Census 2016 - Time to tick NO

What is your particular religious perspective then? Or specific religious perspective, maybe? That may invite too much detail from anti-theists though...

Choochtown wrote: »

But it is hidden as I've pointed out earlier in this thread.
Question 19 "How do you usually travel to work?" First box to tick "Not at work"
Question 20 is the same format.
Questions 7,9,11,27 are all set out in a similar way with choices and they all end with the white boxes for the "other" option
Only in this highly contentious question is:
(1) The second most common answer put last
(2) There is another option after the fill in squares for "other"
(3) There is a presumption for everyone to tick a religion (eg question 14 doesn't presume everyone can speak Irish. It doesn't ask: "How often do you speak Irish?" and then hide away the "I don't speak it" option after 2 rows of white squares.
Added to this is the crazy decision not to put the question in the explanatory notes at the back. THE most contentious question on the census is not even explained at the back!!!
Instead they used that space to explain question 22: "Do you provide regular unpaid help for a friend..?"
They helpfully explain... "If you provide regular unpaid help ... you should mark box 1 (Yes)"
Glad they cleared that one up. No wonder there was no room to explain the religion question!

I thought I answered those questions to your satisfaction the last time you asked, but obviously not I recall you'd decided to attack the CSO for negligence instead, but now you're saying they're crazy for providing information about answering the question online (and personally to respondents via the census enumerators) rather than on the form itself... which do you think really? Negligent or crazy? Can't be both...

Choochtown wrote: »

It is indeed both. I posted that it was a "crazy decision" They are negligent for making such a crazy decision. Their job is to glean accurate information about the population. In this regard for this question they are negligent.

To what degree do you think providing advice on how to complete the question in person and online but not on the back of the form affects the result, and the usefulness of the result? Do you not think that if someone looks for clarification of a question on the form and doesn't find it, they won't look online, or ask their enumerator?
Only I would think it would need to be a fairly significant degree before you could call it negligent. And since there is only advice on the back of the form for twelve of the forty seven questions, was it a negligently 'crazy decision' not to include advice for 75% of the questions? Is it possible there may be a logic to the questions chosen for explanation you're not aware of; like the logic to the construction of potential answers?

"What is your religion" is like the question 'When did you stop beating your wife'

'I never beat my wife' does not answer the question, because it doesn't say when you stopped beating your wife. (incidentally, I still beat my wife, I'm actually beating her right now' is an equally non answer for the same reasons, as is 'I don't have a wife')

You can't answer the question if you've never beaten your wife. You can only challenge the question itself.

Asking a question that pre-supposes an answer is poor methodology.

'Are you religious?' followed by 'If Yes, what religion are you' would get the most accurate results.

Incidentally, religion thrives on presupposing that the 'questions' it asks about reality are valid questions.

What is the 'meaning of life' presupposes that there is one meaning to life

'Why are we here' presupposes that there is a transcendent purpose for our lives

'Where did morality come from' presupposes that morality had to 'come from' anywhere instead of just being a concept invented by humans to explain human behaviour and the unwritten 'rules' governing our social interactions.

The census isn't that important in the grand scheme of things (presupposing that there is a grand scheme) but it would be nice if the next census allowed a proper debate on the religion question so that it can better serve statisticians and policy makers of the future.

They have no excuse. This census is a 'no change' census, the next one has to allow reasonable contributions on the question format.

The campaign for reforming census 2021 should start the day after census 2016 is collected,

Akrasia wrote: »

"What is your religion" is like the question 'When did you stop beating your wife' 'I never beat my wife' does not answer the question, because it doesn't say when you stopped beating your wife. (incidentally, I still beat my wife, I'm actually beating her right now' is an equally non answer for the same reasons, as is 'I don't have a wife') You can't answer the question if you've never beaten your wife. You can only challenge the question itself.

Which is why I think 'No Religion' is logically placed after all the potential answers to the actual question.

Akrasia wrote: »

Asking a question that pre-supposes an answer is poor methodology. 'Are you religious?' followed by 'If Yes, what religion are you' would get the most accurate results

Not if you know from previous censuses that your pre-supposition is an accurate one; in the 2011 Census almost 93% of respondents noted they had a religion. Adding a question to satisfy the grammatical assiduity of 7% of respondents and thereby reducing statistical consistency in the series would seem to be much poorer methodology to me...

Akrasia wrote: »

Incidentally, religion thrives on presupposing that the 'questions' it asks about reality are valid questions. What <..> interactions.

Yes, that definitely seems very incidental, if indicative that your thoughts on the current subject may not be wholly based on simply ensuring our civil servants have access to good statistical data

Akrasia wrote: »

The census isn't that important in the grand scheme of things (presupposing that there is a grand scheme) but it would be nice if the next census allowed a proper debate on the religion question so that it can better serve statisticians and policy makers of the future. They have no excuse. This census is a 'no change' census, the next one has to allow reasonable contributions on the question format. The campaign for reforming census 2021 should start the day after census 2016 is collected,

I imagine the next Census will include any sensible, or needful changes that the CSO arrives at; whether it's sensible or needful to change that particular question is obviously debatable. My own view is it should be changed if the CSO concludes that more useful statistical data can be derived by changing rather than keeping it.

Absolam wrote: »

How exactly is it defeating my own argument? There's certainly an opinion being offered about how valid people think both the question and answers are; but whether or not people think they're trustworthy has no bearing on their consistency.

I love this. You're defending the question, even though it provides inaccurate and useless information on the basis that consistently wrong is better than a question that would provide better information but is not directly comparable to previous census results.

Maybe the census should still ask us about whether we are "deaf, dumb, blind, idiot, imbecile or lunatic" just so we can easily compare todays population with the ireland of the 1901

You can assume that the next time you ask the question, it's likely about 93% will again tick the box 'I never stopped beating my wife' unless something happens to change the answer they will give. The conclusions you may draw from how you look at the figure, such as how many people are still beating their wife, would be up to you, notwithstanding the fact that the particular setup you've chosen might reasonably allow a statistician to assume that in a census following six months later some of those who had answered 'Last year or sooner' might move into the 'last year or greater' category.

You would know that some people admit being reformed wife beaters, you know nothing else about the population. It's a terrible question because you are left with more questions than answers.

I don't think so. What is being is measured is what religion people say they are. Whether they are religious, or what their cultural identity, or adopted faith is, or what their parent beliefs are, is not asked, so any assumptions about that are simply assumptions, not measurements. As you say, they were never asked if they were religious, so why are you trying to match the results against what you think is evidence of being religious?

You're wrong. There is no agreement on what the question is asking, there are no notes to clarify if it means cultural identity, heritage or current belief. Because nobody knows what is actually been asked, the information collected is tainted and unreliable.

Whether or not people are religious is not being questioned; I have said the 93% of the last census justifies the supposition that most people will say they have a religion,

You're arguing that people can have a religion but are not religious? It's as clear as mud, and you're tripping over yourself trying to justify a question that artificially inflates religious belief in Ireland because it suits you to believe that religion is not in terminal decline in modern Ireland. (it would be nice to have official statistics to back up that statement, but the census figures are wrong)

I care about this question because I want the state to become more secular to reflect the secular attitudes of the population, religious people want to leave the question as it is because they want to protect the church state relationship, the religious schools and hospitals and the privileged position that religious leaders have in influencing policy and legislation.

We both have an agenda, but the difference between us, is that I want more accurate information to be recorded, while you want to maintain an illusion of religiosity in the national statistics.

No; the results are straightforward. You're clearly trying to mislead by conflating whether people have a religion with whether people are religious.

Define 'having a religion' and how is it different from being religious. And after you're finished defining 'having a religion' demonstrate that this is an official or universally accepted definition such that it would not cause confusion when interpreting the census data

Akrasia wrote: »

I love this. You're defending the question, even though it provides inaccurate and useless information on the basis that consistently wrong is better than a question that would provide better information but is not directly comparable to previous census results.
Maybe the census should still ask us about whether we are "deaf, dumb, blind, idiot, imbecile or lunatic" just so we can easily compare todays population with the ireland of the 1901

I think your notion of 'defending the question', is pretty odd; it's just a question, it doesn't need defending. I'm saying that contrary to Choochtowns ideas, I can see a logical structure in how it's set out, and with regard to amending it, why statistical consistency in the series would be a reason not to. I think your argument for inaccuracy in the result is based on an inaccuracy in your assessment; trying to use data on what religion people say they are to determine if they are religious is trying to shoehorn one thing into another, very much your previous round pegs and square holes problem. Whether or not it is useless depends on what the CSO uses it for (I'll agree it's useless for in some nebulous fashion persuading the state to become more secular to reflect the secular attitudes of the population, though I'm not convinced of what would be useful for that, nor yet that it's something the CSO would do anyway); you haven't told use why any other question would be more useful to the CSO, you've only put forward ideas for questions which would produce useful results for anti-theists. The Anti Theist Office doesn't operate a census though.

Akrasia wrote: »

You would know that some people admit being reformed wife beaters, you know nothing else about the population. It's a terrible question because you are left with more questions than answers.

In fairness you've only yourself to blame for that; you made up the question and the answers. Based on your own assessment there's good reason to keep you well away from rewriting census questions!

Akrasia wrote: »

You're wrong. There is no agreement on what the question is asking, there are no notes to clarify if it means cultural identity, heritage or current belief. Because nobody knows what is actually been asked, the information collected is tainted and unreliable.

I'm not. There's no disagreement on what the question is asking. "What is your religion?". There's no room for dispute there; the question asks what is your religion. What you think that entails is something else, obviously, and since you don't think it will produce a result you like, you think it's tainted and unreliable. But for the purposes of understanding what religion people say they are, the information collected is perfectly fine.

Akrasia wrote: »

You're arguing that people can have a religion but are not religious? It's as clear as mud, and you're tripping over yourself trying to justify a question that artificially inflates religious belief in Ireland because it suits you to believe that religion is not in terminal decline in modern Ireland. (it would be nice to have official statistics to back up that statement, but the census figures are wrong)

I'm saying there's no reason to think someone can't say they have a religion yet not think they are religious. Again, I'm still not trying to justify the question; I'm pointing out that you're trying to shoehorn unstated information into it. I understand that you'd like information that agrees with your perception and you want to work back to a question that will give it to you, but a census is not the place for that sort of thinking. Just like your perception that it suits me to believe that religion is not in terminal decline in modern Ireland, so you create arguments around it. In fact I've never said any such thing, nor have I tried to justify a question that artificially inflates religious belief in Ireland; I've pointed out the question measures what religion people say they are, nothing more. As religions ebb and flow in Ireland, the current census will measure how many people say they belong to each, and how many say they belong to none. I've no problem with the fact that that is what is measured.

Akrasia wrote: »

I care about this question because I want the state to become more secular to reflect the secular attitudes of the population, religious people want to leave the question as it is because they want to protect the church state relationship, the religious schools and hospitals and the privileged position that religious leaders have in influencing policy and legislation.

Maybe let the religious people offer their own arguments for what they want? However, with regards to what you want, I'd suggest the census simply isn't a means to your end. That'd you'd like it to be doesn't seem to be a really good reason to change it.

Akrasia wrote: »

We both have an agenda, but the difference between us, is that I want more accurate information to be recorded, while you want to maintain an illusion of religiosity in the national statistics.

It's humbling that you can perceive my agenda and set if forth for me without ever even making an enquiry; your psychic skills obviously surpass your ability to create census questions. Though on the other hand, since that's not my agenda, maybe not. But I would suggest that your own agenda is also set out incorrectly; from what you've said it's not more accurate information you're looking to get (or at least you've suggested no means to disarm the interfering numerators, misled dads, bossy mammies, scary flatmates, & no doorbells), but different information; a means of measuring the religiousity of those who are so inconsiderate of your agenda that they say they have a religion.
Personally, I feel the question as stands provides statistical consistency in the series, which is a good reason to retain the format. I wouldn't object to changing it if statistical consistency could be maintained (ie it still measured the same metric), or adding additional questions regarding religion if the CSO had an objective use for the additional information obtained. I'm not inclined to think serving your agenda is an objective use, or even a worthwile use of the CSOs resources though.

Akrasia wrote: »

Define 'having a religion' and how is it different from being religious. And after you're finished defining 'having a religion' demonstrate that this is an official or universally accepted definition such that it would not cause confusion when interpreting the census data

Why? I'm more than happy to stipulate that your idea of having a religion may be different to someone else's, and whilst you seem to think that having a religion in some way necessitates being religious, others may well feel differently. Since we're not setting out to determine how religious people are, or what they think having a religion entails, neither would seem to matter to the question "What religion are you?". The only thing that matters is what religion they say they are....

Cheers for that, just downloaded it! Always hated Irish in school and ever since I left I always wished I'd learned to speak it properly. Now I can remind myself every day how little I know

Absolam wrote: »

I think your notion of 'defending the question', is pretty odd; it's just a question, it doesn't need defending. I'm saying that contrary to Choochtowns ideas, I can see a logical structure in how it's set out, and with regard to amending it, why statistical consistency in the series would be a reason not to. I think your argument for inaccuracy in the result is based on an inaccuracy in your assessment; trying to use data on what religion people say they are to determine if they are religious is trying to shoehorn one thing into another, very much your previous round pegs and square holes problem. Whether or not it is useless depends on what the CSO uses it for (I'll agree it's useless for in some nebulous fashion persuading the state to become more secular to reflect the secular attitudes of the population, though I'm not convinced of what would be useful for that, nor yet that it's something the CSO would do anyway); you haven't told use why any other question would be more useful to the CSO, you've only put forward ideas for questions which would produce useful results for anti-theists. The Anti Theist Office doesn't operate a census though.

The consistency argument doesn't hold up because questions in the census evolve over time. 50 years ago, when the number of outspoken atheists in the country could be counted on the fingers of one hand. And the main social dynamic was catholic vs protestant 'what religion are you' was a useful question. Now, the catholics and protestants have joined forces and the dynamic is secular versus religious, so 'what is your religion' is no longer a useful question when it presumes that everyone is religious.

In fairness you've only yourself to blame for that; you made up the question and the answers. Based on your own assessment there's good reason to keep you well away from rewriting census questions!

It was deliberately written that way to demonstrate how poorly written questions can have useless answers.

I thought that would have been obvious

I'm not. There's no disagreement on what the question is asking. "What is your religion?". There's no room for dispute there; the question asks what is your religion. What you think that entails is something else, obviously, and since you don't think it will produce a result you like, you think it's tainted and unreliable. But for the purposes of understanding what religion people say they are, the information collected is perfectly fine.

There's no room for dispute? Is there room for reality in this discussion, this thread disproves that statement. You're arguing with people about the meaning of this question and in the middle of this long protracted argument, you say that there is no room for dispute about the central disagreement that people are clearly disputing?

What is your religion is clearly a leading question that confuses people who are not religious. The answer for a religious person is obvious, the answer for a non practising/believing person is less obvious, do they tick the religion they were born into, or do they make the conscious effort to become an apostate and abandon their heritage and cultural identity.

If you ask people 'are you religious' loads of people will say no. If you ask people what religion are you, the same people who said no, will often answer 'catholic' or whatever. So the format of the question definitely matters.

I'm saying there's no reason to think someone can't say they have a religion yet not think they are religious. Again, I'm still not trying to justify the question; I'm pointing out that you're trying to shoehorn unstated information into it. I understand that you'd like information that agrees with your perception and you want to work back to a question that will give it to you, but a census is not the place for that sort of thinking. Just like your perception that it suits me to believe that religion is not in terminal decline in modern Ireland, so you create arguments around it. In fact I've never said any such thing, nor have I tried to justify a question that artificially inflates religious belief in Ireland; I've pointed out the question measures what religion people say they are, nothing more. As religions ebb and flow in Ireland, the current census will measure how many people say they belong to each, and how many say they belong to none. I've no problem with the fact that that is what is measured.

I'm trying to shoehorn unstated information?
Me?

All I'm doing is pointing out that the question is flawed. I'm saying that people shouldn't take any information out of that question because it's poorly formed. The only information you can take out of the question, is, as you've said yourself, the answer people gave to the question.

It's as informative as asking people
'Pick a number between 1 and 10' We'll get an answer, but it's not a useful answer.

If you're trying to measure the 'ebb and flow' of religion, you should ask people if they're religious, not just what religion they were born into.

Maybe let the religious people offer their own arguments for what they want? However, with regards to what you want, I'd suggest the census simply isn't a means to your end. That'd you'd like it to be doesn't seem to be a really good reason to change it.

I'm being honest about my agenda, religious people, including you, are being dishonest by pretending that they have no agenda. My agenda is that the census should reflect the true status of religion in Ireland. You don't get to win the debate by lying about your motivations or avoiding the question. I can point out that everyone has a bias, and if you think I am wrong, then deny it and we can discuss that.

It's humbling that you can perceive my agenda and set if forth for me without ever even making an enquiry; your psychic skills obviously surpass your ability to create census questions. Though on the other hand, since that's not my agenda, maybe not.

state your agenda then.

But I would suggest that your own agenda is also set out incorrectly; from what you've said it's not more accurate information you're looking to get (or at least you've suggested no means to disarm the interfering numerators, misled dads, bossy mammies, scary flatmates, & no doorbells), but different information; a means of measuring the religiousity of those who are so inconsiderate of your agenda that they say they have a religion.

I just want a question that gives accurate information about what people believe. You don't.

Personally, I feel the question as stands provides statistical consistency in the series, which is a good reason to retain the format.

Is it a good enough reason to overcome the flaws in the format? Is consistency more important than accuracy?

What is the use of consistently misleading information? Should people have to guess about what people really meant when they answered a question, or should the question be asked in an un ambiguous manner?

I wouldn't object to changing it if statistical consistency could be maintained (ie it still measured the same metric), or adding additional questions regarding religion if the CSO had an objective use for the additional information obtained. I'm not inclined to think serving your agenda is an objective use, or even a worthwile use of the CSOs resources though.

Any change to the questions on the census will impact 'consistency' I don't care if they ask additional questions or change the current question. I want a distinction between belief and heritage to be recorded in the census because that is important information for the development of policy and legislation.

Why? I'm more than happy to stipulate that your idea of having a religion may be different to someone else's, and whilst you seem to think that having a religion in some way necessitates being religious, others may well feel differently. Since we're not setting out to determine how religious people are, or what they think having a religion entails, neither would seem to matter to the question "What religion are you?". The only thing that matters is what religion they say they are....

Why?
When you make assertions, like:

"the results are straightforward. You're clearly trying to mislead by conflating whether people have a religion with whether people are religious"

You should be able to support those assertions with evidence or argument.

If you can't support your assertion that there is an easily understood difference between people having a religion, and people being religious, then you should withdraw that claim.

The results are not straightforward. There is no clear difference between having a religion and being religious. Asking people 'what is your religion' and getting a 93% response indicating a religion clearly implies that 93% of people are religious, and that's in total disagreement with all the other evidence that shows way more than 7% of people are not religious

Absolam wrote: »

No, it holds up for any question for as long as that question doesn't change. That's statistics for you. Whether or not it's a 'useful' question is obviously an entirely different thing; I understand that the questions you favour is one you think would be useful to you, though I'm not sure it's any more useful than the existing one to the CSO. That fact that the question presumes everyone has a religion (not is religious, as you've said) has nothing to do with how 'useful' it is, but the presumption is borne out by the fact that 93% of respondents said they had a religion all the same.

Well done you so.

Actually I said there's disagreement on what the question is asking. It's straightforward: "What is your religion?" It means what is your religion. You may want it to mean something else, you may draw conclusions regarding anything from cultural identity, to adopted faith, to the beliefs of the parents or whoever completes the form etc, but it's only four words. What is your religion. It means nothing other than what it says.
I think there aren't all that many people who aren't religious (or even who don't have a religion) who get so confused by the question what is your religion that they feel compelled to say they have one. And those that do exist probably have much bigger issues than a census form.
So what you're saying is, ask people two different questions, get two different answers. I think I may have pointed that very thing out earlier on.....
Well done. Aside from thinking the question is flawed because you can't derive information from it that isn't put into it you've arrived at where you should have started; the question tells us what religion people say they are and nothing else. Specifically for the purposes of the argument you're advancing, it doesn't tell us if they're religious, or how religious they are, because they're not asked.
That's a pretty useful question if you want to know what numbers people will pick when asked to pick a number between 1 and 10 though. In fact, it's the perfect question, especially if you ask everyone.
Then you'd be measuring the ebb and flow of religiousness. And they weren't asked what religion they were born into, remember? It was a real simple four words; What religion are you?. Nothing about born at all; that's you shoehorning again. So, like I said, the current census will measure how many people say they belong to each, and how many say they belong to none.
So, it's up to you to decide I'm religious, and what my agenda is, and whether I'm honest about it? Don't think so, sorry. Accusing someone of lying just because they're not providing the argument you want is pretty low....
Why do I have to have an agenda? I'd rather do without to be honest, I've been fine up to now without one.
You really need to get over this telling people what they think thing, it's not good for you. I'm happy to have a question that provides information about how people identify themselves that can be measured over time to give an idea how the population is changing. I don't really care to know what they believe; I don't think anything so nebulous can reasonably be captured in a census, I think it probably changes on a day to day basis, and I can't see a lot of use for it.
I don't think there's anything inaccurate or misleading about the current information; I know it's not the information you'd like to have, and using it as if it is would lead to you being accused of being misleading, or even having to guess at what you really think would better serve your purpose, but that's not the fault of the census.
I don't think I like the sound of developing policy or legislation based on beliefs or heritage thanks, so I think I'll pass on your suggestion. Sounds a bit too much like social engineering.

Oh, I don't think I need to define 'having a religion' and how is it different from being religious to make that assertion; you have on this thread conflated being religious with having a religion (over and over and over again), without ever telling us why they should be treated as the same thing.
And yet according to Choochtown 53% of the population claim not to be a religious person, so it would seem that your understanding of what it is to be religious is at odds with a very large proportion of the population.
So either, being religious is not the same as having a religion, or most of the population are very confused. I'm afraid most of the people I meet don't usually strike me as very confused, so I'm inclined to think it's just you, and it is possible to have a religion yet not consider yourself religious. And I've yet to see any reason to measure how religious people actually are, never mind how religious they may say they are.

It's boring to talk to someone who refuses to provide any arguments to support his/her own position

You insist that there is a clear difference between being religious and 'having a religion' but you refuse to define that difference

I keep hammering this point not because i think there is no difference, but because it is confusing and there are multiple interpretations of what it means to 'have a religion'. My position is perfectly reasonable. The census shouldn't ask 'What religion are you' because it's a leading question by every single definition of leading questions. It's clearly and obviously the wrong way to phrase that question without first asking people if they are religious.

If you ask two people who are functionally atheist "What religion are you" and one of them thinks the question means what religion do you practise, and the other person thinks the question means 'what religion were you baptised into before you could crawl' you're going to get two different answers to the same question based on two equally valid interpretations of the same question

That's why it needs to be changed. The two answers are opposite to each other, but are equally valid depending on how each person interprets the question, and without any guidance or notes on what the question means, the two different interpretations are equally valid.

The only argument for keeping the question the same is 'consistency' and it's a stupid argument.

Choochtown wrote: »

For the 2nd time here's the link: http://www.wingia.com/web/files/news/14/file/14.pdf

Ah... and now I see where you posted it the first time, I also see that you pointed out that the Gallup poll did contain a different question, one which you felt was a much much more straightforward one. So we can immediately see why one would generate a result that says 93% of people have a religion, and the other would generate a result showing 53% are not religious; they're different questions with different answers.

Choochtown wrote: »

This figure is so far from the results given by our 2011 Census as to render one or both polls useless. Personally I think the Gallup question is a much much more straightforward one. It doesn't presume any answers unlike the one in our census and it also leaves the onus on the individual to decide "irrespective of... "

I disagree. Each question is different, and has a different result. In fact having both allows us to see that the population clearly doesn't conflate having a religion with being religious; a piece of information we wouldn't have if we didn't have both surveys.

Choochtown wrote: »

Absolam you seem to think that inaccurate answers are "perfectly valid, simply because they are answers. I disagree. The purpose of the CSO is not to collect inaccuracies, otherwise why would they bother with explanatory notes at the back of the Census?"

No, I just think disagreeing with a question, or answer, doesn't make it inaccurate. If you can demonstrate that the 93% who said they are a religion aren't a religion (as distinct from being religious) then we may say the result is inaccurate. Until then, there's no reason to think it is (other than the vagaries of interfering numerators, misled dads, bossy mammies, scary flatmates, no doorbells, and the fact that most catholics just aren't the proper kind of Catholics, obviously) inaccurate.

Choochtown wrote: »

In their words the information gathered is "quality statistical information, which is vital for the formation, implementation and monitoring of policy and programmes at national, regional and local levels in a rapidly changing economic and social environment." It simply is not good enough to have a flawed, ambiguous, misleading, unexplained question in the census that is clearly (as I've pointed out before) set out differently from all the other questions. Particularly given that: 1. the previous results for this exact same question have given such disputable results. and
2. these results help to decide government policy.

Well, you haven't shown it is flawed, or ambiguous, or misleading, or unexplained ( 2 out of 3 ain't bad, as Meatloaf says), and as I pointed out before, they seem to be set out consistently with the potential logic to all the other questions. Sure, certain people want to dispute the results, but it's generally on the basis of how religious they perceive people to be rather than on the basis of what religion people say they are, isn't it?

And as long as the government policy they're used to decide revolves around what religion people are, rather than how religious they are, the answers are perfectly adequate.

Cabaal wrote: »

Having a "religion" and being of the catholic faith specifically are very different things, if people simply want to believe in a god then by all means I have no problem with this. Also not sure where you are pulling the 93% number from, seems pretty baseless and made up.

That doesn't seem terribly relevant to what I said, but I guess you were trying to make some point?
I pulled the 93% from the 2011 census; 92.81% of respondents stated a religion.

Cabaal wrote: »

But right now many people tick catholic blindly even though they don't believe in many catholic beliefs and would laugh or be disgusted at many of them if you explained them to them and said they needed to believe in them.

Sure; people say they are Catholic and you think they shouldn't. I think we've covered that ground fairly thoroughly; at the end of the day it's just not up to you to tell people what religion they should say they are, they're allowed to decide for themselves.

Cabaal wrote: »

At the end of the day the current question format provides inaccurate data, just like the Irish language question does.

Nope, it just provides data you don't like. Bit of a difference.

Absolam wrote: »

Tell you what; you show me where the CSO says what information it wants by asking this question, and I'll see if we get that information from the question, how's that?

From the horses mouth:

People should answer the question based on how they feel now about their religious beliefs, if any. 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.

http://census.ie//the-census-and-you/each-question-in-detail/

Shrap wrote: »

I feel they could have dramatically reduced all the butt hurt here by the simple inclusions of the words "current" and "if any", then.
"What is your current religion, if any?", leaving the answer boxes as is. Seeing as given their explanation, it's clearly what they mean by the question.

Probably; so far that seems like the closest to a sensible option. I have a feeling there might be some debate if the results came in about the same though; then the people who selected 'Catholic' (just as a for instance!) might be up for scrutiny on how well they understood the question still.