survey finds little support for csg -abc news., page-36

I apologise for returning to this rather old thread but it has been troubling me since I first read the media reporting. My concerns specifically relate to the release of the Preliminary report on community perspectives of gas industry developments in Page: Results from an election-day survey at the 2013 federal election prepared by Hanabeth Luke as a PhD Candidate and Dr David Lloyd of Southern Cross University and subsequent ABC Reporting.

You will perhaps forgive my cynicism if I question the motives of the people responsible for the distribution and subsequent reporting of this study including affiliates of Southern Cross University and the ABC.

The release of the report and ABC reporting coincided with a meeting held in Canberra with Ian MacFarlane, Minister for Industry, on Tuesday 22nd to specifically discuss CSG in the Northern Rivers. Due to the “fortuitous” timing of the report’s release, Lock the Gate representatives were able to present it at the meeting and place significant store in the report in their presentations to the Minister.

The release and unquestioning reporting by the ABC further diminishes the ABC’s already tarnished reputation for unbiased journalism and only serves to reinforce concerns that Southern Cross University might be more properly referred to as a centre for unscientific collaboration.

It appears that the release of the report might be in breach of the University’s own Code of Conduct. Part C Section (33) of the Code of Conduct requires:

“Officers and affiliates engaged in research are obliged to achieve and maintain the highest standards of intellectual honesty and appropriate standards of rigour in the conduct of their research as detailed in the Australian Code for the Responsible Conduct of Research.”

As well as including paid staff, the definitions of “Officers and Affiliates” includes students and volunteers contributing to the University.

Amongst other requirements, the Australian Code for the Responsible Conduct of Research Section 4.12.1 states: -

“Discussing research findings in the public arena should not occur until the findings have been tested through peer review. In discussing the outcomes of a research project, special care should be taken to explain the status of the project — for example, whether it is still in progress or has been finalised.”

The report authors at least acknowledge a passing reference to these requirements, presumably by calling it a “Preliminary” report and also the following afterthought placed at the very end of the report.

“The final analysis will be published in a peer reviewed journal, as per Southern Cross University protocol.”

But apparently the authors see no need to comply with this bothersome requirement in relation to their initial releases to the media.

Section 4.12.2 Australian Code for the Responsible Conduct of Research Section goes further and says: -

“To minimise misunderstanding about research outcomes, researchers should undertake to promptly inform those directly impacted by the research, including interested parties, before informing the popular media. “

It would be interesting to know if either Metgasco or the Richmond Valley Council, two parties most obviously “directly impacted by the research”, were advised at all of the research and the report.

Now just to take a look at the report itself.

The most obvious initial problem with the report is that it concludes 65% of the respondents from Richmond Valley voting booths were against CSG.

That this conclusion, based on a selected sampling of respondents, is so at odds with the results of the election in which every voter in the electorate recorded their preferences, does not seem to be at all troubling for the researchers.

Just to repeat those results from the 2013 federal election: -

The Stop CSG Party only received 638 votes out of the 79,267 votes cast in Page, or just over 0.8% of the vote, and
The vote for the Greens, the most strident anit-csg party in the area, dropped by 13.1% in the Senate and 13.5% in the lower house from the 2010 election.

The researchers apparently see no contradiction worth mentioning between the outcomes from their sampled research and the actual binding vote carried out at the same time.

It is difficult to undertake a more detailed analysis of the study as there is a significant lack of detail in the report - for example basic information such as the number of voters surveyed is not provided. While the report does note that “More in-depth results of this survey can be found at http://www.scu.edu.au/environment-science-engineering/index.php/89”, there has been no information provided at this site.

But some obvious questions are raised.

Given the very significant disparity between the results of the survey and the actual vote, there must be considerable doubt about the validity of the question posed to voters. The report notes “The same question posed in the 2012 Lismore City Council poll on CSG was presented to residents at ten polling booths in the Richmond Valley Council local government area (LGA). The question posed was: “Do you support coal-seam gas exploration and production in the Richmond Valley Council area?””

This question was initially prepared by the Southern Cross University at the request of the Lismore City Council, an organisation that is actively engaged in opposing any CSG development. In addition, I don’t believe the SCU received any input to the formation of the question from either of the two CSG companies operating in the area at the time, Metgasco or Arrow.

The report notes that “All the data was entered into an Excel spreadsheet and transferred into the statistical software program, SPSS.” However nowhere in the report is there any statistical analysis of the potential significance or margin of error for the results which such statistical software would routinely generate.

Although not confirmed in the report, the media reports at the time suggested that the survey was based on the responses from 600 respondents. At the time of the last Local Government elections there were 15,576 electors registered in the Richmond Valley Council area. So the sampling represents 3.85% of electors, significantly less than the “almost 6% of all voters in the Richmond Valley LGA” claimed in the report.

One of the more interesting observations in the report is in Figure 4 which shows that approximately 40% of those who said they were undecided did not agree that they would change their mind. In addition, a percentage of those who were undecided said they would NOT change their mind.

There are a number of other inconsistencies. The report notes that “Surveys were completed at ten of the eleven polling booths in the Richmond Valley LGA”; however Figure 1 “Map of polling booths in the Richmond Valley Council electorate” actually shows twelve polling booths. One of these booths, Casino Christian Comm. School, does not appear to have been a booth for the 2013 federal election and the following four booths are not mentioned - Fairy Hill Public School, Bentley Public Hall, Leeville Public School or Clovass-McKees Hill Memorial Hall. Indeed Figure 1 appears to be relevant to the 2012 LGA elections rather than the booths current at the time the survey was performed.

The very high potential for self-selection of respondents is not addressed at all. The report notes that “The 26 survey volunteers were SCU students and staff.” Now SCU staff and students would not be a representative sample of the residents of the Richmond Valley LGA. Presumably the respondents for interview were then self-selected by these staff and students. This results in a very high potential for skewing of the results.

The report also claims that “No external financial support was provided for this study. Instead researchers relied on pro bono contributions of time by SCU students and staff, with printing and logistical support provided through staff research funds from the University’s School of Environment, Science and Engineering.” It is interesting that the authors of the study consider that printing, logistical support and access to the statistical software package, SPSS would be available otherwise than through external financial support.

It will be interesting to see if a peer review of this study is ever made available.