Wednesday, 11 January 2017

School Choice: Abhorrent, Despicable and Morally Bankrupt

In light of The Tyranny of Choice I'd like to revisit another topic of old: the idea of school choice. I'll start off by mentioning that I find this particular notion abhorrent, despicable and morally bankrupt. School choice is completely insidious because it:
  • Forces schools to compete with each other for pupils, which means fewer resources are available to education.
  • Allows "flight" which means that some schools lose rolls (and resultant economies of scale which, for instance, allow greater choice in subject for pupils) and others gain rolls, potentially to the extent that pupils living next door have to commute to a much further away school (going to local schools reduces congestion as it removes needless trips).
  • Combines with the positive association between familial wealth and pupil attainment such that there are fewer pupils who need to be brought to standard and schools are thus less able to reap the benefits of having mixed knowledge levels in the same classroom. (This is not to say that poorer pupils are stupid, it's that the odds are stacked against their being able to fulfil their potential.)
  • Entrenches the light segregation that already exists between wealthy and poor that generally follows through from wealthy and poor locales (as now the wealthy are able to move around).
  • Increases the likelihood of disengagement and disenchantment as the loss of school resources filters through to pupils who begin to feel like they are being left behind... which makes educating them more challenging and if the attitudes flow through to their children...

...all in the name of the benefits of choice mentioned in the earlier post.

Now, the counter-argument to several of these issues is that you can fund them away. Well, that's not a counter-argument at all. Competition for pupils isn't going to go away with funding raises. Perhaps you could legislate against advertising but this would be problematic in practice and I'd rather not set aside money for promotional purposes when it could be spent for educational purposes. The decile system in New Zealand's problems suggest that it is also a difficult proposition to design a funding scheme that accounts for the wealth gap between the home and school environments. Thus, it's really only the economies of scale point that is addressed by funding... after all, if you promise to provide the funding such that you can offer Classical Studies even though only seven pupils actually want to take it out of the school's tiny little roll, it's much the same as if you had two classes of 18 in a larger school. And while I don't think the last bullet point is solved by funding, maybe it is sufficient for the school to have decent resources rather than the vitality that comes with not having a declining roll. But here's the thing... wouldn't it be better if that money was being given to schools anyway? That's why the "funding as fix" argument is bunk. If we have that money to spend, it would be more useful outside the school choice paradigm and you don't get the problems associated with school choice that aren't connected to funding (or are so only tenuously). We should not muck around with the lives of people unless we know we're making changes for the better: "I'm sorry our educational policy reform package's being wrong ruined your life" just doesn't cut it.

But what about what we've been talking about here, in this post? We have to seriously consider that schooling is a dynamic system... that it is complicated, that it changes and that there are a hell of a lot of (contingent) choices to be made. Then there's credence qualities associated with services like education... are parents and pupils really able to evaluate the strength of a school's education anyway? Education is also subject to externalities (external costs and benefits not evaluated by parents and pupils making their private choices). None of these things make education look like the kind of "market" where it makes sense to be suggesting that choice of provider is of critical importance.

The remaining question, of course, is what would the competition for pupils that results from school choice do beyond create new important cost areas? Competition is, I agree, a good thing. It's interesting and it can help spur new ways of doing things. However, competition isn't external. That is, how you choose to compete is an alteration of how you choose to behave when you don't have to compete... it's not something fundamentally new. With regards to school choice, then, where we should start is the character of schools and their members as they stand now (i.e. with limited to no school choice).

When we look at schools, what are the benefits associated with them in personal senses? Sure, public school systems developed (literally) because the absence of them didn't manage to educate people in the way that private individuals and society gradually came to want and that's important. However, I think most teachers get some sense of personal reward from succeeding in teaching but as you can tell from this blog, I'm very self-centred. But, I think, I'm basically right. Notice that people are motivated by experiences with inspiring teachers/education and dissatisfaction with their own education. In other words, some teachers want to change things, whereas other teachers are, perhaps, a bit nostalgic in their ideas of how it ought to be. The critical thing, though, is relationship between why teachers choose to teach and their pupils. Holidays, family members and childcare aside, how the pupils are underlies every single one of those reasons... and the teachers could choose multiple options. The improvements for work/life balances also reflect a connection with pupils, collaboration or people more generally (e.g. less teacher-bashing)... but curriculum change and the like is seen as problematic. Thus, it is necessary to consider the attitudes and values of the pupils. The best a quick search via Google found was this report (which I scanned, pay attention to the methodology, I didn't but should've), which suggests that most pupils are keen on school... that they like collaboration (but independence was also valued, just less so), making-stuff, and discussion but their attitudes toward receiving feedback weren't really looked at (just if they got it and sought it out... about 47-50% seem to do so at least on occasion, pg. 33). So... where to from here?

It seems to me that if you let teachers run wild and just do their stuff, you're going to end up with fairly practical, vocal and collaborative lessons. Teachers think this is a good way of teaching and pupils like it. Teachers also like working with their colleagues. In an ideal world, teacher aides would also be more involved. However, teachers are also going to be relatively less involved outside school hours. This adds to their workload and was one of the negatives identified in that survey we're basing many of our conclusions on. Similarly, we cannot really expect teachers to generate new methods and syllabuses. "No more education changes" was explicitly identified as a problem, but when I looked at why people choose to teach, I concluded that teachers are nostalgic. It is important for a significant number of teachers to replicate the experiences they themselves had. But teachers and pupils aren't really agents in terms of competition. Teachers might be the boots on the ground who do the actual educating but they're not really the supply... and pupils aren't really the demand (even if they almost all like school). This collaborative world that laissez faire principals would probably result in must be tempered with what the competition would actually look like. I mention it, because I think it matters that this is what teachers would move towards if they could. Also, principals are broadly teachers too.

If we assume that principals are connected like their underlings (er, teachers), then we also assume that principals are interested in the same things. Boards of Trustees can be similarly assumed to be connected to the pupils (in particular) and community (in general). So, the questions become what are the interests of these administrators? Above I argued that they're motivated to sustain roll numbers. This makes sense. If they cut back on pupils, they lose funding and have to cut back on teachers. There are two obvious reasons why administrators (assuming I'm correct) would want to avoid this. After all, letting people go involves personally difficult decisions, but we're modelling "caring" not "cold" administrative people so there's hardly likely to be a smiling assassin. Perhaps more importantly, cutting back on staff limits the potential of a school to regain its roll and thus generate more teachers. Opportunities for collaboration, teacher aides and smaller class sizes are reduced with small rolls. There are fewer teachers to collaborate with (and you now compete with other schools for rolls, remember), teacher aides are sustained by pupil-tied funding in the same way as teachers and while small schools need a minimum number of teachers, you do reach a point where you can be too small to continue so smaller class sizes are really only reached if you can afford more teachers than you strictly need. And guess what... the bigger schools? They're parentally wealthier schools. So if we imagine that 10% of parents are willing to donate $1, and we have 100 parents, we'll get $10... and if we have 1000 parents we'll get $100... and it's probably the case that willingness to pay (whatever amount) is also positively associated with school-wealth, which would exacerbate this effect.

But what if principals aren't so constrained by the behaviour of the parents? (If you're getting frustrated by all the "buts" and "ifs" here, imagine how I feel... the number of times I've thought I'd finished this post. This, by the way, is a choice issue of the sort identified in that chapter on behavioural economics.) As it happens, I think that they are (as we shall see soon) but it is important to consider, as we did with teachers, the "theory". From my standing point, the idea of competition is that it creates a cutting edge: if you stay ahead of it, you survive/suffer less but if you fall behind... point is that this destroys complacency and leads to continual improvement. The thing is that these improvements often boil down to cost and revenue dimensions. How do those exist in schools? In an urban area with many schools, you're going to shrink but probably not to the point of unviability (everyone still needs a school, right?). In a rural area, you're really a natural monopoly that has to exist. In some ways this is what we've already mentioned: schools don't really have consumers in the way we usually think about things. And that survey about teachers themselves suggest that "co-creation" of value so familiar to business (not economics) students is very much at the core of what happens in schools... but this value is perceived (differently) both by the pupil and the vicarious consumer (i.e. the parent). It seems to me that the natural/default stance of the educator is to see themselves as the fiduciary of the pupil. This need not be a barrier to this cutting edge model's validity. After all, the naturalness gives a sense of value being produced to the teacher, but this must somehow be conveyed to our (admittedly caring) administrator. Is that possible?

If you look back at our survey of pupils (which is problematic), one of the things that you will notice is that in secondary schools, pupils seem more distant with their teachers. This makes sense. If you spend less time with someone... and I believe this is the explanation the study uses (we're on pp. 18-20 or so). This should translate, right? More teachers (secondaries are bigger) => more distance => "human understandings of value" don't necessarily move easily. But even if they do, how are you meant to translate that to a cutting edge? We know that teachers have ideas about how to improve things, but we know that these are cost increasing measures. And the issue with human understandings is that we're not really sure how they 're reached. In fact, a lot of the time we just "know" that we've seen a bad film (or whatever) and cannot really explain (even to ourselves) what is underlying this. Humans, as it turns out, may as well be mysterious boxes into which we place some input and get some output. Whatever happens inside is, in essence, divine (after all, Man made God in his own image anyway). Thus, there is the need for understandable (i.e. portable) evaluation.

This is where we turn to assessment. And, specifically, performance pay. That, as everyone should know, is an absolute minefield and I've been sidetracked by research enough at this point. What I will say is that standardised tests are absolutely a fine idea, but not all tests are of equal quality (fitness for purpose). What is not fine is "bubbling in" as discussed in The Perfect Score and making assessment (of any kind) the crux of the educational experience. More assessment also doesn't line up well with what teachers think works... and why are we pretending that they're not in a position to know? Especially when to judge the value of a teacher we need to control for the pupil, the pupil's family, the pupil's friends, the pupil's classmates and all sorts of other things but any statistician will be telling you that what we've explicitly identified at this point isn't really going to happen. Performance Pay is just another version of this Divine Box... except now we're putting in some tests and getting outputs that will differ wildly based on the boxes (i.e. people) involved, because we can't control for these variables we know we need to control for but we're choosing to forget. Objecting to Performance Pay has nothing to do with whether or not it could help because it's impossible... and that's where it should begin and end as a proposal.

Now this is not to say that guidelines couldn't be generated to help, well, guide interpretations of teacher quality but it is to say that we can't appeal to Truth, which means we are left with Man. And one thing we know about Man is that we have emotions... they make us humans... so all rewards for performance must be about human relationships. By all means, allow teachers that their peers recognise as good to be rewarded... I don't think that will cause problems (it's a Human Truth, which I think is good enough). But if you try to say "objectively so-and-so is a good teacher" (i.e. appeal to Truth) what everyone recognises is that it's a human judgement, thus subject to righteous indignation (I may be jealous if X is recognised over me, but provided I am not isolated to start with this will spur me on because it's really disappointment). But this is just another cost and/or distraction.

I really don't see what principals are going to do. They can't just cut things which everyone seems to think helps (e.g. teacher aides) and they can't really make their employees compete with each other because they want to collaborate. Taking away less popular subjects is an option, but the relationship is fiduciary and sacking teachers is problematic. But such curricula and timetable aspects of administration are probably the places to anchor any hope of HMS Competition's success. The trouble is how are those areas affected by competition for parents? All those can do is appeal to what parents want. In fact, it's readily becoming apparent that school choice and competition depends entirely on what exactly market share looks like... and the best proxy for this is parents. The problem is that neither parents nor pupils are proper consumers. Neither lines up with the economic assumptions of rationality or perfect information. After all we have started from problematising the ability of people to understand the quality of education and we've literally just shown that we can't really attach signals to teachers. Which means that principals have to think about what will appeal to parents. Which is where you get things like trophies, clean grounds, professional presentations and dressing up academic performances. In other words, the improvements that competition will bring have next to nothing to do with efficiency or social welfare. Principals will look to make their schools as attractive as possible. And this would manifest in allowing teachers to do their things that they think will help. But it also means that you need to be able to sell what's good about the school to parents, Basically, principals are incentivised to waste money on things that aren't the school's main concern (education) in order to keep up the roll numbers that basically facilitate allowing teachers to do their things. There is no behaviour of the principal, without the parents(And I really wish these ideas are read by someone who understands them and puts them more coherently... sorry.)

What about parents? Well, there are many reasons to believe that parents are genuinely interested in their children... that is, the vicarious consumer we've assumed the whole time is a reasonable assumption (or, perhaps, inference is a better word). To this end, we can trust the argument that parents will seek out schools that they perceive as being better. Even better it is a well known truth that parents fail miserably to live up to assumptions that what I guess we can call choice theory make. I am referring, of course, to the idea that deciles indicate school quality. They do not. Possibly this is an attempt to find some "signal" of quality but as we've established, there is no real means of doing this... there is only ever parental perceptions. Deciles are negatively associated with funding levels in an attempt to make up for the resource gap between rich and poor but they are associated with performance due to the poor levels of resilience in NZ's education system, i.e. its severe inequality issues. I personally think that this resilience issue is what parents perceive to indicate school quality. This doesn't necessarily work. Sure, maybe teachers prefer to teach at certain schools so some schools are able to be more selective in hiring practices than others (that we can't objectively determine which teachers are better doesn't mean that all teachers are equally good... and this doesn't mean that all teachers aren't sufficiently good) and maybe these schools are generally higher decile schools. And maybe the higher donations that wealthier parents and/or just more parents bring enables schools to offer something more, but it may be the case that parents who are wealthy enough to afford tutors, better materials (e.g. internet access, study guides) and provide relatively more secure and stable home environments represent a reason to think that their children gain as much from a good teacher as an okay one... i.e. higher decile schools aren't better, they're just performing where they'd be expected to... same as lower decile schools. And I believe there is evidence that parents do send their children to higher decile schools more than should be the case based on the number of pupils in New Zealand.

When we look at this graph, what should we be noticing? Firstly, that there are substantial numbers of pupils that don't go to schools with (known) deciles. Secondly, that it is only higher decile schools that have pupil numbers that track with the overall number of pupils in New Zealand. However, the four mid deciles manage to increase in the mid-late nineties, whilst low decile schools held constant. It is only after school choice was curtailed in 2000 that we start to see those levels dropping off and mid-decile schools stagnate. According to what I have been outlining, parents shouldn't really have been able to cause this to happen without "choice". But what is going on? It's possible that we're looking at artefacts of decile calculation changes (as is the case between 2014 and 2015), but my suspicion is that parents began to move more locally. You get some movement from low to mid deciles and some movement from mid decile to high decile school zones. It is possible that, in an attempt to keep roll numbers steady, higher deciled schools expanded their zones somewhat, whereas the low decile schools (which I believe don't usually have zones due to excess capacity) did not. And once the school populations started to drop, they kept dropping (as my ideas above predict) because small rolls don't look good. But how do they have the ability to move around without school choice? Very possibly the reason is that school zones manage to preserve school choice... but also mean that gross miscarriages of justice that see pupils from next door turned away are impossible. It seems people manage to have success in getting around zones.

Artefact? Were some decile ten schools reduced to 8 or 9? There was a census about then... same formula, number shift?
But what about that period up until 2000? It's hard to be sure and, in truth, I don't think just looking at these time series is sufficient (we need to account for changes in the demographics of the meshblocks themselves rather than just tracking the overall numbers) but I think it's nothing. After all, what we're seeing is a period where the overall number of NZ pupils was expanding. In 1996 there were 696335 pupils, in 1998 724579 pupils and in 2006 760745. That is, in two years there was a 4.1% increase and in ten years there was a 9.2% expansion in the total number of pupils. I think things stayed fairly constant, then, because there were enough new pupils coming in and even with school choice, there were enough new parents who were satisfied with (I assume) their local schools. Once this supply of new parents dried up, the rolls for low decile schools began to drop and kept dropping right up until the number of pupils altogether began to increase again. And drops are self-perpetuating for the reasons I describe. Perhaps the most convincing argument is that there had been four years or so for the roll shifts to have taken already by the time the data I have starts.

Percentage Change From Previous Year

But when you look at this... maybe I am a bit doom and gloom about school choice. Where is the spiralling? The data that I have suggests that as long as there are enough new pupils coming in, school rolls lower down can be sustained. The first question is how did they manage that? Well, maybe the answer is because it had already bottomed out... limited by capacity constraints at the top. The other question is, of course, how does some form of school choice seem to exist with school zones? After all, a large part of my argument is that school choice is bad because the absence is better... but zones don't seem to prevent it. Is it enough for low-decile = bad parents to be able to move from a low-decile school to a mid-decile school? (Not all schools have zones.) I don't have any data that we can use to validate the "marketing budget" argument... but I do have this. Indeed, nor have I considered whether or not there has been some widespread cultural change that needs to be taken into account. For instance, just because people "know" about deciles now, doesn't mean they knew about them pre-2000... deciles were, in fact, introduced in 1995 and by at least 2003 people were complaining about the quality thing. People still have terrible trouble understanding NCEA, so it's entirely possible that people took a while to "learn" what they meant (you can learn wrong things). I tried to look into this but I didn't really find anything. However, I did find a 2003 study called "Parental Choice or School Choice: Who Benefits from the Removal of Zoning?" I think it substantiates the bottomed out idea... which suggests that all my theory was right... school choice deserves a doom and gloom reputation. I will end with some choice selections from that study (sadly I forgot to note the page numbers).

The Evils of Marketing Budgets:

Roll Uncertainty (bad for decision making => zoning creates inefficiencies)

School Size (sceptical of big is good, whereas I embrace the philosophy, also capacity)

Hurting the Most Vulnerable (the Spiral of Decline in Low Decile Schools)

Racism, Classism and the Roll of the State (see what I did there?)

The Cutting Edge (Absence of Support in the Spiral and Loss of Collaboration... except by cartel)

Dumping Grounds? (The failure of market ideas in practice... funding reform the solution?)

Returning to Zoning: Operation and Problems

Uncertainty: Exactly Why it Reduces Innovation (says me)

Ah, finally... it's over. No, sorry, I shouldn't sound like that... this is a very worthwhile topic and I think I have said something worthwhile... it's just that this was taxing to write.

No comments:

Post a Comment