Bryce Edwards: Political round-up: Setting standards

National Standards data has been published online today. Photo / Thinkstock

National Standards data has finally been published online today by the Ministry of Education. You can see the information at the Education Counts website. However, this official release has been overshadowed by the decision of Fairfax Media to go for the big scoop by publishing some initial data from over 1,000 schools a week ago - which you can see here: School Report website.

The Fairfax project was a major effort, accompanied by many interesting articles and an acknowledgement of the problems with the raw data - see John Hartevelt and Clio Frances' How New Zealand schools rate. Project co-ordinator John Hartevelt writes that the journalists expected criticism (they surely got it) and felt the need to defend their decision to publish - see: The School Report project: a balanced view. Unsurprisingly various Fairfax newspaper editorials have been written in support of the publication of the data - see the Press' editorial Raising the standard and the Dominion Post's A parental right to know.

But just being open about the data's shortcomings is not enough says Keith Ng: 'You have missed the point of all those people pleading with you not to release the data. It's not that people like me think people like you are unable to draw the right conclusions from the data. It is that, if subjected to standards of statistical rigour, there are aren't many right conclusions that can be drawn by anyone' - see: Because Statistical Rigour. Both Hartevelt and Ng will go face to face on Russell Brown's Media3 programme this Saturday morning on TV3. Brown also takes other reporters to task in Standards Showdown.

Of course, all the qualifying statements about the 'ropey' data didn't stop most from looking for some conclusions anyway. You don't have to be a statistical genius to see the very direct correlation between the raw results and decile rankings of schools - see Danyl Mclauchlan's Chart of the day, almost as if there's some sort of relationship edition. As one commenter noted, it seems the national data is an expensive and clumsy system for measuring exactly what the decile ratings produce already. If that becomes one of the major themes then the policy could seriously backfire on the government. The persistent 'tail' of under achievement, in what is internationally recognised as a high performing education system, has been the main stated targets of national standards. If the results actually show that this 'tail' is largely the result of growing deprivation and inequality then it may be poorly performing politicians rather than bad teachers and schools in the firing line.

Headmaster of Wellington College, Roger Moses, doubts that this underachievement is the result of 'abysmally poor' teaching but rather of: 'Intergenerational unemployment, a widening gap between the haves and have-nots, the challenges of redress for Maori, and the integration of new immigrants to our country are all major challenges that our country is facing. Inadequate teaching, I would suggest, is not the cause of these deep-seated issues, any more than inspired teaching can be the sole panacea' - see: NZ education is among the world's best. Similar arguments are made by educationalist Terry Crooks in Standards no help to 20pc of children who struggle but ex-Secretary of Education Howard Fancy argues that the education system will be much better off having the information - see Better teaching and learning.

If anyone thought statistics would resolve the arguments over indentifying the problems and solutions, then they may have to think again. David Farrar gets excited by the possibilities for analysis, pointing to work by Luis Apiolaza looking at the variations within deciles - see: Good data analysis. However Danyl McLauchlan has a graph based on the same data to measure the same thing but appears to use a slightly different calculation and so draws a different conclusion - see: The fantasy and reality of national standards. Similarly, the Herald on Sunday declared National Standards shock: Big classes work, only to be taken to task by both Danyl Mclaughlan in Well below standard in analysis and Keith Ng in Re: Education in a 'textbook' rebuttal.

Eric Crampton details some of his serious number crunching and concludes that while class size does seem to relate to results there are probably other factors at play. Crampton shows that decile ratings and ethnicity are clearly the strongest determinants but variations within deciles need further explaining - they may or may not be related to the quality of the teaching see: Education regressions.

The Emil Demesne blog makes a very good case against school league tables and standardised testing, but for the collection and other use of the data: National standards: this data is not for ranking. This issue is one where the blogosphere really added value to the debate. Shallow or self-serving analysis will be ruthlessly examined by people who know their way around a scientific calculator and/or a classroom.

There will no doubt be a fresh round of analysis next week based on todays official release but, as David Farrar points out, at this stage the information from National Standards is just a snapshot of unmoderated data, and more value will be added over time as trends are identified. Opponents probably have a hopeless cause trying to put this genie back in the bottle, arguing for the suppression of information about a subject that much of middle New Zealand is obsessed with - see Vaimoana Tapaleao's Public wants school data: poll. The media simply won't be able to resist.

Herald on Sunday deputy editor Jonathan Milne thoughtfully but somewhat sadly reflected on the British experience of similar policies put in place during the Blair years - see: Lessons of the motherland and soberly reported that they had decided not to publish ranked league tables. However, he then went on to rummage through the 'ropey' data for the shock headline on class sizes.

Elsewhere on the web today:
Justice Neazor's report on the illegal spying on Kim Dotcom is being labeled a whitewash and an anti-climax by a variety of commentators and politicians. Perhaps the most damning coverage is in John Armstrong's column, A washout of an inquiry, and an apology that could easily backfire, in which the report is described as setting 'a new low' in avoiding issues of accountability and dishing out a mere 'slap over the wrist with the old wet bus ticket'. Tracy Watkins' column stresses 'John Key's evisceration' of the 'inept' GCSB, but also slips in an alternative theory about how the illegal action could have come about: 'the other, more sinister, explanation - that the GCSB spied on Dotcom knowing it was unlawful but, because it is so heavily protected by secrecy, believed it would never be caught. It was only rumbled because a policeman tripped over his words in court' - see: PM doesn't hold back on inept spies.

Much of the commentary in Parliament, the media and the blogosphere is fairly scathing of Justice Neazor's report. At the extreme end, The Standard blogsite has the angriest commentary - see, for example, Key fails to guard the guardians and What does Key have to gain by lying? Similarly, No Right Turn points out what the report omits in The Neazor report. For the best explanation of why the Inspector-General's report is inadequate, see Gordon Campbell's On the failures of the Neazor report. In general, however, the main reaction seems to be the call by all and sundry for a full and independent inquiry - see Adam Bennett and Claire Trevett's Pressure grows for spy fiasco inquiry.

David Farrar provides a useful contrast to this consensus in his blog post on The Dotcom case. He gives a thoughtful and lengthy evaluation of both the case and Mr Dotcom himself. Nonetheless, there are plenty of others thoroughly enamoured with the internet tycoon. All sorts of fanciful and playful ideas are now being suggested in social media - Could Kim Dotcom eventually stand for Parliament? Could he start his own political party to replace Act? See for example, an architecture blogger calling for Dot Com for PM. These are signs of just how much the Dotcom fiasco has caught the public's imagination.