Never Say Never … Or the Value of Context in Political Communication Research

The need for context is so obvious in everyday life that we could ask why is it so often overlooked in political communication comparative research? The fact that political communication deals with things as bound to circumstances and therefore as context-dependent as communication and politics further substantiates the need to pay attention to context and to its implications in theory-building. The examples put forward by Rojas and Valenzuela (2019) show that without proper contextualization, research results can easily be misinterpreted and consequently misguide the ensuing theorizing efforts. The relation between social media news exposure and political polarization is a prime example of how context mediates effects and any attempt at generalization would result in flawed conclusions if the conditions under which occur the proposed relationships among variables are not examined thoroughly.

Most comparative research in political communication has been guided by an overarching objective of generalization that entails high levels of conceptual abstraction and is often accompanied by a resistance in acknowledging local specific contextual factors that if included would introduce a qualitative component in the research design. Such an approach could be perceived by some quantitavists as a fundamental weakness, but is central to qualitative approaches that perceive all meaning as a local construction, and thus grounded in a specific context. Hallin and Mancini (2017) consider that there is a methodological imbalance in political communication comparative research, that is, more quantitative than qualitative approaches. If we consider that qualitative and quantitative approaches to study democracy, for example, often stress different elements, this is not a minor detail. Powers and Vera-Zambrano (2018) refer instead to an epistemological imbalance and to the need for ‘contextualism,’ namely accounts that adequately explain the diversity of contexts and the mechanisms that determine the prevalence of differences or similarities across cases.

Fifty years ago, Sartori (1970) was already warning against conceptual stretching. According to him, the development of comparative studies meant “gains in extensional coverage matched by losses in connotative precision” (1970, p. 1035). To borrow from Sartori, often the option has been to cover more by saying less. There is a perceived gap between political theory and empirical studies (Fung, 2007) and the debate between those arguing for the need of concepts that can be applied across borders and those advocating the idea that “concepts do not travel” (e.g., Hantrais, 2009; Jenson, 1978) continues, but the role of context has been increasingly emphasized.

Such debates are particularly meaningful for comparative research that includes countries beyond the western world or when cross-national approaches include samples composed of a wider variety of countries. It could be claimed that when concepts travel they should always travel with company (i.e., contextualization). Context is key in different stages: to test the analytical framework, to select the sample, to operationalize concepts, and to interpret the analysis results. Theoretical concepts may have different meanings and gradations across countries, or in some instances reality may be subject to specific temporary circumstances that need to be acknowledged and explained, otherwise the research results might not be accurate. The long-standing goal of working with abstract concepts that are empirical testing-proof needs to be dealt with great caution, as a primary commitment to abstraction might result in overlooking other important aspects in comparative research.

Different Methodological Approaches, Geographical Areas, and Media Environments: Similar Conclusion?

Numerous examples of empirical research developed with both qualitative and quantitative approaches support the importance of further knowledge production about the mediating role of context. Recent research has been drawing further attention to the several layers of complexity that, for example, social media have added to media and political environments. In addition to structural factors, several contextual factors do influence the type and level of media impact on politics and on audiences (for recent accounts, see e.g., Park et al., 2018; Rossini, Hemsley, Tanupabrungsun, Zhang, & Stromer-Galley, 2018; Vu et al., 2018). Contextualization thus became even more relevant with the Internet and in increasingly complex media environments. There are not only new forms of media production, distribution, and reception, but there are also new types of political participation, new participants, and new forms of political expression, arguments and debates. The effects of these new communication environments are still largely undetermined, but they are influenced by context.

Lusophone Africa

Research on the role of the media in the democratization processes of the Lusophone African countries (Angola, Cape Verde, Mozambique, Sao Tome and Principe) has explicitly drawn attention to the importance of contextual knowledge (Salgado, 2014, 2018). Drawing on interviews, fieldwork and media analysis, my research revealed that, although the Internet has made possible the access to information from other countries (at least for citizens with access and skills), which could potentially suggest a trend toward homogenization, the Lusophone African countries displayed considerable differences in their understandings and adaptations of democracy, despite their common language area, with shared patterns of cultural influence and where all kinds of interchanges were thus facilitated.

The actual functions of online media were also dependent on context. Salgado (2014, 2018) developed a typology that enhances how different levels of freedom and democratic development cause differentiated media effects. Depending on the greater or lesser levels of freedom and journalistic independence, the online media fulfilled different main roles: as ‘alternative’ to mainstream media, as ‘complement,’ or as ‘reinforcement.’ In the environments with less freedom (Angola and Mozambique), the online media functioned mainly as an ‘alternative.’ There were several journalists and citizens publishing their own means of expression (e.g., blogs, online papers), on which many people relied to get information, as mainstream media were controlled by the State or by political elites. ‘Reinforcement’ occurred predominantly in Sao Tome and Principe, where there are higher levels of freedom, but very low journalistic professionalization and independence. The few online media that were available reinforced the agenda and frames of mainstream media, thus contributing to strengthening consonance. ‘Complementary’ was manifest in Cape Verde, where both freedom and journalistic independence were higher. In this country, the development of the Internet allowed the emergence of the only daily newspaper and because online media outlets have significant readership, they also had some influence in setting the agenda, framing issues, and giving voice to different sources of information and opinion. In terms of distribution and reach, the online media were extremely important for diaspora communities and in parts of the country where newspaper distribution was too expensive. Their main function was thus predominantly of complementarity.

Placing the Internet in the wider media environments, we can see that the Lusophone African countries’ political settings did shape in a differentiated manner the media effects and that context thus mediate these effects. The research has also demonstrated that concepts as democracy, freedom, or citizenship were subject to different interpretations in each of these countries and were often tailored to fit particular political settings. For example, in Sao Tome and Principe, the word democracy was directly linked to the level of economic development and measured by the construction of new buildings and roads.

Europe, Crisis and Elections

Based on a different methodological approach (quantitative content analysis), research on European countries has reached similar conclusions. Examining the media coverage of the first elections (in 2015 and 2016) after the Euro Crisis in the countries that were most affected by it (Greece, Ireland, Portugal, Spain), Salgado (2019) concluded that the attempt to achieve high levels of generalization through the empirical testing of political communication concepts, such as personalization, populism, negativity, or political news framing, was often hampered by the singularity of these countries in many respects. It was not possible to identify clear patterns of influence of the crisis. What stood out were some country specificities in particular regarding populism and political news framing.

Although some transnational trends in the news media coverage of politics and elections (as suggested in literature, e.g., Swanson, 2004; Swanson & Mancini, 1996) have emerged in the analysis, there was always an outlier or exceptions that pointed to the importance of the countries’ specific political, social and cultural context. This was, for instance, the case of Greece and Portugal in the dominant news frame: their 2015 election news coverages were more focused on issues than on strategy. The percentage of issue news frame was 57% in Portugal and 54% in Greece, while in Spain and Ireland it was 38% and 27%, respectively. As Greece and Portugal were severely affected by the Euro Crisis, it was not surprising the salience that the crisis as an issue had in their elections and news coverage. What was puzzling was the fact that the Euro Crisis issue was not important in Ireland and Spain. This is possibly explained by the different framing (and management) that the crisis had in Spain and to the fact that Ireland had exited the financial assistance program more than 3 years before the 2016 election, in 2013.

Although election news coverage has been increasingly framed as strategy rather than as a discussion of issues and candidates’ proposals (e.g., Patterson, 1993), crisis situations may influence the election news framing, but the specific national context may also hold some influence over the type of coverage, for example, due to journalistic cultures and/or the special circumstances of the crisis in each country.

The manifestation of populism in election news corroborates the mediation of context, which could help explain why, contrary to Spain and particularly Greece, Ireland and Portugal had a negligible prevalence of populism. Additionally, the most prevalent features of populist discourse in each country do not seem to be strictly a consequence of the crisis situation. In Portugal, the most common feature was the ‘reference to the people’, in Ireland anti-elitism, while in Spain and Greece it was the exclusionary discourse, substantiated in the ‘us and them’ division.

These examples of research focused on different issues and carried out in different places with different methodologies lend support to the idea that the relations between media and politics are better grasped within the economic, political, and media context. Context should ensure a better understanding of the possible variations that occur in the relations between variables. And this holds true in qualitative and quantitative empirical approaches, and in research that deals with western and non-western settings.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Fundação para a Ciência e Tecnologia, Portugal (IF/01451/2014/CP1239/CT0004).

Notes on contributors

Susana Salgado

Susana Salgado is FCT Research Fellow at the Institute of Social Sciences, University of Lisbon, Portugal.