Measuring differences in the Chinese press: A study of People’s Daily and Southern Metropolitan Daily

Abstract

The development of the market has produced a differentiation inside the Chinese press between an ‘official’ press with traditional propaganda functions on behalf of the Communist Party and a ‘commercial’ press whose objective is to maximise revenue. Scholarly opinion has differed over whether marketization undermines Party control and whether new forms of journalism have arisen that lead to conflicts. These discussions have rested on little evidence as to the practises of Chinese journalism. This article presents empirical data on the extent of the differentiation, reporting on a content analysis of the national news in People’s Daily and Southern Metropolitan Daily. These titles are popularly believed to represent the polar opposites of official, orthodox journalism and commercial, liberal journalism. The evidence presented here demonstrates that while there are indeed significant differences in the journalism of the two titles, there remains a substantial overlap in their choice of subjects, their use of sources and the degree to which news is presented ‘objectively’. Southern Metropolitan Daily does display some ‘popular’ features and does contain more ‘watchdog’ journalism, but it shares with its official cousin an emphasis upon the party as the source for news.

Keywords

China journalism newspapers

Introduction

A major consequence of the opening of China since the late 1970s was a shift in the economic model of the media from state subsidy to reliance on their own revenues. Before the late 1970s, advertising was unknown in the media of the New China but since then it has become the staple form of finance for both broadcasting and the printed press. The rapid increase in advertising revenue, which accompanied the fast-growing market sector of the economy, has led to an expansion in the number, scale and quality of all media outlets.

It has been argued that the shift towards the market has led to differentiation between newspapers concerned with the straightforward propagation of the Communist Party of China’s (CPC) “party line” and others concerned with reaching a wider audience in order to maximise both subscription and advertising revenue. The implications of this differentiation have been much debated, particularly with regard to the potential of a conflict between the propaganda needs of the CPC and the commercial needs of media outlets. Much less contentious has been the recognition that the shifting economic model has resulted in new forms of journalism. The more market-oriented newspapers, it is generally agreed, cover different topics using different journalistic techniques than were current in the Chinese press before the reform.

Empirical evidence in support of these opinions has been relatively scanty. This article seeks to remedy that deficiency by providing quantitative evidence of the differences between two Chinese newspapers. It presents a detailed comparison of the national news coverage in People’s Daily (Renmin Ribao, hereafter PD), which is the national mouthpiece of the central Party leadership, published in Beijing, and Southern Metropolitan Daily (Nanfang Dushi Bao, hereafter SMD), which is a commercially oriented newspaper in the province of Guangdong in southern China. This evidence is derived from a content analysis of national news in the two titles for selected dates in 2012 and 2013.

The article begins with a review of the structure of the Chinese press. It then discusses the provenance of the data used here, which was gathered as part of an international comparative project designed to explore aspects of journalistic professionalism. The research instruments developed for that purpose had the side-effect of allowing the comparison of aspects of journalistic practise between different newspapers. The article then presents the findings of the content analysis and concludes with a discussion of the implications of the results for our picture of the Chinese press today.

The impacts of marketization

The newspapers press in mainland China has undergone an economic transformation, beginning with the economic opening in the late 1970s and continuing more or less uninterrupted ever since (Chu, 1994; Polumbaum, 1990; White, 1990). With the active encouragement of the Party, newspaper managements have moved away from state subsidy towards becoming self-sustaining and, if possible, profitable (Zhao 1998, Zhao 2008, Lee, Zhou and Yu 2007, Lee, He and Yu 2006). As the conditions under which this shift was taking place have been favourable, with an urbanising population, rising living standards and a rapidly developing advertising industry, the marketization of the press was extremely successful (Lee, 1994).

While this process of marketization has continued up to the present, it has only been one part of the overall picture. Party control has remained a constant, although its mechanisms have changed somewhat (Esarey, 2006; Hassid, 2008, pp. 145-168). Private ownership of newspapers remains forbidden, all publications require a licence from an appropriate Party or state body, and all are supervised by the Party Propaganda Departments of the Communist Party of China (CPC; Hu, 2003).

Within this structure, a division of labour has developed. ‘Official’ Party publications, like PD, are primarily concerned with interpreting the world from the point of view of the CPC: they remain dedicated to the classic propaganda functions of the press. There are also commercially oriented publications that, while they retain a political function, are more oriented on circulation building and advertising sales than on propaganda. It is often claimed that members of this latter category constitute ‘tabloid’ newspapers. The category includes evening papers, which experienced a revival in the 1980s, and ‘metro’ papers like SMD, mostly founded during the headlong marketization of the 1990s (Zhao, 1998, pp. 134-152, Zhao, 2002).

During the early days of marketization, many commentators argued the forces of the Party’s political control and the market-driven needs of circulation and advertising would lead to conflicts and, potentially, a rupture between the Party and the media (Chan, 1993; Chu, 1994; Pei, 1994). These writers tended to assume a correlation between markets and democracy, and to assume that the former would lead inexorably to the latter, bringing with it the potential of a new and free press (Levy, 1994, p. 3). Others argued that, although a market orientation allowed a space for new, and even critical, kinds of journalism, the drive to marketization was initiated by the CPC leadership and subject to their close supervision (Zhao, 1998, p. 157). While the pressures of market and Party might lead to occasional spats, sometimes with serious consequences for the journalists involved, there was no fundamental incompatibility that would necessarily lead to a systematic conflict between the media and the CPC (Zhao, 1998, p. 159).

In the course of the 1990s, the pace of change was such that the evidence for one view or another was inconclusive and some commentators thought they had identified an unstable system that was constantly in danger of collapse (Lynch, 1999). The main beneficiaries of the new situation were the commercial newspapers that gained revenues at the expense of the official party newspapers (Chen & Guo, 1998; Zhao, 2004, p. 178). In the new century, however, it has become clear that, at least in the medium term, the CPC has developed mechanisms that allow it to co-exist with the market. In the newspaper press, one of the key mechanisms was the grouping of different titles in to ‘conglomerates’, headed by official party publications but also containing titles with a commercial orientations (He, 2007; Hu, 2003; Zhao, 2000). The party-facing and market-facing papers were thus under the same overall management and, as an added benefit, the revenues of the commercial paper could be used to subsidise the political paper (Volland, 2012, pp. 114-115).

There remain debates as to the effectivity of the current arrangements. Some writers see the new system as representing a model as effectively closed as the old Stalinist system (Brady, 2008, 2012). Others continue to argue the dynamics of the system will lead to liberal outcomes (Esarey, 2006; Rosen, 2003; Shirk, 2011, pp. 237-239). Some retain the belief that there is an inherent contradiction between party and market that is constantly on the point of creating an explosion (Qian & Bandurski, 2011). Overall, however, there is general agreement that the mechanisms currently in place have survived for more than two decades and cannot be considered a temporary and unstable formation (McCormick, 2002/2003).

The fact that the CPC monopolises all of the levels of power in China does not, however, mean that there are no substantial political conflicts. There are many bitter conflicts between the CPC and its business allies on one hand and peasants and workers on the other. Sometimes, particularly on ecological issues, there are conflicts between the elites and the mass of the population of a particular city. There are also, however, conflicts within elite groups. Just as in any other one-party system, such political conflicts take place within the party rather than between parties since the uneven development of the economy leads to differences between different geographical sections and different levels of the Party apparatus. Some of these conflicts find expression in the media responsible to particular party sections. This can lead to differentiation in newspaper content from one province to another and to direct struggles between different levels of the party over editorial direction (Cho, 2014; Sun, 2012, pp. 183-185; Tong, 2010).

The overall picture of the Chinese press in the early years of this century is one that can be described as ‘diversity in unity’. The CPC retains direct control over the press, which ensures that, on major issues of domestic and international politics, all titles will follow the same approach. However, different titles concentrate on different audiences and prioritise different kinds of content even within the same press group, and different press groups, responsive to different Party committees, permit different degrees of diversity in the content and approach of their titles (Zhao, 2003, p. 51).

Measuring the variety of reporting

All of the above is relatively well-established but what is less clear is the nature and extent of the diversity that is permitted within this unity. Although there have been some important analyses of the content of different newspapers, these have mostly focused on particular issues and the extant English-language literature has tended to be dominated by surveys and interviews with journalists and news executives. The aim of this article is to provide a first attempt at measuring how and what ways content differs in the Chinese press.

Given the established differences between the official press and the commercial press, the following research questions suggest themselves:

Are there significant differences between the kinds of journalism practised by the different types of newspapers, as expressed in various indicators like different average article lengths or emphasis on covering different topics?

Are there significant differences in the degree of dependence upon government between the different types of newspapers, as expressed through different patterns of sourcing news stories?

Are there significant differences between the degrees of ‘objectivity’ in the different kinds of newspapers? Are, for example, journalists on official newspapers more likely to insert opinions or interpretations in to their stories than those on commercial papers?

Are there significant differences in the overall models of journalism present in the different types of newspapers? Do official papers tend to devote more weight to the classical propaganda functions of the Chinese media while commercial papers devote more weight to functions closer to the western idea of ‘watchdog’ journalism?

There are two other, rather less precise, questions upon which this article casts light. The first is whether market-facing titles in China can meaningfully be considered ‘tabloid’ (Zhao, 2002). Given that there is no internationally agreed definition of a ‘tabloid’ newspaper, and it means different things in Europe and the United States, it is impossible to give a conclusive answer. It is, however, possible to examine some of the indicators of the broad category – for example, the presence or absence of ‘serious’ news about politics and economics, the extent of ‘infotainment’ coverage, the length of stories and degree to which visual images are present – to draw tentative conclusions about the ‘tabloid’ press in China.

The second question concerns the decline in investigative journalism. It is argued that this kind of journalism flourished in the later 1990s but has been in decline since around 2003. Other writers have argued that while some of the more confrontational forms of journalism have been discontinued or limited, other topics, for example, environmental investigations, have become more prominent (Tong, 2015). The difficulty in giving an answer again lies in the lack of definitional clarity, in this case over what might constitute ‘investigative reporting’. For some writers, it necessarily involves digging out facts that ‘the party’ would rather remain hidden. Others argue that, given the differences within the party, it is not always clear what ‘the party’ would rather hide: higher-level party organisations indicate to publications responsible to them that a lower level organisation is worth investigating and that exposing its wrongdoings would be welcomed (Sun, 2010). As it is not possible to be certain which kinds of reporting are genuinely ‘pushing the red line’ and which are following directions from the responsible party body, it is difficult to give an exact answer as to whether investigative journalism remains healthy, but it is possible to measure the extent to which activities associated with the more independent kinds of investigative journalism, often dubbed the ‘watchdog’ role of the media, are present.

Methodology

To answer these questions, a content analysis of a representative sample of two newspapers was undertaken. The papers chosen were selected as ‘most different cases’ to illuminate the range of possible difference. The first of these choices was PD. This is the most official political daily newspaper in China. As it said about itself in 2013:

Renmin Ribao (People’s Daily) is the official newspaper of the Central Committee of the Communist Party of China … For over 60 years, the People’s Daily newspaper has adhered to the correct direction, and has actively promoted the party’s theory, line, principles and policies, and has actively promoted the central major decisions and plans, timely dissemination of information in various fields at home and abroad, in order that the Communist Party may unite and lead the people across the country to triumph in the revolution, construction and reform. (Remin Ribao, 2013)

PD has been the subject of many studies going back over at least 40 years and there have been several relatively recent scholarly analyses of its content from a variety of perspectives (W. Wang, 2013; Dong, Chang, & Chen, 2008; Parsons & Xu, 2001; S. Wu, 2014; W. Zhang, 2012; Z. Zhang, 2014). Although all of these studies demonstrate that PD is a mouthpiece of the Party leadership, even here there has been a move towards what some authors describe as ‘democratization’ and greater openness (Huang & Chen, 2009; Q. F. Zhang, 2014a; Q. F. Zhang, 2014b).

The second choice was SMD, which is not only a highly commercial title but it is widely recognised as the most ‘liberal’ daily newspaper in China. As a mass-circulation newspaper in the press grouping headed by the official Southern Daily, it has enjoyed a real, if limited, measure of freedom in its reporting and commentary (Zhao and Xing, 2012). SMD has a reputation for investigative journalism, which has sometimes transcended the official limits on reporting and occasionally led to confrontations with the state (Lu, n.d.; Tong, 2007, 2011, 2013; Tong & Sparks, 2009; H. Wang, 2010). In a 2011 survey of investigative reporters in China, SMD was identified as the most respected domestic medium, scoring three times as highly as PD, which was the least respected of the titles identified (Shen & Zhang, 2014, pp. 44-45). The contrast with the official political press could not be more clearly drawn (D. Wu, Huang, & Liu, 2015).

This content analysis is part of an international project aimed at comparing journalistic role conceptions and journalistic role practise in different countries (Journalistic Role Performance, n.d.). The coding frame was designed to allow the classification of the content of newspapers into one of six different models (Mellado, 2014). To construct these models, a range of variables were coded, recording differences in story length, story topic, sourcing and so on.

The models are composites of these different variables. The first, the ‘Interventionist’, concerns how the narrative of a story is organised, and in particular the presence of the journalist in the story. As Mellado (2014) wrote in defining the six models used in this study:

Unlike the other five dimensions of role performance that we will identify, the intervention dimension is not context based, but is always present in the news content’s narrative logic. The more presence or absence of the journalistic voice in the news story always interacts with the other dimensions of role performance used by the journalists in their daily work. (p. 9)

This model, uniquely, has two dimensions depending upon the presence or absence of these indicators. The other models are ‘one dimensional’, measuring the extent to which indicators of different media functions (‘Watchdog’, ‘Loyal Facilitator’, ‘Service’, ‘Infotainment’ and ‘Civic’ models) are present. For this article, the categories which concern the attitude of the title towards political and economic power, the ‘Watchdog’ and ‘Loyal Facilitator’, are particularly interesting. The ‘Watchdog’ model, for example, contains variables as to the extent to which the story contains questioning or criticism of elite groups, and the extent to which it embeds investigative journalism. The ‘Loyal facilitator’ model contains variables as to the extent of support for official activities and government policies, presenting a positive image of elite groups, emphasis on progress and national success and so on, which fit closely with the propaganda duties of the Chinese press.

Because the models are indicated by different numbers of variables, an index was constructed by adding all of the occurrences of indicators of each model and dividing the sum by the number of indicators in each model in each paper and the number of coded cases in that paper. This index was used as the basis for comparing the presence of the different models in the two papers, although to compare the articulations of particular models required the use of the original data.

A distinction was made between those stories generated by SMD staff and those from outside sources like wire services. SMD is a provincial newspaper with fewer reporters dedicated to national reporting than PD, the percentage of stories originating from outside sources is much higher in the former. Although such stories contribute to the overall character of the journalism present in the newspaper, it is also important to measure whether the ‘native’ journalism of SMD is significantly different from the other types in the sample.

These data yields no information as to the substance of stories. It is not possible to determine whether PD and SMD cover the same story in the same way. This partially insulates the findings from changes in the political climate. There has been an increase in direct political control over public discussion since the advent of the Xi presidency in 2013, so might be anticipated that there would be greater uniformity of reporting in different newspapers. If this analysis was comparing contents, a change in the political climate would mean that there was a shift in the relationship between the different titles. The fact that this survey is concerned only with the form and not the substance of any given story allows it some independence from such considerations, and suggests that the findings presented here retain validity despite the changed political climate.

The project analysed 28 days organised into randomly constructed weeks, two in each year for each paper in 2012 and 2013, which provides a representative sample of the content of a Chinese newspaper (Song & Chang, 2012). Only national news stories were considered, so local news, business and finance, opinion piece and so on were not coded. The sample yielded 2034 news items, 1301 from PD and 733 from SMD. Coding was undertaken by graduate students who had undergone a period of prior training. The inter-coder reliability scores measured by Krippendorf’s alpha were above .7, with a range from 0.71 to 0.88. Throughout this article, all reported statistics are significant at least at the .05 level unless otherwise stated.

Findings

The stories in SMD were shorter than those in PD. Although PD had a higher percentage of brief stories than SMD (5.8% vs. 1.2%), 91% of SMD stories were classified as ‘articles’, as against 79.9% in PD. Conversely, the latter had a higher proportion of the longer forms of ‘feature’ and ‘reportage’ journalism (14.4%) than did SMD (7.8%). SMD also had a different focus to its journalism. For both PD and SMD the category of ‘politics’ was by far the most prominent (43.7% and 40.4%, respectively). Although the difference is statistically significant, in both cases this is by far the largest category in both. SMD, however, had a larger proportion of stories concentrating on the categories of ‘police/court’ and ‘accident/disaster’ than did PD (21.9% vs. 5.9%). PD carried a far higher percentage of stories on the ‘economy’ than SMD (17.1% vs. 7.9%) as well as more on the ‘environment’ (3.7% vs. 2.3%). Stories in the two newspapers had a different pattern of sourcing, corresponding to the balance of resources noted above. Fewer than half (42.4%) of SMD stories were attributed to ‘reporter/internal staff’ while 31.0% were solely from wire services; for PD, the corresponding figures were 65.3% and 13.3%.

Although SMD carried more photographs than PD, both titles carried a majority of article without any photographic support: 84.2% of the stories in PD were without photographs, while for SMD the figure was 75.2%. Including all types of images, the percentage of stories with images was 20.7 for PD and for SMD, 29.9. The number of stories in which the images were the preponderant feature was significantly higher for SMD (8.5% as opposed to 6.5%) but in both cases the number was small.

The average number of sources cited per article was 2.87, with a range from 0 to 23 sources. As Table 1 shows, 40% of stories had one source or less and only 19% had more than five. The highest proportion of single sourced or unsourced stories was in PD (44%) and the lowest among SMD staff (28.6%), who also produced the highest proportion of stories with five or more sources. The average for SMD staff is higher than for SMD overall, and very much higher than for PD. The tendency for SMD to use relatively more sources was particularly strong in housing, where its average was more than three times as high as that for PD.

Table 1.

Number of sources used.

	All(n = 2034)	PD(n = 1301)	SMD(n = 733)	SMD staff(n = 409)	SMD non-staff(n = 324)
Mean	2.87	2.69	3.19	3.68	2.57
SD	2.942	2.976	2.855	3.124	2.336
Maximum	23	20	23	23	19
Percentage ⩽1	40.0%	44.1%	32.7%	28.6%	38%
Percentage ⩾5	19.0%	16.9%	22.6%	31.1%	12%

PD: People’s Daily; SMD: Southern Metropolitan Daily.

Multiple sources do not necessarily mean a diversity of views. As Table 2 shows, 74.2% of stories had only one kind of source present and 8.5% quoted no sources at all. The use of one source type was a characteristic of stories in all of the categories, although SMD overall (66%), and SMD staff in particular (58.9%), were less likely to use this source strategy than PD or than the stories in SMD sourced from outside the newspaper. In all cases, the type of source cited was most likely to be from either the state or a political party (i.e., the Communist Party) and these were the most likely to be the main source (overall, 72.0% of cases were from this category, and in 63.1% of cases they were the main source). SMD staff were more likely to use ordinary people, civil society and expert sources than other categories, but in all cases these were in a small minority compared with official sources.

Table 2.

Distribution of sources.

	All(n = 2034)	PD(n = 1301)	SMD(n = 733)	SMD staff(n = 409)	SMD non-staff(n = 324)
No sources present	173 (8.5%)	134 (10.3%)	39 (5.3%)	17 (4.2%)	22 (6.8%)
Only one source present	1510 (74.2%)	1026 (78.9%)	484 (66.0%)	241 (58.9%)	243 (75%)
More than one source present	351 (17.3%)	141 (10.8%)	210 (28.6%)	151 (36.9%)	59 (18.2%)

PD: People’s Daily; SMD: Southern Metropolitan Daily.

Table 3 shows the results for the model index calculated as described above. The ‘Interventionist’ model is by far the most prevalent. One striking feature of both newspapers is the absence of any indicators of the ‘Civic’ model of journalism. The ‘Service’ model, although present, was only weakly represented in both titles, but was more prominent in PD (an index of 0.4) than in SMD (an overall index of 0.1, which does not differ between stories sourced internally and those sourced externally).

Table 3.

Distribution of journalistic models in PD and SMD.

	All(n = 2034)	PD(n = 1301)	SMD all(n = 733)	SMD staff(n = 409)	SMD non-staff(n = 324)
Interventionist	22.9	24.5	20	22.7	16.5
Watchdog	1.1	0.5	2.1	3.2	0.8
Facilitator	3.4	4.5	1.3	1.4	1.2
Service	0.4	0.4	0.1	0.1	0.1
Infotainment	2.3	2.5	2.1	2.6	1.4
Civic	0	0	0	0	0

PD: People’s Daily; SMD: Southern Metropolitan Daily.

More important are the indices for the ‘Loyal facilitator’ and ‘Watchdog’ models. Stories in PD are more likely to display indicators of the ‘Loyal facilitator’ model than are those in SMD (an index of 4.5 as opposed to 1.3). SMD staff-sourced stories are marginally more likely to adopt this model than are those from outside, although the difference is very small. Conversely, SMD stories are much more likely to display indicators of the ‘Watchdog’ model than are those in PD (an index of 2.1 as opposed to 0.5), and here there is a difference between stories sourced from SMD staff and those sourced from outside (3.2 as opposed of 0.8). The ‘Infotainment’ model is more evenly distributed between titles, with PD, showing an index of 2.5, slightly more likely than SMD (an index of 2.1) to display this model. If only those stories sourced from SMD staff are considered, however, the index rises to 2.6.

Although both titles have a high index for the ‘Interventionist’ model, this conceals differences in the ways it is expressed. There are five indicators of the ‘Interventionist’ model: ‘Presence or absence of the journalist’s opinion’, ‘Presence or absences of the journalist’s interpretation’, ‘Presence or absence of a proposal or demand’, ‘Journalist’s use of adjectives’ and ‘Journalist’s use of the first person’. Table 4 shows how frequently the indicators of this model appeared in the two papers. Overall, SMD carries fewer indicators of the interventionist model than does PD. SMD staff contributions are more interventionist than are those originating from non-staff sources, but the former are not different from those of PD.

Table 4.

Number of indicators of the Interventionist Model in stories in PD and SMD.

	All(n = 2034)	PD(n = 1301)	SMD all(n = 733)	SMD staff(n = 409)	SMD non-staff(n = 324)
Journalist’s opinion	52 (2.6%)	48 (3.7%)	4 (0.5%)	2 (0.5%)	2 (0.6%)
Journalist’s interpretation	1046 (51.4%)	677 (52.0%)	369 (50.3%)	244 (59.7%)	125 (38.6%)
Proposals or demands	21 (1.0%)	15 (1.2%)	6 (0.8%)	2 (0.5%)	4 (1.2%)
Use of adjectives	1047 (51.5%)	718 (55.2%)	329 (44.9%)	203 (49.6%)	126 (38.9%)
Use of the first person	160 (7.9%)	135 (10.4%)	25 (3.4%)	14 (3.4%)	11 (3.4%)

PD: People’s Daily; SMD: Southern Metropolitan Daily.

Table 5 records the distribution of these indicators. Compared with PD, SMD staff contributions display fewer journalist’s opinions, adjectives and first person usage, while offering more interpretations. The other two differences are not statistically significant. Differences as to the presence of journalistic opinions in stories about politics are not statistically significant, although they are low across all categories. For the indicator of the journalistic interpretation of the story there are significant differences. SMD as a whole (39.3%) and SMD staff (43.9%) are more likely to offer interpretations of political stories than PD (34.7%) and less likely to offer interpretations of stories about the economy. SMD as a whole show such indicators in 11.1%, and SMD staff showing 9.8%, as against PD showing 24.2%. SMD is similarly more likely to use the first person for political stories than PD, with SMD as a whole showing 52.0% of such indicators against 31.1% for PD, while SMD staff showed 57.1%. Again, the balance was the opposite with economic stories, with PD showing 24.4% as against 8% for SMD as a whole and 7.1% for SMD staff.

Table 5.

Presence of indicators of the Interventionist Model in PD and SMD.

Number of indicators	All(n = 2034)	PD(n = 1301)	SMD all(n = 733)	SMD staff(n = 409)	SMD non-staff(n = 324)
0	686 (33.7%)	404 (38.1%)	282 (38.5%)	130 (31.8%)	152 (46.9%)
1	500 (24.6%)	314 (24.1%)	186 (25.4%)	100 (24.4%)	86 (26.5%)
2	737 (36.2%)	487 (38.4%)	250 (34.1%)	172 (42.1%)	78 (24.1%)
3	93 (4.6%)	79 (6.1%)	14 (1.9%)	57 (1.7%)	7 (2.2%)
4	17 (0.8%)	17 (1.3%)	0	0	0
5	1	0	1 (0.1%)	0	0

PD: People’s Daily; SMD: Southern Metropolitan Daily.

Neither the ‘Watchdog’ nor the ‘Loyal facilitator’ models are as strongly present, but there are important differences in the ways they are implemented. The ‘Watchdog’ model was measured by seven variables: ‘Questioning by journalist’, ‘Questioning by others’, ‘Criticism by journalist’, ‘Criticism by others’, ‘Denunciation by journalist’, ‘Denunciation by others’ and ‘Investigative reporting’. The first six were broken up into stories concerning the government or political elite, those concerning the business and economic elite and those concerning civil society actors. The seventh category was divided in to investigative reporting by journalists and by others. Table 6 presents the breakdown of this analysis, although given the distribution of the data it is not possible to test for significance except in the case of ‘Others questioning the government’, where the differences are indeed significant. Bearing that limitation in mind, it can be noted that both papers’ preferred mode is to report the comments of others rather than to commit themselves to an independent view. Although SMD is more likely to engage in this kind of journalism, and its own staff is the source of most of this material, it, too, prefers to quote others rather than take its own stance. The same reticence is clear in the reporting of investigations, where reports of investigations by other people outnumber those of the journalists themselves.

Table 6.

Presence of indicators of the Watchdog Model in PD and SMD.

Indicators	All (n = 2034)	PD (n = 1301)	SMD all (n = 733)	SMD staff (n = 409)	SMD non-staff (n = 324)
Journalist questioning government	8 (0.4%)	5 (0.4%)	3 (0.4%)	2 (0.5%)	1 (0.3%)
Journalist questioning business	4 (0.2%)	3 (0.2%)	1 (0.1%)	1 (0.2%)	0
Journalist questioning civil society	1 (0.0%)	1 (0.1%)	0	0	0
Others questioning government	45 (2.2%)	9 (0.7%)	36 (4.4%)	33 (8.1%)	3 (0.9%)
Others questioning business	7 (0.3%)	2 (0.2%)	5 (0.7%)	2 (0.5%)	3 (0.9%)
Others questioning civil society	3 (0.1%)	2 (0.2%)	1 (0.1%)	1 (0.2%)	0
Journalist criticising government	7 (0.3%)	5 (0.4%)	2 (0.3%)	2 (0.5%)	0
Journalist criticising business	1 (0.0%)	0	1 (0.1%)	0	1 (0.3%)
Journalist criticising civil society	0	0	0	0	0
Others criticising government	38 (1.9%)	6 (0.5%)	32 (4.4%)	32 (7.8%)	0
Others criticising business	7 (0.3%)	3 (0.2%)	4 (0.5%)	2 (0.5%)	2 (0.6%)
Others criticising civil society	2 (0.1%)	0	2 (0.3%)	2 (0.5%)	0
Journalist denouncing government	1 (0.0%)	0	1 (0.1%)	1 (0.2%)	0
Journalist denouncing business	0	0	0	0	0
Journalist denouncing civil society	0	0	0	0	0
Others denouncing government	8 (0.4%)	2 (0.2%)	6 (0.8%)	4 (1.0%)	2 (0.6%)
Others denouncing business	2 (0.1%)	0	2 (0.3%)	2 (0.5%)	0
Others denouncing civil society	0	0	0	0	0
Investigation by journalist	3 (0.1%)	1 (0.1%)	2 (0.3%)	2 (0.5%)	0
Investigation by others	30 (1.5%)	11 (0.8%)	19 (2.6%)	14 (3.4%)	5 (1.5%)

PD: People’s Daily; SMD: Southern Metropolitan Daily.

The ‘Loyal facilitator’ model is shown in Table 7. PD stories carry indicators of this model more frequently than do those of SMD as a whole in every case, with a particularly strong discrepancy in the variables emphasising progress and success and a positive image of the country. Within SMD, stories originating from SMD staff are more likely to display these indicators than those sourced from outside in the cases of support for official policies, presenting a positive image of the political and economic elites and emphasising progress and success. In most of these cases, however, the absolute size of the number of stories involved, and the range of differences, remains small. Again, due to the distribution of the data, only the results for the following indicators can be considered statistically significant: support for official activities, support for government policies, presenting a positive image of political elites and emphasising success/progress.

Table 7.

Presence of indicators of the Loyal Facilitator Model in PD and SMD.

Indicators	All(n = 2034)	PD(n = 1301)	SMD all(n = 733)	SMD staff(n = 409)	SMD non-staff(n = 324)
Support official activities	177 (8.7%)	149 (11.5%)	28 (3.8%)	14 (3.4%)	14 (4.3%)
Support government policies	121 (5.9%)	105 (8.1%)	16 (2.2%)	11 (2.7%)	5 (1.5%)
Positive image of political elite	104 (5.1%)	80 (6.1%)	24 (3.3%)	17 (4.2%)	7 (2.2%)
Positive image of economic elite	19 (0.9%)	17 (1.3%)	2 (0.3%)	2 (0.5%)	0
Positive image of the country	15 (0.7%)	14 (1.1%)	1 (0.1%)	0	1 (0.3%)
Emphasise progress/success	139 (6.8%)	131 (10.1%)	8 (1.1%)	5 (1.2%)	3 (0.9%)
Emphasise national triumphs	5 (0.2%)	4 (0.3%)	1 (0.1%)	1 (0.2%)	0
Comparison to rest of the world	14 (0.7%)	12 (0.9%)	2 (0.3%)	0	2 (0.6%)
Expression of patriotism	22 (1.1%)	19 (1.5%)	3 (0.4%)	1 (0.2%)	2 (0.6%)

PD: People’s Daily; SMD: Southern Metropolitan Daily.

Discussion

It is possible to give answers to the formal questions posed above. There are marked and significant differences between the kinds of journalism practised in PD and in SMD, and within SMD there are differences between stories generated by staff journalists and those from outside. Overall, the ‘native’ journalism of SMD is more different from that of PD than is the overall paper. Although PD carries a larger number of very brief stories, in general it carries more substantial articles than does SMD. PD is more likely to concentrate upon ‘hard news’ stories about politics and, particularly, economics than SMD which, for its part, concentrates more on ‘softer’ news in the categories ‘police and courts’ and ‘accidents and disasters’. Differentiation, however, is of limited scope: in all cases, by far the largest single category of stories is about politics.

Both papers had a tendency to cite only one source in their stories, and for both that was likely to be related to the CPC. SMD was, however, more likely to quote more than one source, and SMD staff were even more likely so to do. In none of these cases was there strong evidence of the use of alternative sources. SMD journalists were more likely to quote non-official sources like ordinary people, but these quotations were marginal compared with the presence of official sources. Overall, although there is some evidence of SMD, and SMD staff in particular, being slightly more independent of official definitions of events, this is not strong enough to support the contention that there is any real difference in the extent to which its reporting was much more ‘independent’ than that of PD.

Both PD and SMD were likely to run stories in which journalistic opinions and other forms of ‘interventionism’ were present. PD and SMD staff were the most likely sources of these stories, and SMD non-staff sources were, by a considerable margin, less likely to adopt this model of journalism. The presence of journalistic interpretations and the use of adjectives in description were the most common indicators of intervention. SMD journalists were most like to offer an interpretation of their subject matter and non-SMD staff were least likely. Although the differences in the extent of ‘objectivity’ are significant, and the proportions of the indicators differ between titles, overall these differences are not large and there is little ground for arguing that either paper represents a different kind of journalism. In fact, the most ‘objective’ stories were those from non-SMD sources. One factor in this may be that, as is common with the ‘wholesaling’ of news through wire services elsewhere, these stories tend to be constructed in this way since they are likely to be used by a wide range of ‘retail’ news organisations. An alternative view is that SMD distinguishes between different kinds of stories: those that it generates itself are permitted a high degree of interpretation and other forms of journalistic intervention, while those from outside are edited to minimise such characteristics.

Although the ‘Interventionist’ model of journalism is the most common, there are significant differences between the titles as regards the ‘substantial’ models. PD more likely to display indicators of the ‘Loyal facilitator’ model, while SMD, and particularly SMD staff-generated stories are more likely to display indicators of the ‘Watchdog’ model. These are not the central characteristics of the journalism examined here, but they do demonstrate the presence of different models of journalistic practise in the two titles. The basis of the popular view of the two papers is confirmed by the evidence.

With regard to the ‘Watchdog’ model it is notable that both titles go to some lengths to present it as the work of others rather than as an artefact of their independent activity. They are more likely to report the opinion of someone else about the government or the business elite than they are to mount their own criticisms or ask their own questions. The stories that depend upon investigation in both papers are much more likely to result from work done outside the organisation than the efforts of their own journalists. Although there is indeed watchdog journalism, particularly authored by SMD staff, it is presented evasively as the result of activity outside of the paper, which simply reports rather than initiates criticism.

The question of ‘tabloidization’ is harder to answer. It is clear that SMD does not correspond with to the US or the UK model of the tabloid newspaper. SMD has much more hard news content than the standard US ‘supermarket tabloid’ (Bird, 1992). It also has more hard news content, and a lower ratio of images to text, than do the British tabloid newspapers. However, it does have relatively shorter stories, and relatively more images, than does PD and, if the term ‘tabloid’ is understood as a relational, rather than an absolute, category, then the evidence supports the use the term in this. Indeed, since this study is concerned only with national news, and excludes categories like sport and entrainment news, then the extent to which SMD corresponds at least to the Scandinavian model of the tabloid newspaper is probably greater than appears from this evidence.

It is also clear that investigative journalism continues to be present, more strongly in SMD than in PD, although in both cases it is a relatively marginal category. For the reasons discussed above, it is impossible on the basis of this evidence to determine the substance of such journalism. This study did not gather the kind of data that would permit a conclusion as to whether this investigative journalistic content is pushing the boundaries of the possible or whether it is simply acting as a watchdog for the party against errant members and rogue businesses. To give a more satisfactory answer to this question, it would be necessary to conduct a study which tracked the substance of investigative reporting longitudinally from before 2003 up to the present.

Conclusion

This study has demonstrated that there is an empirical substance to the claim that the contemporary Chinese press is best characterized as ‘diversity in unity’. Politics, and party sources, remain the staple of national news both the official and the market-oriented press. The claims that more than 30 years of market orientation have transformed any segment of the Chinese press and set it on a course that is inimical to the propaganda functions demanded by the party are clearly mistaken. However, the suggestion that a combination of political interventions and material rewards have defused any critical functions is equally mistaken. Within the overall picture of a press that is dominated by official politics and official voices there are clear, if limited, indications that the party-oriented press devotes more time and energy to the classical tasks of propaganda than does the market-oriented press. Conversely, the latter is more likely to give voice to questions and criticism of the political and economic elite, albeit as cautiously as possible.

As with all studies, this research has some limitations. The first of these is that, since SMD is a provincial level newspaper, it is possible that its distinctive voice might be more clearly visible in coverage of news originating in Guangdong. It would, however, be impossible to compare this provincial coverage with the national coverage that dominates PD. Future research might overcome this problem by comparing provincial-level coverage in the “liberal” SMD with that of a commercial title from a more “orthodox” province, or indeed, if sufficient resources were available, analysing such coverage across all of the provincial press groups.

A second limitation arises from the changing overall situation of the Chinese newspaper press. The situation of the media in general has changed in substantial ways over the last 30 years, and a longitudinal study might very well reveal periods in which the differences between the two titles would be greater or smaller than those discovered in our sample. Unfortunately, in order to participate fully in the comparative project, we were obliged to select samples from 2012 and 2013 along with all the other participants. This is a limitation which means that it is difficult to generalise, not only about the situation in previous years but also the subsequent period. One obvious factor is that, in common with the press in the developed world, Chinese newspapers have begun to experience the effects of the diffusion of digital technologies (Sparks, et al. 2016). A fall both in advertising and circulation revenues has provoked important changes in editorial strategies, particularly in the commercial press (Wang and Sparks 2018). The sample used in this study dates from a relatively early phase of the developing situation, and it may well be that a more recent sample would produce somewhat different results. We think, however, that the changing situation in China would not, overall, have resulted in greater differentiation between the kinds of journalism practised in the newspaper press, and that the main outlines of the conclusions drawn here are likely to be repeated, with some significant modifications, in any subsequent study.

Footnotes

Funding

This research was supported by the Hong Kong Research Grants Council’s General Research Fund grant 12406914.

ORCID iD

Colin Sparks

Author biographies

Haiyan Wang is Associate Professor in the School of Communication and Design at Sun Yat-Sen University. Her main research interests lie in journalism studies and women in media. Currently she is working on a project looking at the impact of digital media on Chinese journalism.

Colin Sparks is Professor of Media Studies in the Department of Journalism at Hong Kong Baptist University. His main research interests are in the political economy of the media and in comparative media systems. He is currently working on the impact of digital media on mainland Chinese journalism.

Yu Huang is Professor in the Department of Journalism at Hong Kong Baptist University. He is Dean of the School of Communication.

References

Bird

S. E.

(1992). For enquiring minds: A cultural study of supermarket tabloids. Knoxville: University of Tennessee Press.

Brady

A.-M.

(2008). China. Lanham, MA: Rowman & Littlefield.

Brady

A.-M.

(2012). Market friendly, scientific, high tech, and politics-lie. In Brady

A.-M.

(Ed.), China’s though marketing dictatorship: Propaganda and thought-work in contemporary management (pp. 1–8). London, England: Routledge.

Chan

(1993). Commercialization without independence: Trends and tensions of media development in China. In Yu-Shek Cheng

Broseau

(Eds.), China review 1993 (pp. 251–252). Hong Kong: The Chinese University Press.

Chen

Guo

(1998). Opposites apart: Explaining differential revenues between China’s party and mass-appeal newspapers. Asian Journal of Communication, 8, 70–99.

Cho

L.-F.

(2014). The development of investigative reporting and journalistic professionalism in Southern Weekend. In Svensson

Sæther

Zhi’an

(Ed.), Chinese investigative journalists’ dreams: Autonomy, agency and voice (pp. 181–198). Lanham, MA: Lexington.

Chu

L. L.

(1994). Continuity and change in China’s media reform. Journal of Communication, 44, 4–8.

Dong

Chang

T.-K.

Chen

(2008). Reporting aids and the invisible victims in China: Official knowledge as news in People’s Daily, 1986–2002. Journal of Health Communication, 13, 357–374.

Esarey

(2006). Caught between state and society: The commercial news media in China (Unpublished PhD dissertation). Columbia University, New York, NY.

10.

Hassid

(2008). Controlling the Chinese media: An uncertain business. Asian Survey, 48, 414–430.

11.

(2007). Media control in China. In Hom

Mosher

(Eds.), Challenging China: Struggle and hope in an era of change (pp. 32–45). New York, NY: The New Press (Human Rights in China).

12.

(2003). The Post-WTO restructuring of Chinese media industries and the consequences for capitalization. Javnost/The Public, 10, 19–36.

13.

Huang

Chen

J.-P.

(2009). Discursive democratization in mainland China: A diachronic study of the New Year’s editorials in People’s Daily. Journal of Asian Pacific Communication, 19, 199–217.

14.

Journalistic Role Performance. (0000). Journalistic role performance around the world. Retrieved from http://www.journalisticperformance.org/

15.

Lee

C.-C.

(1994). Ambiguities and contradictions: Issues in China’s changing political communication. In Lee

C.-C.

(Ed.), China’s media, media’s China (pp. 3–20). Boulder, CO: Westview.

16.

Lee

C.-C.

Huang

(2006). Chinese party publicity inc.’ conglomerated: The case of the Shenzhen press group. Media, Culture & Society, 28, 581–602.

17.

Lee

C.-C.

Huang

(2007). Party-market corporatism, clientelism, and media in Shanghai. Harvard International Journal of Press/Politics, 12, 20–42.

18.

Levy

(1994). Editor’s note. Journal of Communication, 44, 3–4.

19.

(n.d.). Construction of media public sphere in social transformation: A study of news commentary column in Southern Metropolis Daily. Retrieved from http://journalism.uoregon.edu/~IAMCR-PolEcon/Huan.html

20.

Lynch

(1999). After the propaganda state: Media, politics and “thought work” in a reformed China. Stanford, CA: Stanford University Press.

21.

McCormick

(2003). Recent trends in mainland China’s media: Political implications of commercialization. Issues and Trends, 38-39, 175–215 (Original work published in 2002).

22.

Mellado

(2014). Professional roles in news content. Journalism Studies, 16, 596–614.

23.

Parsons

(2001). News framing of the Chinese embassy bombing by the People’s Daily and the New York Times. Asian Journal of Communication, 11, 51–67.

24.

Pei

(1994). From reform to revolution: The demise of communism in China and the Soviet Union. Cambridge, MA: Harvard University Press.

25.

Polumbaum

(1990). The tribulations of Chinese journalists after a decade of reform. In Lee

C.-C.

(Ed.), Voices of China: The interplay of politics and journalism (pp. 33–68). New York, NY: The Guildford Press.

26.

Qian

Bandurski

(2011). China’s emerging public sphere: The impact of media commercializaiton, professionalism, and the internet in an era of transition. In Shirk

(Ed.), Changing media, changing China (pp. 38–76). Oxford, UK: Oxford University Press.

27.

Ribao

(2013). People’s Daily Profile. Retrieved from http://www.people.com.cn/GB/50142/104580/index.html

28.

Rosen

(2003). Chinese media and youth: Attitudes towards nationalism and internationalism. In Lee

C.-C.

(Ed.), Chinese media, global contexts (pp. 97–118). London, England: Routledge Curzon.

29.

Shen

Zhang

(2014). The identities and demographics of investigative journalists in China. In Svensson

Sæther

Zhi’an

(Eds.), Chinese investigative journalists’ dreams: Autonomy, agency and voice (p. 33). Lanham, MA: Lexington Books.

30.

Shirk

(2011). Changing media, changing China. In Shirk

(Ed.), Changing media, changing China (pp. 1–37). Oxford, UK: Oxford University Press.

31.

Song

Chang

T.-K.

(2012). Selecting daily newspapers for content analysis in China: A comparison of sampling methods and sample sizes. Journalism Studies, 13, 356–369.

32.

Sparks

Wang

Lü

Wang

Huang

. (2016). The impact of digital media on newspapers: Comparing the US and China. Global Media and China, 1(3), 186–207.

33.

Sun

(2010). Alliance and tactics among government, media organizations and journalists: A description of public opinion supervision in China. Westminster Papers in Communication and Culture, 7, 43–55.

34.

Sun

(2012). Localizing Chinese media: A geographic turn in media and communication research. In Sun

Chio

(Eds.), Mapping media in China: Region, province, locality (pp. 13–27). London, England: Routledge.

35.

Tong

(2007). Guerrilla tactics of investigative journalists in China. Journalism, 8, 530–535.

36.

Tong

(2010). The crisis of centralized media control theory: How local power controls media in China. Media, Culture & Society, 32, 925–942.

37.

Tong

(2011). Investigative journalism in China: Journalism, power and society. London, England: Continuum.

38.

Tong

(2013). The importance of place: An analysis of changes in investigative journalism in two Chinese provincial newspapers. Journalism Practice, 7, 1–16.

39.

Tong

(2015). Investigative journalism, environmental problems and modernization in China. London, England: Palgrave Macmillan.

40.

Tong

Sparks

(2009). Investigative journalism in China. Journalism Studies, 10, 337–352.

41.

Volland

(2012). The CPC’s “reforms of the cultural structure.” In Brady

A.-M.

(Ed.), China’s thought management (pp. 107–121). London, England: Routledge.

42.

Wang

(2010). How big is the cage? An examination of local press autonomy in China. Westminster Papers in Communication and Culture, 7, 56–72.

43.

Wang

Sparks

. (2018). Marketing credibility. Journalism Studies. doi:10.1080/1461670X.2018.1513815

44.

Wang

(2013). The framing analysis of the media representation in 7.23 motor car accident—A study of People’s Daily, Nanfang Metropolis Daily, People’s Daily online and Phoenix new media. China Media Report Overseas, 9,70–80.

45.

White

(1990). All the news: Structure and politics in Shanghai’s reform media. In Lee

C.-C.

(Ed.), Voices of China: The interplay of politics and journalism (pp. 88–110). New York, NY: The Guildford Press.

46.

Huang

Liu

(2015). Competing and hybridized discourses in Chinese news reporting: Case studies of three key newspapers in Southern China. Journal of Multicultural Discourse, 10, 67–84.

47.

(2014). The revival of confucianism and the CPC’s struggle for cultural leadership: A content analysis of the People’s Daily 2000–2009. The Journal of Contemporary China, 23, 971–991.

48.

Zhang

(2014a). Representation of homoeroticism by the People’s Daily since 1949. Sexuality & Culture, 18, 1010–1024.

49.

Zhang

(2014b). Transgender representation by the People’s Daily since 1949. Sexuality & Culture, 18, 180–195.

50.

Zhang

(2012). New assertiveness and new confidence? How does China perceive its own rise? A critical discourse analysis of the People’s Daily editorials and commentaries on the 2008 Beijing Olympics. International Journal of China Studies, 3, 1–23.

51.

Zhang

(2014). Lei Feng reports’ historical discourse analysis-semiotic analysis: Lei Feng reports in People’s Daily. China Media Report Overseas, 10, 64–68.

52.

Zhao

(1998). Media, market, and democracy in China: Between the party line and the bottom line. Urbana: University of Illinois Press.

53.

Zhao

(2000). From commercialization to conglomeration: The transformation of the Chinese press within the orbit of the party state. Journal of Communication, 50, 3–26.

54.

Zhao

(2002). The rich, the laid-off, and the criminal in tabloid tales: Read all about it. In Link

Madsen

Pickowicz

(Eds.), Popular China: Unofficial culture in a globalizing society (pp. 111–135). Lanham, MD: Rowman & Littlefield.

55.

Zhao

(2003). “Enter the world”: Neoliberal globalization, the dream for a strong nation, and Chinese press discourses on the WTO. In Lee

C.-C.

(Ed.), Chinese media, global contexts (pp. 32–56). London, England: Routledge Curzon.

56.

Zhao

(2004). The state, the market and media control in China. In Thomas

Nain

(Eds.), Who owns the media? Global trends and local resistance (pp. 179–212). Penang, Malaysia: Southbound.

57.

Zhao

(2008). Communication in China: Political economy, power, and conflict. Lanham, Maryland: Rowman & Littlefield.

58.

Zhao

Xing

(2012). Provincial papers, national power: The scaling up of the Nanfang Daily media group. In Sun

Chio

(Eds.), Mapping media in China: Region, province, locality (pp. 31–46). London, England: Routledge.