2004). A comparison between these two counties may help us uncover the prevalent discourse and the similarities and differences between countries with different political and economic systems and traditions, and those that have experienced far-right and populist political movements in the 21st century. Moreover, this study can lead to more comparisons, including countries from the global south and north.
Materials and methods
Our study is based on a systematic review of 47 peer-reviewed articles published in indexed journals over the past two decades (2001-2020) in either English or Portuguese. The articles focus on references to immigration and/or refugees in both legacy and digital media. We followed PRISMA guidelines when collecting, mapping, and systematically reviewing the literature (Moher et al., 2015). The articles were collected from the Web of Science Core Collection, Scopus, and B-on databases.
Search and selection procedure
We systematically searched for articles from the past two decades that focus on the representation of immigrants and refugees in Brazil and the UK. First, we conducted a search of all databases using a Boolean search string that combined the terms: "content analysis"; "discourse analysis"; "media analysis"; "media"; "coverage"; "news"; "migrat*"; "immigrat*"; "migrant*"; "refugee"; "Brazil", and "the United Kingdom" 1 . We performed the search in English for research related to the UK and in Portuguese and English for research related to Brazil. In the case of Brazil, we also included the B-on database, since only a very small number of Brazilian communication articles was found in the Web of Science Core Collection and Scopus databases (five and three articles, respectively). We therefore decided to add the Portuguese database B-on to our criteria, since this database searches Scopus, Web of Science, and other important academic indexers in Latin America, such as Latindex, Sumários.org, and DOAJ (Costa, 2015; Vasconcelos, 2004). We opted for including B-on to find results beyond the major western databases, hence avoiding the exclusion of important academic literature in Brazil and Latin America.
We used the following inclusion criteria (CR) when searching for papers: (CR1) peer-reviewed articles related to the UK and Brazil; (CR2) peer-reviewed articles related to discourses about immigration in mass media, political campaigns, social media, or cultural productions; (CR3) peer-reviewed articles written in English or Portuguese. We found 40 articles in Web of Science (35 on the UK and five on Brazil), 26 in Scopus (three on Brazil) and 76 on the B-on platform (all related to Brazil). From the total of 142 articles, 10 were excluded because they were duplicates. Of the remaining 132 articles, 85 more were excluded because they did not adhere to the abovementioned criteria. The final corpus of analysis thus included 47 peer-reviewed articles published in scientific journals, 19 related to Brazil, and 28 to the UK (Figure 1).
When analyzing the literature, we identified the object of analysis; research question(s); method(s); and findings. Additionally, we used the SPSS software for quantitative analysis. We added variables, including methods (quantitative, qualitative, mixed); types of study (comparative or single-country), and tone of representation in media analyzed in the studies (negative, neutral, and positive). We quantified the methods that were used by coding them as: 1 (quantitative); 2 (qualitative); or 3 (mixed). The types of study were coded as: 1 (comparative); or 2 (single-country). When analyzing the tone of representation, we coded this as: -1 (negative), 0 (neutral), and 1 (positive). Articles that did not specifically mention the tone of representation were excluded from this analysis.
Analysis and findings
Methods and objects of analysis
We found some important cross-national differences when examining the use of mixed, quantitative, or qualitative methods in the two countries. Table 1 shows that quantitative methods were most prevalent in the UK (67.9%), followed by qualitative (17.9%), and mixed approaches (14.3%). Studies that included Brazil were mostly qualitative (94.7%), with only one quantitative example. The Fisher's exact test demonstrates significant differences (p<.001) between Brazil and the UK in terms of these studies' use of mixed, quantitative, or qualitative methods.
The most frequently used method in UK studies was quantitative content analysis, which was applied in 39% of the reviewed articles (Amores et al., 2019; Aydemir & Vliegenthart, 2018; Hughes, 2019; Kaleda, 2014; Koopmans, 2004; Langer & Gruber, 2021; Masini, 2019; Masini et al., 2018; Shan-Jan, 2019; Walter, 2019). Other studies used a combination of methods to answer their research questions, such as semi-structured interviews, focus groups, social network analysis, and panel surveys (Balabanova, 2019; Froio & Ganesh, 2019; Hamlin, 2016; Scalvini, 2016; Wirz et al., 2018). Discourse analyses were carried out in 15% of the articles (Bates, 2017; Gibson & Booth, 2017; Lams, 2018; Pruitt, 2019). A minority of papers used other methods, such as an experimental approach to media effects, topic modeling, sentiment analysis, thematic analysis, case studies, and public opinion polls (Evans & Mellon, 2019; Harper & Hogue, 2019; Heidenreich et al., 2019; Risam, 2018; Ryan & Reicher, 2019; Schumann et al., 2020; Vautier, 2009).
The vast majority (94%) of the papers related to research from Brazil used qualitative methods (Aragão & Santil, 2018; Moreira et al., 2019; Silva, 2020; Zanforlin & Cogo, 2019); by contrast, quantitative methods were prevalent in the UK (accounting for 67% of all articles reviewed). Most articles that included Brazil (31%) combined different methods, such as interviews and questionnaires (Brignol & Costa, 2018b; Javorski & Brignol, 2017; Theodoro & Cogo, 2020). The second most frequently used method (15%) was discourse analysis (Mizga & Trovão, 2018; Moreira et al., 2019; Sacramento & Machado, 2015). Other methods included a case study approach, narrative analysis, experience reports, framing analyses, reception studies, and exploratory research. Most of the UK studies analyzed newspapers (75%). Some (19%) analyzed newspapers alongside other data sources, broadcast transcripts, social media, official immigration rates, longitudinal monitoring surveys, and in-depth interviews (Evans & Mellon, 2019; Hamlin, 2016; Langer & Gruber, 2021; Ryan & Reicher, 2019; Wirz et al., 2018). A few used other media sources, such as Twitter, photos, political speeches, or questionnaires (Amores et al., 2019; Froio & Ganesh, 2019; Schumann et al., 2020; Vautier, 2009). Similarly, most research on Brazil (35.3%) analyzed newspapers (Moreira et al., 2019; Ranincheski & Uebel, 2018; Sacramento & Machado, 2015). However, other media outlets were also investigated, and a significant portion of the studies (35.3%) examined a variety of media (Cogo, 2018; Silva, 2020; Zanforlin & Cogo, 2019). Some articles (11.8%) studied television broadcast material (Javorski & Brignol, 2017; Mizga & Trovão, 2018); others (11.8%) analyzed social media (Brignol & Costa, 2018b; Morais & Santos, 2017); and, lastly, a single study focused on literary books related to migration and refugees in Brazil (Tonus, 2018).
Table 2 provides an overview of which types of study were most frequently seen in each country. In total, 57.1% of studies that included the UK were comparative, while in Brazil, there was only a single comparative study. In Brazil, most studies (94.7%) focused only on national media, in contrast to the sample from the UK (42.9%). The chi-square test yielded a significant difference (p<.001) between Brazil and the UK when considering comparative versus non-comparative studies (Table 1).
Of the 16 comparative studies including the UK, 12 (75%) focused on European countries (Aydemir & Vliegenthart, 2018; Mancini et al., 2019; Masini, 2019), and of those 12, only three (25%) compared countries from eastern and western Europe (Balabanova, 2019; Heidenreich et al., 2019; Lams, 2018). Four studies (25%) compared the UK with non-European countries, such as Canada, Hong Kong, Kenya, the United States, and Taiwan (Kaleda, 2014; Lawlor, 2015; Risam, 2018; Shan-Jan, 2019). The analysis shows that, even when a study made an attempt to adopt a more cross-national perspective, this primarily focused on comparisons between countries in the global north.
One study that did conduct a comparative, cross-national analysis of media framing of immigration in the UK, United States, Hong Kong, and Taiwan (Shan-Jan, 2019) showed higher levels of negative tone toward immigrants in the UK and United States than in Hong Kong and Taiwan. According to this study's findings, most media frames in the United States and the UK portray immigrants as threats to the economy or national security, while the media in Hong Kong and Taiwan adopt a cultural integration frame which presents the assimilation immigrants into their host culture. In addition, studies which compared the UK and Canada showed that negative representation and an emphasis on crime and security frames prevailed in both (Lawlor, 2015); however, Canada adopted a less negative tone by using an economic frame on immigrants' contributions to economic growth.
Comparative studies that included the UK and the United States found that reporting on refugees were characterized by suspicion, lack of sympathy, and uniformization (Risam, 2018; Kaleda, 2014). Further studies have found that most of the actors quoted directly or indirectly in legacy media reports on refugees were state authorities or actors in favor of or against immigrants, but not the immigrants themselves (Masini, 2019; Masini et al., 2018). Alternative media, on the other hand, usually gave immigrants a voice by quoting refugees or immigrants directly (Kaleda, 2014).
In contrast to the prevalence of comparative studies in the UK, only one article in the Brazilian corpus compared different countries, both in Latin America (Silva, 2020). Most research focused on the national level and analyzed important Brazilian newspapers, such as O Globo and Folha de S. Paulo (Morais & Santos, 2017; Moreira et al., 2019; Ranincheski & Uebel, 2018; Sacramento & Machado, 2015). Findings related to national media demonstrate a process of silencing immigrants and refugees, who are rarely quoted or referenced as sources of information and are not presented as the protagonists in media stories (Moreira et al., 2019). In addition, articles also diminish immigrants by portraying aspects of their home countries as negative, contributing to the process of othering and the separation between "us" and "them" (Ranincheski & Uebel, 2018). When newspapers show the terrible conditions and difficulties faced by immigrants, these immigrants are also presented as a threat to public health (Sacramento & Machado, 2015) or as victims (Morais & Santos, 2017). When comparing national and regional television news, it is clear that the theme of immigration has little visibility in Brazil and that international reporting trends are reproduced on the national level (Javorski & Brignol, 2017). Results from an analysis focusing on southern Brazil also show that the negative tone is observed locally, with news items focusing on regional problems and immigrants' difficulties, again using a victim frame (Almeida & Brandão, 2015).
Patterns in the representation of immigration
Table 3 shows that the tone of representation in media analyzed in the studies that included Brazil and the UK is mainly negative (50%), followed by neutral (32.5%), and then positive (17.5%). When comparing the research on both countries, no major differences were found. In both Brazil and the UK, the aggregate media representation from the data analyzed in the literature we reviewed was primarily negative (50%), followed by neutral (33.3% and 31.3% for Brazil and the UK, respectively), and then positive (16.7% and 18.8% for Brazil and the UK, respectively).
A majority (78.5%) of studies including the UK focused on the general category of immigration and refugees (Amores et al., 2019; Lawlor, 2015; Mancini et al., 2019; Masini, 2019), while in Brazil, most studies (79%) focused on specific groups of immigrants, such as Senegalese, Haitians, or Venezuelans (Almeida & Brandão, 2015; Brignol & Costa, 2018a; Morais & Santos, 2017; Sacramento & Machado, 2015). The results reveal that representation differs depending on the nationality of the immigrants in question, with European immigrants receiving more positive representation in Brazil (Cogo, 2018; Dadalto, 2013) and immigrants to the UK who originate from Muslim countries being the worst represented (Froio & Ganesh, 2019; Scalvini, 2016). Studies in both countries lack intersectional perspectives, uniformizing the figure of the immigrant or refugee. For example, only one study about immigrants in Brazil references invisibility dynamics in the LGBTQI+ diaspora (Theodoro & Cogo, 2020).
As found in one study, conducted by Benson (2013), this systematic literature review revealed standardization and repetition of the threat, victims, and hero frames. The hero frame is usually related to neoliberal values, presenting stories that describe immigrant success stories, helping to shape the idea that outsiders have to fight for their achievements (Zanforlin & Cogo, 2019). Media also represent refugees and immigrants as victims, for instance, in articles that show overcrowded spaces, describe conflicts with authorities, or mention a lack of resources, creating the idea of a uniform mass of victims (Lima et al., 2016; Mizga & Trovão, 2018). Finally, another frequent issue is presenting these groups as a threat by associating them with words such as "invasion," stating that there is a necessity to control them, and mentioning state-sanctioned benefits for migrants (Cogo & Silva, 2015; Lima-Pimentel & Castro-Cotinguiba, 2014; Ranincheski & Uebel, 2018). The results also showed that the terms "refugees" and "immigrants" are generally used interchangeably (Brignol & Costa, 2018a; Pereira-Lima & Faria-Campos, 2015).
Of studies including the UK, only a minority (16.7%) of results showed a positive tone (Balabanova, 2019; Bates, 2017; Langer & Gruber, 2021; Ryan & Reicher, 2019). In some cases, positive representation was the result of special circumstances, for instance, the United Nations' Refugee Agency campaign that sought to increase positive sentiment related to refugees (Balabanova, 2019). Other examples of positive representation were in response to scandals related to hostile behavior against immigrants or newspapers portraying immigrants positively in opposition to human rights violations against them (Bates, 2017; Langer & Gruber, 2021; Ryan & Reicher, 2019). In rare cases (4.3%), studies in both countries referred to positive discourses that were linked to hero frames (Ryan & Reicher, 2019; Zanforlin & Cogo, 2019). Nevertheless, hero frames portray immigrants positively when immigrants have been accepted by and adapted to fit the culture of a more welcoming country (Shan-Jan, 2019). Positive representation is also found in alternative media, as some articles related to the Brazilian context show (Aragão & Santil, 2018; Brignol & Costa, 2018b; Cogo, 2018). In these media outlets, some prevalent themes are citizenship, political and social participation, integration, culture valorization, exposing racism, and helpful information for those groups.
In political discourses, liberal immigration programs are presented as a "'fair" option founded on meritocratic methods, provided these programs dictate which immigrant groups should be accepted (Gibson & Booth, 2017). Research shows connections between the growth of the United Kingdom Independence Party (UKIP), anti-migration and anti-European Union (EU) sentiment, the EU referendum in 2016, and Brexit (Evans & Mellon, 2019). UKIP regularly used populist language to appeal to "ordinary" people, diminishing elites and evoking the differences between British people and others (Hughes, 2019) in a bid to boost its success. Politically, Brazil faced a similar situation. Until the election of President Jair Bolsonaro, Brazilian politicians tended to use pro-immigrant and refugee discourses (Cogo, 2018; Dadalto, 2013); Bolsonaro, on the other hand, portrays immigrants as the enemy (Silva, 2020). Like UKIP, Bolsonaro gained visibility by adopting a populist discourse that separated "us" from "them," criticized the system and the establishment, condemned corruption, and adopted a nationalistic perspective.
This systematic literature review of peer-reviewed studies examining media representations of immigrants and refugees in Brazil and the UK revealed similarities and differences in how research was conducted in these two countries. Studies focusing on Brazil were predominantly local, while in the UK, they usually adopted a comparative approach that incorporated other countries. However, these comparative studies were mainly conducted in European countries and the global north, focusing on WEIRD countries. The results also show differences in the methods chosen for studies based on the country it took place in. In Brazil, a qualitative approach was more popular, while quantitative methods were more prevalent in the UK. One possible explanation for the difference in which methods were used could be the different research traditions and inequity in terms of access to research infrastructure. These differences also demonstrate how varied the research traditions are in each country. Brazil, for instance, is influenced by Anglophone, Latin-American and French communication research and focuses on critical approaches to media and communication. In addition, WEIRD countries in the global north have more significant resources to devote to research, while universities in Latin America, including Brazilian communication faculties, often lack access to news databases and statistical software (Skoric, 2014; Zelenkauskaite & Bucy, 2016).
Different studies in this review also focused on different media platforms. Studies focusing on Brazil looked at a diverse range of media, including newspapers, television, social media, or a combination of two or more. In contrast, research focusing on the UK concentrated mainly on newspapers; the prevalence of newspaper articles in these studies is most likely due to the fact that they are easily accessible via digital archives, while collecting data from broadcast media is more difficult. The results also demonstrate that research in both Brazil and the UK tends to focus primarily on legacy media, with alternative communication platforms rarely being studied. The emergence of new voices and their possible participation in social media has gone almost unnoticed. This is most likely a result of the significant influence that legacy media has at the agenda across society (Langer & Gruber, 2021).
Although both countries have different media and political systems and significant cultural differences, the results show that discourses on immigration are predominantly negative. In line with previous systematic reviews showing the dominance of political elite sources in legacy media (Consterdine, 2018; Eberl et al., 2018), our findings demonstrate that immigrant groups are usually underrepresented and they are rarely included as sources in these discourses. Instead, other groups communicate for them, such as national authorities and pro- or anti-immigration groups or representatives. The studies revealed that immigrants and refugees are better represented in alternative media, which offers them a platform to express their opinions with less stereotypical framing in comparison to mainstream media (Aragão & Santil, 2018; Cogo, 2018; Kaleda, 2014).
Refugees are framed as vulnerable, suffering, helpless victims, and as a mass of people (Kaleda, 2014; Morais & Santos, 2017). In some cases, concepts are used interchangeably, with "immigrants" and "refugees" being used synonymously, or the categories overlap, generating and reinforcing stereotypes (Mizga & Trovão, 2018). Both immigrants and refugees are represented as threats to culture, the welfare state, and public health (Hamlin, 2016; Pruitt, 2019; Sacramento & Machado, 2015). Metaphors describing refugees and immigrants as an "invasion" that is "flooding" a country promote in the public imagination a sense that these groups are dangerous or threatening, increasing fear toward them. These findings are in line with previous studies, in which immigrants and refugees are often framed as cultural, economic, or criminal threats (Consterdine, 2018; Eberl et al., 2018). The presence in these studies of the threat, victim, and hero frames mirrors findings by Benson (2013), which suggest a standardization of frames in both countries.
Furthermore, this type of framing also contributes to the process of otherization, in which a society is divided into "us" and "them." These stereotypical frames are particularly prevalent in populist discourses, designed to create groups of others, or strangers, who are seen as dangerous enemies of the nation. Negative stereotyping is also articulated with hate speech (Paz et al., 2020), and a past study showed that forms of hate speech are more common today in traditional media (Winiewski et al., 2017), which often reproduces insults directed to minority groups through others' declarations, such as the ones from populist leaders. However, even before the election of populist leaders, there existed frames dividing ingroups from outgroups and an association between immigration and invasion. This makes it clear that frames prevail across many years and that, even if they can be dynamic, they create common cognitions and produce societal bonds that are collective and resistant to change. Previous results demonstrate that people more exposed to hate speech are more likely to use hate speech and to support repressive measures directed to immigrants and refugees, such as closing frontiers. Moreover, the increase of hate speech is also associated with the desensitization of people towards this type of discourse (Winiewski et al., 2017).
Although negative representations are prevalent, the results also revealed some positive discourses in both Brazil and the UK. Positive representations were usually associated with exceptional and unique events, such as illegal or controversial deportations, issues of discrimination, or amnesty for immigrants; or with memorable dates, such as World Refugee Day (Balabanova, 2019; Bates, 2017; Langer & Gruber, 2021; Ryan & Reicher, 2019; Zanforlin & Cogo, 2019). Another frequent positive representation is the hero frame, which encapsulates a neoliberal notion of immigration in which refugees or immigrants are responsible for their own success, with the state bearing no responsibility for it.
The aim of this systematic literature review was to analyze research concerning media representations of immigrants and refugees in Brazil and the UK, given the absence of comparative studies related to immigration and media in these two countries. This review first sought to provide answers to the following questions: which research methods were used in these studies, and how the representation varied between Brazil and the UK in the findings of the scholarly literature. This article evaluates patterns that exist in studies including both countries and highlights the primary differences and similarities between the two by systematically reviewing the existing scholarly literature on media representations of immigration in both countries. This is the first comparative examination of its type, and it aims to pave the way for further empirical comparative studies that incorporate the global south and global north by comparing WEIRD and non-WEIRD countries.
Our study also revealed differences in terms of which methods were used in research focusing on Brazil versus the UK. For studies related to the UK, we would suggest using additional methods, including mixed-method approaches and comparative studies that incorporate the global south. Research focusing on Brazil may benefit from more quantitative approaches. In all cases, our results revealed a focus on legacy media and newspapers, to the exclusion of alternative communication platforms and other media outlets. Thus, future research should also analyze alternative media, compare legacy and alternative media, and focus on other media outlets. Alternative communication platforms may also offer an invaluable perspective into immigration coverage for those who might wish to craft an ideal ethical model for covering the topic.
We demonstrated that comparative studies, including those from the UK, tend to concentrate on WEIRD countries, while there is a lack of comparative studies including Brazil and other countries in the global south. We would like to emphasize that comparative studies incorporating both the global south and north are essential for a better understanding of the similarities and differences at play in relation to media and immigration. The limitations of this research lie on the fact that it concentrates on only two countries. Further research focusing on additional nations from the global north and global south can offer more insight into other similarities and differences. We would also recommend further research comparing different countries across Asia, Africa and Latin America with countries from the global north. Strengthening and internationalizing research that is produced in the global south is valuable for a broader, decolonizing perspective, since scientific production is often westernized.
Finally, a systematic literature review has by its very nature several limitations; a keyword search, for example, can sometimes exclude essential research results. Further systematic literature reviews and meta-analyses using additional research criteria may be able to point to other important avenues of research. We searched for terms such as migration, immigration, and refugees. Additional studies can include keywords such as alien, illegal aliens, asylum seekers, international students, or expatriate.
This study contributes to our understanding of the representation of immigration in Brazil and the UK media analyzed in the studies. This paper reveals the similarities and differences between studies that examine media and migration in both the UK and Brazil. It highlights a standardization of framing, the prevalence of a negative tonein media analyzed in the studies, the silencing of migrant and refugee voices, and differences in applied research methods. Our research and its results cover countries from both the global north and the global south; it is evident that there is a repetition of frames related to immigration issues in both. Legacy media usually represents the topic negatively, focusing on aspects such as threats to the economy, security, or safety. Hence, the findings show that mass media generally contributes to negative stereotyping, which is shown to contribute to the increase of hate speech and violence towards immigrants and refugees. To improve the representation, mass media should emphasize immigrants' and refugees' contributions to increase positive associations and adopt a balanced view that highlights different angles.