School deciles
New Zealand has a national school funding system that allocates money to schools based on socio-economic data about students. This isn’t self-reported individual-level data, but is at the level of Census meshblocks (details here.) Schools are divided into ten deciles, and more funding given to lower-decile schools. Despite the higher funding, lower-decile schools, on average, have poorer results on standardised assessments. You can see good visualisations of this from Luis Apiolaza,
There are less-good ones at Stuff: dot plots aren’t ideal for this, and it really is better to look at cumulative categories (‘at standard or better’) rather than individual categories (‘at standard but not better’). Unfortunately, these graphs tell you almost nothing about the policy question of whether there are better ways to target the funding. There might be; there might not be.
One advantage of the current system is its automatic stabilisation. The Herald, earlier this week, had a good story about changing ethnic profiles of schools, with the sort of combination of data and individual stories it would be nice to see more often. It turns out the low-decile schools are seeing fewer students of European ethnicity, and more Māori and Pasifika students. The phrase ‘white flight’ was used, but because of the funding system this isn’t the same sort of problem as the original ‘white flight’ from US inner cities.
In the US, a lot of public school funding comes from local government. When more-affluent families leave an area, the government funding for education goes down. In New Zealand, when more-affluent families leave an area, the government funding for education goes up. There’s still a concern about diversity, but not the same sort of vicious circle that was seen in the US.
Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »
I’ve wondered about the idea that school decile is “based on socio-economic data about students”. I presume it is based on household socio-economic status not the students themselves, but are adjustments made to only include households with school aged children? Doing this properly would seem to require access to the base data so that you could make such adjustments from a position of knowledge. Otherwise, if all you have is meshblock level it might take lots of indirect methods to make an approximate adjustment.
10 years ago
Yes, it’s based only on households with school-age children. The page I linked says:
So it looks like Stats NZ comes up with the meshblock-level scores based on individual-level Census data.
10 years ago
The funding issue is a huge problem. My opinion is that decile funding is probably the fairest as long as there is adequate money for RTLB, ESOL and kids with disabilities. (I think a lot of people would say there isn’t at the moment.)
Hekia Parata probably wants to go to funding based on national standards.
However, if she funds based on high scores she rewards the already well resourced schools.
If she funds based on lows scores than the incentive for teachers is to get their kids to perform badly.
If she funds based on growth than national standards data is not really detailed enough to measure it (kids can move a long way without crossing a threshold) and incentivises teaching to the kids who can move the most i.e. those just below the break points, the kids who can move multiple levels. The “above” standard kids, especially if they are well above standard, will be left to fend for themselves because, from a funding point of view, there is no point wasting resources on them if there is nowhere higher to move them too.
10 years ago
“White flight”/socio-economic flight actually benefits the govt. As kids leave low decile schools the proportion of kids in low decile school goes down relative to high decile schools – their decile 1 school would have got money for them being there but there decile 10 school doesn’t. Although 10% of schools are decile 1, <10% of students are from decile 1 schools.
It's pretty hard for decile 1 school to compete with a decile 10 school which can call on a large group of non-working mums with big SUVs.
10 years ago
My opinion is that decile funding is probably the fairest as long as there is adequate money for RTLB, ESOL and kids with disabilities.
To add another layer of complexity, there’s some evidence that parents use decile as a proxy for ‘good’ (I haven’t read the Herald article but presumably that might be part of the story). If this is true, then even if decile funding is fair, there might be reason to change it if it undermines some of the other policy objectives of education.
10 years ago
I don’t think they use it as a proxy for “good”, I think they use it at a proxy for “what type of kids will my kids be playing with”.
10 years ago
Yes this is real. At a friends Decile 1 school, even though they are a high performing school, when surveyed the teachers all said they wouldn’t send their own kids there.
At the very bottom of the heap the other kids are not very inspiring and have lots of behavioral baggage also.
10 years ago
Isn’t that what I said?
Seriously, I’m sure parents consider educational quality when evaluating what is a good school, but I’d be surprised if ‘what type of kids will my kids be playing with’ isn’t in the mix too. And like I said, that might undermine the some of the broader objectives of the decile system, even if its a fair way of distributing funding.
10 years ago
I’ve wondered if the lower decile schools are also smaller and higher decile schools are bigger. As decile is at a school level not student level the lower decile schools are even more under priviledged than their decile indicates as the slope per student is weighted to the top end. And the weight of funding per student is averaged up to the weight of numbers at the top.
Anyone have insight into if this is a real factor?
10 years ago
This is an interesting discussion, so I went back and dug up the data I scraped from NZQA a while back.
The data only goes up to 2011 (NZQA changed how they release data, the format is now much nicer, but a lot of detail is lost, including decile information), and is only for Scholarship data, but it still shows some interesting patterns.
The graphs were produced automatically, so some of them are nonsensical, but it took me about 30 seconds to produce 66 graphs, so I’m not complaining.
Scholarship Entries by Year, Decile, Ethnicity and/or Gender:
https://www.stat.auckland.ac.nz/~joh024/Research/BlindfirePlotter/NZQA_Scholarships_BFP_Entries.pdf
Scholarship Success (including Outstanding achievement) by Year, Decile, Ethnicity and/or Gender:
https://www.stat.auckland.ac.nz/~joh024/Research/BlindfirePlotter/NZQA_Scholarships_BFP_Success.pdf
Scholarship Success scaled by Entries, by Year, Decile, Ethnicity and/or Gender:
https://www.stat.auckland.ac.nz/~joh024/Research/BlindfirePlotter/NZQA_Scholarships_BFP_Success_Scaled.pdf
10 years ago