Hey there! Sign in to join this conversationNew here? Join for free
    Offline

    3
    ReputationRep:
    (Original post by mrITguy)
    What is the most accurate way of conducting polling? and yes I bet I could as I would be independent and free of political agenda's.

    Furthermore, I would randomly target area's of people from a list and do more than one mini poll's at the same time before analysing the results. I would then take out all anomaly's and proceed to look at trends.
    Well, all methods have inherant inaccuracies but the point is that they are all weighted. The results aren't just published as is (although they do publish raw data). An example of inaccuracies are that old people are under-represented in online polls and young people and those who removed themselves from the directories on phone polls. Difficulties only arise though when you have a ridiculously small representation from a articular group (like if you had less than 30 people over 65) and providing they do get large enough samples for all groups, you can accurately weight results to provide a picture of what the country thinks.

    As for what you'd do that you think is different, this is how phone polls are conducted, randomly and if you've ever had a phone call from a pollster, you'd find that generall speaking, they ask you multiple questions for multiple polls. They remove anomalies too but statistical analysis has to be done on whether the result is acually anomalous. Polls are much more rigourous than people think, they do show actual results.
    Offline

    9
    ReputationRep:
    (Original post by tomlam)
    Well, all methods have inherant inaccuracies but the point is that they are all weighted. The results aren't just published as is (although they do publish raw data). An example of inaccuracies are that old people are under-represented in online polls and young people and those who removed themselves from the directories on phone polls. Difficulties only arise though when you have a ridiculously small representation from a articular group (like if you had less than 30 people over 65) and providing they do get large enough samples for all groups, you can accurately weight results to provide a picture of what the country thinks.

    As for what you'd do that you think is different, this is how phone polls are conducted, randomly and if you've ever had a phone call from a pollster, you'd find that generall speaking, they ask you multiple questions for multiple polls. They remove anomalies too but statistical analysis has to be done on whether the result is acually anomalous. Polls are much more rigourous than people think, they do show actual results.
    But all those issues that you have just stated more often than not, happen. Also as I said before most of these polls are not independent and have a political agenda.
    Offline

    3
    ReputationRep:
    (Original post by mrITguy)
    But all those issues that you have just stated more often than not, happen. Also as I said before most of these polls are not independent and have a political agenda.
    Perhaps you could tell me how with real evidence then? I have''t seen your degree in statistical analysis.
    Offline

    9
    ReputationRep:
    (Original post by tomlam)
    Perhaps you could tell me how with real evidence then? I have''t seen your degree in statistical analysis.
    Wow.....are you really arguing with me about this...............

    Go and look into some facts about how these polling's organisations are funded and if you cannot see how biased or margin for error affects the results then why should I bother to explain it to you?

    Fact: statistics can be manipulated so that it suggests a certain view.

    Also I don't need a degree in statistics as I know the basics, I did a GCSE in statistics along time ago as well as GCSE Maths.
    Offline

    2
    ReputationRep:
    Polls are generally fairly accurate to within 2%.
    The sample size of 2000 is good enough.
    Offline

    1
    ReputationRep:
    Both are utter *****
    Offline

    3
    ReputationRep:
    (Original post by mrITguy)
    Wow.....are you really arguing with me about this...............
    Go and look into some facts about how these polling's organisations are funded and if you cannot see how biased or margin for error affects the results then why should I bother to explain it to you?
    Fact: statistics can be manipulated so that it suggests a certain view.
    Also I don't need a degree in statistics as I know the basics, I did a GCSE in statistics along time ago as well as GCSE Maths.
    Whilst I admit that headlines can be twisted to make the statistics seem to point a certain way, the polls themselves remain unbiased. For god's sake, they publish the raw data. If you have an issue with a poll in particular, look at the raw data to see if it's actually biased.

    Look at all the polling organisations in the last few years, YouGov, Ipsos Mori, Ashcroft, ICM etc. Most can be accused of spin but that doesn't mean that the data that they produce is incorrect.

    Look at this article from YouGov for example,
    https://yougov.co.uk/news/2016/06/06...tting-through/
    It accuses the stay campaign of inciting fear in the title, suggesting that that's the way that the article was going but you have to realise that the statistics in the article are unbiased and they published the raw data so you with your oh-so-important GCSE Maths and stats can analyse. But no, you'll just call bias without even evaluating whether the statistics are correct or not. Strikes me as a tin-foil-hat sort of thing to say.

    Look at this article from the Independant:
    http://www.independent.co.uk/news/uk...-a7075131.html
    This, you could argue is a biased article with biased stats but again, whether or not the article is biased is irrelevant, the raw data is published so statisticians such as yourself with such prestigious qualifications can criticise the data collection, the weighting etc.

    Again, for your wealth of knowledge of statistics, you haven't actually re-evaluated the results of any poll and told me exactly how the results of these polls are biased. They're not biased. The results reported are the genuine results of the polls whether they've been spun or not is irrelevant, they are still statistically accurate results.
    Offline

    9
    ReputationRep:
    (Original post by tomlam)
    Whilst I admit that headlines can be twisted to make the statistics seem to point a certain way, the polls themselves remain unbiased. For god's sake, they publish the raw data. If you have an issue with a poll in particular, look at the raw data to see if it's actually biased.

    Look at all the polling organisations in the last few years, YouGov, Ipsos Mori, Ashcroft, ICM etc. Most can be accused of spin but that doesn't mean that the data that they produce is incorrect.
    Oh dear Oh dear,

    You say these organisations can be accused of spin, yet you say there statistics must be reliable? lol

    (Original post by tomlam)
    Look at this article from YouGov for example,
    https://yougov.co.uk/news/2016/06/06...tting-through/
    It accuses the stay campaign of inciting fear in the title, suggesting that that's the way that the article was going but you have to realise that the statistics in the article are unbiased and they published the raw data so you with your oh-so-important GCSE Maths and stats can analyse. But no, you'll just call bias without even evaluating whether the statistics are correct or not. Strikes me as a tin-foil-hat sort of thing to say.

    Look at this article from the Independant:http://www.independent.co.uk/news/uk...-a7075131.htmlThis, you could argue is a biased article with biased stats but again, whether or not the article is biased is irrelevant, the raw data is published so statisticians such as yourself with such prestigious qualifications can criticise the data collection, the weighting etc.Again, for your wealth of knowledge of statistics, you haven't actually re-evaluated the results of any poll and told me exactly how the results of these polls are biased. They're not biased. The results reported are the genuine results of the polls whether they've been spun or not is irrelevant, they are still statistically accurate results.
    I don't need sarcasm thanks and ye YouGov who (like most media) is trying to present the case to "stay in the EU". As its "the government policy".

    Its a businesses at the end of the day, they do this to make money. Therefore whoever side profits the most is who the media takes.

    Yes the Raw data is a little better if you poll organisation has conducted it reliably, but if these polls were soooo accurate then it would give us a clear result/trend etc. Which it does not, and instead if you are like me I use trusted sources that are truly independent (free from political funding).

    Example? this site http://www.migrationwatchuk.org/briefing-paper/250

    So you are saying whether the owners of these organisations are spinning the results during and after the results are collected is irrelevant and doesn't affect the overall result? lol.
 
 
 
Reply
Submit reply
TSR Support Team

We have a brilliant team of more than 60 Support Team members looking after discussions on The Student Room, helping to make it a fun, safe and useful place to hang out.

Updated: June 12, 2016
  • See more of what you like on The Student Room

    You can personalise what you see on TSR. Tell us a little about yourself to get started.

  • Poll
    Brussels sprouts
    Useful resources

    Groups associated with this forum:

    View associated groups
  • See more of what you like on The Student Room

    You can personalise what you see on TSR. Tell us a little about yourself to get started.

  • The Student Room, Get Revising and Marked by Teachers are trading names of The Student Room Group Ltd.

    Register Number: 04666380 (England and Wales), VAT No. 806 8067 22 Registered Office: International House, Queens Road, Brighton, BN1 3XE

    Quick reply
    Reputation gems: You get these gems as you gain rep from other members for making good contributions and giving helpful advice.