Reporting Diversity
Television News 2007 Study

Gail Phillips, Murdoch University;
May 2008

Method & Design

The survey looked at the flagship prime-time nightly news bulletin of the 3 commercial networks (Seven, Nine and Ten) and the two public broadcasting services (the Australian Broadcasting Corporation and the Special Broadcasting Service). 

The period surveyed was two weeks (14 sequential days) from 30 April to 13 May 2007.  As with the previous survey the time period was randomly selected with the aim of capturing a news period unlikely to be impacted upon by unusual events.  As it happened the 2005 survey coincided with the introduction of the new counter-terrorism laws by the Howard government and the counter-terrorism raids in Sydney and Melbourne.  In 2007 the news agenda was more anodyne and in line with what might be considered ‘normal’ in the domestic context.  The question was: would this show a return to the story selection and reporting patterns observed in the original 2001 survey of the pre-9/11 era (see Phillips and Tapsall 2007a)? 

Because of the format and content changes at weekends, the analysis in this paper isolates the weekday bulletins in both weeks and focuses on the main news content.  The Sports bulletins are not dealt with here, though Sports stories are included when they have been integrated into the main bulletin.

The study used the same multi-method approach employed in the previous two surveys combining both quantitative and qualitative analysis.  A specially devised database allowed the researchers to capture quantitative data (duration and percentages of stories from different categories) and qualitative data relating to the nature of the reportage.  Both the 2005 and 2007 studies were done in partnership with the Department of Immigration and Citizenship as part of its Living in Harmony initiative.  The Department was particularly interested in assessing whether news coverage differed depending on the ethnic composition of the local population.  Hence the 2005 study covered Sydney as a major metropolitan city, Perth as a smaller capital city, and the Victorian regional centre of Shepparton, which the Department had identified as having a particularly diverse mix. According to the 2001 census – the latest data available for 2005 - 10.8% of its population was born overseas with the main countries of origin being Italy, England, New Zealand, Iraq and Turkey (ABS 2001). In 2005 it also welcomed 10 African refugee families as the inaugural site for the Federal Government’s Regional Humanitarian Settlement Pilot project.  By the time of the 2007 study the number of people in Shepparton born overseas had risen to 11.4%  with Albania now added to the mix (ABS 2006a).  The 2007 study was further expanded to include a fourth centre – Townsville in Queensland - also identified by the Department as having a diverse ethnic mix. According to the 2006 census 11.6% of the population was born overseas with the main countries of origin England, New Zealand, Papua New Guinea, the Philippines and South Africa (ABS 2006b).

The news bulletins were coded by a four-person team working under the supervision of the research assistant who had overseen the previous coding exercises and had helped refine the coding methodology which was now captured in a formal coding manual. Intercoder agreement was assisted by the training of coders which included participating in communal coding sessions under the research assistant’s supervision.  These allowed for further clarification as the coding progressed.  After this training period the coders and the researchers had regular meetings to discuss other issues as they emerged. In order to do the analysis the chief researcher had to examine both the recorded material and the coding database very closely herself, which added another layer of scrutiny which could rectify any inconsistencies.

As in the previous study stories were identified according to origin (local, national, international) and analysed according to both the topic and the storytelling convention used.  There were 20 content categories which were subsequently grouped together into six story types that gave a better sense of the kind of news treatment they received:

Courts, Crime and Disasters: Emergencies/Disasters; Courts/Justice; Crime.
Clever Country: Education/Schools; Technology/Science.
Fun and Games: Arts/Culture; Leisure/Tourism; Personalities/Entertainment; Sports News.
Money and Work: Business/Finance; Work/Industry
Power and Policy: Politics; Military/Diplomatic; Media/ Communications;  Transport Issues
Social Matters: Social Issues; Environment; Health/Medicine; Consumer Affairs; Religion/Faith.

The quantitative data gave an idea of relative proportions of news content within and between news services identifying:

  • total bulletin times, and duration and percentages of stories from different categories (to be compared with the 2005 three-city data, and the baseline Perth data from 2001 where possible)
  • duration and percentages of stories with a potential multicultural impact  (compared with the 2001 study where possibly significant changes are noted)

The data from the other studies was recalibrated to reflect the same weekday-only content as for the 2007 survey.

The qualitative analysis focused on the detail of the reportage itself: the type of talent used, the way the talent was used (for example, speaking or silent), the way the talent was presented (for example, in formal or informal settings).  Additionally story tone was coded against a 9-point scale (adapted from Media Monitor’s Tone Ratings system) ranging from Extremely Negative to Highly Positive to get a sense of the impression left on the viewer by the way a story was reported.  

The coders were required to note which stories featured people from ethnic minority (EM) groups.  Following van Dijk  (2000) the researchers used this terminology to distinguish the diverse range of ethnic groups from the ‘(white) elites’ (39) which define mainstream Australia.  While this sort of identification is highly subjective, on the basis that the coders were viewers like anyone else they were asked to note where the people featured could be assumed by the audience to be identified as non-Anglo, whether through appearance, dress, accent, name or title.  Using this group of stories it would then be possible to examine:

  • What sort of stories ethnic minorities appeared in; 
  • What types of issues these stories were associated with;  
  • The ways in which people from non-Anglo backgrounds were portrayed;
  • The types of talent used;
  • The tone adopted in the presentation;
  • The role of pictures and graphics on the overall impact of the story.

The talent was identified in this way for both domestic and international stories. Reflecting the focus of the DIAC Living in Harmony program stories were also rated on the basis of the possible impact on community harmony to the extent that they might have a negative, neutral or positive impact, or reinforce a stereotype.


Member : Murdoch UniversityMember: Griffith UniversityMember: University of South AustraliaMember: Media MonitorsMember: SBSMember: University of CanberraMember: Journalism Education AssociationMember: University of Western Sydney
Department of Immigration and Citizenship