Journalistic performance in Latin America: A comparative study of professional roles in news content
In the past decade, the interest in journalists’ conceptions of their professional roles, missions and functions as well as the contexts that shape them has increased (Christians et al, 2009). Frequently, studies on professional roles have focused on the conceptions of journalists based on single-country cases, types of media or news beats. More recently, scholars have compared journalists’ role conceptions cross-nationally at a global scope (Hanitzsch, 2011). This comparative research has contributed enormously to the field by mapping differences and similarities across countries, as well as by empirically supporting the argument of context-bound journalism cultures, particularly in developing countries and transitional democracies.
However, despite this burgeoning literature,, there is still insufficient cross-national research on three key aspects that might help to better understand role variation across countries and how or why these roles are put into practice. First, while a consistent body of research dealing with comparative reporting styles and media content confirms that content variations are connected to national journalistic cultures, such studies deal with professional roles studies only indirectly. In fact, fewer studies address the way in which professional role conceptions materialize in news content through specific indicators of “role performance” (Mellado, 2015). Highlighting the limitations of comparing journalistic cultures based solely on journalists’ conceptions of professional roles, some scholars has long insisted on the need to examine to what extent professional roles materialize in actual new content (Tandoc, Hellmueller and Vos, 2012; Mellado and Van Dalen, 2013; Mellado, 2015; Carpenter, Bohemer and Fico, 2015; Phil-Thingvad, 2015). As Mellado suggests (2015: 596-97) while the study of role conceptions deals with the analysis of journalistic culture at the abstract level, the study of role performance deals with “behaviors (actions, process or gathering of reactions) influenced by reference groups, which leads to performing a task or function in a specific group, organization or society…. a collective outcome of dynamic negotiations.”
Second, most comparative studies on role conception –and the few dealing with role performance— confirm that national level variables such as journalistic culture do vary across countries and therefore have a strong context-bound element to it. The studies in question mainly confirm existing differences at the media system-level, further visibilizing the strength of literature about Western media systems. However, there are still research opportunities in comparative research about how professional roles vary—or not—across less-studied countries with hybrid media systems and no clear-cut journalistic traditions, such as those in Latin America and other transitional democracies.
Finally, while there exists vast research on the impact of organizational, individual or national factors in shaping role conception, this is not yet the case for studies on role performance.
This study tackles the aforementioned issues by going a step further and comparatively anayzing journalistic role performance in a still under-researched area in the English-language and global literature. Taking five countries of Latin America as case studies –most of which share key similarities in their hybrid media systems—this paper examines professional role performance in national newspapers with two aims: first, analyzing how professional roles materialize in news content and map the variation across countries. Second, exploring whether other organizational and media variables such as newspaper’s political orientation, newspaper type or news topic predict the performance of models of role performance.
The study comprises the Latin American leg of a larger, cross-national comparative study set to analyze how different professional roles materialize in news content in different political/societal systems. Inspired by the work of Mellado (2015), Mellado and Van Dalen (2014), Tandoc et al (2013) and Vos (2005), among others, this project develops six models of journalistic role performance – all of which are comprised by different indicators of professional practice – based in three main domains: a) the presence of the journalistic voice, b) the relationship that journalism holds with institutional and the facto powers; and c) the way journalism approaches the audience.
In the journalistic voice domain we find a first model that measures the ‘interventionist’ or active stance of journalists versus their passive stance in their reporting.
In the second domain related to journalists’ position towards institutional power, two further models can be identified: the ‘watchdog’ and the ‘loyal-facilitator.’ The first one is the monitoring, often antagonistic position that journalism takes in order to hold institutional and the facto powers accountable, bringing attention to wrongdoing. The ‘loyal-facilitator’ model, meanwhile, materializes in two ways. Either journalism cooperates with those in power, or focuses on the nation-state helping to strengthen and visibilize national triumphs and prestige.
Finally, the third domain measures the different ways journalism approaches the audience, identifying three models of role performance: the ‘service’, the ‘infotainment’, and the ‘civic.’ The service model of journalism views the public as a client, providing information, knowledge, and advice about goods and services that are useful for their daily lives. The ‘infotainment’ model of journalism addresses the public as spectators, relying on different stylistics, often dramatic or sensationalist narrative and/or visual discourses to entertain and thrill the public. Finally, the ‘civic’ model pursues democratic goals and deems the public as citizens whom ought to be empowered, educated and informed on complex and controversial topics.
From journalistic role conception to journalistic role performance
Survey research on journalists’ role conception has one of the most established traditions within the journalistic field. It initially focused on single-country cases such as the United States, and on statements about the functions of journalism that reflect the debates between “subjectivity versus objectivity, detachment versus advocacy, and observer versus watchdog” (Johnstone, Slawski and Bowman, 1972: 522). Follow-up surveys by Weaver and colleagues gradually added more roles to capture the complexities and changing nature of media industries and the workforce (Weaver and Wilhoit, 1996; Weaver et al, 2007; Weaver, 1998, Weaver and Willnat, 2012). Their work subsequently served as a template for survey research on role conception not only in other countries, but to improved versions of role conception surveys in recent cross-national research (Hanitzsch, 2011). In Western Europe, where contrasting distinctive media systems and reporting cultures coexist (Hallin and Manicini, 2004), comparative research on role conception has specifically focused on the partisan vs. objective polarities, and in the position and tone that journalists take in the face of institutional actors and events (Patterson and Donsbach, 1996; Donsbach and Patterson, 2004).
With respect to the materialization of roles in news content –although indirectly addressed as role performance— various studies on comparative content analysis found that reporting styles do tend to vary across countries, reflecting their respective political and journalistic cultures (Alexseev and Bennett, 1995; De Vreese et al, 2001, Benson and Hallin, 2007; Esser 2008).,A study by Van Dalen (2012) exploring how journalists covered the Executive and Legislative branches of the government in Denmark, Germany, the UK and Spain found that variation in news reporting was better explained by nation and journalistic culture than by media type or the medium’s political orientation. Another study specifically connecting role conception with role performance by Van Dalen et al (2012) again found that role conception and role performance vary more across countries than within countries. Similarly, statistical tests carried out by Esser and Umbridtch (2014: 236) in their comparative study of objective vs. interpretive newspaper reporting in six European countries, confirmed that effect sizes were greater between media systems than newspaper types.
This literature confirms the defining importance of “country” to explain variation in role conception across countries, where organizational influences, news beats and market orientation are also important variables (Hanitzsch and Mellado, 2011; Skovsgaard and van Dalen (2012; Relly, Zanger and Fahmy; 2015; Scherr and Baugut, 2016). One of the few studies of role performance in Latin America is the work by Mellado and Lagos (2014), who found a direct relation between news beat and the materialization of roles in Chilean newspapers. For example, the infotainment, the service and the civic models (See Mellado and Van Dalen, 2016, for a theoretical and methodological discussion) are present in different thematic beats.
While the vast literature has also shown that news beat, political orientation or media type are all important definers of professional role conceptions and performance, it has been a line of research underexplored outside traditional Western countries with distinctive media systems (Mellado, 2015; Wahl-Jorgensen and Hanitzsch 2009).
In Latin America, survey research on journalists’ roles has been less consistent than in other countries, and has generally focused on single-country samples. Combined findings found no clear-cut regional models of role conception as the endorsement of watchdog, propagandist or missionary functions vary across countries (Arroyave and Barrios, 2012; Herscovitz, 2004; Mellado, et al, 2011; Mellado, 2012; Oller et al, 2015; Oller et al, 2016). The work by Mellado et al (2012) is one the first to analyze role conception cross-nationally, finding contrasting differences in role conception across Chile, Brazil and Mexico. For their part, qualitative studies also support the idea of a lack of clear-cut journalistic models across the region, as typical Western roles are not always adopted straightforwardly, but instead, adapted, re-interpreted (Mellado and Humanes, 2012), re-signified or challenged through the local contexts and conjunctures (Waisbord, 2000; Márquez Ramírez, 2012; de Albuquerque and da Silva, 2009).
In summary, the combined results of Latin American research on role conceptions show no clear or consistent results across the region. Nevertheless, to our knowledge, no comparative studies have been carried out comparing role performance cross-nationally.
Latin America media systems, journalistic cultures and political contexts
The local contexts of Latin American countries have been crucial to the shaping of media systems in the region (Fox and Waisbord, 2002; Waisbord 2013), and may also help explain role performance variation across countries. External factors that may account for the context in which professional roles materialize in news content are, for example, deep social inequality, decay of political institutions, weak rule of the law, human rights violations, widespread corruption, raising crime and violence in many countries, as well as widespread social protests. Other media system-characteristics that could also help explain at the contextual level the variations of role performance across countries are unique to Latin America, such as high media concentration, high levels of clientelism, high political parallelism and high levels of instrumentalization of journalists (Hallin and Papathanassopolous, 2002). In few other regions, private media corporations flourished under the legal, economic and political protection of dictatorial and authoritarian regimes, and later grew in gigantic proportions with the wave of economic liberalization and political democratization. This resulted in neo-conservative media elites that continue to negotiate their collusion and alliances with actors in power in exchange for economic benefits and governmental advertising contracts, like in Mexico, Chile or Brazil (Mastrini and Bolaño, 2000; Rockwell and Hanus, 2002; Guerrero and Márquez Ramírez, 2014). Depending on the political context, the press-state relations across the region may become volatile and even antagonistic. Proposals of media reform posed by more-socialism-oriented rulers in Argentina, Bolivia, Ecuador, Venezuela or Nicaragua –with the alleged purpose of countering media concentration and fostering diversity and real public service— have raised a great deal of social polarization and antagonism between the government and some media, amid fears of the use of ‘public’ service media for propagandist aims or the use of legislation to punish or reward enemies or allies (Lugo-Ocando, 2008; Waisbord, 2013; Guerrero and Márquez Ramírez, 2014).
With this in mind, at the hypothetical level, the political and social context may prompt watchdog and civic journalism, whereas either collusive or antagonistic relations between political and media elites, as well as the political instrumentalization of journalists, the growing attacks to the press, or the compromised press freedom in some countries (Hughes and Lawson, 2005) may curtail both roles. A selective or enforced version of political parallelism and varying formality of state intervention–to borrow Hallin and Mancini’s (2004) concepts— may eventually favor either a more disseminative or interventionist type of reporting, as well as a loyal role model of performance. Likewise, the concentrated nature of some media systems and the private, pro-market orientation in others might also facilitate the materialization of the service and infotainment role performance models.
In the context of a hybrid media system that combines features of a market-led type of liberal system –private property of the media but with elite press– with those of a pluralist-polarized one in Hallin and Mancini’s (2004) seminal categorizations –medium to high political parallelism and instrumentalization of journalists, as well as formal and informal means of State intervention—plus the exogenous factors of weak rule of the law, clientelism and violence, we can expect that the actual role performance be ambiguous or tense.
However, at the regional level there is not standardized operationalization neither reliable index that enables the cross-country comparison of Hallin and Mancini’s variables such as market development, political instrumentalization, political parallelism or State intervention across countries. For that reason, these aspects will be taken into account as means to contextualize the region media system and interpret our results, but not as empirical and measurable variables.
Due to the lack of comparative literature of professional roles in the region and the scarcity of literature on role performance at the global level, there is not enough evidence to predict the prevalence of specific models in each country. The study has therefore set the following research questions and general hypothesis:
RQ1: What are the journalistic role performance models that predominate in Latin America and how do role indicators vary across countries?
RQ2: What variables best explain variance of role performance across countries?
We conducted a content analysis of print news stories published in the most important newspapers from Mexico, Chile Ecuador, Cuba and Brazil during 2012 and 2013, based on standardized operationalizations of role performance models of journalism. Countries included here are those also participating in the global study mentioned above. Despite their overall similarities, they do uphold a variety of contextual factors at the national and cultural level that make them different from others: Brazil, for example, has the longest tradition of adoption of US values and investigative journalism (Herscovitz, 2004; Matos, 2008). Cuba has a State-media system with salient propagandist roles and interventionist-type of missions (Oller et al, 2016). Ecuador is the country undergoing extensive media reform and social polarization due to antagonist press-state relations at the time of the study, which has impacted in the role conception of journalists (Oller et al, 2015). Mexico has a very concentrated media system and a long tradition of varying forms of authoritarianism that has resulted in variation of role conceptions across newsrooms (Hughes, 2006). Chile was the first neoliberal laboratory under dictatorial rule, is the best-ranked democracy in the region, and has the least diverse print press market, but also a history of interventionist and tendencies towards infotainment (Mellado and Lagos, 2014)
Three to four general-interest newspapers focused primarily on national news were content-analyzed in each country. They are most influential, better known or representative of the niche market. Despite the prevalence of other media formats, newspapers still tend to set the political agenda, assign more financial and human resources to the coverage of a wider range of news content, and in theory provide more diverse and nuanced coverage of controversial issues. Table 1 shows the sample distribution per newspaper, as well as their political and market orientations.
— Table 1 —
The timeframe for the content analysis was January 2nd 2012 to December 31st 2013. By using the constructed week method a stratified-systematic sample was selected for each of the newspapers. Our sample included editions from every day of the week and month, to reach the maximum diversity: in total, two constructed news weeks were sampled per newspaper per year. In total, our sample consists of 9,841 articles, of which 2,582 news articles are from Chile, 3,009 from Mexico, 627 from Cuba, 874 from Ecuador and 2,749 from Brazil.
The unit of analysis was the news story, but only those published in sections associated to the national or main desks, leaving international, regional, science, sports, lifestyle, entertainment and celebrity news stories aside. Coding teams in each country received training in the application of a common codebook translated from English into Spanish and Portuguese. Different coder-trainer tests were performed to ensure that they had a similar understanding of the codebook. Intercoder-reliability post-tests based on Krippendorff’s alpha (Ka) was .78 for the case of Chile, .76 for the case of Mexico, .75 in the case of Bazil, .72 in the case of Ecuador, and .79 in the case of Cuba.
The coding manual included variables to operationalize the six models of role performance included in this study, based on previous research on the topic (Mellado, 2015; Mellado and Van Dalen, 2014, 2016).
The six models of journalistic role performance were examined through the explicit presence of specific indicators. Five indicators where used to measure the presence of the ‘interventionist’ model: opinion, interpretation, proposal/demands, adjectives, and the use of first person on the part of the news author. Four indicators measure the ‘service’ model: impact on everyday life, tips and advice (grievances), tips and advice (individual risks) and consumer advice. Six indicators measure the ‘infotainment’ model: personalization, private life, sensationalism, scandal, emotions and morbidity. Nine indicators measure the ‘civic’ approach: citizen perspective, citizen demand, credibility of citizens, education on duties and rights, background information, and local impact.
Ten indicators measure the ‘watchdog’ model: information on judicial/administrative processes, questioning on the part of journalist, questioning on the part of others, criticism on the part of journalists, criticism on the part of others, denouncement on the part of journalist, denouncement on the part of others, reporting of external investigation, reporting of conflict, or investigative reporting. Finally, nine indicators measure the ‘loyal’ model: defense/support activities, defense/support policies, positive image of the political elite, positive image of the economic elite, emphasis on progress/success, comparison to the rest of the world, emphasis on national’s triumphs, promotion of the country’s image, and patriotism. Each of these indicators required the coding of manifest content, on a presence (1) or absence (0) basis. The items were recoded, so that higher compulsive scores of all items combined (range: 0-1) would result in a final score for each model for each article. A higher score thus expressed higher presence of those journalistic role performance models, and vice versa.
For descriptive purposes we calculated the raw scores (sum of points divided by the total of items in each model). Meanwhile, we used factor scores to test for differences in presence of the six models of roles performance depending on country and organizational level variables.
Prior to our main analyses we conducted a confirmatory factor analysis (CFA) per each analyzed domains: journalistic voice, power relations and audience approach. CFA was performed using Mplus 7.0. Within that framework, we empirically tested competing measurement models.
Within the journalistic voice domain, the interventionist model showed a very satisfactory fit with the data (χ² = 6256.3, p < .001, RMSEA = .018 [90% CI = .010: .025], CFI = .997, TLI = .996, WRMR = 1.032). Within the power relations domain, the watchdog and the loyal-facilitator models provided a good fit with the data (χ² = 1620.3, p < .001, RMSEA = .031 [90% CI = .030: .034], CFI = .957, TLI = .952, WRMR = 1.102).
Finally, within the audience approach domain, the model composed by the service, infotainment, and civic approaches showed a satisfactory good fit with the data: χ² = 1023.2, p < .001, RMSEA = .025 [90% CI = .023: .028], CFI = .977, TLI = .971, WRMR = .984. We identified each solution as providing a better account of the data, than other competing solutions.
To address our first research question we compared the presence of the six analyzed professional role models, looking at the mean scores for each role performance model, as well as the percentages of indicator presence of each model. The significance of differences across countries was determined through analysis of variance (ANOVA).
Voice of the journalist: Interventionist model
Our data shows that within the journalistic voice domain, Latin American journalism is more passive tan active, with the interventionist role present to much lesser degree in the press than the disseminator role in all five countries (M= .16, SD= .20 for the interventionist role). Nevertheless, the presence of the interventionist model significantly varies in comparative terms among the analyzed countries (F= 177,743; df= 4; p= .000). For this model, the differences in the national context accounted for more than 8% of the overall variance (η2)
The data shows that in general, the interventionist role model is comparatively more present in Chile than the rest of the countries, closely followed by Cuba and much farther by Mexico. In the opposite pole of the spectrum, Brazil and Ecuador are the countries with the lowest presence of journalistic voice and intervention in the print press (see Table 2).
Specific indicators for the interventionist model tell us a more nuanced story. The use of adjectives, interpretation and opinion were the most frequent indicators across countries. In contrast, the use of first person, and the proposal or demand for change on the part of the journalist is significantly less present. The Chilean press significantly peaks in the use of adjectives. Cuban journalists rank higher in the use of opinion, first person –which is generally low across countries—and interpretation, giving credit to their historical tradition of opinion and militant journalism. Finally, the “proposal and demands” indicator is generally low in all countries, with slightly more presence in Chile, Mexico and Brazil. (See table 2).
Power relations: Watchdog and loyal-facilitator models
Considering all Latin American data, journalists from the region tend to give almost similar importance in their reporting styles to both the watchdog (M= .059; SD= .112) and the loyal facilitator (M= .052; SD= .137) role models. It is an important finding because it reflects the myriad paths that press in undergoing: while some stories and events do appear to merit deep scrutiny, others still display reporting practices oriented towards publicising the official agenda and spin.
Nevertheless, when comparing national contexts, significant differences emerge in both the watchdog (F=348,482; df= 4; p= .000) and the loyal journalism model (F=736,771; df= 4; p= .000), fully supporting our main hypothesis The effect size is quite large in the case of both the loyal (η2= .230), and the watchdog role (η2= .125) accounting for the 23% and the 13% of the overall variance, respectively.
Brazil and Mexico display the highest presence of the watchdog model, although mainly through the use of indicators such as third party questioning or criticism; while in Chile, the model is practically non-existent (See Table 3). These results resonate with the study of Mellado et al (2102) on Latin American role conceptions, wherein Brazilian and Mexican journalists gave significantly more support to acting as a watchdog of the government than Chileans did. Also, Brazilians have a more steady tradition of watchdog and investigative journalism (Waisbord, 2000; Herscovitz, 2004), while in Mexico, some newspapers are known critics of the governing party (Hughes, 2006). Meanwhile in Ecuador there is a low presence of the watchdog model in general but its press does peak at two key indicators: questioning on the part of the journalist and reporting overt conflict between journalists and politicians, which could be a manifestation of the press’ antagonist relations with the government (Oller et al, 2015). As expected, a control-led press system like the one prevailing in Cuba is less likely to exert watchdog journalism, whereas for the case of Chile, there has been historically practices of censorship and self-censorship, and as a result, little investigative journalism in newspapers (Lagos, 2009; Leon- Dermota, 2003).
— Table 3 —
The results also show that the loyal-facilitator model is, by far, most represented in the Cuban press, as expected (Oller et al, 2016). In fact, as we shall see in the following section of this article, the variable “country” is significantly higher to predict this model in Cuba. Cuban journalism, due to its high level of instrumentalization and political parallelism is at the top of all the indicators for this model, while in Chile, as reported, the model practically disappears, despite its high political parallelism. This result is consistent with Mellado and Lagos’ (2014) claims that the absence of an explicit manifestation of the loyal-facilitator role masks the existence of more implicit or indirect forms of loyal reporting, such as reliance on official sources.
— Table 4 —
Audience approach: the service, infotainment and civic models
Latin American journalism overall, performs the civic role (M= .055; SD= .128) slightly higher than the infotainment (M= .049; SD= .121) or the service role (M= .042; SD= .130). Nevertheless, when comparing by country significant differences emerge in the presence of these three models. The effect sizes were small but still relevant for the service role (F= 90,142; df= 4; p= .000; η2= .035), the infotainment role (F= 94,824; df= 4; p= .000; η2= .037), and for the civic role (F= 40,334; df= 4; p= .000; η2= .016).
The service model is the least present in the Latin American press, but its presence is higher in Ecuador, in all of its indicators, followed by Brazil, Chile, Mexico and finally, Cuba (See Table 5). The predominance of the service role model in Ecuador might reflect a strategy on the part of the journalists to avoid confrontation with the government and at the same time to comply with new legislation that enforces the media to provide content that proves itself useful to audiences (Oller et al, 2015).
— Table 5 —
The infotainment model is much more evident in Chilean journalism, followed by the Brazilian and Ecuadorian press. Mexico is further beyond the trend while in Cuba the model is nearly non-existent (see Table 6). However, specific indicators are much more varied. For example, Chile is in the top of sensationalism, the focus on private life and the use of emotions in the news. However, personalization is more frequent in the Brazilian press, while scandals are mostly found in the Ecuadorian press, and morbid information in the Mexican press, the latter a reflection of the inclusion of a known tabloid that prints front-page stories and pictures of corpses and crime victims.
— Table 6 —
Finally, results show the greatest presence of the civic model in the Mexican and the Ecuadorian presses, followed closely by Cuba (See Table 7). In Brazil and especially in Chile, the model is, comparatively, less existent. Specific indicators for this model also vary across countries, as has been the trend with other models. While the focus on citizen perspectives, local impact of news and the addressing of citizen questions are indicators more frequently found in the Ecuadorian press, news that incorporate citizen demands, provide background information, educate on duties and rights, provide information on citizen activities and support citizen movements are more present in the Mexican press. The results for Ecuador again might reflect new legislation on journalists’ public service mission, while Mexican results can be explained by the fact that two newspapers in the sample are perceived to be pioneers of civic journalism in the country (Hughes, 2006), and by the context of questioned presidential elections during the period of study, as well as drug-cartel violence and human right abuses. Cuban journalism, in turn, strongly focuses on educational and cultural values as its mission (Oller et al, 2016)
— Table 7 —
Modeling journalistic role performance in Latin America
Our second research questions aims to better explain how organizational and internal-level factors help to predict and explain variation of journalistic role performance, besides country. To address this research question we conducted a multiple linear regression analysis with each model as a dependent variable, and the following variables as predictors: country, media type, media political orientation and item topic.
— Table 8 —
The first column presents the results for the interventionist model, with most of the variables having significant effects, explaining the 10% of the total variance. Controlling for the other variables included in the model, the strongest predictor is “country.” Specifically, Chilean newspapers are more likely to include the interventionist model in their reporting style. Likewise, the interventionist model tends to be more absent in center and right-leaning media, whereas media type is also a significant predictor, where the popular press tends to use this model more than the elite press. Finally, social affairs and especially miscellaneous news tend to display an interventionist type of reporting more than political news.
The second column presents the results for the watchdog model, with all the variables having significant effects and explaining the 21% of the variance. Again, the most important predictor of watchdog journalism is the country, confirming the most developed tradition of watchdog journalism in Brazil. However, the medium’s political orientation is also a very important predictor, where right and center-leaning newspapers are less likely to put this model into practice than left media. Finally, in comparison to political news, topics such as economy and business, as well as miscellaneous news are les likely to display the watchdog model, while court and police and crime news are more likely to put this reporting style into practice than political news.
The third column presents the results for the loyal model, with all the variables having significant effects and explaining 26% of the total variance for the model. In this case, the most important predictors are country, and media’s political orientation. Controlling for the other variables, Cuba and Ecuador, two countries with very interventionist types of government in their media systems, tend to be more loyal in their news media than Brazil, Chile and Mexico. Likewise, the right and center-oriented newspapers are less likely to be loyal to the government in their reporting style than the left–leaning media. As for news beat, the loyal model tends to be more present in political news than any other topic–with the exception of economy and miscellaneous news, with not significant differences. Likewise, the popular press tends to use this model significantly less than the elite press.
The fourth column presents the results for the service model, explaining the 14% of the total variance for the model. The most important predictor is the media’s political orientation. Left-leaning media tend to use this model of performance much more than the right and center-oriented newspapers. Regarding media topic, economy/business news tend to incorporate this model of reporting style and focus on audiences specific needs more than political news do, while the popular press tends to be more service oriented than the elite press. Country is also a significant an important predictor. Controlling for the rest of the variables, Ecuadorian press tend to be more service oriented than Brazil, while all the rest of the analyzed nations tend to be less service oriented than Brazil.
The fifth column presents the results for the infotainment model, explaining 14% of the total variance for the model. The most important predictor for this model is media type, where expectedly the popular press tends to use it much more than the elite press. In the same line, the left-wing press is more prone to the infotainment model than the center and right-leaning press. Country is also an important predictor controlling for all the other variables included in the model, with news from Cuba, Ecuador and Mexico less prone to entertainment than Brazilians, and Chilean papers slightly higher than Brazil in their use. Finally, the infotainment model is more present in news about police and crime, miscellaneous and social affairs, than on political news. However, political news are more likely to display infotainment elements than news on court or economy and business.
Finally, the last column shows the results for the civic model, explaining 14% of the total variance for the model. Overall, the most important predictor is the political orientation of the media, with left-wing oriented media expectedly more likely to include this model of performance in their news coverage. Likewise, topics like social affairs are more likely to be related to this type of reporting style than political news, while news on court are less likely to include this journalism model. Media type, meanwhile, is not a significant predictor for this model. Finally, controlling for all these variables, the civic model still tends to be more present in Brazilian news than in the news published in Chile, Cuba and México.
The aim of this study was twofold. On the one hand, we wanted to test how the performance of different journalistic roles varied across five Latin American countries. On
the other hand, the article explored whether variables such as country, media type,
media’s political orientation, and news beat predicted the performance of different
roles and which predictors explained them the most. The results show significant differences in the performance of the six analyzed roles and their indicators across countries, which seem related to the countries’ particular contexts and histories, specifically to their hybrid media systems and changing political environments. Our first finding is that there is no clear, prevalent regional model of journalism. Despite their highly instrumentalized nature and political parallelism, Latin American journalism tends to be highly disseminative and more passive than interventionist in nature. The interpretive, militant type of journalism often found in countries of the polarized-pluralist model like Spain and Portugal is not quite replicated across the ocean, in their former colonies, as we found little evidence of a clear-cut advocacy press overall, except in
Cuba where the expectations were met regarding its highly intervened and controlled
media system. This was the country that mostly manifested the ‘loyal’ role and key
indicators of the ‘interventionist’ role. Despite this result and apart from Cuba, Chile
still has the highest presence of the ‘interventionist’ role – a likely trace of the overtly
partisan press that once existed.
Likewise, while the ‘watchdog’ and ‘civic’ roles have an overall low presence across
countries, in comparative terms they rank as the second and third among all the six analyzed dimensions of role performance, specially in Brazil and Mexico, probably reflecting the nature of their presses but also the political and social contexts of both countries.
The ‘service’ role is very low across countries, which is nevertheless mostly present in
the Ecuadorian press, presumably reflecting the demands of new governmental reforms
on the media that want to ensure public service content is met. Finally, the ‘infotainment’ role is likely to be more present in popular newspapers, miscellaneous as well as police and crime news, and also in Chile.
The results from our regression analyses helped to support the idea that the nationlevel
does play a very important part in shaping role performance, specially those that
involve the relation between journalism and established powers, such as ‘loyalty’ vis-avis
‘watchdog’ news, and also when talking about the presence of the journalistic voice
in the news story. However, other types of news, which are more audience-oriented, tend
to be better explained by either the type of media, the topic of news, or the political orientation of the news outlet.
We believe these findings shed some light on the dynamics of journalistic performance
and the multiple factors that shape it. While several organizational variables are
important in the prediction of the performance of roles, the article confirms that the
‘country’ variable continues to be a good predictor of democratic roles.
Nevertheless, there are challenges and limitations that need to be addressed for future
research. First, there is a need for theorizing and systematizing media system variables
beyond Western countries. Second, the lack of reliable, systematic country-level indexes
that can help measure and quantify media system variables – such as those proposed by
Hallin and Mancini (2004) – is a limitation to develop more nation-level indicators of
variance, and generally a limitation to comparative research across countries. Third,
more longitudinal research is needed to ascertain whether or not variation across countries in the performance of some roles is due to the specific conjunctures of the time
period comprising the study or are steady trends that are consistent over time. Finally,
future research should also study role performance on different news platforms, to see
how this new variable predicts and explains variation.
Alexseev, I. and W. L. Bennett (1995) For whom the gates open: News reporting and government source patterns in the United States, Great Britain, and Russia, Political Communication, 12(4): 395-412.
Arroyave, J. and M.M. Barrios (2012) Journalists in Colombia, in Weaver D and Willnat L (Eds) (2012) The Global Journalist in the 21st Century. New York: Routledge.
Benson R and Hallin DC (2007) How states, markets and globalization shape the news: The French and US national press, 1965–97. European Journal of Communication 22(1): 27–48.
Lugo-Ocando J (Ed) (2008) The media in Latin America. Maidenhead: McGraw-Hill-Open University Press
Carpenter, S., Boehmer J and Fico F (2015) The Measurement of Jounalistic Role Enactments: A Study of Organizational Constraints and Support in For Profit and Nonprofit Journalism, Journalism and Mass Communication Quarterly. Doi:10.1177/1077699015607335
Christians CG, Glasser TL, McQuail D, Nordenstreng K and White RA (2009) Normative theories of the media: journalism in democratic societies. Urbana: University of Illinois Press.
De Alburquerque A and Roxo de Silva MA (2009) Skilled, loyal, and disciplined: Communist journalists and the adaptation of the model of the American model of independent journalism in Brazil. The International Journal of Press/Politics 14(3): 376-395.
De Vreese C, Peter J and Semetko H (2001) Framing politics at the launch of the Euro: a cross-national comparative study of frames in the news, Political Communication 18(2):107–122
DiStefano C, Zhu M and Mindrila D (2009). Understanding and using factor scores: Considerations for the applied researcher. Practical Assessment, Research & Evaluation, 14(20), 1-11.
Donsbach W and Patterson T (2004) ‘Political news journalists: Partisanship, Professionalism and Political Roles in Five Countries’, In: F. Esser and B. Pfetsch (eds.) Comparing political communication: Theories, cases, and challenges. Cambridge: Cambridge University Press
Dooner P (1989) Periodismo y política: La prensa política en Chile, 1970-1973 [Journalism and politics: the political press in Chile, 1970-1973]. Santiago: Andante.
Esser F (2008) Dimensions of Political News Culture: Sound Bite and Image Bite News in France, Germany, Great Britain, and the United Kingdom, The International Journal of Press/Politics 13:401–28
Esser F and Umbricht A (2014) ‘The Evolution of Objective and Interpretative Journalism in the Western Press: Comparing Six News Systems since the 1960s, Journalism and Mass Communication Quarterly, 91(2): 229-249
Fox E and Waisbord S (Eds.) (2002) Latin politics, global media. Austin: University of Texas Press
Guerrero MA and Márquez-Ramírez M (2014) Media Systems and Communication Policies in Latin America, London: Palgrave Macmillan
Hallin D and Mancini P (2004) Comparing Media Systems: Three Models of Media and Politics. New York: Cambridge University Press.
Hallin D and Papathanassopoulos S (2002) Political clientelism and the media: Southern Europe and Latin America in comparative perspective. Media Culture & Society 24 (2): 175-195.
Hanitzsch T (2011) Populist Disseminators, Detached Watchdogs, Critical Change Agents and Opportunist Facilitators: Professional Milieus, the Journalistic Field and Autonomy in 18 Countries. International Communication Gazette 73(6): 477-494.
Hanitzsch T and Mellado C (2011) What shapes the News Around the World? How journalists I Eighteen Countries Perceive Influences on their Work? International Journal of Press-Politics, 16(3): 404-426
Herscovitz H (2004) Brazilian journalists’ perceptions of media roles, ethics and foreign influences on Brazilian journalism. Journalism Studies 5(1): 71-86.
Hughes S (2006) Newsrooms in conflict. Journalism and the democratization of Mexico. Pittsburgh, PA: University of Pittsburgh Press.
Hughes S and Lawson C (2005) The barriers to media opening in Latin America. Political Communication 22(1): 9-25.
Johnstone J, Slawski EJ and Bowman W (1972) The professional values of American newsmen, Public Opinion Quarterly, 36 (4): 522-540
Lagos C (2009) El periodismo de investigación y el fortalecimiento de la democracia. La excepción a la regla [Investigative journalism and the strengthening of the democracy. The exception to the rule]. Légete 12(1), 121-132.
Leon-Dermota K (2003) And well tied down: Chile´s press under democracy. Westport, CT: Praeger.
Matos C (2008) Journalism and political democracy in Brazil. Lanham, MD: Lexington Books.
Márquez-Ramírez M (2012) Valores normativos y prácticas de reporteo en tensión: Percepciones profesionales de periodistas en México. Cuadernos de Información 30: 97–110.
Mastrini G and Bolaño C (2000) Globalización y monopolios en la comunicación en América Latina. Buenos Aires: Biblos.
Mellado C (2012) The Chilean journalist. In: Weaver D and Willnat (eds) The Global Journalist in the 21st Century: News People Around the World. New York: Routledge, 382-399.
Mellado C (2015) Professional roles in news content: Six dimensions of journalistic role performance. Journalism Studies 16(4): 596-614.
Melado C and Humanes M (2012) Modeling perceived professional autonomy in Chilean journalism. Journalism (13)8: 985-1003
Mellado C and Lagos C (2014) Professional Roles in News Content: Analyzing Journalistic Performance in the Chilean National Press. International Journal of Communication 8: 2090–2112
Mellado C and Van Dalen A (2013) Between rhetoric and practice. Explaining the gap between role conception and performance in journalism. Journalism Studies 15(6): 859-878.
Mellado C and Van Dalen, A (2016) Challenging the Citizen–Consumer Journalistic DichotomyA News Content Analysis of Audience Approaches in Chile, Journalism and Mass communication Quarterly. Doi:10.1177/1077699016629373
Mellado C, Lagos C, Moreira S, and Hernández ME (2012) Comparing Journalism Cultures in Latin America: The case of Chile, Brazil and Mexico. Gazette 74 (1): 60–77.
Oller M, Chavero P, Carrillo J and Cevallos P (2015) La autopercepción de los roles profesionales de los periodistas en Ecuador. Revista Quórum 12(1): 157-187.
Oller M, Hernández D, Argüelles L and Olivera D (2016) La ideología y la cultura profesional de los periodistas en Cuba. La Habana: UPEC
Patterson T and Donsbach W (1996) News Decisions: Journalists as Partisan Actors. Political Communication 13 (4): 455–468
Pihl-Thingvad S (2015) Professional ideals and daily practice in journalism, Journalism: theory, Practice and Criticism, 16(3): 392-411
Relly J, Zanger M and Fahmy S (2015) Professional Role Perceptions amonf Iraqui KKurdish journalists from a ‘state within a state’ Journalism, 16 (8): 1085-1106
Scherr S and Baugut P (2016) The Meaning of Leaning: The Impact of Journalists’ Political Leaning on Active Role Perception and Satisfaction With Audiences and Editorial Policy, Jounalism and Mass Communication Quarterly, 93 (1): 142-163
Skovsgaard M and Van Dalen A (2013) The Fading Public Voice: the polarizing effect of commercialization on political and other beats and its democratic consequences, Journalism Studies, 14(3):371-386
Tandoc E, Hellmueller L and Vos T (2013) Mind the Gap: Between Role Conception and Role Enactment. Journalism Practice 7 (5): 539–554.
Van Dalen A (2012) Structural Bias in Cross-National Perspective: How Political Systems and Journalism Cultures Influence Government Dominance in the News, in The International Journal of Press/Politics, 17 (1): 32-55
Van Dalen A, de Vreese C and Albaek E (2012) Different roles, different content? A four-country comparison of the role conception and reporting style of political journalists, Journalism, 13(7): 903-922
Vos T (2005) Journalistic Role Conception: A Bridge between the Reporter and the Press. Presented at International Communication Association (ICA) conference, Journalism Studies Division, New York, May 29.
Waisbord S (2000) Watchdog Journalism in South America. New York: Columbia University Press.
Waisbord S (2013) Vox Populista: Medios, Periodismo, Democracia. Barcelona: Gedisa
Weaver D and Wilhoit C (1996) The American Journalist in the 1990s: US Newspeople at the End of an Era, Mahwah, NJ: Lawrence Elbaum Associates
Weaver D, Beam R, Brownlee B, Voakes P and Wilhoit C (2007) The American Journalist in the 21st Century: US Newspeople at the Dawn of a New Millenium
Weaver D (1998) The Global Journalist: News People Around the world. Creeskill, NJ: Hampton Press.
Weaver D and Willnat L (Eds) (2012) The Global Journalist in the 21st Century. New York: Routledge.
 The authors wish to thank their country collaborators and/or research assistants for their invaluable help: Mexico: Cedric Gall, Luis Lemini, Patricia Maldonado, Lucano Romero, Rubén Delgadillo and Eréndira Martínez
 The six Latin American countries participating in the project are: Argentina, Brazil, Chile, Cuba, Ecuador and Mexico. However, the Argentinean team was till in data-collection stage by the time this regional study was conducted and had to be excluded from the sample.
 Since each model represents a latent variable, the factor score is technically a better measurement of this variable on having weighted the items or indicators, according to how much each item contributes to that latent variable (DiStefano, Zhu & Mindrila, 2009). In other words, factor scores are linear combinations of the observed variables, which consider what is shared between the item and the factor. This way of building overall scores, helping to address the following problems: First it assigns more weight to some items than to others; it avoids the problems of distortion in raw punctuations when variables have a very low or a very high mean (ceiling and floor problems), equalizes the relative difficulties of different indicators to within a specific dimension, and generates standardized scores similar to a Z-score metric, where values range from approximately -3.0 to +3.0.
 The model fit was assessed using the following criteria: chi-square value (χ2), the root mean square error of approximation (RMSEA) (smaller than .05), the comparative fit index (CFI) value (greater than .90), the Tucker–Lewis Index (TLI) value (greater than .90), and the weighted root mean square residual (WRMR) (less than 1.0) (Schreiber, Nora, Stage, Barlow, & King, 2006; Yu & Muthén, 2002).
 Two indicators were excluded since their factor loadings were non-acceptable nor significant for the model (“scandal,” “support of citizen movements”)