Legal Beagle by Graeme Edgeler

Read Post

Legal Beagle: A little known story of the Māori seats

59 Responses

First ←Older Page 1 2 3 Newer→ Last

  • Graeme Edgeler,

    If anyone can tell me how to change this so the footnotes and footnote references link to each other, that would be cool :-)

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Stephen Day,

    Wow. Fascinating!

    Wellington • Since Dec 2006 • 9 posts Report Reply

  • Graeme Edgeler, in reply to Stephen Day,

    Wow. Fascinating!

    Thanks! I've been a little worried that this might be a little nerdy, even for Public Address :-)

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Tristan,

    so in a nut shell i don’t answer the decent question but answer the following (hypothetical genralising here)

    low income
    speak Maori
    christian
    low education

    then i might get assigned as Maori..


    Is this right thing to do? for me it depends on what the other questions they use. The census forms are here

    http://www.stats.govt.nz/Census/about-2013-census/2013-census-forms-guidenotes.aspx

    clearly if you answered the language question with Maori, the hapu question and the ethnicity question then you have a pretty good shot at being Maori (i wonder how many people answered all those questions but not the decent question)

    but when you start going down income, employment status etc then its starts to get interesting form a political point of view. isn’t that racial profiling?

    Auckland • Since Nov 2006 • 221 posts Report Reply

  • Graeme Edgeler, in reply to Tristan,

    so in a nut shell i don’t answer the decent question but answer the following (hypothetical genralising here)

    low income
    speak Maori
    christian
    low education

    Not quite. They don’t use income, language, religion or education as these are not well enough correlated with Māori descent status.

    The four factors they use are (as noted in the footnote – sorry):

    1. the Māori descent responses of others in the household (responses which are connected by a dwelling form);
    2. the inclusion of a valid iwi in the iwi question;
    3. whether the answer to the ethnicity question included Māori as an ethnicity; and
    4. age

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Graeme Edgeler, in reply to Tristan,

    but when you start going down income, employment status etc then its starts to get interesting form a political point of view. isn’t that racial profiling?

    Of course, they did try a bunch of other things to see if they were correlated, including income and Labour force status (they just weren't, or weren't very well).

    If you’re interest in the technical detail, the article I cite from the Australian & New Zealand Journal of Statistics (which is also on the Statistics New Zealand website) has it. Page 13 has the information we’re discussing.

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Tristan,

    Ah thanks for that. It seems a sensible approach. Is there any rules on using the decent question or could they have just used the ethnicity question?

    I'm presuming a chief statistician decided to separate out the question in the first place and another one decided that wasn't good enough.

    Auckland • Since Nov 2006 • 221 posts Report Reply

  • Tristan,

    also interesting that speaking Maori didn’t have strong enough correlation. The optimist in me says its because so many non Maori are speaking Maori now..

    update
    bugger kept reading…not enough people speak Maori to make any difference to the results

    Auckland • Since Nov 2006 • 221 posts Report Reply

  • Graeme Edgeler, in reply to Tristan,

    Is there any rules on using the descent question or could they have just used the ethnicity question?

    For the purposes of the Electoral Act, there's a requirement that the information used is about Māori descent, which means you must have a direct ancestor who is/was (New Zealand) Māori. Statistics NZ defines this as someone who is of the Māori race.

    I'm presuming a chief statistician decided to separate out the question in the first place and another one decided that wasn't good enough.

    I'm not sure that's the case. The first time this 'mattered' was the boundary calculation after the Māori option in 1994. This was a special Māori Option brought about by the change to MMP. It will likely have been under more time pressure than usual (it didn't immediately follow the census, which was held in 1991), and there simply may not have been time for someone to do things differently.

    The decision that was made to do things differently after the next Māori Option was made in advance of the census to which it related. I don't know, but I suspect that it was the same Government Statistician who made both calls.

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Tristan,

    This report is interesting (well more interesting than working) this from the report

    After seeking and receiving legal advice, the Government Statistician decided that those not answering Yes
    or No to the question on Mäori descent in the 1996 Census should be allocated to one of these two categories
    for electoral purposes.

    so you could look at it from the other direction and say the chief statistician wanted to make sure the correct amount of people were allocated to the general role. It seemed to be his view that everyone had to be one or the other he wasn't able to leave that question 'unanswered' and there by leaving those people with no representation

    right time to let someone else have ago. Thanks for the post Graeme!

    Auckland • Since Nov 2006 • 221 posts Report Reply

  • Graeme Edgeler,

    Interestingly, because people can now complete census forms on-line (and a lot did), this may change the scope of the calculation - on paper forms, it was possible to tick two boxes for the Māori descent option. On the on-line form it wasn't possible to select two options (can someone confirm whether you could leave questions on the on-line form blank?)

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Raymond A Francis,

    Thanks for doing the hard yards (again) so that we can all get some insight into this mysterious world of stats

    45' South • Since Nov 2006 • 578 posts Report Reply

  • Tim McKenzie, in reply to Graeme Edgeler,

    If anyone can tell me how to change this so the footnotes and footnote references link to each other, that would be cool :-)

    Are you writing the code for the links in HTML? If so, it looks like you've got some tags like <a href="#_ftn1">[1]</a> and <a href="#_ftnref1">[1]</a>. If you change them to <a name="_ftnref1" href="#_ftn1">[1]</a> and <a name="_ftn1" href="#_ftnref1">[1]</a>, I think they'll link to each other.

    Lower Hutt • Since Apr 2007 • 126 posts Report Reply

  • Graeme Edgeler, in reply to Tim McKenzie,

    Are you writing the code for the links in HTML?

    I copy pasted from a Word document, and was pleasantly surprised it worked at all.

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Graeme Edgeler, in reply to Tim McKenzie,

    I think they'll link to each other.

    And they do! Thanks!

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Brent Jackson, in reply to Tristan,

    then i might get assigned as Maori

    I think this is the important thing to stress. A certain percentage of you is assigned as Māori based on your other answers.

    My understanding is that they look at the answers to the four questions that most strongly correlated with the results in people who did answer the question :

    1. the Māori descent responses of others in the household (responses which are connected by a dwelling form);
    2. the inclusion of a valid iwi in the iwi question;
    3. whether the answer to the ethnicity question included Māori as an ethnicity; and
    4. age

    If 20% of the people who answered the question "Yes" answered these questions the same as you, then you had a 20% chance of being assigned as Māori. Or, to put it another way, of every 100 people who answered the question unclearly, and answered these four questions the same way as you, 20 would be considered to have answered it "Yes", and 80 would be considered to have answered it "No".

    It's not used to allocate a single person one way or the other, but more to split groups one way and the other.

    Hope this helps explain it for those who were unclear.

    Auckland • Since Nov 2006 • 620 posts Report Reply

  • BenWilson, in reply to Graeme Edgeler,

    Thanks! I've been a little worried that this might be a little nerdy, even for Public Address :-)

    Nah, this is some good research, sir, very interesting, most informative. Going to have to cogitate on this for a bit, but my initial feeling about these questions:

    It also raises important questions. Was the decision proper (in a legal sense)? And perhaps most importantly, is it right?

    is "IANAL" and "Yes". It seems like some basic data error fixing that is the realm of statisticians. But I'm not firm either, because you say:

    And most importantly, it is a choice that those who don’t know whether they have Māori ancestry, or who won’t clearly answer the question, will never get to make. People who cannot or will not give a clear “yes” to the question ‘are you of Māori descent?’ cannot enrol on the Māori roll.

    Can you clarify this point? Are you saying that the way they answered the census will affect their ability to enrol on the Maori Roll? Or are you saying that being unable to answer is what excludes them, something you are then carrying forward when it comes time to decide which roll to be on? I ask because maybe they might answer the question differently when choosing to enrol. Choosing to exercise democratic rights might cause people to buy in much more than the more vague and (to a lot of people) somewhat sinister aspect of the census "collecting data on us".

    Auckland • Since Nov 2006 • 10657 posts Report Reply

  • Graeme Edgeler, in reply to Brent Jackson,

    It’s not used to allocate a single person one way or the other, but more to split groups one way and the other.

    I believe that, for the purpose of drawing boundaries, it is actual individuals (albeit in an anonymised way) who are randomly assigned as either Māori descent, or non-Māori descent. It's peripheral to the points I'm making, but interesting to note that it's possible that boundaries could be different if a different set of random numbers was used.

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Brent Jackson, in reply to Graeme Edgeler,

    Hmm. Interesting. I guess it would have be be done that way it order to assign the boundaries. However, a few dozen people one way or the other is not going to affect boundaries as much as choosing appropriate rivers / roads to draw a line along.

    Auckland • Since Nov 2006 • 620 posts Report Reply

  • Graeme Edgeler, in reply to BenWilson,

    Can you clarify this point? Are you saying that the way they answered the census will affect their ability to enrol on the Maori Roll? Or are you saying that being unable to answer is what excludes them, something you are then carrying forward when it comes time to decide which roll to be on?

    The answer on the census does not effect the eligibility to enrol and take part in the Māori Option.

    That turns on the question asked on enrolment forms:

    Are you Māori?

    With the information provided:

    Only New Zealand Māori, or descendants of New Zealand Māori, may answer by ticking the "YES" circle. All other people must tick the "NO" circle.

    It is of course possible that someone may answer the Māori descent question on the census differently from the Māori descent question on the enrolment form (although lying on either is an offence).

    In respect of people who honestly select "don't know" on the census form, they are instructed to tick "no" on the enrolment form. Similarly, people who leave that question on the enrolment form blank are not recorded as being of Māori descent for electoral purposes (such as the Māori Option).

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Rich of Observationz,

    I'd assume that Stats NZ's mission in this is to provide an accurate number.

    There are a few ways people can provide an uncategorised answer to the question:
    - they genuinely don't know
    - they don't understand the question
    - they don't want to fill the census out properly

    There is also an aspect of Māori having been discriminated against variously in the history in NZ, which would most likely result in people choosing to hide their ancestry from their children, sadly.

    The approximations used by stats do seem reasonable. (For instance, declaring an iwi affiliation mostly indicates that one is of Māori descent, as does having a Māori parent).

    It's true on the face of it that someone who doesn't give an unequivocal yes can't honestly register on the Māori roll, but it's also possible that they could learn more about their ancestry and later make that decision.

    Back in Wellington • Since Nov 2006 • 5550 posts Report Reply

  • Graeme Edgeler, in reply to Brent Jackson,

    However, a few dozen people one way or the other is not going to affect boundaries as much as choosing appropriate rivers / roads to draw a line along.

    Probably not a great difference usually, although at the margins I imagine it occasionally could. And, of course, nearly half-a-million people are having Māori descent (or lack of it) imputed to them in this way.

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Graeme Edgeler, in reply to Rich of Observationz,

    I’d assume that Stats NZ’s mission in this is to provide an accurate number.

    Yes, but an accurate number of what? Statistics NZ are charged with using "the total number of ordinarily resident persons of New Zealand Māori descent as determined by the last periodical census", the number they give out in their publicly-available census results is different.

    This leaves the question: why is this better, more accurate number not used for anything else?

    And in respect of Māori enrolment: why does this imputation increase the number of people of Māori descent, but not increase the number of people of Māori descent who are on the general roll? If we are counting people who don't know, or won't say they are Māori in the first stage of the calculation, why are we assuming there are more Māori for the next stage of the calculation?

    Wellington, New Zealand • Since Nov 2006 • 3215 posts Report Reply

  • Rich of Observationz,

    why is this better, more accurate number not used for anything else

    Indeed. Possibly Stats prefer the general ethnicity datasets for all other purposes (it would be interesting to see how they align?)

    The census and electoral registration forms are different: the census is yes/no/don't know while the registration form is simply yes/no. I'd suspect that when people who answer "don't know" are given a binary choice, they might tend to split between yes and know (this must have been researched...)

    Back in Wellington • Since Nov 2006 • 5550 posts Report Reply

  • Paul Campbell,

    I have a question (not stirring here, just my ignorance) electorates are all supposed to be the roughly same size why not hold open enrollments for the Maori roll for 4 months, then look at the sizes of the rest of the electorates (after the Maori roll electees have been removed) and then allocate enough Maori seats so that the Maori seats are roughly an equivalent size? (ie based on the number enrolled rather than the population size of those who might enroll)

    Does the size of an electorate depend on the total number of people of any age - if say Stuart Island was suddenly populated by 25,000 (or whatever the appropriate number is) 13 year olds would it create an electorate in which no one was allowed to vote? (apart from the whole lord of the flies thing that would invariably result) .... or more seriously do populations that have bigger families get more representation (per voter)?

    Dunedin • Since Nov 2006 • 2623 posts Report Reply

First ←Older Page 1 2 3 Newer→ Last

Post your response…

Please sign in using your Public Address credentials…

Login

You may also create an account or retrieve your password.