Wednesday 30 July 2014

Who wins the Education Debate ?: UMR and Herald-Digi Polls on Quality Teachers vs Class Size




Herald-DigiPoll

The Herald have just released further results from a Herald-DigiPoll (part of their
Mid July political poll), which finds that "New Zealanders would rather money was
spent on improving teaching standards" - ostensibly National's position - "than on reducing class sizes" -  allegedly the core of Labour's policy. (see article here
 http://www.nzherald.co.nz/nz/news/article.cfm?c_id=1&objectid=11297568 )

The Herald-DigiPoll asked respondents: 'Labour has promised to lower class sizes by hiring 2000 more teachers instead of spending $359m National would spend on trying to improve teaching standards. Which of the following best fits your view ? (1) The money
is better spent on cutting class sizes or (2) The money is better spent on trying to improve teaching standards.'

The result:

Cut Class Sizes 34.5%

Improve Teaching Standards 61.3%

Don't Know/Refused 4.2%

Understandably, Labour's education spokesman, Chris Hipkins, has rejected the whole premise of the Poll question, pointing out that a focus on improved teacher training and quality was also a central facet of the policy package announced at the Labour Party Congress a few weeks back. Amongst other features, for instance, Labour announced it would use pre-screening to raise the standard of entry into teaching courses, would establish a comprehensive school advisory service to share best practice and act as mentors and advisors to teachers and would redirect resources currently spent on enforcing National Standards into teacher development programmes.

While The Herald would no doubt argue that the Parties' respective policy platforms needed to be greatly simplified in order to make them more easily comprehensible for poll respondents, there seems little doubt that Herald-Digi have asked a leading question, here (some have described it as push-polling) - framing things in a highly reductive Labour = Smaller Classes / National = Higher Teaching Standards narrative. And the problem goes much further than simply the exclusion of Labour's improved teacher training proposals. As a regular commentator on The Standard Blog has pointed out (  http://thestandard.org.nz/tuwhera-mike-23072014/#comment-853451 ) "What do they think the point of smaller classes is if not higher teaching standards ? Surely higher teaching standards is the desired result of both approaches.... But they're comparing the method on Labour's part with the desired goal on National's".  Some have suggested a more accurate wording of the poll question would have been something along the lines of: Labour and National both say they want to improve the quality of teaching and outcomes for students. Do you prefer Labour's policy of seeking to achieve this by training more teachers in order to reduce class size OR National's policy of seeking to achieve this by financially-rewarding and utilizing executive principals and selected high-performing teachers ?

It's perhaps not all that surprising then that, on the simple and somewhat inaccurate terms in which the debate has been rendered by the MSM, most voters do, indeed, prefer National's strategy.


The Dominion Post Weighs-In

Nor is it surprising when one considers the broader MSM framing of Labour's education policy over recent weeks. Labour leader David Cunliffe announced the policy in a keynote address to the Labour Congress on Sunday July 6. And on the Monday immediately following, my local daily, The Dominion Post, seemed to want to nip any resulting cut-through with voters in the bud with its Front Page headline: Cuts only Half the Story - Educators.

"Labour's proposal to reduce class sizes", Dom Post journalists Michael Forbes and Olivia Wannan assured readers, in what seemed like an explicit, pre-prepared, in-your-face rebuttal, "has failed to win a universal gold star, with experts saying the small cuts without improving teaching would do little to raise the bar of student achievement."

Click on Read More for full Stats and Analysis



Let's put aside for one moment the fact that, as Hipkins and others have made clear, Labour had, in fact, proposed a policy package that would both reduce class sizes and upgrade professional teacher training, while eschewing the current competitive, bonus-driven environment and the specifics of National's so-called super teacher policy.

The more immediate question is: precisely who were these "experts" (in the plural) ?  After all, the relationship of reduced class sizes to improved student achievement is a matter of some contention in education circles, with the experts far from unanimous in their views. How, then, was the Dominion Post able to present this as some sort of expert consensus ?

Five people were quoted in this lead article, only three of whom could in any way be considered  objective, politically-neutral experts. (the fourth was  Education Minister, Hekia Parata herself and the fifth was "Miramar mum Lesleigh Norrington").

The views of the first, Professor John O'Neill, of Massey University's Institute of Education, could certainly be said to broadly concur with the Dom Post's reductive rendition of the debate. O'Neill suggests Labour's proposal wouldn't achieve much without changes to teaching itself, arguing that recent research suggested making classes slightly larger or smaller did not greatly alter the achievement levels for average students. He did, however, agree that class size could make a big difference for children "on the margins", arguing that Labour's policy would be effective if it targeted lower decile schools. O'Neill went on to suggest that Labour's approach would see the $359m spent on children rather than teachers, which he considered "a pleasing aspect."

The second expert quoted was Wellington College headmaster Roger Moses, the Principal of what is, arguably, one of Wellington's more conservative, establishment schools. Perhaps not quite Christ College or Wanganui Collegiate, but not necessarily all that dissimilar either. And yet Moses sounds rather more enthusiastic about Labour's policy than the Dom Post's summary of expert opinion would have us believe. While he emphasises that changes to class sizes would require close consultation with schools beforehand to ensure teachers were actually giving students more one-on-one time and that "it's a more complex issue than simply making classes smaller", he went on to emphasise that "I  know from experience that if you go from having to mark 30 exams to 20 exams then it frees up a lot more time to spend with students. Common sense would suggest there are advantages to smaller class sizes."

And the third and final expert quoted in the report, New Zealand Educational Institute President Judith Nowotarski, was positively glowing in her praise for Labour's new policy, suggesting smaller class sizes would make a huge difference to the quality of teaching and learning, that international studies showed the benefits went well beyond the school gate and that children who benefit from being in smaller class sizes are more likely to have better reading skills, complete their education, and have lower likelihood of unemployment.

So only three people who could in any way be characterized as experts. One, Judith Nowotarski,  positively glowing in her praise of Labour's policy. A second, Roger Moses, largely positive, albeit with one or two provisos. And only the third, John O'Neill, seeming to fit the Dom Post's characterisation of expert opinion.

How many readers, though, would have taken the time to think this through ? The headline, the tone, the inaccurate rendering of the issues at hand, the core message of the Front Page article (essentially: Experts prefer National's policy over Labour's) would each, no doubt, have had some effect in shaping perceptions of Labour's new policy and of the wider debate in general.


UMR Research Poll

There is, however, some important polling evidence that suggests a majority of voters do, indeed, favour smaller class sizes. In late June, the PPTA commissioned UMR Research to conduct a poll (750 respondents over the age of 18, nationwide) on attitudes to class size. Specifically, people were asked to identify what they thought should be the maximum number of students in a secondary school class in order to facilitate learning and higher achievement and for their views on current class sizes.

In terms of the latter, a plurality of poll respondents (46%) believed that class sizes are generally Too Big in New Zealand, 8 points higher than the proportion believing class sizes are either About Right (36%) or Too Small (2%). The remainder were unsure.

On the question of preferred maximum number of students per class:

- An overwhelming majority believed the number should be 25 or fewer - 83% of the entire sample (86% of those who were parents, and 90% of women with children).

- Within this overwhelming majority, 45% of the entire sample (48% of parents, 52% of women with children) felt the maximum should be 20 or fewer.

- The results crossed Party lines, with not only 87% of Labour voters and 84% of Green voters choosing 25 or less, but also 78% of National voters.

It's fair to conclude, then, that these UMR Research results do offer at least a tacit endorsement of Labour's broad approach over National's. The National Party is essentially comfortable with the status quo (secondary school classes continuing to have more than 25 students) and emphasises instead the idea of quality (empowering selected highly-paid teachers deemed to be high-performing) over quantity (more teachers and smaller classes). Indeed, many voters still, no doubt, associate National with the desire for even larger classes - public mobilisation a couple of years ago forcing Education Minister Hekia Parata to back down from a policy that would have resulted in the slashing of teacher numbers and a concomitant rise in teacher-pupil ratios. Labour, of course, have come out strongly in favour of reducing class sizes with the aim of getting student numbers down to 23 per class by 2018.

Clearly, most New Zealanders agree that smaller class sizes are a good thing.

To explore this tacit support for Labour's position further, I've set out the detailed demographic breakdowns for both questions below.


Attitude to Current Class Sizes

In Table (1), Big=Too Big, Right=About Right, Small=Too Small and AR+TS=About Right and Too Small combined.

The colour-coding provides a rough idea of which demographics incline towards Labour's position (and to what degree) and which are happier with the status quo and therefore lean towards National's approach (and to what degree).

Specifically, demographics with figures rendered in purple denote social groups where the proportion choosing Too Big significantly outstrips the combined About Right+Too Small figure (by more than 10 points). So, for example, 52% of women feel class sizes are currently too big and this outstrips the AR+TS figure by 22 points (hence the figures are rendered purple).

Statistics in Red denote those demographics in which Too Big mildly outstrips the aggregate AR+TS figure (by 1-10 points).

Light Blue figures are used to highlight demographics mildly favouring AR+TS over
Too Big (by 1-10 points), while Dark Blue denotes demographics where AR+TS significantly exceeds the Too Big figure (by more than 10 points).

-----------------------------------------------------------------------------------------
(1) Class Sizes in New Zealand
Question: Based on what you have seen, read or heard, do you think class sizes in New Zealand state secondary schools are generally Too Big, Too Small or About Right ?
   
Demographic   Big   Right   Small   AR+TS    Unsure   Gap   Base
                                 
All....................46.......36...........2..........(38).........16.........+8......750

Male................38.......45............2..........(47).........15.........+9......359
Female............52........29............1..........(30)........18.........+22.....391

Dependent Children
Yes..................54.........34............2..........(36)........10........+18.....270
No...................41.........38............1..........(39).........20........+2......480

Gender and Children
Malewith dependent...44........43............4..........(47).........9..........+3......131
Femalewith dependent63.......26............0..........(26).........11........+37.....139

Age Group
18-29................33.........53............2.........(55)..........12........+22.....157
30-44................53.........30............3.........(33)..........14.........+20....225
45-59................56.........28............2.........(30)..........14.........+26.....196
60+...................37.........37............0.........(37)..........26...........=.......172

Location
Auckland.........38.........40............3.........(43)...........19........+5......241
Wellington.......44.........36............0.........(36)...........20........+8.......77
Christchurch...52.........39............0.........(39)............9.........+13.....82
Provincial........49.........33............2.........(35)...........16.........+14....350
Rural................44.........38............2.........(40)...........16.........+4......150

North Island.....44.........36............2........(38)...........18.........+6......568
South Island.....51.........36............1........(37)...........12.........+14.....182


Party Support
National............43..........42...........1.........(43)..........14...........=.......269
Labour..............47..........39...........2.........(41)..........12..........+6......174
Green................52..........31...........4.........(35)..........13..........+17.....65
NZ First............47..........39...........3.........(42)..........11..........+5.......24

Personal Income
Less than 15k...47...........40...........2..........(42).........11..........+5......82
15-30k...............48...........35...........1..........(36).........16..........+12...146
30-50k...............43...........36...........1..........(37).........20..........+6.....137
50-70k...............52...........30...........3..........(33).........15..........+19...120
More than 70k.44...........42...........2...........(44)........12...........=......152

Education
Secondary.........43...........39...........2...........(41)........16...........+2.....237
Polytech/
Trade.................51...........27...........3...........(30)........19..........+21....171
University..........44...........40...........1...........(41)........15..........+3......303

Ethnicity
Pasifika..............35............45...........9...........(54)........11.........+19......38
Maori.................64............25...........1...........(26)........10.........+38......83
Euro/Asian........44............37...........1...........(38)........18..........+6.......632
----------------------------------------------------------------------------------------

As the above table shows, some demographics express unusually strong sentiment (figures in purple) that class sizes are currently too large. These involve an interesting mix of (1) demographics that, according to most party support breakdowns, disproportionately support the political Left (women (especially women with dependent children), the younger wave of the middle-aged (30-44 year olds) and Maori) and (2) certain demographics that are currently (or historically) less inclined to support Labour and the Left Bloc (people in later middle-age (45-59 year olds), those with an education up to Trade or Polytech level, people in the higher income bracket of 50-70k, along with voters in Christchurch, the Provincial Centres and the South Island as a whole).

More generally, the poll suggests there are quite a few demographics where the proportion favouring Labour's argument that class sizes are currently too large significantly outstrips support for Labour and the Left (basing the latter figure on breakdowns from recent Herald-Digi and Fairfax-Ipsos polls). Large minorities, for instance, of Men (38%), Men with dependent children (44%), rural New Zealanders (44%), the wealthy (also 44%) and even National Party supporters (43%) favour Labour's argument on class size.

The UMR Research Poll, therefore, suggests there exists at least a latent potential for Labour to both consolidate its existing support bases and make inroads into some of National's key demographics. All the more so when you consider Education regularly tops various polls on what voters consider to be the most important issues for the coming Election. A February One News Colmar-Brunton poll, for instancefound that education was the top election issue, with 40% of the sample mentioning it (ahead of Health 37%, Jobs 30%, Child Poverty 27% and a range of other issues chosen by smaller percentages). What's more, a particularly large proportion of both Undecideds (44%) and National supporters (45%) saw education as the key issue. There exists, then, at least the potential for Labour's education policy to resonate with what some commentators would call "middle New Zealand".





Preferred Maximum Class Size

By far the most important thing to say about the poll's findings on maximum class size is that an overwhelming majority felt the number should be 25 or less (ie lower than current secondary school class sizes). This is true of all of the demographics - even those with the lowest 25 or less figure (Men 77%, National voters 78%). Clearly, even National's core support-base disagrees with its relaxed stance on class size.

Hence, in Table (2), I use colour-coding simply to highlight those demographics that diverge (whether significantly or mildly) from the average. In contrast to the colour-coding in Table (1), therefore, figures rendered here in Light Blue or Dark Blue do not denote social groups leaning towards National's stance but rather simply highlight those where the overwhelming majority in favour of 25 or less is somewhat smaller than for the sample as a whole.

But, first, a word on the columns in Table (2). The options given to respondents were that the maximum class size should be either 10, 15, 20, 25, 30, More than 30 or Unsure. In order to keep things easily digestable, I've simplified things a little. The figures for each demographic below the first column (ie 15) denote the combined percentage choosing either 10 or 15 as their preferred maximum. The figures under the second column (20) denote those within each demographic choosing either 10, 15, or 20 (in other words, 20 or less). Hence, the percentage under the 15 column is included within the percentage under the 20 column (for example, the 11% of women who chose 15 or less (first column) are included within the 47% of women who chose 20 or less (second column)). Similarly, the percentage figures in the third column (25), include the figures from the first two columns (ie 88% of women chose either 10, 15, 20 or 25 as their preferred maximum). Or, to put it all another way, 15 = 15 or less, 20 = 20 or less, and 25 = 25 or less.


I take the 25 or less figure (ie percentages under the 25 column) to represent those leaning towards Labour's argument on class size and the proportion under the 30 or more column to be those comfortable with National's stance. Obviously, respondents choosing 20 or less (the 20 column) can be considered particularly supportive of Labour's view.

The colour-coding: Figures in purple denote those demographics where support for a particular maximum class size is 8 points or more higher than average (for any or all of the first three columns - 15, 20, or 25) or 8 points or more lower than average for the 30 or more column. In other words, those demographics that particularly lean towards Labour's stance. Figures rendered in red denote the same thing for demographics mildly leaning towards Labour's position. Mildly = 4-7 points from average.

Figures in dark blue are the polar opposite of those in purple and figures in light blue are the polar opposite of those in red.

I use colour-coding for each column. So, for example, it's quite possible for a particular demographic to have figures in both red (or purple) and light blue (or dark blue). This would suggest a demographic where views tend a little more toward the 'extremes'. 18-29 year-olds, for example, are significantly more likely than average to choose 15 or 20 as their preferred maximum (hence those figures are rendered in purple). But they are also mildly more likely than average to choose 30 or more (hence that figure is rendered in light blue).

--------------------------------------------------------------------------------------
(2) Maximum Class Size
Question: In terms of supporting learning and ensuring students do as well as possible, what do you think should be the maximum number of students in a secondary school class ?

Demographic        15        20         25        30or more   Unsure    Base

All.........................10........45.........83...........14..............3...........750

Male......................9..........42.........77...........20..............3...........359
Female.................11.........47.........88............9...............3...........391

Dependent Children
Yes........................12.........48.........86...........12..............2...........270
No.........................10.........44.........82...........14..............4...........480

Gender and Children
Malewith dependents.......10.........44..........83...........16..............1..........131
Femalewith dependents...13.........52..........90............8...............2..........139


Age group
18-29......................18.........53..........81...........18.............1..........157
30-44......................11.........48..........86...........12.............2..........225
45-59.......................5..........43..........84...........12.............4..........196
60+..........................9..........37..........81...........15.............4..........172

Location
Auckland...............11.........38..........80...........18..............2.........241
Wellington..............8..........45..........81...........15..............4.........77
Christchurch..........5..........43..........82...........16..............2.........82
Provincial..............12..........51.........86............11.............3.........350
Rural.......................9..........45..........82...........18..............0.........150

North Is.................11..........44..........83...........15..............2.........568
South Is..................8...........48..........84...........12..............4.........182

Party Support
National..................5...........35..........78............20..............2........269
Labour...................14..........49..........87............10..............3........174
Green.....................10..........52...........84............13.............3.........65
NZ First.................29..........65...........89............3...............8.........24

Personal Income
< 15k........................9...........43...........89.............9..............2.........82
15-30k.....................15..........43...........82............15.............3.........146
30-50k.....................12..........51...........80............18.............2.........137
50-70k......................8...........48...........87............10.............3.........120
>70k.........................7...........39...........81............17.............2.........152

Education
Secondary...............12..........45...........82............15.............3.........273
Polytech/
Trade......................10...........45...........82............15.............3.........171
University................9............45...........84............13.............3.........303

Ethnicity
Pasifika...................17...........47...........95.............5..............0.........38
Maori......................16...........60...........93.............7..............0.........83
Euro/Asian..............9............43...........81............16.............3.........632
--------------------------------------------------------------------------------------

There's not necessarily all that much one can say about the 25 or less figures, given the overwhelming proportions in each and every demographic. Except, perhaps, to note that support for a 25 or less maximum is nigh on universal for women (particularly women with children), for Labour and NZ First supporters, for certain income groups and, above all, for Maori and Pasifikas. (the latter's almost 100% support for a 25 or less maximum, incidently, is in striking contrast to the response to Question (1), where Pasifikas were conspicuous in their unusually weak support for the 'Classes are currently Too Big' option). The young (18-29 year-olds) display a similarly 'conflicted' view (unusually strong in their support for the view that class sizes are currently About Right (Question (1)), but also among the most supportive of maximums of 20 or less).

More interesting are the 15/20 or less figures. These highlight demographics that can be considered particularly supportive of Labour's argument. And, once again, as in the first question, you can see at least the potential for Labour to both consolidate and expand its support-base. Absolute majorities of traditional Labour or Left-leaning social groups - women with children (52%), Maori (60%), the young (53%) - prefer a maximum class size of 20 or less. But so too do voters in the currently National-leaning Provincial Cities (51%) and middle income groups (51%), along with NZ First supporters (a whopping 65% - albeit on a very small base of just 24 respondents).

And, again, you can even detect some potential for Labour in demographics that are not unusually supportive of a 20 or less maximum. For instance, the sizeable minorities of men (42%), men with children (44%), people in late Middle-age (43%), Rural New Zealanders (45%) and the two wealthiest socio-economic groups (39% and 48%) who choose 20 or less are all far larger than the proportions in those demographics telling pollsters they intend to cast their vote for Labour (or even, indeed, for the Left Bloc as a whole). One might even point out that a healthy minority of National supporters (35%) prefer the Labour-friendly 20 or less option.


Conclusion

If I was to critique the broad contours of my own argument, here, I'd suggest something along the following lines: Well, yes, obviously an overwhelming majority of New Zealanders do, indeed, prefer smaller class sizes, but (1) almost half of them are clearly unaware of it because, while 83% chose "25 or less" to the second question of the UMR Research Poll, only 46% answered "Too Big" to the first question, (2) the majority for "25 or less" may be overwhelming but given the large majority choosing National's education policy over Labour's in the Herald-Digi and, more generally, given the size of National's lead in the polls, clearly the strength of feeling on class sizes isn't there - as with Asset Sales, most people may be opposed but not to the degree that they're unwilling to vote National once again, (3) the only poll to directly compare the two major parties' policies on education (the Herald-Digi) came down decisively on National's side, and that (4) maybe, just maybe, what the Herald-DigiPoll is telling us has nothing to do with education policy at all and everything to do with the voters' 'phone being off the hook' to Labour. That is, if the policies were suddenly reversed tomorrow, a large majority of poll respondents would suddenly choose "Cut Class Sizes" because what these polls are actually (inadvertently) measuring is strength of commitment to the Key Government.

That's basically me trying to anticipate what a National-friendly critic or perhaps someone playing devil's advocate would argue.

And I don't entirely rule out these propositions. But to counter, I would argue the following:

That spectacular U-turn a couple of years ago by Education Minister, Hekia Parata, on her policy to increase Class sizes arguably tells us a good deal about the public's strength of feeling on the issue. The mobilisation and anger clearly went well beyond just the teaching profession. As various sections of the media pointed out at the time: National were backtracking because of  "widespread opposition" (Fran  O'Sullivan, NZ Herald), "a public backlash" (Morning Report, RNZ), it had "sparked immediate controversy" (John Hartevelt, Dominion Post), a "deeply unpopular" plan (Audrey Young, NZ Herald), the "resurrection" of "voters' doubts" (John Armstrong, NZ Herald), a "backlash" (Tracy Watkins et al, Dominion Post), the "damage done to (National's) relationship with women voters" (Tim Watkin, Pundit),  "teachers, their unions, and principals began to unify, and more worringly for the Government, parents joined in" (Sunday Star-Times),  and even the "rumbling of protest on the floor of their own (Northern Regional National Party) conference" (Young, NZ Herald).

And as Parata, herself, conceded at the time: "...clearly, it has not been a trade-off that parents were prepared to accept...the level of anxiety that parents were feeling...it's clear and genuine - and I respect that - that parents do not feel comfortable with that trade-off and that's why we've reversed that part (of the policy plan)...I have heard what parents have said." (Morning Report, RNZ) and that it had caused "a disproportionate amount of anxiety for parents" (Dominion Post). Meanwhile, John Key conceded in the Dominion Post that "It's clear that parents don't see it (the alteration of class sizes) as modest and in the end perception is reality." and that it had become "blindingly obvious" that parents would not wear the policy.

Indeed, well before National's humiliating backdown, Listener columnist Ruth Laugesen had highlighted the immediate public backlash that greeted Treasury's call for bigger class sizes back in its February post-2011 Election briefing. The signs of broad opposition throughout the Country were obviously already there.

What were these "blindingly obvious" signs that Key had noticed ? Well, for one thing, a May 2012 One News Colmar-Brunton Poll, taken at the height of the furore, had found that almost 80% of New Zealanders opposed bigger class sizes, with less than 20% claiming to be in favour. For another, National had taken quite a big hit in the polls from the backlash, suffering a 5-point fall in May 2012 (monthly average), before enjoying a slight revival after the backdown in June.

It's reasonable, too, to assume that National's all-important focus-group research was telling it pretty much the same thing. Initiated when Key himself started contracting the controversial Australian political tacticians, Crosby/Textor, to manage both his public image and the Party's political strategy on becoming National's leader in 2006,  the broad consensus among commentators has long been that the Nats barely make a move without detailed analysis of what these highly-expensive monthly focus-groups are telling them about any particular issue. It's striking, then, that National made such a rapid reversal on their class size proposals in 2012, while blithely ignoring opposition to asset sales, spy legislation and other core policies.

The general consensus, therefore, is that the policy inflicted some heavy damage on the Key Government and this "deafening outcry" (Dominion Post, Editorial) suggests a broad, visceral opposition to larger class sizes.

A National-friendly critic might argue that the public in 2012 were reacting, first and foremost, to National's plans to alter the status-quo by increasing class sizes. This wasn't, that is, a spontaneous public mobilisation positively in favour of smaller class sizes and thus against the status-quo. Maybe. But the consensus, both then and now (excepting coverage surrounding the Herald-DigiPoll ), seems to be that "every parent believes the smaller the class, the more of the teacher's attention their child will receive" (NZ Herald, Editorial) , that "school class sizes are a subject that can move votes" (NZ Herald), that smaller class size is "an issue that goes to the heart of parents' concerns" and is likely to have "much greater cut through" (Tracy Watkins, Dominion Post), "intuitively has plenty of appeal" (Dominion Post), is "likely to appeal to a wider swathe of voters" (Dominion Post), and "has an obvious appeal to middle New Zealand" (Dominion Post, Editorial). What's more, it's a fact that one of the defining moments in the furore that erupted in May 2012 was the surfacing on-line of a 2005 John Key interview in which he indicates that he'd chosen to put his own children through private schooling because he believed the class sizes there would be much smaller.

The objection that the MSM in general, and the Herald-DigiPoll in particular, have framed the education debate within a false dichotomy of 'class size vs teacher quality' still stands. UMR Research suggests at least a tacit endorsement of Labour's emphasis on smaller class size but obviously only when it's not erroneously contrasted in a zero-sum way with higher teaching standards.

Despite the potential cut-through for Labour, however, the reality is that the Party's poll support has slightly declined since Cunliffe's Party Congress announcement (or, more precisely, dipping a little in early July, before reviving in mid July - remembering, of course, that relatively minor movements in support may be entirely explained by sampling error). Clearly, Labour didn't quite receive the boost it wanted. This, no doubt, had more to do with the on-going sense of crisis that MSM attack's on Cunliffe's leadership had created through June and early July than anything to do with education policy. However, we should bear in mind the critique by long-time Labour Party activist, Patrick Leyland, who has taken a swipe at the lack of follow-up by Labour's external comms team in the immediate wake of the education policy announcement.  http://theprogressreport.co.nz/2014/07/17/how-not-to-release-policy/

Given the obvious strength of public feeling over recent years, however, that latent potential remains for Labour to harness.