Chapter 4
News content analysis
Basis of analysis
What news did Australians get out of their newspapers in 2005? What was the favourite news topic? What sources were used in the stories and how many? What gender were the sources and how many stakeholders played a role in the delivery of news? How were the stories framed? A total of 2,448 articles was analysed from 14 newspapers across the country, to find out what constitutes news in the print media and what kinds of influences were brought to bear on that news. The newspapers surveyed are set out in Appendix 2, Methodology, later in this Report.
Although similar to the analysis of the presentation of the news in the print section of an American study, The State of the News Media, an Annual Report on American Journalism, 2005, the Australian analysis contains more Sunday papers because of the importance of their circulation here.
The stories were analysed from the first five news pages over 28 days randomly generated by computer from January to December 2005. The newspapers included were identified on their potential to reflect, as accurately as possible, the Australian media array. Only The West Australian, the sole metropolitan daily in Perth, Western Australia, declined to participate.
Topics in the News
The adage "bad news sells" was borne out. Taken across all newspapers, crime featured as the most favoured topic constituting 14.5 per cent of all stories analysed, with federal politics in second place because of its importance in metropolitan newspapers. This differed from the 2005 US study which featured government stories as the most popular, perhaps reflecting the coverage of the US election in 2004.
Federal politics was of much more interest to the metropolitan newspapers (14.5 per cent) than Sunday newspapers (10.1 per cent), their regional cousins (4.75 per cent) or any other newspapers.
Crime stories also featured on page one in Australian newspapers more prominently than any other topic. The Australian study backed up the US study in finding that crime was fundamentally a local story that regional newspapers particularly focused upon.
Entertainment shared that space with Federal politics (both 12.1 per cent of all newspapers) as the next most popular topic but as might be expected it was the most popular topic in the Sunday newspapers (15.9 per cent) with crime lagging marginally behind at 15.2 per cent.
| Content | % of All (2448) | % of Page 1 (299) | % of Metro (1573) | % Regional (442) | % Sundays (435) |
|---|---|---|---|---|---|
| Federal politics | 12.10 | 11.70 | 14.60 | 4.75 | 10.10 |
| State politics | 8.05 | 7.36 | 10.20 | 4.07 | 4.14 |
| Local Politics | 2.37 | 3.01 | 0.57 | 10.90 | 0.23 |
| Crime | 14.50 | 15.7 | 13.90 | 16.1 | 15.20 |
| Foreign Affairs | 2.33 | 4.01 | 2.48 | 1.36 | 2.76 |
| Military | 0.41 | 0.33 | 0.32 | 0.45 | 0.69 |
| Terrorism | 4.17 | 5.69 | 4.70 | 0.90 | 5.52 |
| War | 0.57 | 0 | 0.57 | 0.68 | 0.46 |
| Entertainment | 12.10 | 7.36 | 10.80 | 13.10 | 15.90 |
| Lifestyle | 6.25 | 3.34 | 5.98 | 9.50 | 3.91 |
| Business | 7.60 | 6.69 | 8.26 | 7.01 | 5.75 |
| Science | 1.67 | 0 | 1.53 | 1.36 | 2.53 |
| Health | 3.72 | 2.34 | 3.56 | 4.52 | 3.45 |
| Accident/ Emergency | 10.70 | 10.70 | 9.35 | 15.80 | 10.60 |
| Sport | 5.84 | 14.00 | 5.34 | 3.39 | 10.10 |
| Weather | 2.12 | 2.68 | 2.23 | 2.71 | 1.15 |
| Other | 4.86 | 4.68 | 5.15 | 2.71 | 5.98 |
For a country that defines itself by its sport, this proved one of the most popular topics for Sundays which devoted almost double the amount of space to sports stories compared with metropolitan newspapers although accident and emergency stories were almost equally as popular a choice for Sunday newspapers.
Fig.9. Content of stories 2005
Source: Van Heekeren M, & Simpson L, research for this publication
The least favoured topic was war although this may be because three separate categories were devoted to war: military, terrorism and war. Metropolitan newspapers ran significantly higher Federal politics stories than any other newspaper, selecting this over crime as their main topic. What was the major front-page agenda? Crime again featured well ahead of the other topics across all newspapers with military stories being the least popular12. Sports stories were less likely to be run on page one in regional newspapers but 17 per cent of all newspapers ran sports on page one.
Length of newspaper stories
News stories tended to be under the 500-word category in length in the majority of newspaper articles surveyed, over 60 per cent being in this category. The next highest category were stories of less than 100 words (19.4 per cent), suggesting a preference for news briefs (this finding might have altered if more than the first five news pages had been analysed, as feature-length pieces are more likely to be published further back in the newspaper).
Almost 10 per cent of stories were between 500 and 1000 words with only 0.2 per cent of stories running at more than 1000 words.
Metropolitan newspapers were more likely to run stories between 100 and 500 words (73.5 per cent) compared with regionals (65.6 per cent) and Sundays (64.4 per cent).
Overall the paucity of reporting in depth (i.e., articles longer than 500 words) in metropolitan newspapers, while possibly reflecting the times and preferences of readers, is cause for thought.
Fig.10. Article length by type of newspaper 2005
Source: Van Heekeren M, & Simpson L, research for this publication
Sources
Newspapers are known as the media that offer the most in-depth coverage of news. One way of measuring this is to examine the range of views offered. As the US study states, these are essential elements in trying to assess the quality of reporting.
To break this down, we looked at sourcing in several ways. A source was identified as a person or a statement or "an anonymous source'. First we measured how many sources a story contained. Then we measured the types of sources according to particular categories, to ascertain the number of viewpoints and stakeholders or interest groups a story contained.
Taking these components of sourcing one at a time, at first glance the most alarming finding was that more than 40per cent of all stories cited only one source with only 9.14 per cent of all stories citing four sources and 5.25 per cent citing more than five sources.
The strong correlation of the sources data with the length of articles data showed that very short stories rarely quoted more than one source. Notwithstanding that correlation, more than one source is relatively rarely quoted even in longer articles. It may have implications for assessments of fairness and balance.
Fig.11. Number of sources quoted 2005
Source: Van Heekeren M, & Simpson L, research for this publication
The source data was in marked contrast to the US study which found that nearly half of all newspaper stories (48 per cent) identified four or more sources and, including opinion stories, this figure rose to 51 per cent. 39.4 per cent of all page one stories contained only one source. Regional newspapers contained the highest number of stories with only one source (47.1 per cent).
The predominant source in Australian stories was government (a politician or person employed by a government department) which accounted for 23.9 per cent of all sources.
Fig.12. Type of source by newspaper type 2005
Source: Van Heekeren M, & Simpson L, research for this publication
Metropolitan newspapers cited more government sources than any other section of newspapers (26.9 per cent). There was also considerable gender imbalance in the sources used. More than 70 per cent of all sources cited were male (76.1 per cent) with 23.9 per cent females sourced, perhaps reflecting the number of males in senior positions within government and corporations.
Leaving aside government sources, which in an ideal world might not be accused of spin, but often are, there was a surprisingly small number of other PR sources cited, accounting for 4.9 per cent of all newspaper stories. Part of the success of public relations is not to be obviously attributable in a story13.
A member of the public was the second highest source (20.4 per cent) with NGOs accounting for 10.5 per cent of all sources cited.
For 9.2 per cent of all stories the sources were anonymous, with metropolitan newspapers citing more anonymous sources (8.2 per cent) than any other category of newspaper. Of page one stories, 8.7 per cent in all newspapers, had unidentified sources. This corresponded with 7 per cent of all stories in US newspapers and 13 per cent of front page stories containing anonymous sources.
Framing
The original intention was to parallel framing categories with those used in the US study. Too many, such as the designation of a "reality check" frame, were too steeped in US vernacular to use unchanged so a mix of frames was developed14 with a strong bias toward traditional news values.
How were stories framed? Human interest was the most popular framing for newspaper stories with 16.2 per cent of all stories. Sunday newspapers favoured this category above other newspapers with 19.7 per cent of stories in the human interest category, with regionals 18.3 per cent of all stories printed representing human interest stories. Metropolitans had fewer with 14 per cent of human interest stories.
Proximity (stories close to home) was favoured by regionals over other categories (21.9 per cent) reflecting local bias whereas Sunday newspapers had 4.8 per cent of stories in the proximity category.
11.3 per cent of all stories involved conflict and 12.4 per cent dealt with a problem to be solved. 12.7 per cent of stories involved consequence (impact).
Coders were instructed to identify a single frame where it clearly dominated over others. Where several frames were used without a dominant frame, stories were deemed to be multi-frame.
The exploratory framework used needs some further refinement of the method to assist in determining the best categories of framing to be used15.