On decile ratings of schools

Written By: - Date published: 11:50 am, August 21st, 2012 - 20 comments
Categories: education, national, schools - Tags: ,

Interesting piece on Stuff this morning:

ERO drops decile ratings from reports

The decile rating of schools has been scrapped from Education Review Office reports. ERO chief review officer Dr Graham Stoop made the surprise announcement yesterday in an effort to “correct the stereotype that a school’s decile equals performance”.

Schools are given a decile rating of one to 10, reflecting the proportion of students from low socio-economic communities. About 10 per cent of schools are in decile one and have the highest proportion of pupils from low socio-economic communities. Lower deciles are allocated higher rates of funding by the Education Ministry but deciles are no reflection on the quality of a school.

There have been suggestions that some parents have been treating deciles as a reflection of quality, with a “white flight” recorded of tens of thousands of Pakeha children away from decile one, two and three schools in the last 10 years.

Prime Minister John Key has said some parents assume the decile ranking is “a proxy for the quality of a school” which could be “very unfair”.

Dr Stoop yesterday said taking the decile rating off ERO reports would “help remove this element of confusion and correct this misconception”.

Right, so the government ignored most expert advice, international precedent, and the overwhelming majority of schools and teachers, to bulldoze through national standards. The data from the standards is nonsense, but it is going to be published anyway because the Nats’ claim that the good parents of [insert region here] are desperate to see it. But – they are excising decile ratings because they are confusing, unfair and misleading? White flight – what the hell do they think is going to happen with the national standards data?

These two “policies” are so completely incoherently at odds with each other that it’s hard to believe (typical for Nat “education” policy I’m afraid). A cynic might suggest (and several did in Open mike this morning) that the dropping of decile ratings is really an attempt to obscure the link between poverty and educational underachievement. Mmmmm.

All that said, no one is a big fan of decile ratings. The piece above continues:

Teacher union the New Zealand Educational Institute (NZEI) said clear information about the social and economic context of schools should be published in place of the decile ratings, which were “crude”. It suggested including data on student transience, the number of children with special needs or English as a second language and the number of children attending breakfast clubs.

Decile ratings shouldn’t be dropped, they should be replaced by better data. And if “ropey” national standards data is to be published, these economic measures should be included with it. The only reason that the Nats would argue that parents want some facts but not others is political game-playing.

20 comments on “On decile ratings of schools ”

  • Comments are now closed

  1. Lanthanide 1
    21 August 2012 at 12:32 pm

    They had two head masters interviewed on the radio this morning.

    They were saying it was good that decile ratings were being dropped, because it had nothing to do with educational achievement. A question or two later, one of them said that when the National Standard’s results were published, they would in some way mirror what the decile rating showed. This means that, implicitly, decile ratings did actually reflect student achievement, despite the fact that they both said that it didn’t.

    Really the problem is that there’s no easy and clear way to differentiate student achievement from school achievement, and in many regards they are the same thing.

  2. fabregas4 2
    21 August 2012 at 1:18 pm

    Doesn’t sat that all – it means that decile rating reflect socio economic status of the school community – low socio economic status leads to it being harder to be successful at school.

  3. Dave Kennedy (bsprout) 3
    21 August 2012 at 2:13 pm

    Lanthanide-Margret Wu is an internationally respected academic who has assisted with the development of the PIRLS assessment used to compare different nations’ education achievement. She often explains that of all the determining factors of a child’s academic success (including family and socioeconomic background) a teacher only contributes around 10%.

    Schools and teachers can and do make a difference but cannot work miracles over the deficits that many children bring with them through the school gate. While school achievement will generally align with the decile rating it cannot be used to ascertain the quality of teaching and learning within a school. It just so happens that you are less likely to have children in a community full of migrants who speak English as a second language and have low incomes who will do well in assessments based on English. Whether they are academically able or struggle with learning there should be little difference in their attainment no matter what decile school they attend. There is also some evidence that the quality of teaching in higher decile schools can be below that of low decile because it is much easier to teach in such schools. Able, motivated students from supportive, academic families often succeed regardless of the teaching they receive. When you say “…there’s no easy and clear way to differentiate student achievement from school achievement, and in many regards they are the same thing”, I would have to disagree. You are implying that if a child scores badly in a National Standards it just reflects the level of teaching in the school.

    A mother of a child with autism told me recently that a “good” school for her child would be one that is inclusive of all children, provides a variety of learning contexts that would excite and inspire her child and one where her child will feel happy and supported in their learning. A decile number or a National Standards assessment wouldn’t be able to discover that for her.

    Using the either flawed National Standards data or the decile rating to ascertain the quality of a school is equally problematic.

    • Lanthanide 3.1
      21 August 2012 at 5:30 pm

      “Using the either flawed National Standards data or the decile rating to ascertain the quality of a school is equally problematic.”

      I don’t disagree.

      What I was pointing out are the following logical positions:
      1. It was stated that decile rankings do not have *anything* to do with educational achievement
      2. It was stated that when National Standards rankings are brought in, some of the decile ranking will be reflected in the National Standards rankings; presumably he meant that schools with lower National Standards rankings would tend to be the ones that had lower decile rankings and vice versa.

      If we assume that National Standards do measure student achievement in some way (perhaps a bad assumption?), from statement 2 we should conclude that statement 1 is in fact wrong, in that lower ranked decile schools perform worse on National Standards, therefore lower decile schools are transitively indicators of lower performance. I’m not saying that lower decile schools cause lower performance, just that’s it’s correlated. This should be obvious.

      ““…there’s no easy and clear way to differentiate student achievement from school achievement, and in many regards they are the same thing”, I would have to disagree. ”

      Ok, if there is an “easy and clear way” to differentiate student achievement from school achievement are you asserting there is, then surely you can come up with good measures, better than National Standards, to do this.

      Because at the moment National Standards will be used to measure school achievement, based on individual achievement, and IMO the system simply will not work at all for all the obvious reasons (kids in lower decile schools will likely have lower National Standards scores, but that’s more to do with the kids attending the schools than the actual school performance). It seems you agree.

      • Georgecom 3.1.1
        21 August 2012 at 6:27 pm

        Decile ratings don’t tell you very much about the quality of a school or its ability to raise achievement levels for particular students from the point the student eneterd that school.

        What decile ratings might reflect, broadly, are some of the socio-economic advantages or disadvantages that children bring with them to that school which will have an inpact on academic achievement levels.

        A schools finds children as they walk in the school gate, including such things as decile, and then goes to work from there to create learning, the actual achievement rates that schools can control.

        Reporting data against National Standards might also reflect some of the socio-economic advantages or disadvantages that children bring with them to that school. That is, the ‘readiness for learning’ state that students enter the school gate may well be reflected in reporting under, at or over National Standards achievement rates for students.

        What the NS don’t reveal is the level of learning growth that each particular child has made in a set period of time, measured against a preceding set period of time. The NS data is far too blunt a measurement to capture contextual learning improvement for individual students.

        So, decile and reported NS data might both reflect the particular socio-economic data that students bring with them through the school gate. Decile ratings won’t describe the quality of learning that occurs within the school and neither will, really, reported data against National Standards.

        • Lanthanide 3.1.1.1
          21 August 2012 at 6:57 pm

          Yes, I fully agree with you, and in fact that is my point.

          NS (purport) to show the performance of an individual: do you know what 2+2 is or not? What it doesn’t show (very well) is how much progress an individual has made. NS attempts to take a measure of individual ability and then apply it in aggregate as if it is a measure of the performance of the school they attend. But because NS isn’t directly measuring a change in attainment, it isn’t measuring whether a school does a good job at improving attainment or not, which is the purpose to which it is going to be used.

          • Dave Kennedy (bsprout) 3.1.1.1.1
            21 August 2012 at 8:50 pm

            It seems like we are generally in agreement, Lanthanide, and what it all really means is that while decile rankings and NS do measure things that have some basis in fact, they are blunt and limited measures and cannot and should not be used to make judgements about something as complex as the teaching and learning process.

            • Dv 3.1.1.1.1.1
              22 August 2012 at 8:05 am

              To say that NS measure anything is a bit of a stretch.
              Measurement needs to be reliable, and valid.
              Can anyone guarantee that a standard measure in a school in Northland is the same as a std in southland.
              Will it give the same results over time and with the same and different pupils who are at the same level?
              Will it give the same result over time?

              The ‘standards’ are neither national nor standard.

              • Georgecom …
                22 August 2012 at 9:59 pm

                More than that, decile ratings were developed with a logical purpose in mind and have some logical and consistent methodology behind gathering of data.

                National Standards? Hmmmm, well.

                Shambles?

              • Dave Kennedy (bsprout) …
                23 August 2012 at 12:34 pm

                National Standards are overall teacher judgments (OTJs) on a child’s attainment in literacy or numercy. I have no doubt that the information they provide would be useful for parents and supporting learning, but they only focus on a fraction of the curriculum and do no more than touch the total package of what a school provides to support teaching and learning. DV, they do measure something but I agree, they are neither national nor standard.

  4. Ianmac from Vietnam 4
    21 August 2012 at 2:14 pm

    Many low decile schools are hugely successful. Not necessarily in academic standings but in adding huge value from where the kids start. Learning enough words to be able to communicate with more than grunts. Being able to share. Being able to play fair. Being able to respect the rights of others. Being able to handle books properly. Many kids from deprived homes have these hurdles to overcome before they can begin to learn the academic stuff. It must tear the teachers apart to be regarded as low level schools because of the labels like decile ranking. As for National Standards…….

  5. Georgy 5
    21 August 2012 at 3:20 pm

    Lanthanide – you don’t appear to know what decile rankings actually are.

  6. Georgy 6
    21 August 2012 at 3:29 pm

    To see what how effective a school is you get a better idea by looking at data from the “native cohort”

    When a school separates the achievement data in literacy and numeracy into two sets –

    [a] children who started at the school and are still there and
    [b] children who have enrolled since starting school [ie have been to other schools]

    then the data for set [a] shows quite a different picture from set [b] – obvious to schools but not necessarily so to others. A significant number of children in the second cohort tend to be transient and usually come from “poorer” homes.

    As schools are forced to submit data to the Ministry of Educ they should send the data from set [a]

    The MoE will misuse it for league tables and pai websites but there may be more integrity in the set [a] data as a reflection of the school learning programme.

    While the decile rank will reflect the community the school sits, it does not reflect the quality of the teaching programme.
    Many schools will tell you that set [a] data shows very high levels of attainment against any criteria.

    When all the data is collated, the effect of the set [b] data will show the school as being “less successful’ in the eyes of the community – a very unfair way of measuring.

  7. Mike Steinberg 7
    21 August 2012 at 5:49 pm

    *** obscure the link between poverty and educational underachievement. Mmmmm.***

    This obscures the main causes of educational achievement are motivation and intelligence. Unless you are talking extreme malnourishment (or perhaps exposure to toxins in utero), cognitive ability is not going to be significantly depressed. As Professor Steve Hsu notes:

    “It is sometimes claimed that IQ is just a proxy for SES (Socioeconomic Status): high IQ kids are merely the beneficiaries of better home environments, and correlations between IQ and life outcomes are merely a proxy for correlations between childhood environment and outcomes. Of course, this claim does not address the significant variations in IQ within families. Does IQ have predictive power once SES is controlled for? The answer is obvious from anecdotal experience: we all know siblings who, by definition, shared the same family SES, but with different IQs and life outcomes….

    SES does not cause SAT (weakly at most).
    SES does not predict college success, SAT does.

    http://infoproc.blogspot.co.nz/2010/03/ses-and-iq.html

    • Urban Rascal 7.1
      22 August 2012 at 11:45 am

      Although from an economic and statistical analysis it seems that things such as merely having the presence of books in the house as a child can drive better schooling outcomes, as well as mother’s age at birth and a raft of other correlated factors.
      – Based of the analysis in the book Freakonomics.
      So I’d have to say you argument has a fairly well known counter. As these common factors are far more evident in lower socioeconomic areas.

  8. Ianmac in Ho Chi Minh city 8
    22 August 2012 at 1:59 am

    An afterthought. Without decile rankings will it mean that when national Standards are published, there will be no more guidance that a “lower performing” school draws from lower socio economic shoeless English as a second language transient kids. Just a bald percentage. This would confirm the government’s argument that some schools are failing. Ugh!

  9. Georgy 9
    22 August 2012 at 10:21 am

    Poverty, Deciles and Achievement

    Poverty is one dimension of a bigger and more complex situation. Many families who are “at the bottom” of the heap socio-economically move a lot for a variety of reasons. This directly impacts on learning. Also many children from these families have a high absentee rate, as well as often arriving at school late.

    We know that learning is a wholistic process and in school terms begins the moment a child arrives at school until the moment they leave. The first 15 minutes of class time are crucial and set the tone, the direction and timetable for the day. Children who arrive late miss the informal events and social interaction before the bell and the first part of the formal stuff – and are then on the back foot for the rest of the day. And if they have missed days then whats’s going on is a mystery. It is easier then to turn off and not really try. Input zero, output zero, development low, attitude negative.

    Many schools try various strategies to overcome this but it is a major challenge – and if the transient rate is 25% or more the school faces a lot of problems. the families also don’t value education, have language deprived homes, don’t have books and don’t have the routines that children need to support their education.

    This dimension of poverty is very real for schools and impacts hugely on individual achievement thus the collation of school wide data will reflect this. So, low decile schools will reflect their community in terms of achievement. What this won’t show is the incredible job a lot of teachers do to get children ahead. Many Reading Recovery teachers will report that they just get children going and the family moves again. Many Special Ed teachers will report the same. Many children can attend anywhere between 5-12 schools during their primary school years, often with long breaks between schools while the family shifts, finds a house, buys new clothes, sorts out benefits with the new winz office, waits for the benefit to come through to get money to put on their phone….

    The link between poverty and school achievement is very real for many teachers throughout nz.

  10. fabregas4 10
    22 August 2012 at 9:47 pm

    This move also raises questions about whether the powers that be want to forget about the context that schools work in. Maybe not talking about the socio economic deprivation that is Decile 1-3 areas
    means that the govenemnet can expect the same results from these kids as those for decile 8,9, 10 schools?