Page 89 of 89 FirstFirst ... 3979878889
Results 2,641 to 2,646 of 2646

Thread: Brexit Begins

  1. #2641
    Senior Member RandBlade's Avatar
    Join Date
    Jan 2010
    Posts
    14,957
    Quote Originally Posted by Aimless View Post
    Not only are you wrong about the ONS data, but you also seem to have trouble keeping track of your own claims.

    You claimed that about a million people had moved to the UK since the referendum ie. in just over two years, and your later posts suggest that you truly believe about a million new people have settled in the UK since the referendum. You claim that the ONS data supports your position, but this is incorrect because the ONS does not track this information.
    Yes it does. They've recorded immigration at over 600k per annum. Which is 1.4 million since the referendum.
    What they do is conduct interviews at various ports of entry where they record (among other things) the number of people who intend to stay in or leave the UK for at least a year. Based on a small sample of such interviews they attempt to estimate the real number of people who intend to stay in or leave the UK for that length of time. That means that if I go to the UK with the intention of staying there for over a year, and state this in an interview, but then leave before the year is up, I'll still be counted as a long-term immigrant--there is no follow-up to ensure that I actually stayed for a year.
    That's nonsense. You're challenging the very basis of survey-based statistics in which case you can't use the survey stats for any other claim - including that net migration is decreasing.
    If I instead leave more than a year after the referendum, but before two years, both the ONS and you will still believe that I'm in the UK two years after the referendum. If I'm interviewed as I leave the UK after a year, I might be included in the emigration statistics, but the ONS data will not indicate that the same person who migrated TO the UK is now migrating FROM the UK. For these reasons, you cannot use the ONS immigration data to support your claim about a million new people moving to the UK. The claim might be true, but your data does not support it.
    Except that as I already explained from the maths that simply isn't possible. You'd have had to have been counted entering, leaving, re-entering, re-leaving, re-entering and re-leaving again for a year each time within the last 2 years. As would every other immigrant and emigrant. With zero people actually emigrating.

    Net immigration of 750k plus a further immigration of 650k can't possibly be less than 1 million unless the statistics are bullshit. In which case don't use them.
    In addition, you forget an important bit of context, which is that the situation for people migrating to the UK from EU27 was very unclear after the referendum, but people had been promised that nothing would change in practice. For all practical purposes, Brexit's impact on EU27 migration should have been expected to be small--at least on people moving due to accepting a long-term job, studies, research, family reasons etc--for months after the ref. as the UK continued to dissemble.
    Indeed and it has been small. In fact it has been miniscule. All we have seen is a reversion to mean effect after the spike caused by the accession of Romania and the removal of transition controls. In 2016 a record high net migration rate included over 100k Romanians moving here within the year, that was never likely to be maintained. The UK's net migration rate now and EU net migration rate now are still higher than in any year before the accession of Romania.
    An analogy to this situation is election polling. Taking the recent Swedish election as an example, all major pollsters underestimated the Sweden Democrats' prospects when attempting to forecast the 2014 election. Most people therefore concluded that polls were underestimating their support this time around as well. In reality, those polls who were the closest the last time around--and who predicted the largest gains for SD this election--greatly overestimated SD's share of the votes. If your reasoning re. the census and the ONS data were accurate and generally applicable, SD would've gotten much better results. But your reasoning is neither accurate nor generally applicable, because, just as SD's support is simply difficult to assess either way using current polling methods, EU migration is difficult to assess either way using the ONS's methods.
    No my reasoning wasn't the census though if you're using the census and polling nature to claim the statistics are useless then stop using them to imply there's a problem.

    My reasoning is that the maths shows [if you believe the statistics] that there's well over a million more immigrants here, while if you don't believe the statistics then your claims that they show anything is built on foundations of sand.
    Quote Originally Posted by Ominous Gamer View Post
    Being upset is understandable, but be upset at yourself for poor planning, not at the world by acting like a spoiled bitch during an interview.

  2. #2642
    SEÑOR Member Aimless's Avatar
    Join Date
    Jan 2010
    Posts
    13,908
    Quote Originally Posted by RandBlade View Post
    Yes it does. They've recorded immigration at over 600k per annum. Which is 1.4 million since the referendum.
    Again, no, they do not. They record an intention to stay in or leave the UK for 12+ months. Based on this, they have estimated immigration at 600k or so. Estimation is not what's at issue however--it's what the data actually says. If I state in my interview that I intend to stay in the UK for 2 years but leave the day after my interview, for the purposes of the ONS's data, I am a long-term immigrant. That is what the data says. Their estimates are based on people's statements about their intention to either leave or stay for a specified minimum length of time, and the estimates hold provided their assumption holds that people do as they intend. The analysis of the discrepancy between ONS data and the 2011 census showed that this assumption is flawed.

    That's nonsense. You're challenging the very basis of survey-based statistics in which case you can't use the survey stats for any other claim - including that net migration is decreasing.
    No, you are once again incorrect. First of all, this is a survey on intentions, and it must be treated as such, just like surveys on voting intentions. You can use this data to determine whether intentions have changed. You can also use the ONS's information to get an idea about trends (although even this is risky) even though you cannot use it to determine that 1 million different people have moved to the UK over the past 2 years or so. Again, you might be right, but your data does not show that.

    Except that as I already explained from the maths that simply isn't possible. You'd have had to have been counted entering, leaving, re-entering, re-leaving, re-entering and re-leaving again for a year each time within the last 2 years. As would every other immigrant and emigrant. With zero people actually emigrating.
    No, you would only have to have been counted as entering. If a million people are recorded by the ONS as intending to stay in the UK for over a year, and then they all leave the next day, how many remain in the UK? From the ONS's perspective, one million; in reality, zero. That is the core issue. Recording or failing to record the same person multiple times only modifies this core problem.

    Net immigration of 750k plus a further immigration of 650k can't possibly be less than 1 million unless the statistics are bullshit. In which case don't use them.
    The statistics are reliable enough to be used in a discussion about trends. However, if you object to even that, then don't make unfounded claims and try to justify them with data that does not say what you want it to say.

    Indeed and it has been small. In fact it has been miniscule. All we have seen is a reversion to mean effect after the spike caused by the accession of Romania and the removal of transition controls.
    Based on sudden and grave difficulties w/ recruiting personnel in several sectors not greatly affected by the accession of Romania, this claim is simply not very credible.

    No my reasoning wasn't the census though if you're using the census and polling nature to claim the statistics are useless then stop using them to imply there's a problem.
    You said:

    As your link shows the last census showed the ONS had UNDERestimated migration so much that 350k people extra were here according to the census.
    The implicit claim here is that, if anything, the ONS is underestimating net migration, so it must be even higher than we believe based on ONS data. This is false. The ONS can overestimate net migration for the same reasons it underestimated net migration: because it can't accurately track EU migration. The ONS statistics are useless if you want to arrive at an exact figure--as you yourself note, they were off by an average of net 35k per year over a decade. The ONS statistics, can, however, be used to detect trends, because their methods typically do not change substantially from one year to the next.

    My reasoning is that the maths shows [if you believe the statistics] that there's well over a million more immigrants here, while if you don't believe the statistics then your claims that they show anything is built on foundations of sand.
    My point is that you're applying the maths to data that you do not understand. If you apply maths that you do understand to data that you do not understand, you will not be more informed by the result.
    “Humanity's greatest advances are not in its discoveries, but in how those discoveries are applied to reduce inequity.”
    — Bill Gates

  3. #2643
    Senior Member RandBlade's Avatar
    Join Date
    Jan 2010
    Posts
    14,957
    Quote Originally Posted by Aimless View Post
    Again, no, they do not. They record an intention to stay in or leave the UK for 12+ months. Based on this, they have estimated immigration at 600k or so. Estimation is not what's at issue however--it's what the data actually says. If I state in my interview that I intend to stay in the UK for 2 years but leave the day after my interview, for the purposes of the ONS's data, I am a long-term immigrant. That is what the data says. Their estimates are based on people's statements about their intention to either leave or stay for a specified minimum length of time, and the estimates hold provided their assumption holds that people do as they intend. The analysis of the discrepancy between ONS data and the 2011 census showed that this assumption is flawed.
    Except that if you believe that the survey can be relied upon then it should have on average caught you leaving too. Unless you think somehow it is statistically catching people arriving but not catching them leaving, maybe there's a tunnel we aren't aware of smuggling people out of the country?
    No, you are once again incorrect. First of all, this is a survey on intentions, and it must be treated as such, just like surveys on voting intentions. You can use this data to determine whether intentions have changed. You can also use the ONS's information to get an idea about trends (although even this is risky) even though you cannot use it to determine that 1 million different people have moved to the UK over the past 2 years or so. Again, you might be right, but your data does not show that.
    Actually that's all you CAN use it to determine. If the survey is believable then roughly 600k people arrived and roughly 300k left and the remaining is net change. Now of course if there's any error then the more derivatives you make with the statistic the greater the error is multiplied. Garbage in, garbage out. The net figure has the error on arrivals compounded with the error on leaving (it is why opinion pollsters always say they have more confidence in a parties share than the lead) while comparing different years net figures has 4 errors involved.
    No, you would only have to have been counted as entering. If a million people are recorded by the ONS as intending to stay in the UK for over a year, and then they all leave the next day, how many remain in the UK? From the ONS's perspective, one million; in reality, zero. That is the core issue. Recording or failing to record the same person multiple times only modifies this core problem.
    Yes if they all morphed into flying pigs and left they wouldn't have been seen that way either. Unless you have evidence that a million people in reality left the next day and for some reason weren't caught leaving in the survey then your claim is just bullshit and undermines everything from this statistic.
    The statistics are reliable enough to be used in a discussion about trends. However, if you object to even that, then don't make unfounded claims and try to justify them with data that does not say what you want it to say.
    No they're not. Not if you deny what the statistics say in the first place. You're the one making unfounded claims, I'm quoting what the data actually says. The data says that the gross immigration has been 600k per year for the last few years, and 1 million is a very conservative figure for 2.25 years of over 600k.
    Based on sudden and grave difficulties w/ recruiting personnel in several sectors not greatly affected by the accession of Romania, this claim is simply not very credible.
    There aren't sudden and grave difficulties, but OTOH we are a nation currently experiencing full employment. Furthermore migrants are no more needed to fill job vacancies (unless there is a very specific skill shortage) than "they steal our jobs". Migrants overall increase both the supply and demand and thus are neutral for net employment.
    The implicit claim here is that, if anything, the ONS is underestimating net migration, so it must be even higher than we believe based on ONS data. This is false. The ONS can overestimate net migration for the same reasons it underestimated net migration: because it can't accurately track EU migration. The ONS statistics are useless if you want to arrive at an exact figure--as you yourself note, they were off by an average of net 35k per year over a decade. The ONS statistics, can, however, be used to detect trends, because their methods typically do not change substantially from one year to the next.
    Yes which is why I said a million not 1.4 million. Your rebuttal of my claim only stands if somehow there is magically a 500k error in the statistics but then you are quibbling over a change of tens of thousands while waving away hundreds of thousands.
    My point is that you're applying the maths to data that you do not understand. If you apply maths that you do understand to data that you do not understand, you will not be more informed by the result.
    I understand it, you don't. If you understood it you wouldn't jump up and down over a change within the standard deviation for the statistics while making claims that something that is miles outside the standard deviation could be true.
    Quote Originally Posted by Ominous Gamer View Post
    Being upset is understandable, but be upset at yourself for poor planning, not at the world by acting like a spoiled bitch during an interview.

  4. #2644
    SEÑOR Member Aimless's Avatar
    Join Date
    Jan 2010
    Posts
    13,908
    Quote Originally Posted by RandBlade View Post
    Except that if you believe that the survey can be relied upon then it should have on average caught you leaving too. Unless you think somehow it is statistically catching people arriving but not catching them leaving, maybe there's a tunnel we aren't aware of smuggling people out of the country?
    No. First of all, if that were truly the case, the ONS wouldn't have underestimated net migration by 350k over a decade. Secondly, just because someone agrees to an interview once doesn't mean they'll agree to be interviewed again the next day or whenever. In some cases they are less likely to participate multiple times, in other cases more. Thirdly, even if you're interviewed both as you enter and leave, you will be recorded as one individual entering and one individual leaving, with no indication that it is the same individual--that means that you can get a reasonable idea of net migration, but that you cannot determine whether or not your claim of one million different people moving to the UK is true.

    Actually that's all you CAN use it to determine. If the survey is believable then roughly 600k people arrived and roughly 300k left and the remaining is net change. Now of course if there's any error then the more derivatives you make with the statistic the greater the error is multiplied. Garbage in, garbage out. The net figure has the error on arrivals compounded with the error on leaving (it is why opinion pollsters always say they have more confidence in a parties share than the lead) while comparing different years net figures has 4 errors involved.
    Again, no. What you CAN estimate with this data is people's intentions to remain or leave long term, not whether they actually do so.

    Yes if they all morphed into flying pigs and left they wouldn't have been seen that way either. Unless you have evidence that a million people in reality left the next day and for some reason weren't caught leaving in the survey then your claim is just bullshit and undermines everything from this statistic.
    Again, you misunderstand what the ONS records, and it seems you're beginning to forget your own claims. If a million people move to the UK and another million people leave, your claim would still be correct, because a million people will have moved to the UK. However, the ONS data doesn't show that a million people actually move to the UK--only that a million people intended to move to the UK. If those who intended to move to the UK long term instead leave the next day, the ONS will record that people moved to the UK and people left the UK. Are they the same people or different people? We do not know.

    No they're not. Not if you deny what the statistics say in the first place. You're the one making unfounded claims, I'm quoting what the data actually says. The data says that the gross immigration has been 600k per year for the last few years, and 1 million is a very conservative figure for 2.25 years of over 600k.
    This is once again not what the data shows. The data shows that 600k/year intended to move to the UK for at least a year. Whether or not they remain in the UK is not shown by the data.

    There aren't sudden and grave difficulties
    This is plainly a lie. Businesses in many sectors, as well as employers in healthcare etc, have reported much greater difficulties recruiting personnel since shortly after the referendum. This is true for both qualified and less qualified work.

    but OTOH we are a nation currently experiencing full employment. Furthermore migrants are no more needed to fill job vacancies (unless there is a very specific skill shortage) than "they steal our jobs". Migrants overall increase both the supply and demand and thus are neutral for net employment.
    This is simply a misconception because the migrant population differs greatly from the native population. Migrants don't steal jobs, but they are necessary to compensate for deficiencies in the local market. Specifically, the UK has been heavily reliant on migrants to reduce staffing shortages in healthcare, hospitality, construction and seasonal agricultural labour. The healthcare sector, for example, is not net neutral--the migrant population is younger and healthier than the native population, so their healthcare workers represent a net positive.

    Yes which is why I said a million not 1.4 million. Your rebuttal of my claim only stands if somehow there is magically a 500k error in the statistics but then you are quibbling over a change of tens of thousands while waving away hundreds of thousands.
    There's no need to involve magic. You're correct in principle but incorrect about the magnitude of error required for you to be wrong. According to ONS, immigration was around 572k in the year (ending June) after the referendum, and around 610k in the following year (ending March so covering some of the previous data). If we're generous we can still assume at least 1.27 million immigrants according to ONS. For you to be wrong, the ONS would have to have overestimated all long-term immigration--Brits, EU27, non-EU, students, everything--by around 270k or just over 27%. That sounds absurd but the analysis of the 2011 census showed that the ONS had underestimated net immigration by 20-26% for several years following the 2004 enlargement of the EU and a year after the 2007 enlargement, lending credence to the notion that the ONS has difficulties keeping up with sudden and dramatic changes.

    Moreover, a recent ONS analysis showed that their methods may have made it difficult for them to assess non-EU immigration as well, as indicated by HO data showing that 60% of non-EU immigrants with 12-month visas end up leaving before 9 months or so.

    I understand it, you don't. If you understood it you wouldn't jump up and down over a change within the standard deviation for the statistics while making claims that something that is miles outside the standard deviation could be true.
    This is in fact a succinct summary of all your absurd and unrealistic positions re. the impact of Brexit on various aspects of the UK's economy, and your callous dismissal of opinions and analyses from industry, from trade experts and from academics. Thank you, I could not have put it better myself
    “Humanity's greatest advances are not in its discoveries, but in how those discoveries are applied to reduce inequity.”
    — Bill Gates

  5. #2645
    Senior Member RandBlade's Avatar
    Join Date
    Jan 2010
    Posts
    14,957
    In case you still don't understand how confidence intervals work, the statistics are available to download in an Excel spreadsheet.

    XLS File: https://www.ons.gov.uk/visualisations/dvc527/data1.xls
    Full Article containing link to file: https://www.ons.gov.uk/peoplepopulat...ort/august2018

    For the most recent statistics the following are the most recent statistics at a 95% Confidence Interval:
    Immigration: 614k +/- 38k.
    Emigration: 344k +/- 26k.
    Net Migration: 271k +/- 46k.

    As I said the error range on net migration is considerably bigger than the error range on immigration.

    That means immigration per annum is in a range of 576k to 652k.
    Net migration per annum is in a range of 225k to 317k.

    Even if we take the absolute minimum immigration figure from their confidence interval (576k) and multiply that by 2.25 years it massive exceeds 1 million still.

    On the other hand if we take into account confidence intervals it dismisses your hysteria about falling net migration. Net migration potentially is 317k. The absolute peak of net migration was in 2016 (while we were having the post-transition controls surge of Romanians entering the country) and it peaked at 336k +/- 40k.

    Thus potentially in 2016 net migration was only 296k and currently it is potentially 317k.

    We don't just still have incredibly high net migration well above recent norms, we are still within the confidence interval range of having the highest net migration rate EVER. That is what the statistics do tell us. We are well above the confidence range for having more than one million immigrants. In order for there to be under one million immigrants there would have to be something catastrophically wrong with the statistics that takes us well, well outside of the 95% Confidence Interval.
    Quote Originally Posted by Ominous Gamer View Post
    Being upset is understandable, but be upset at yourself for poor planning, not at the world by acting like a spoiled bitch during an interview.

  6. #2646
    Senior Member RandBlade's Avatar
    Join Date
    Jan 2010
    Posts
    14,957
    Incidentally the ONS makes my own point on their report. You're claiming you can discern a fall in net migration and banging on about it, they're explicitly saying that's not the case.
    1. We do not recommend users make comparisons year on year and instead look at the broader evidence and longer time series, which allows a better assessment of trends.
    2. None of the changes are statistically significant at the 5% level.
    The broader evidence and longer time series is that migration is in recent years (and still is) running at the highest levels ever since records began. There has been no statistically significant change since the referendum.
    Quote Originally Posted by Ominous Gamer View Post
    Being upset is understandable, but be upset at yourself for poor planning, not at the world by acting like a spoiled bitch during an interview.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •