Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

Census 2022 question on religion

Options
  • 17-01-2022 2:00pm
    #1
    Registered Users Posts: 40,247 ✭✭✭✭


    So I seen the image below elsewhere. Is the question badly phrased or am I being too literal. It asks "what is your religion, if any". To my mind you only tick a box if you don't have a religion. That AI need to put an explainer telling people they need to tick the "No religion" box backs up my position I think.




«1345678

Comments

  • Moderators, Category Moderators, Arts Moderators, Sports Moderators Posts: 48,587 CMod ✭✭✭✭magicbastarder


    i would assume that it's because if you tick a box, you've explicitly answered the question. if you don't tick a box, it's not clear whether that is in itself an answer, or that you've simply not answered it.



  • Registered Users Posts: 6,568 ✭✭✭Allinall


    Not sure why the question needs any explanation.

    Looks perfectly clear to me.

    Are AI just looking for attention, as usual?



  • Registered Users Posts: 40,247 ✭✭✭✭ohnonotgmail


    in that case the "if any" should not be there.



  • Moderators, Society & Culture Moderators Posts: 24,401 Mod ✭✭✭✭robindch


    Is the question badly phrased or am I being too literal. It asks "what is your religion, if any"

    The census bureau is highly - like, really, really, amazingly - reluctant to change the format or text of questions on the census form, as it makes it hard to distinguish, over successive censuses, the impact of changes to questions from changes to the replies. The census bureau, after all, is primarily concerned with documenting society's statistics accurately and, where possible, in such a fashion as to allow direct, reliable comparisons between data sets to be made.

    The change above is fairly minor, but does resolves some issues with the previous format and is likely, IMHO, to lead to a non-trivial increase in the number of people declaring that they have "no religion". Frankly, I'm surprised to see that "No religion" is now first on the list, and whatever lobbying which led to this placement was certainly worth it.

    Anybody willing to bet on what the figures might be when they're released later in the year?



  • Registered Users Posts: 40,247 ✭✭✭✭ohnonotgmail


    except they have changed the text of the question. And not for the better. The addition adds ambiguity.




  • Advertisement
  • Moderators, Category Moderators, Arts Moderators, Sports Moderators Posts: 48,587 CMod ✭✭✭✭magicbastarder


    i'd say there's an argument to be made that a 'what is your religion?' question implies a faint presumption that you have one, even if there's an option to state otherwise.

    in the same sense as 'what brand of car do you drive?' could be more accurately phrased as 'do you drive a car, and if so what brand?'



  • Registered Users Posts: 40,247 ✭✭✭✭ohnonotgmail


    ye but phrasing the question with "if any?" at the end might cause somebody to not fill any boxes if their answer to the question was negative. To use your example, if the question was "What car do you drive, if any?" then somebody who does not drive any car may just skip over the question and not pick any box.



  • Registered Users Posts: 1,086 ✭✭✭Michael Nugent


    I hope everyone is well since I last posted here, and as many as possible have survived the pandemic unscathed. Here's a brief summary of Atheist Ireland's lobbying on the census issue. As an overview, we think the religion question has been slightly improved, but it is still flawed.

    During the consultation process, the Census office argued for a minimalist approach to changing the religion question, because they said it was important to compare results from census to census. We had argued that it was more important to get accurate results, rather than to be able to compare one set of flawed results with another.

    The old question was ‘What is your religion?’ That was clearly a leading question, that resulted in a higher figure for religion, by assuming that the respondent is a member of a religion. The new question will be ‘What is your religion, if any?’ This is slightly better, but is still implicitly a leading question, as it assumes that religion is the default, and that not having one is an aberration.

    The new question is also slightly better than the old one, in that ‘No religion’ will be the first option of the check boxes, instead of the last one which was hidden beneath the write-in box for other religions. This means that people are more likely to see it before they have ticked the check-box based on their childhood religion or cultural identity.

    We had argued for the check-boxes to be removed entirely, and for a return to the write-in answer that had been used for a century until the check-boxes were introduced in 2002, and given to the five highest religions from the previous census. These check-boxes overstate those religions that are given a check-box. Church of Ireland, Presbyterian, and Methodist all dropped consistently from the 1960s to the 1990s. When given check-boxes, they all increased in 2002. Methodist doubled when given the check-box in 2002, then halved when it was taken away in 2011.

    As the census date approaches, Atheist Ireland will continue to publicise the slight improvement and the underlying flaws in the census question on religion. We are confident that, despite these flaws, the 2022 census will yet again see a significant rise in the number of nonreligious people in Ireland.



  • Registered Users Posts: 40,247 ✭✭✭✭ohnonotgmail


    We had argued for the check-boxes to be removed entirely, and for a return to the write-in answer that had been used for a century until the check-boxes were introduced in 2002, and given to the five highest religions from the previous census. These check-boxes overstate those religions that are given a check-box. Church of Ireland, Presbyterian, and Methodist all dropped consistently from the 1960s to the 1990s. When given check-boxes, they all increased in 2002. Methodist doubled when given the check-box in 2002, then halved when it was taken away in 2011.

    Is it not more likely that having no checkbox for a particular religion will lead to an understatement of their number rather than the provision of a checkbox leading to an overstatement of their number?



  • Moderators, Category Moderators, Arts Moderators, Sports Moderators Posts: 48,587 CMod ✭✭✭✭magicbastarder


    there was a short podcast (a segment on BBC's 'more or less' IIRC) where they discussed how important phrasing of questions can be in surveys. they were investigating a claim which you often hear, along the lines of '5% of british people don't believe the holocaust occurred', and it was a nice short primer as to how badly structured surveys can lead to inaccurate results.





  • Advertisement
  • Registered Users Posts: 1,086 ✭✭✭Michael Nugent


    I suppose you could interpret it either way. If there is no tick box, you have to consciously select your religion, and you might describe it differently to other members of the same religion. If there is a tick box, you might be prompted to tick a childhood religion that you don't actually identify with. In practice, the tick boxes are the privilege not the default as you can't have a tick box for every religion in the country.



  • Moderators, Category Moderators, Arts Moderators, Sports Moderators Posts: 48,587 CMod ✭✭✭✭magicbastarder


    maybe some people's faith is so shallow that having to write the name of the faith out is too much work..



  • Registered Users Posts: 40,247 ✭✭✭✭ohnonotgmail


    given that

     Methodist doubled when given the check-box in 2002, then halved when it was taken away in 2011.

    it seems to me that providing a checkbox leads to a more accurate count. I find it unlikely that half of them had a faith so shallow that having to write it in was too much work in 2011. I think it more likely that they did write it in but it either it wasn't read properly or wasn't written precisely in whatever form the census was expecting.



  • Moderators, Category Moderators, Arts Moderators, Sports Moderators Posts: 48,587 CMod ✭✭✭✭magicbastarder


    i would be curious about how other faiths may have changed based on the switch; and map that against what is believed to be a background noise of error.

    i.e. it's easier to make a mistake ticking a box than writing out the name of the faith; and a faith with say 1% representation when the overall error rate is say 1% will see a much bigger swing in how they are represented as a result.



  • Registered Users Posts: 40,247 ✭✭✭✭ohnonotgmail


    you think 10000+ plus people accidentally ticked methodist in 2002 and entered their religion correctly in the next census?



  • Registered Users Posts: 24,737 ✭✭✭✭breezy1985


    No it doesn't add any ambiguity you are just trying very hard to see something that is not there.

    The old "what is your religion" is a stupid question because a lot of people don't have one.



  • Registered Users Posts: 24,737 ✭✭✭✭breezy1985


    I don't know about other religions but the lack of a "no religion" definitely leads to higher catholic numbers. I know people who just tick catholic out of laziness because they were baptized and don't want to have to write in an explanation for anything. I also know people who before the AI campaigns thought it was illegal to say atheist if they were baptized as it would be lying to the government



  • Registered Users Posts: 40,247 ✭✭✭✭ohnonotgmail


    "No religion" has been an option since at least 2011.



  • Moderators, Category Moderators, Arts Moderators, Sports Moderators Posts: 48,587 CMod ✭✭✭✭magicbastarder


    i don't know, was just speculating. let's say the error rate for choosing methodist is close to zero when written, but 0.1% of people ticked the wrong box - the methodist box - by accident. that'd account for the extra 5k nationally. i'm not saying that's what happened, but i suspect the CSO have inklings of what the error rate is.



  • Registered Users Posts: 40,247 ✭✭✭✭ohnonotgmail


    i think you underestimate the error rate for written answers and vastly overestimate the error rate for ticking the wrong box especially when practically all the people who ticked the wrong box would have to make the same mistake and pick Methodist.



  • Advertisement
  • Registered Users Posts: 24,737 ✭✭✭✭breezy1985


    Since before that and the numbers have been on a steady increase. The AI campaign letting people know that they don't have to tick birth religion is an important and slowly showing to be a successful one



  • Registered Users Posts: 40,247 ✭✭✭✭ohnonotgmail




  • Moderators, Category Moderators, Arts Moderators, Sports Moderators Posts: 48,587 CMod ✭✭✭✭magicbastarder


    to call my guesses 'estimates' would be a mistake!



  • Registered Users Posts: 24,737 ✭✭✭✭breezy1985


    I don't even know what you are on about at this stage.

    The whole premise of the thread is nonsense. You are clearly just upset that the atheists got something they wanted.

    Funny that if I started a similar thread in the catholic section I would be shut down pretty quick on that little bubble



  • Registered Users Posts: 26,166 ✭✭✭✭Peregrinus


    But that effect will be countered by the fact that the very first box is "no car".

    The objection to the old question, "What is your religion?", is that it implies that everyone has a religion, or at least that it suggests a norm in which everyone is expected to have a religion. Other "What is . . .?" questions on the form in 2016 included "what is your ethnic or cultural background?", "What is your date of birth?", "What is your place of birth?" and "What is your current marital status?" — all referring to characteristics which everybody has and to which the answer "none" is logically impossible. "What if your religion, if any" suggests that having a religion and not having a religion are equally normal.

    We'll know if your fears are correct when the census results are compiled, if there is a rise in the proportion of respondents who fail to answer this question.



  • Registered Users Posts: 26,166 ✭✭✭✭Peregrinus


    There was a "no religion" box on the census form in 2016; nobody had to write in "no religion". The change this year is that the "no religion" will be the first option, rather than the last, and that the option will be pointed to in the question.

    I struggle to believe that more than a tiny, statistically irrelevant number of people thought it was illegal to say "atheist" if they had been baptised. That's a gobsmackingly bizarre belief.



  • Registered Users Posts: 24,737 ✭✭✭✭breezy1985


    I mentioned no time or year or ever said this will be the first time it was on the census.



  • Registered Users Posts: 26,166 ✭✭✭✭Peregrinus


    I know. But the thread is about the changes to be made in the 2022 census form, so I thought it was worth making the point that this isn't such a change.



  • Registered Users Posts: 24,737 ✭✭✭✭breezy1985


    Thanks for making the point that was made already.

    As for the comment on people feeling obliged to tick catholic it was enough of a problem that the CSO had to release a statement on their website in 2016.

    “The question is asking about the person’s current religion or beliefs and not about the religion the person may have been brought up with”

    There is a bit on it here which also covers the legal obligation to participate and the antiquated "head of the household" rule.




  • Advertisement
  • Registered Users Posts: 40,247 ✭✭✭✭ohnonotgmail


    pedantic the thread may well be your accusation is nonsense. Going by the comments from Michael Nugent it seems they didn't get what they wanted. If anything it has been made worse but maybe that is all in my pedantic head.



Advertisement