Estimating Media Slant in Australia

Joshua Gans and I have a new paper out today on media slant. Here’s the abstract (click the title to see the full paper).

How Partisan is the Press? Multiple Measures of Media Slant
Joshua S. Gans & Andrew Leigh
We employ several different approaches to estimate the political position of Australian media outlets, relative to federal parliamentarians. First, we use parliamentary mentions to code over 100 public intellectuals on a left-right scale. We then estimate slant by using the number of mentions that each public intellectual receives in each media outlet. Second, we have independent raters separately code front-page election stories and headlines. Third, we tabulate the number of electoral endorsements that newspapers give to each side of politics in federal elections. Overall, we find that the Australian media are quite centrist, with very few outlets being statistically distinguishable from the middle of Australian politics. It is possible that this is due to the lack of competition in the Australian media market. To the extent that we can separate content slant from editorial slant, we find some evidence that editors are more partisan than journalists.

Across 27 media outlets, we find only two that stand out as partisan in either direction. Using our first approach, only ABC TV appears slanted (towards the Coalition – sorry, Mr Costello). And using the second and third approaches, only The Age appears slanted (towards Labor). On the whole, Australian journalists seem a centrist bunch; their editors a little less so.

Update: See also discussions about our methodology by Sinclair Davidson and Andrew Norton, plus a response from Joshua. Mark Bahnisch has more too.

Advertisements
This entry was posted in Media. Bookmark the permalink.

11 Responses to Estimating Media Slant in Australia

  1. Pingback: Estimating Media Slant in Australia « Andrew Leigh | australianews

  2. wilful says:

    well quelle surprise.

    Though I do wonder about this, I would be sure that the Australian would come out as strongly coalition biased.

  3. ChrisPer says:

    Overall, centrist? Amazing. The mean is near the median. However, the bias leftward in public intellectuals is not compensated by putting some hacks from the Liberal Students Club, or whatever right-wingers in the list, because they still won’t rate points in your count of parliamentary mentions.

    That the ABC heads slightly coalition-wards startles me, and makes me want to study the methodology very closely. In reading the first part of the paper I am struck with uncertainty that what is being measured is actually helpful in understanding bias.

    In restricting the period, you cover without mention the ‘three years public hate’ period 1996-1999 of the gun control campaign (when the Coalition was strongly supported by media) yet deny interest in liberty as an issue in Australia as evidenced by gun control, abortion et al. You are correct that liberty is not a driver for most media people’s values, I think.

    However analysing the framing of issues and choice of what stories to cover is the key to rating bias. I don’t see that discussed in your paper at all. The presumptions of ‘nice educated peoples values’ create a strongly negative framing of stories on key Coalition positions that somehow was toned out when Labor was being examined even though Hawke/Keating maybe gave us more practical economic rationalism than the Libs. The automatic balance airtime to ratbag activists frames them as equals to Government ministers, just because ‘he said she said’ journalism is so common.

    For framing, look at Bolt and Blair posts about Kerry O’Brien interviews and Media Watch target selection to see many instances of how ‘right wing’ framing differs from standard instituional biases. Another Eg would be the IPA article on the media framing of the waterfront conflict in the late nineties, and the sidebox on their old paper of ABC bias in which Katharine Betts lays out with clarity how ‘moral status display’ forms a kind of competition that is not related to which millionaires happen to own shares in the media outlet. Also her book ‘The Gret Divide’ is very pertinent.

    Those those references cover part of your research period, and ten years ago contributed to the understanding of how bias works in Australian public discussion, but they didn’t make your reference list.

    Would it be fair to say that you found the idea of non-cash motives for competition irrelevant?

  4. ChrisPer says:

    Curse my splling.

  5. Philips Adams a right-winger? lol

  6. Pingback: Andrew Norton » Blog Archive » Can public intellectuals be used to assess partisan media slant?

  7. John Simmons says:

    The entire idea of constructing the same tests of bias that have been done in America (while throwing in a couple of other biased tests) is a poor one.

    As a consumer of a fairly vast amount of news across an entire spectrum of media, I have to inform you that neither reportage nor parliament are influenced by public intellectuals in any significant way. This observation is confirmed by your finding that negative mentions of public intellectuals in the media were rare. Wouldn’t media slant necessarily mean more often attacking the ideas of intellectuals who hold antithetical views? It seems strange that media bias would purely be reflected through the positive mention of intellectuals whereas political slant is reflected in both negative and positive mentions.

    In the US, as noted in your research, the relatively larger size and number of think tanks makes the US method meaningful, since raw mentions of think tanks necessarily mean the media is advocating a think tank’s position. This is not so true with public intellectuals.

    Indeed this is confirmed by looking at the table of intellectuals and the dearth of mentions they had in nearly 10 years of parliament. Weighting public intellectuals according to the number of mentions means this research is highly reliant on a very small number of public intellectuals. This kind of method is likely only to reveal which public intellectuals are the best self-promoters in the media.

    More nonsensically you compiled the list of public intellectuals from… a newspaper! A newspaper which, you have discovered reflects a left-ward bias. Then you provide an ad-hoc correction of the bias by adding minor right-wing researchers from right-wing think tanks.

    Your second method also fails to give a meaningful result. If the independent “coders” for the study were obtained from within ANU and randomly assigned, you would still obtain a political bias in your findings. It is further bizarre that you confirm your results via the bias-o-meter of a media outlet (which would, one might imagine, be biased itself) by looking at a correlation to your findings! Once again this appears to reflect the difficulty in finding non-biased means to analyse the results.

    Your third method has some validity. However I would not rely on endorsements as ipso facto revealing media slant.

    Producing such politically charged research does not do economics research a great service. Even though it appears the slants you discovered were quite small (indeed sometimes within the SE you measured), it is open to be reported very sensationalistically.

  8. ChrisPer says:

    Whoops sorry you already pointed it out Andrew.

  9. anon says:

    Choose a poor set of measures? CHECK

    Reach a surprising conclusion that goes against most peoples’ intuition? CHECK

    Argue the results are capturing the underlying phenomenon of interest rather than just acting as a poor set of instruments? CHECK

    Yep, my robustness checks find this paper sits just about at the mean of econometrics papers in the post-freakonomics world.

  10. Pingback: Is Gerard Henderson a leftie? – Corporate Engagement

Comments are closed.