Sky poll demonstrates support for leaving the single market

[sasguru]

Sas,

you're wasting your time with this conversation
.

You're probably right, but hope springs eternal.

[GB9]

Sas,

you're wasting your time with this conversation

He thinks Sky "Full Package" Subscribers who react to an unsolicited SMS are a representative sample of the electorate to use for drawing sentiment conclusions of the whole population of the country regarding Brexit

You don't think that Sky "Full Package" Subscribers who react to an unsolicited SMS are a representative sample of the electorate to use for drawing sentiment conclusions of the whole population of the country regarding Brexit

That's it.

Milan.

[vetran]

Sas,

you're wasting your time with this conversation

He thinks Sky "Full Package" Subscribers who react to an unsolicited SMS are a representative sample of the electorate to use for drawing sentiment conclusions of the whole population of the country regarding Brexit

You don't think that Sky "Full Package" Subscribers who react to an unsolicited SMS are a representative sample of the electorate to use for drawing sentiment conclusions of the whole population of the country regarding Brexit

That's it.

Milan.

Why do you think the respondents of the Poll are 'Full Package' subscribers? No mention of that in the story or the figures there is however: Region,Age & social grade.


http://interactive.news.sky.com/SMSXLI_EU_170117_FP.pdf

Do you think people who agree to be interviewed by MORI are a representative sample?

[milanbenes]

"Sky "Full Package" Subscribers who react to an unsolicited SMS are a representative sample of the electorate to use for drawing sentiment conclusions of the whole population of the country regarding Brexit"

makes me laugh too :-)

Milan.

[milanbenes]

Why do you think the respondents of the Poll are 'Full Package' subscribers? No mention of that in the story or the figures there is however: Region,Age & social grade.


http://interactive.news.sky.com/SMSXLI_EU_170117_FP.pdf

Do you think people who agree to be interviewed by MORI are a representative sample?

Hi Vetran,

I added the "Full Package" bit to make the story more interesting.

Worked a treat don't you think

Milan.

[NotAllThere]

EU supporting Sky News poll shows vast majority support leaving the single market.[/URL]

You clearly don't read Sky News as it is pro-EU. The article (which you also clearly haven't read) describes the 10% majority in its poll as 'narrow'.

Looks like you didn't read it either. Those in favour are 51%. Those against are 39%, those neither or undecided are 10%. The majority is narrow (as you managed to get right second time around), and it is 1%.

Do you wonder why you are frequently being accused of being thicker than mince, when you can't even do basic arithmetic, never mind statistics.

[vetran]

That wouldn't be you.

Don't express your stupidity to all and sundry.

The representativeness of a sample is not related to its size.

no as per link.

However as I pointed out all populations that pollsters have access to have bias.

The trick as per the link is to pick a representative sample from the population you have. If you have a bigger population then it is easier to pick a representative sample. Something like Sky TV is neither class based or Education limited so apart from innate prejudice why would you believe it doesn't hold a decent sample?

seems they know more about their users than most others.

https://www.skyadsmart.co.uk/audience-selection/

Sky AdSmart has clocked up nearly 500 clients, 69% new to TV or Sky Media – Videonet

[vetran]

Hi Vetran,

I added the "Full Package" bit to make the story more interesting.

Worked a treat don't you think

Milan.

The Wail are hiring ! It could be you

[GB9]

sorry started writing an hour ago so some of it has been updated in this thread.





I think he is saying a survey of sky customers is not representative of the UK just like a survey of Guardian readers (178,758) or Daily Mail (1,657,867) or maybe those willing to be polled by Mori/Youguv (<600,000 I suspect) etc.

However as Sky have 12 Million customers to choose from and has details of their class values I suspect that it is less of a concern than he believes.


As always precision is a key part of the AssGuru experience his link of course gets you straight to the answer




So moving swiftly on lets try listening to someone who knows what they are talking about.

https://www.ipsos-mori.com/newsevent...Questions.aspx

Thanks for this.

I was hoping he would state it himself but he obviously realised what a dick it would make him look which is why he hasn't done it.

As he won't call it out explicitly, SAS's assumption is that all 12m plus Sky subscribers are Leave voters and therefore the sample can't be correct.

As you rightly point out, his link was to a wiki search page. He didn't even bother linking to the sample page.

Sky's pro EU stance is also well known and complained about.

Regardless of that, the survey sample is statistically relevant to a 3% margin of error.

Sky balances the sample using several demographic variables. A link to the full set of tables is included at the end of the article.

The following statement is also at the end of the article:

"Sky Data interviewed a nationally representative sample of 1,068 Sky customers by SMS on 17 January 2017. Data are weighted to the profile of the population."

That last sentence tells you that the fact that they are Sky customers is irrelevant.

So, SAS has proven he knows nothing about statistics. He couldn't state what was statistically incorrect you because there wasn't anything. His whole argument comes down to "it's Sky so it must be wrong".

Thanks SAS.

[sasguru]

Thanks for this.

I was hoping he would state it himself but he obviously realised what a dick it would make him look which is why he hasn't done it.

As he won't call it out explicitly, SAS's assumption is that all 12m plus Sky subscribers are Leave voters and therefore the sample can't be correct.

As you rightly point out, his link was to a wiki search page. He didn't even bother linking to the sample page.

Sky's pro EU stance is also well known and complained about.

Regardless of that, the survey sample is statistically relevant to a 3% margin of error.

Sky balances the sample using several demographic variables. A link to the full set of tables is included at the end of the article.

The following statement is also at the end of the article:

"Sky Data interviewed a nationally representative sample of 1,068 Sky customers by SMS on 17 January 2017. Data are weighted to the profile of the population."

That last sentence tells you that the fact that they are Sky customers is irrelevant.

So, SAS has proven he knows nothing about statistics. He couldn't state what was statistically incorrect you because there wasn't anything. His whole argument comes down to "it's Sky so it must be wrong".

Thanks SAS.

Even by your standards that is a pile of utter wan k
I've highlighted the more egregious errors.