BC Teachers Getting Screwed...Again

[Pct2017]

what makes it garbage?

Well, in actual fact, all opinion polls are garbage. That has been proven without a doubt in the last two provincial elections out west. But, to be specific, The Tyee is a very left wing website that panders to it's NDP audience accordingly. They posted a write-in poll which young Socialist (notice how I don't call you Plink a Plink anymore since WCR convinced you to drop the neo-lib crap) has bought as biblical in nature. So, your audience is all lefties, and you have a write in poll, what do you think the results are going to be?

To put this in prospective, would you see a write in poll by the Fraser Institute which would no doubt show that the "public" is 95% supportive of the government's actions as anything but garbage?

[Michael Hardner]

Well, in actual fact, all opinion polls are garbage.

That's a strange claim to make. What does "garbage" mean ? They're useful in predicting outcomes to elections, for one thing.

But, to be specific, The Tyee is a very left wing website... So, your audience is all lefties, and you have a write in poll, what do you think the results are going to be?

To put this in prospective, would you see a write in poll by the Fraser Institute which would no doubt show that the "public" is 95% supportive of the government's actions as anything but garbage?

You are close, but not exactly right. It doesn't matter what the audience is, it matters how the sampling is done. Any news organization worth its salt should show methodology and strive to present the real news, not a skewed version.

Did The Tyree do that ? No, they did not. But not for the reasons you stated. The reasons why the Tyree poll is not representative was explained right beside the poll results:

Please note that Tyee Barometer polls are only intended as a quick and engaging non- scientific snapshot of our readers’ opinions on various topics that fit with The Tyee's very broad editorial mandate. They are not intended to be seen as a representative sampling of BC opinion

[cybercoma]

Please note that Tyee Barometer polls are only intended as a quick and engaging non- scientific snapshot of our readers’ opinions on various topics that fit with The Tyee's very broad editorial mandate. They are not intended to be seen as a representative sampling of BC opinion

This is specifically about the Tyee poll and is not generalizable about "all public opinion polls" either. So it's a strange claim to make that "all public opinion polls are garbage." Well what makes them garbage? Poor, non-random, and unrepresentative sampling is one problem, amongst many that could be pointed to. So the Tyee poll is representative only of their readership, not generalizable to the BC public. But wasn't there another poll in there?

[Hydraboss]

The other poll was the one I posted which had an attached methodology. The one from Tyee is "garbage" because of the reasons stated earlier - it's like polling the three wolves and the sheep about what's for supper.

If anyone can post a link to any other "validated" poll (in favor of either side), I'd like to view it.

[cybercoma]

The poll I'm talking about is the Angus Reid poll that says the public overwhelmingly supports the teachers. You can see that poll summary here: http://www.angusreidglobal.com/polls/48983/bc-teachers-strike-parents-support-teachers-over-government-as-job-action-begins/

Detailed results with methods here: http://www.angusreidglobal.com/wp-content/uploads/2014/05/ARG-BCTF-Strike-May-2014.pdf

That poll was a random sample.

[Michael Hardner]

So it's a strange claim to make that "all public opinion polls are garbage." Well what makes them garbage?

Yes, I thought that was strange as well. I asked Pct to elaborate on that.

[Bryan]

That's a strange claim to make. What does "garbage" mean ? They're useful in predicting outcomes to elections, for one thing.

How useful were they in predicting the last BC election?

[overthere]

Well, the public is pissed off at Clark's government over there under funding of public education.

Oh, so now you agree it's about the money.

Careful you don't get whiplash with all those sudden turns.

[overthere]

How useful were they in predicting the last BC election?

and failed badly with the Alberta provincial election too.

Not surprisingly, all the pollsters who were so confident on TV and radio interviews before the election were 'unavailable for comment' after they had failed so miserably.

[GostHacked]

Polls are useless. This is why. They are all different.

[cybercoma]

How useful were they in predicting the last BC election?

What do election predictions have to do with an opinion poll of this nature? They're fundamentally different in the sense that election polling often results in a very large number of "undecided" respondents. Moreover, the election campaign matters. People opinions change as the election moves on. A poll conduct Monday through Thursday doesn't capture the controversy that happens on the Friday and the subsequent slip or boost in the polls.

What we're talking about here is an opinion poll asking people how they feel about a particular situation that has been in the news. There isn't going to be a large group of people who think it's taboo to discuss who they're going to vote for before an election. If you look at the Angus Reid poll that I posted, you can see their numbers and how many people they polled. You can unpack their methodology.

There is a criticizable issue with it, but I'll leave it up to you to find the problem and explain how it might have affected this poll. When you find it and are able to elaborate on it, I'll tell you why it might not be a problem at all. Then we can pretend we're having a real discussion, where we're both bringing something of substance to the table.

[Michael Hardner]

How useful were they in predicting the last BC election?

I have no idea. If you have a cite, please go ahead and post it.

The fact of the matter is that polls are a longstanding and proven tool, so to dismiss them - even on a single election - is flippant.

[Pct2017]

I have no idea. If you have a cite, please go ahead and post it.The fact of the matter is that polls are a longstanding and proven tool, so to dismiss them - even on a single election - is flippant.

Last year, all polling in BC had the common prediction of a landslide NDP victory. Not even close. And those polls stayed consistent right up to May 13. It made for some of the funniest tv that I have ever watched during the results. You had all of these talking heads discussing who would be in Adrian Dix's cabinet and who should replace Clark, but meanwhile scrolling below them were results that showed the

Liberals leading by substantial margins.

You have questioned why I stated that all opinion polls are now garbage. Well, the BC election and the Alberta election have proven that statement without a doubt. I can appreciate that if you do not live out here then you would not be as aware of this.

The bigger question is what happened to the effectiveness of polling. In the aftermath of last May, the general consensus is that the older demographic simply does not participate in opinion polls whereas younger ideologues still do. So your population to sample from is no longer representative.

Can I assume that you live in Ontario? If so, look at the correlation between the upcoming election results and the opinion polls. I would bet dollars to donuts that there will be very little relevancy.

As for the idea that polls done on specific issues like the one being discussed here are somehow more accurate, well for the reasons stated above that is incorrect.

[Michael Hardner]

Well, the BC election and the Alberta election have proven that statement without a doubt.

I actually do have a doubt. One bad result doesn't necessitate throwing out decades of proven utility.

[TimG]

What we're talking about here is an opinion poll asking people how they feel about a particular situation that has been in the news. There isn't going to be a large group of people who think it's taboo to discuss who they're going to vote for before an election. If you look at the Angus Reid poll that I posted, you can see their numbers and how many people they polled. You can unpack their methodology.

In most cases these kinds of polls are meaningless push polls because they don't provide the context that would allow someone to make an informed decision.

For example, if a poll reminded people that the government cannot meet the teachers demands without raising taxes or cutting healthcare spending then you would likely see support for the teachers plummet.

OTOH, if a poll presents the nonsensical interpretation pushed by the BCTF on SCC ruling on class sizes then the support for teachers would rise. If the poll then explained what the SCC really said and pointed out how the BCTF is completely misrepresenting the ruling you would see support for the teachers drop.

[Michael Hardner]

OTOH, if a poll presents ...

I think we're smart enough on MLW to look at the methodology together and decide - within a range of likelihood - how valid the polling is. The Tyree poll shouldn't have even been done, it's misleading IMO.

[cybercoma]

I think we're smart enough on MLW to look at the methodology together and decide - within a range of likelihood - how valid the polling is. The Tyree poll shouldn't have even been done, it's misleading IMO.

Exactly.

But nobody's addressing the Angus Reid poll that used a random sample, instead of a convenience sample of Tyree readers. The fact is the Angus Reid poll gives a snapshot that is most likely accurate enough to understand the general feelings of people in BC about the education issues in the province.

[TimG]

The fact is the Angus Reid poll gives a snapshot that is most likely accurate enough to understand the general feelings of people in BC about the education issues in the province.

The poll itself appears to be an unbiased survey of opinions, however, part of the issue is the government has not been explaining its position well. On top of that you have teachers misrepresenting the issues when talking to kids in classroom (I know this because I have to explain to my kids why what their teacher told them is deceptive nonsense).

[cybercoma]

So the public is being deceived, but you happen to see through all that? Is that about right?

[TimG]

So the public is being deceived, but you happen to see through all that? Is that about right?

I happen to pay attention to what SCC rulings actually say and when they get misrepresented for political purposes I am aware of the misrepresentation. In the case of the BCTF the court ruled that the government could not make it illegal for the the BCTF to ask for class sizes in the contract, however, the government is under no obligation to accept such terms. So the suggestion that the SCC requires the government to agree to class size limits is complete nonsense. That said, the recently suspended ruling was related to restoring the original contract prior to negotiation which is a silly move by the BCTF because the government can impose whatever contract it wants with back to work legislation so the BCTF cannot credibly argue that it could have kept those terms even if the government had not tried to make such terms illegal. Edited May 29, 2014 by TimG

[Pct2017]

Exactly.But nobody's addressing the Angus Reid poll that used a random sample, instead of a convenience sample of Tyree readers. The fact is the Angus Reid poll gives a snapshot that is most likely accurate enough to understand the general feelings of people in BC about the education issues in the province.

Maybe I am not explaining very well just what it was to what went wrong in the polling here in BC, and I suspect the same problem happened in the Alberta election polling which was similarly completely wrong. In order for a polling company such as Angus Reid to conduct a statistically accurate poll, they have to be able to get a representative sample of the entire population. That is Statistics 101. After the BC election, the polling companies hypothesized that what went wrong is that they were not able to access the older voters while the younger voters were more willing to participate in opinion polls. So, by natural extension, the polls moved from right to left. But when the actual election happened, not only did the older voters come out to vote, the younger voters tended to stay home.

This explanation is not a theory made up by yours truly. It was the best explanation provided by both the polling companies as well as the news media.

By extension, any polls done to elicit opinion regarding the teachers will end up with the same built in bias as discussed above.

BTW Michael, BC and Alberta are separate provinces, so that would make two massive failures in the polling industry.

But, as I have stated all along, who cares about any of this vis a vis the BCTF. Public opinion is not going to be a deciding factor, even if you could measure it accurately.

[Pct2017]

Buggy whips and Olivetti typewriters were also longstanding and proven tools.

[cybercoma]

Maybe I am not explaining very well just what it was to what went wrong in the polling here in BC, and I suspect the same problem happened in the Alberta election polling which was similarly completely wrong. In order for a polling company such as Angus Reid to conduct a statistically accurate poll, they have to be able to get a representative sample of the entire population. That is Statistics 101. After the BC election, the polling companies hypothesized that what went wrong is that they were not able to access the older voters while the younger voters were more willing to participate in opinion polls. So, by natural extension, the polls moved from right to left. But when the actual election happened, not only did the older voters come out to vote, the younger voters tended to stay home.

This explanation is not a theory made up by yours truly. It was the best explanation provided by both the polling companies as well as the news media.

By extension, any polls done to elicit opinion regarding the teachers will end up with the same built in bias as discussed above.

BTW Michael, BC and Alberta are separate provinces, so that would make two massive failures in the polling industry.

But, as I have stated all along, who cares about any of this vis a vis the BCTF. Public opinion is not going to be a deciding factor, even if you could measure it accurately.

Your close but not quite there.

The Angus Reid poll is from a random sample of online panelists. They're more likely to be younger.

In this case, which has absolutely nothing to do with elections forecasting (iow, you can drop that red herring now), that means it's more likely to be younger people surveyed. That means people of child-bearing age to whom education issues would be most relevant.

The Angus Reid poll is perfectly fine and generalizable about people's opinions of the education issue, particularly the people that matter most: those who have children or could potentially have children in school. So the skew towards younger people with internet polling is not as relevant here as it would be in a poll about an election.

So give up the "all polls are crap" nonsense because this poll clearly shows that people are on the side of the teachers, but nobody wants to address that clear fact. You keep repeating that everyone's against the teachers, yet the researchers who've actually gone out and asked people in BC found out that your claim is wrong. Most people do support the teachers. Even people who don't have children in school.

So you can stop making the claim that the teachers have no support from the public.

[Pct2017]

Your close but not quite there.The Angus Reid poll is from a random sample of online panelists. They're more likely to be younger.In this case, which has absolutely nothing to do with elections forecasting (iow, you can drop that red herring now), that means it's more likely to be younger people surveyed. That means people of child-bearing age to whom education issues would be most relevant.The Angus Reid poll is perfectly fine and generalizable about people's opinions of the education issue, particularly the people that matter most: those who have children or could potentially have children in school. So the skew towards younger people with internet polling is not as relevant here as it would be in a poll about an election.So give up the "all polls are crap" nonsense because this poll clearly shows that people are on the side of the teachers, but nobody wants to address that clear fact. You keep repeating that everyone's against the teachers, yet the researchers who've actually gone out and asked people in BC found out that your claim is wrong. Most people do support the teachers. Even people who don't have children in school.So you can stop making the claim that the teachers have no support from the public.

Actually, I believe that I have stated very, very clearly that it matters naught who is "winning" public support in this dispute, but maybe I used big words that were hard to understand.

As for your assertion that it is ok to sample only part of a population to get a statistically valid poll, all I can do is shake my head and wonder. Parents are not the only stakeholders, that is unless you support the notion that education should be run by user fees. If not, then all tax payers should be considered the population to sample from, and if part of that population chooses to not participate, then the sample is invalid and statistically rendered useless. It is the conundrum that all polling companies are wrestling with.

Point of curiosity, did you ever take any university level Statistics courses? And if you did, did you pass?

[cybercoma]

As for your assertion that it is ok to sample only part of a population to get a statistically valid poll

i have no interest in discussing things with someone who argues against fabrications. I never said this at all. And the fact that you resort to insults tells me all I need to know about your interest in an honest discussion. Edited May 29, 2014 by cybercoma