The 2018 Census

Forums › Off topic › The 2018 Census

1 | ... | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18

3646 posts

Uber Geek
+1 received by user: 4554

ID Verified

#1971172 8-Mar-2018 12:35

Seems to me there’s a lot of preciousness, combined with the usual media hype, about extreme minorities' or individuals' situations having been excluded from the questions.

This is statistics, gathering percentages.

Take the gender question for instance. The intent is no doubt to get an understanding of the male/female ratio in the various parts of NZ. I’d hazard a guess that more than 99.5% of those who completed the forms would have ticked either male or female. Obviously then the 0.5% who left it blank identify themselves as neither or as something else. Statistically irrelevant. (I can’t recall if it was possible to leave it blank. Even if it wasn’t and people abandoned the form in protest those numbers would still be statistically irrelevant.)

Sometimes I use big words I don't always fully understand in an effort to make myself sound more photosynthesis.

networkn

Networkn
32871 posts

Uber Geek
+1 received by user: 15464

ID Verified

Trusted

Lifetime subscriber

#1971177 8-Mar-2018 12:48

floydbloke:

Seems to me there’s a lot of preciousness, combined with the usual media hype, about extreme minorities' or individuals' situations having been excluded from the questions.

This is statistics, gathering percentages.

Take the gender question for instance. The intent is no doubt to get an understanding of the male/female ratio in the various parts of NZ. I’d hazard a guess that more than 99.5% of those who completed the forms would have ticked either male or female. Obviously then the 0.5% who left it blank identify themselves as neither or as something else. Statistically irrelevant. (I can’t recall if it was possible to leave it blank. Even if it wasn’t and people abandoned the form in protest those numbers would still be statistically irrelevant.)

Nicely put. I agree entirely.

MikeAqua

8031 posts

Uber Geek
+1 received by user: 3820

#1971208 8-Mar-2018 13:29

floydbloke:

Seems to me there’s a lot of preciousness, combined with the usual media hype, about extreme minorities' or individuals' situations having been excluded from the questions.

This is statistics, gathering percentages.

Take the gender question for instance. The intent is no doubt to get an understanding of the male/female ratio in the various parts of NZ. I’d hazard a guess that more than 99.5% of those who completed the forms would have ticked either male or female. Obviously then the 0.5% who left it blank identify themselves as neither or as something else. Statistically irrelevant. (I can’t recall if it was possible to leave it blank. Even if it wasn’t and people abandoned the form in protest those numbers would still be statistically irrelevant.)

The male female question seem largely irrelevant to me. For planning purposes it can be assumed to be 50:50. More likely that the info is used for gender issues.

Mike

Geektastic

18009 posts

Uber Geek
+1 received by user: 8465

Trusted

Lifetime subscriber

#1971252 8-Mar-2018 14:44

floydbloke:

Seems to me there’s a lot of preciousness, combined with the usual media hype, about extreme minorities' or individuals' situations having been excluded from the questions.

This is statistics, gathering percentages.

Take the gender question for instance. The intent is no doubt to get an understanding of the male/female ratio in the various parts of NZ. I’d hazard a guess that more than 99.5% of those who completed the forms would have ticked either male or female. Obviously then the 0.5% who left it blank identify themselves as neither or as something else. Statistically irrelevant. (I can’t recall if it was possible to leave it blank. Even if it wasn’t and people abandoned the form in protest those numbers would still be statistically irrelevant.)

Logic suggests that if it's statistically irrelevant it could simply be left out...

kiwi_64

276 posts

Ultimate Geek
+1 received by user: 142

Subscriber

#1971291 8-Mar-2018 16:09

I don't think he means that the question is statistically irrelevant, but rather the "0.5%" who might object to answering the question

UHD

655 posts

Ultimate Geek
+1 received by user: 298
Inactive user

#1971298 8-Mar-2018 16:19

MikeAqua:

Something that is interesting to do is use an add-in to see what is running in the background on the Census website. I did this when my partner was completing the census.

There are >10 links to other sites. Most seemed to be for functionality. However, the list also includes google analytics, google ads and double click.

I'd be interested to know why these are being used, in particular the latter two, as there were no adds on the census site.

Do Google Analytics commonly collect data entered into forms?

Dell

Shop now for Dell laptops and other devices (affiliate link).

freitasm

BDFL - Memuneh
80657 posts

Uber Geek
+1 received by user: 41065

Administrator

ID Verified

Trusted

Geekzone

Lifetime subscriber

#1971342 8-Mar-2018 17:40

No, it does not. Also it is against the TOS to link personal identifying data to Google Analytics records.

Referral links: Quic Broadband (free setup code: R587125ERQ6VE) | Samsung | AliExpress | Wise | Sharesies

Support Geekzone by subscribing (browse ads-free), or making a one-off or recurring donation through PressPatron.

DaveB

1139 posts

Uber Geek
+1 received by user: 456
Inactive user

#1971363 8-Mar-2018 18:40

MikeAqua:

floydbloke:

Seems to me there’s a lot of preciousness, combined with the usual media hype, about extreme minorities' or individuals' situations having been excluded from the questions.........

The male female question seem largely irrelevant to me. For planning purposes it can be assumed to be 50:50. More likely that the info is used for gender issues.

I have highlighted some important words here, but at the end of the day MikeAqua possibly points at the real question to be asked. Issues is the word here. How many people have an issue with their gender? Is it 0.5%, 1% or more. Do the other 99% really care? And of that 0.5% or 1% or even 10%, how much is this driven by social media wannabes?

We have always had majority and minority (and then the REAL minority). Totally irrelevant in my view and a waste of tax payer money trying to make sense of it.

Nil Einne

469 posts

Ultimate Geek
+1 received by user: 35

#1971489 9-Mar-2018 03:32

nunz:

Severely under whelmed by the online stats process.

2 - dumb logic - my barely 15 son is too young to legally smoke, be searching for work or be allowed to work and answering those dumb questions when logic of being in high school full time should preclude answering those.

3 - gender - lgbqt*** are going to be pissed - male / female and ... no other options.

Neither 2 or 3 are online specific. The paper form also required you to answer questions those questions if you were 15 or older. Frankly I thought 15 was a little old at least for smoking. The census isn't supposed to be just collecting info on stuff that is legal. If you are here illegal, or hiding your income etc it's intended that you report that hence why there are protections on anyone else being able to use the information. (Well I don't think the census actually asked you on your legal residency status did it? But at least you're supposed to fill in the census and accurately even if you're not supposed to be here, or not supposed to be working here.)

How many people actually do that I don't know but that's the intention. With children, the fact that parents may see the answers will often be something which will discourage truthful answers although the online form may actually increase the chance of honest answers since AFAIK, there was no way for someone else to see the answers unless they monitored as the answers were being entered (either electronically or physically). I'm not just thinking smoking here, areas where it could come that I remember off hand is religion and of course gender. (Even more so they do ever sort out gender and sexuality questions.)

Anyway, as for the gender thing, I think others have already mentioned this but it came up last year and multiple times since then. They did try and come up with questions on this but found that they didn't work https://www.stuff.co.nz/the-press/news/95692743/Gender-issues-left-out-of-New-Zealands-next-census Not everyone were happy with the outcome or agreed with their POV obviously. Well I guess one thing with the online form is that since it was automatic, there was no way to answer the question about how many children you gave birth to if you answered male AFAIK. With the paper form you could have done that although I suspect it would have just been ignored.

I'm not personally interested in debating those issues. I do wonder if people are confusing those issues with the purpose of the various questions. While they are often useful in isolation, they also generally aren't just intended to be used in isolation and that includes the gender one. For example it enables things like this to be studied http://archive.stats.govt.nz/browse_for_stats/people_and_communities/Women/women-occupation.aspx but basically anything else combining the different answers.

E.g. since it came up earlier, is there a difference between the number of male and female smokers? What about languages spoken? Ethnicity?Religion? Non paid work? Even looking at the gender ratio is probably of interest since it tends to start off with a minor bias in favour of males and ends up being a more major bias in favour of females with increasing age. And it doesn't just have to be two factors, you could look at how the number of female and male smokers changes with age. And provided the next/previous census also collected that, how it has changed can also be looked at. The later incidentally is one reason why there is great reluctance to change stuff too easily.

What is actually studied, I'm not sure. I've looked at the public released stuff on occasion, but not in depth. Once you have the data, you could go back and study it if it's felt there is a need for some reason although, intentionally, this isn't something which can be done willy nilly. You may or may not consider studying those sort of things useful, again I'm not particularly interested in debating that but suffice to say, some people do feel it's useful.

lmnop

22 posts

Geek
+1 received by user: 8

#1971554 9-Mar-2018 09:29

freitasm: No, it does not. Also it is against the TOS to link personal identifying data to Google Analytics records.

No government department should be using Google Analytics. Stop being lazy and run in-house stats. (oh the irony)

They shouldn't be running ANY off-site javascript or other files.

mrdrifter

589 posts

Ultimate Geek
+1 received by user: 294

ID Verified

Trusted

#1971560 9-Mar-2018 09:40

freitasm: No, it does not. Also it is against the TOS to link personal identifying data to Google Analytics records.

Being not super familiar with Google Analytics, does it not capture the IP address of the person visiting the site/page?

AliExpress

Shop now on AliExpress (affiliate link).

freitasm

BDFL - Memuneh
80657 posts

Uber Geek
+1 received by user: 41065

Administrator

ID Verified

Trusted

Geekzone

Lifetime subscriber

#1971577 9-Mar-2018 10:25

mrdrifter:

freitasm: No, it does not. Also it is against the TOS to link personal identifying data to Google Analytics records.

Being not super familiar with Google Analytics, does it not capture the IP address of the person visiting the site/page?

But that is not personally identified.

Back on 2018 Census, daughter is back from school camp and she didn't fill the Census form because no one distributed those amongst the students, different from what they said it would happen.

Referral links: Quic Broadband (free setup code: R587125ERQ6VE) | Samsung | AliExpress | Wise | Sharesies

Support Geekzone by subscribing (browse ads-free), or making a one-off or recurring donation through PressPatron.

DarthKermit

5346 posts

Uber Geek
+1 received by user: 3317

Trusted

#1971621 9-Mar-2018 10:34

It makes me wonder how Stats NZ estimate NZ's total population, given that a small percentage of people either won't or don't fill in their census info each time.

mrdrifter

589 posts

Ultimate Geek
+1 received by user: 294

ID Verified

Trusted

#1971624 9-Mar-2018 10:37

freitasm:

But that is not personally identified.

Under GDPR an IP address can be classified as Personally Identifiable Information and there have been a number of recent EU court cases where the ruling has backed this.

nunz

1421 posts

Uber Geek
+1 received by user: 314
Inactive user

#1971689 9-Mar-2018 10:53

MikeAqua:

floydbloke:

Seems to me there’s a lot of preciousness, combined with the usual media hype, about extreme minorities' or individuals' situations having been excluded from the questions.

This is statistics, gathering percentages.

Take the gender question for instance. The intent is no doubt to get an understanding of the male/female ratio in the various parts of NZ. I’d hazard a guess that more than 99.5% of those who completed the forms would have ticked either male or female. Obviously then the 0.5% who left it blank identify themselves as neither or as something else. Statistically irrelevant. (I can’t recall if it was possible to leave it blank. Even if it wasn’t and people abandoned the form in protest those numbers would still be statistically irrelevant.)

The male female question seem largely irrelevant to me. For planning purposes it can be assumed to be 50:50. More likely that the info is used for gender issues.

49:51 in favour of females across world averages.

1 | ... | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18