Australian educational research on PISA

Since publication of PISA's first results in 2000, a significant body of literature has emerged. The main drivers of this research have been the many articles in the print and television media and the 2009 release of Australia's poor and declining scores. In the last 3 years alone (2015–2017), there have been eight peer-reviewed articles with a sole focus on Australian students’ PISA performance.

Table 1 summarises the eight articles on PISA and the three main themes: government and non-government school funding, education inequality and teaching quality. There is a lack of research exploring Indigenous performance in PISA. We explore the three themes to understand what educational issues have attracted more academic interest than Indigenous PISA performance.

Table 1.

The eight peer reviewed journal articles published since the 2015 round of PISA

The first theme is the difference in funding between government and non-government schools. Education academics have reported on the level of public funding for non-government schools (Dowling, Reference Dowling2008; Forsey et al., Reference Forsey, Proctor, Stacey, Koinzer, Nikolai and Waldow2017) and mostly agree that it is a flawed, complicated system (Connors, Reference Connors2007) that continues to undermine the government schooling system in Australia. The funding of schools is an important determinant of student educational performance. The literature has also argued that the growing divide between government and non-government school funding in Australia has wider implications for segregation and multiculturalism (Butler et al., Reference Butler, Ho and Vincent2017; Rowe and Lubienski, Reference Rowe and Lubienski2017), the socioeconomic status background of each sector (Watson and Ryan, Reference Watson and Ryan2010) and student educational outcomes (Perry and McConney, Reference Perry and McConney2013).

The second theme is inequality in the Australian schooling system (Kenway, Reference Kenway2013; Yates, Reference Yates2013). This theme has appeared repeatedly in the literature, while the Federal Government has introduced educational policy interventions with the intention to reduce the causes of Australia's educational inequality. For example, one policy is the promotion of ‘school choice’ under neoliberal ideology and practices. Many academics have examined the problematic nature of ‘school choice’ (Windle, Reference Windle2014; Angus, Reference Angus2015) and shown that the Australian schooling system has fallen short in fairness and equity, which has further been linked to the decline in Australian PISA results (Lingard, Reference Lingard2016). Surprisingly there is little, or no explicit, discussion of the iniquitous ‘gap’ between Indigenous and non-Indigenous students.

The third theme is teacher quality. There is concern that teacher quality may have fallen in recent years, and this correlates with the decline of Australia's PISA performance (Leigh and Ryan, Reference Leigh and Ryan2008; Treagust et al., Reference Treagust, Won, Petersen and Wynne2015). Leigh and Ryan (Reference Leigh and Ryan2008) found the academic aptitude of new teachers had fallen considerably over the past two decades. Recent initiatives to ensure quality teaching in Australian schools include the introduction of professional standards for teachers and the evaluation of teachers against those standards (Mayer et al., Reference Mayer, Mitchell, Macdonald and Bell2005). However, many academics have questioned the effectiveness of initiatives in teacher professional development (Bowe and Gore, Reference Bowe and Gore2017), teacher accountability (Thompson, Reference Thompson2014) and an oversupply of teachers (Dinham, Reference Dinham2013), which are combining to work against teacher quality (p. 92).

School funding, issues of inequity and teacher quality are also dominant in academic research on PISA. However, there is a notable lack of attention to, and neglect of, the performance of Indigenous Australian students in PISA. This neglect is surprising as the lowest performers are disproportionately from an Indigenous background (Thomson, Reference Thomson2016); there have been numerous policies to address Indigenous underperformance (Gray and Beresford, Reference Gray and Beresford2008); and Indigenous education research has increased for almost two decades (Mellor and Corrigan, Reference Mellor and Corrigan2004).

Indigenous Australian PISA results

Since the initiation of PISA in 2000 only four empirical research studies have investigated the achievement gap between Australian Indigenous and non-Indigenous students in PISA (McConney et al., Reference McConney, Oliver, Woods-McConney and Schibeci2011; Woods-McConney et al., Reference Woods-McConney, Oliver, McConney, Maor and Schibeci2013; Song et al., Reference Song, Perry and McConney2014; Sikora and Biddle, Reference Sikora and Biddle2015). McConney et al. (Reference McConney, Oliver, Woods-McConney and Schibeci2011) analysed science literacy performance in PISA 2006 and found that Indigenous students’ interest in science was equal to their non-Indigenous peers, while not having the same educational outcomes as other Australians. The researchers argued school science curriculum should acknowledge and publicly value Indigenous Australian science knowledge, and incorporate it into professional learning for teachers. Woods-McConney et al. (Reference Woods-McConney, Oliver, McConney, Maor and Schibeci2013) examined and compared factors associated with science literacy and with science engagement for Indigenous and non-Indigenous 15-year-old students in Australia and New Zealand and found consistent patterns of influence on engagement in science for Indigenous and non-Indigenous students in both countries. Socioeconomic status, time spent on science lessons and study and the character of science teaching explained variations in science literacy.

Song et al. (Reference Song, Perry and McConney2014) found the achievement gap between Indigenous and non-Indigenous students in PISA 2009 was related to unequal allocations and differences in the rates at which home and school resources are converted into positive educational outcomes. The study also found the achievement gap is substantially larger in Australia than in New Zealand, and is related to greater inequity in the allocation of school resources. Sikora and Biddle (Reference Sikora and Biddle2015) used PISA 2006 and 2009 data to investigate educational and occupational expectations of Indigenous and non-Indigenous youth. They found similarities in school-related factors that boost ambition of Indigenous and non-Indigenous boys and girls. In contrast, maternal and paternal role models influence Indigenous and non-Indigenous youth differently.

These four studies on Indigenous PISA results are largely comparative in their approach, and examine the extent to which the achievement gap between Indigenous and non-Indigenous students can be measured or attributed to various influencing factors both within and outside the school environment. There has been very little academic, empirical attention on Indigenous performance in PISA. This paper examines this neglect by analysing PISA discourse in the Australian media and the attention on Indigenous student performance.

Thematic analysis

Only 10 newspaper articles on Indigenous students in PISA formed the sample for this analysis, with seven of the articles published by Fairfax Media newspapers, two by News Corp Media newspapers and one by the Seven West Media newspaper (see Table 4). One newspaper, the Canberra Times, had three of the 10 articles.

Table 4.

The title given to the Australian Indigenous PISA newspaper articles published across the sample between 2001–2015 (n = 10)

It might be expected the mass media would specifically report on the large discrepancy between Indigenous student PISA scores and their non-Indigenous peers. While this does occur, not all articles address this issue with in-depth insight or analysis. Rather the articles, recognising the need to address the issue of Indigenous PISA performance, provided limited in-depth discussion that focused on embarrassing educational dilemma. From the limited discourse of only 10 articles there are three specific themes (see Table 5):

Table 5.

Outline of the central concept and boundaries attributed to each of the three themes for this thematic analysis

A 2004 article in The Sydney Morning Herald titled ‘Aboriginal performance tarnishes result’ documented the issue associated with poor Indigenous student results and the effect it was having on the overall Australian PISA results. For the author, Doherty, the results themselves were not problematic for Indigenous student populations, but for the impact on the nation's academic prestige. She notes ‘students are performing above the national average in reading, mathematics, science and problem-solving but share with the rest of the country a poor record on Aboriginal education’ (Doherty, Reference Doherty2004). The language used shows that for the author, Aboriginal education is tarnishing and letting down the rest of the country's otherwise positive results in the 2003 PISA.

The first theme argues that a lack of teacher quality in remote and rural schools is having a major influence on the Indigenous student PISA performance.

The focus on quality teaching in both metropolitan and national newspapers is interesting, as it has been identified as a cause for concern across the teaching profession in Australia. Reasons for the lack of quality teaching in remote and rural school settings are largely undiscussed in the media discourse. However, some discourse identifies the need for government and education authorities to identify and rectify a lack of quality teaching in remote and rural schools. One journalist's reflection outlines the potential future benefits of addressing this issue in remote and rural schools:

The language creates a negative ideological discourse towards students in rural and remote schools (which are highly populated by Indigenous students): without increased monetary support, rural and remote students will be left with a very poor future outlook. This shows how the media is capable of a bias: that students in these schools are potentially problematic, particularly in relation to future ‘costs’ to the rest of society if the issue is not addressed.

Similar to the first theme, the second theme argues for increased government funding, explicitly to address the Gonski report recommendations. Two metropolitan newspaper articles in 2012 and 2013 argued that the latest PISA results confirm serious inequity within Australia's education system, and therefore the need to commit to the most recent recommendations from the 2011 Gonski Review of Funding for Schooling—Final Report. However, both articles describe the federal government as uncommitted on the proposed funding recommendations:

This discourse has again highlighted the issue surrounding the first and second themes, where current funding is either inadequate or misdirected and seen as having a direct effect on the Indigenous educational dilemma. While the Gonski report might present a case for radical restructuring of the school education funding model, analysis of the media discourse shows its implementation remains unclear (Keating and Klatt, Reference Keating and Klatt2013).

The third theme focuses on the achievement gap between Indigenous and non-Indigenous students, and is referenced in five newspaper articles. While this represents half of the sample, the discourse provides little examination of the two and a half year achievement gap existing between Indigenous and non-Indigenous students according to PISA. One article, by journalist Trevor Cobbold, provides some critical dialogue on the various achievement gaps in Australia's 2012 PISA results:

The analysis also highlighted how the language and tone used delivers an ominous outlook for Australia's disadvantaged students, consistent with the negative outlook apparent in the first theme. The Indigenous PISA media discourse in the last 15 years has pitched a single message to its Australian readers—there is a stark educational disparity between Indigenous and non-Indigenous students according to PISA indicators that has not improved and warrants further attention in terms of teacher quality and monetary investment. However, while this is an important message, it becomes almost inconsequential when less than one article a year is published on Indigenous PISA performance, and some newspapers failed to publish a single focused article. In comparison, on average nearly 20 articles a year were published on non-Indigenous PISA issues with a yearly high of 57 non-Indigenous PISA articles in 2013. There is almost a complete paucity in media reporting that is not relative to the magnitude of this educational outcomes issue, or the social implications that it has for Indigenous youth, communities and Australian society.

The analysis highlights the lack of attention to Indigenous PISA education performance, six newspapers failed to address Indigenous performance in any way: The Brisbane Times, The Courier Mail, The Daily Telegraph, The Mercury, The Australian Financial Review and The Northern Territory News. News Corp newspaper articles have been silent on Indigenous PISA reporting. There are only 10 Indigenous PISA articles in the sample over 15 years, during which Australian educational policy has clearly defined Indigenous education as a priority area and the Council of Australian Governments (COAG) has produced numerous policies and practices to address the alarmingly high levels of Indigenous inequality and entrenched disadvantage (see e.g. COAG Reform Council, 2010). While Indigenous inequality and disadvantage has become more pronounced and evident in the trend of Australian PISA results, the media shows bias and neglect on this issue.

Indigenous PISA performance is one of the outstanding features of the Australian national performance profile and therefore deserved more attention by the media. This finding suggests mass media bias in what they do and do not print, that is consistent with other research. For example, other research has shown that the majority of newspaper articles involving Indigenous Australians are focused predominately on Indigenous athletes in sport (Hallinan and Judd, Reference Hallinan and Judd2009).

In the Northern Territory, where Indigenous Australians are 25.5% of the population, the highest proportion of any state or territory (Australian Bureau of Statistics, 2016), News Corp's Northern Territory News did not produce a single article that focused on Indigenous PISA results between 2001 and 2015 (see Table 4). In comparison Fairfax Media's Canberra Times produced 30% of the articles that focused on Indigenous PISA results between 2001 and 2015, from a territory that has the lowest proportion of Australia's Aboriginal people (Australian Bureau of Statistics, 2016). If the reporting were proportional to the minority status we would expect in the Northern Territory an article count that is comparable with the state's Indigenous population. This comparison further suggests that the lack of reporting on Indigenous PISA performance is not only inherent to News Corp newspapers, but is across all three owners of print media in Australia.

The media neglect of Indigenous PISA performance results is even more evident given the high number of articles on general Australian PISA performance. For example, between 2001 and 2015 the two nationally distributed Fairfax Media and News Corp newspapers, The Australian Financial Review and The Australian, published 96 articles on national Australian PISA performance. Of these articles, only one from The Australian examined the PISA performance for Indigenous students in PISA.

Here again the analysis showed the language in many media articles attempts to deflect responsibility from the governments upon which responsibility rests. While there is a strong feature critiquing teacher quality and the achievement gap, which is a central issue in the larger proportion of articles, there remains little critical analysis of the issues that have been raised in the thematic analysis. This lack of critical media discourse and attention to the Indigenous PISA performance points to the existence of bias in Australian mainstream media. As shown in previous studies that investigated Indigenous representation in the media, there have been continuous and unequivocal attempts to isolate and misrepresent Indigenous Australians through the controlled media discourse.

Media bias

Media bias, defined as: any form of preferential or unbalanced treatment by personnel within the mass media, in the selection of events and stories that are reported, and how they are covered (Lee, Reference Lee2005, p. 45) can help understanding of the lack of Indigenous PISA reporting in Australian media. Bias can happen in a number of ways: which stories are selected, selection bias (White, Reference White1950), how much attention is given to a story, coverage bias (Stempel, Reference Stempel1969), and how a story is reported, statement bias (Hofstetter, Reference Hofstetter1976). Media bias is well documented and continues to cause considerable international debate in the current media landscape i.e. the ‘fake news’ debate (Allcott and Gentzkow, Reference Allcott and Gentzkow2017). Our analysis suggests that all three biases are evident in the reporting examined.

The theory of liberal bias suggests that the media supports liberal causes and thinking and thereby influences readers to prefer liberal over conservative positions and beliefs (D'Alessio and Allen, Reference D'Alessio and Allen2000, p. 134). For example, Lichter et al.’s (Reference Lichter, Rothman and Lichter1990) United States study combined survey data with content analysis and argued that journalists project their shared, predominantly liberal views onto media coverage. Numerous studies have found there is a larger portion of media personnel who espouse their liberal attitudes to media coverage (Kohut, Reference Kohut2004; Groseclose and Milyo, Reference Groseclose and Milyo2005).

Studies have also investigated the conservative bias of the media, arguing that the media shapes public opinion in a manner that preserves the hegemony of society's ruling interests (Lichter, Reference Lichter, Kenski and Jamieson2017, p. 403). Bagdikian (Reference Bagdikian2004) argued that the concentration of the media is increasing and proposed that media owners and advertisers both shape the news directly and indirectly, through biases linked to media production. These biases include professional routines that define news to favour the rich and powerful, such as through overreliance on official sources to marginalise opposing opinions (Lichter, Reference Lichter, Kenski and Jamieson2017, p. 406). Similarly our analysis provides evidence for the marginalising practices of the media with nine of 10 Indigenous PISA newspaper articles failing to include Indigenous sources in their coverage.

Debate over media bias includes societal functions, including racial bias within mass media. Studies in the United States and Europe have argued that the media has promoted racial stereotyping and also regularly marginalised non-white voices (van Dijk, Reference Van Dijk1991; Jakubowicz et al., Reference Jakubowicz, Goodall, Marin, Mitchell and Seneviratne1994). For example, Henry and Tator (Reference Henry and Tator2002) provided a critical discourse analysis of case studies in Canadian and English-language press, showing how the media produces, reproduces and disseminates racist thinking. They argue that these kinds of racial bias are not isolated, and reflect a set of core assumptions, hypotheses and world views held by many who work in the mass media (p. 4). In our analysis this is evident in both extent and nature of media coverage on Indigenous PISA performance.

Australia has one of the most concentrated media ownership structures in the world (Newman et al., Reference Newman, Fletcher, Kalogeropoulos, Levy and Nielsen2017, p. 116) with Rupert Murdoch's News Corp having owned roughly 70% of the newspaper industry, and Fairfax Media with most of the rest (Griffen-Foley, Reference Griffen-Foley2002). Few researchers have investigated bias in the wider media, although there are studies on the ideological stance of newspapers (McKnight, Reference McKnight2010; Hobbs and Owen, Reference Hobbs and Owen2016), and their coverage of environmental issues (Beder, Reference Beder and White2004), minority groups (Kabir, Reference Kabir2006) and religions in Australia (Richardson, Reference Richardson1996).

While much of the available literature is on the American mass media and its political allegiances, there is research worldwide. It is therefore likely that in Australia, with its concentration of media ownership one of the highest in the world (Newman et al., Reference Newman, Fletcher, Kalogeropoulos, Levy and Nielsen2017), media bias and control not only exists but has substantial influence on issues reported on by mass media. Certainly our analysis here identified trends in relation to media ownership and the issues it chooses to ignore and report on.

Indigenous Australians in the Media

Our findings are consistent with widespread research on the connection between Indigenous Australians and the mass media. The general trend of media coverage defines Indigenous people as a threat or problem, and almost always negative and associated with anti-social activities (Jakubowicz and Seneviratne, Reference Jakubowicz and Seneviratne1996; Meadows and Molnar, Reference Meadows and Molnar2002). Historically research has also found there has been little change in the low level of use of Indigenous sources in media stories about Indigenous affairs (Hippocrates and Meadows, Reference Hippocrates and Meadows1996; Meadows et al., Reference Meadows, Hippocrates and van Vuuren1997). Another study by Bullimore (Reference Bullimore1999) found there is a considerable lack of Aboriginal voices in two of Australia's major newspapers, The Sydney Morning Herald and The Australian: ‘when Aboriginal voices did occur, they were mostly outnumbered by the voices of elite actors (such as academics, government officials etc.), or were mediated by white voices that appear on behalf of, and instead of, Aboriginal voices’ (p. 75). While this study does not refer to bias operating in the Australian media, it shows that the media makes a legitimate attempt to marginalise the views on Indigenous peoples through their underrepresentation in the reporting of Indigenous affairs. As stated earlier we have found similar evidence reflected in our analysis with all but one article from the three major news corporations failing to provide Indigenous sources in their reporting.

Many studies have attempted to further understand the role, nature and processes used by the mass media to help shape and construct dominant ideas and assumptions about Indigenous Australians and their place in society. Simmons and Lecouteur (Reference Simmons and Lecouteur2008) used a discourse-analytical approach to explore the ways in which a particular formulation, ‘the possibility of change’, was repeatedly implicated in descriptions of two ‘riots’ that received widespread media attention in Australia: one involving Indigenous, and the other involving non-Indigenous community members (p. 667). Their analysis of the media discourse demonstrated how, for the event involving Indigenous Australians, ‘change’ was repeatedly represented as an unachievable outcome. In contrast, descriptions of problems within the non-Indigenous community regularly represented ‘change’ as an achievable outcome. The study labelled this practice ‘as a modern form of racism, which uses more subtle language techniques so racist discourse may be plausibly denied’ (p. 684). This example shows that while bias is not mentioned, the media and its systems of representation are not unequivocally neutral, and the messages delivered within a discourse are invariably complex in nature and implications. As our analysis has shown, articles on Indigenous PISA performance use language that displays Indigenous education in a negative light, either as ‘challenging’ or ‘problematic’ for the nation.

Indigenous Australians have been further affected by how the media frames stories (McCallum, Reference McCallum2016; McCallum and Waller, Reference McCallum, Waller, Budarick and Hans2017). McCallum (Reference McCallum2007) performed a news frame analysis of seven Australian newspapers, from January 2000 to June 2006, reporting on incidents of ‘Indigenous violence’. Over the 6-year period, stories on Indigenous violence featured at regular intervals through stories of substance abuse, domestic violence and child abuse in remote communities. McCallum (Reference McCallum2007) suggested that narrow and stigmatising portrayals of Indigenous Australians were used by the media, who ‘sought to actively drive political agendas in Indigenous affairs through the construction of mediated public crises’ (p. 1). Again while bias is not mentioned, this study argued that seven newspapers played both a regular and active role in delineating chosen messages to their public audience.

While the words ‘media bias’ do not appear, it is clear the media decides what to report. Biases manifest in various ways, and although subtle, are clearly an observable phenomenon. Ewart (Reference Ewart1997) interviewed journalists from a regional newspaper and importantly found that the journalists were largely uninterested in reporting on Indigenous Australians and Aboriginal affairs, as they weren't considered news by the newspaper's editorial hierarchy (p. 112). When an issue was deemed newsworthy, Ewart (Reference Ewart1997) states that ‘race or colour are usually mentioned where Indigenous Australians are involved in issues with negative slants, but seldom are subjects of stories described as Anglo Saxons’ (p. 115). The self-described practices of these representative journalists show that their reporting is undoubtedly manipulated by higher editorial powers, showing how the discourse can be easily forfeited to inherent biases of the media. Our analysis has shown that across a 15 year period a mere 10 newspaper articles reported on Indigenous PISA performance, illustrating a clear disregard for the educational issue across the Australian mainstream media.