Why not save the kitten and write that, "for any p outside the interval, the probability of observing this sample, conditional on p being the true probability is less than 5 per cent". But God would probably kill you instead
An interesting article which, in my view, contains explanations that it would be useful for many to see.
Politicians, when questioned about polls, often answer that they don't pay great attention to them for "the only poll that matters is that on election day." Well, I don't believe that they don't pay attention to polls prior to election day because many actually refer to them whilst campaigning. Nor is it correct that the poll on election day is the only one that matters for the reality is that those prior to it have already had an effect, no matter whether they are reliable indicators of valid opinions or intentions of voters. - So, those pre election day polls do matter. Indeed, when the influence of the wealthy and other influencers, for instance media and celebrities, is taken into account, and that they can skew poll results, there is no doubt, (at least in my mind), that polls are a concern and can certainly influence election results.
Given also that the election is, in effect, a poll - yet another problem exists and that is the process used in the election. Issues such as whether voting is mandatory; whether all candidates are treated equally and whether all voters are treated equally, need also to be considered. The reality is that, generally speaking, they are not.
So, particularly in a modern world where every Tom, Dick, Mabel, & Sue can add their 10c worth to social information and debate, and where major media companies are owned by a handful of, inevitably, overly wealthy individuals with corresponding views on government, what is for sure is that the 'average' person (whatever that is), really has little influence on who is elected or how the nation is governed, managed, conditioned, legislated or anything else.
In other words, put simply, the whole electoral process is so fraught with opportunities for error and skewing that it being seen as an indicator of democracy is, at best, extremely debatable. This is particularly true where there is basically only one of two parties that has any real chance of holding government. In almost every cases, there is likely to be a result where probably at least 40% but often far greater a number of voters will have voted against whoever wins.
Polling is 80% art and 20% science some pollsters only have the art in certain geographical location (e.g. Ann Zelzer in IA) and don't have any way to replicate that "knob twiddling" outside of a single location. The difficulties in finding pollsters that are good is that the business of political polling is in large part a way to funnel campaign money to the right people, and serving the public true information is less important.
Then you have situations where someone who isn't even a pollster takes a bunch of other pollsters outputs as a black box and puts it into his own black box and pretends that this is useful somehow at the level that it can calculate the probability of winning the election at exactly 54% (eyeroll).
As a crosscheck of the current polls, based on the voting demographics of recent election cycles, Harris needs to be up several points nationally to be ahead in the battleground states. Either the national polls are wrong by several points, or she is well behind in the EC. I guess we'll find out soon enough.
How can a poll be anything other than disinformation? It's an attempt to influence an election based on highly problematic data, which is sanitized and made "legitimate" by numeric manipulation. There's a good reason why they speak of Democratic and Republican pollsters.
Makes sense, but do you think it'd be more defensible that changes in polling numbers, if using the same sampling strategy and the same post-corrections, contain some useful signal?
And to state the obvious, assuming the poll responses are unbiased and it's a representative sample, a forecast where the Expected Prob (X beating Y) = 80% reflects a strong likelihood but not certainty. If Y wins, this outcome does not necessarily discredit the poll, as events with a 20% probability sometimes occur. However, since elections are one-time, non-repeatable events, the poll is a snapshot in time, and its accuracy depends on factors such as sampling error, late shifts in opinion, or methodological biases. While probabilistic models account for the occasional occurrence of lower-probability outcomes, the unique nature of each election means such an outcome provides only limited insight into the overall reliability of the polling methods.
There is no punishment in getting election predictions wrong. There is no punishment when Goldman predicted we will have a 100% chance of falling into a severe recession in the next 6 months (this was made last year and we have not had a severe recession yet i don’t think or nowhere near 2008 bad). Nate Silver has Trump - Harris at 60%-40%, 538 has Trump - Harris at 40%-60%, the betting market has Harris at a slightly higher percentage than Trump. And for all we know, what do we learn from it? Tbh, I’m not sure lol. Just let the experts do their thing and we do our part by voting. I do find polling results entertaining to read about though.
Arguably the trust is established through incentives: pollsters and media companies will get skewered online if their polls turn out to be inaccurate. But this incentive exists mainly for exit polls, and to a lesser degree for polls immediately preceding an election. There is no incentive for truth-telling in September, because there is no ground truth in September. There is only an incentive to obtain "publishable" numbers.
Why not save the kitten and write that, "for any p outside the interval, the probability of observing this sample, conditional on p being the true probability is less than 5 per cent". But God would probably kill you instead
An interesting article which, in my view, contains explanations that it would be useful for many to see.
Politicians, when questioned about polls, often answer that they don't pay great attention to them for "the only poll that matters is that on election day." Well, I don't believe that they don't pay attention to polls prior to election day because many actually refer to them whilst campaigning. Nor is it correct that the poll on election day is the only one that matters for the reality is that those prior to it have already had an effect, no matter whether they are reliable indicators of valid opinions or intentions of voters. - So, those pre election day polls do matter. Indeed, when the influence of the wealthy and other influencers, for instance media and celebrities, is taken into account, and that they can skew poll results, there is no doubt, (at least in my mind), that polls are a concern and can certainly influence election results.
Given also that the election is, in effect, a poll - yet another problem exists and that is the process used in the election. Issues such as whether voting is mandatory; whether all candidates are treated equally and whether all voters are treated equally, need also to be considered. The reality is that, generally speaking, they are not.
So, particularly in a modern world where every Tom, Dick, Mabel, & Sue can add their 10c worth to social information and debate, and where major media companies are owned by a handful of, inevitably, overly wealthy individuals with corresponding views on government, what is for sure is that the 'average' person (whatever that is), really has little influence on who is elected or how the nation is governed, managed, conditioned, legislated or anything else.
In other words, put simply, the whole electoral process is so fraught with opportunities for error and skewing that it being seen as an indicator of democracy is, at best, extremely debatable. This is particularly true where there is basically only one of two parties that has any real chance of holding government. In almost every cases, there is likely to be a result where probably at least 40% but often far greater a number of voters will have voted against whoever wins.
Representative democracy? I think not.
Polling is 80% art and 20% science some pollsters only have the art in certain geographical location (e.g. Ann Zelzer in IA) and don't have any way to replicate that "knob twiddling" outside of a single location. The difficulties in finding pollsters that are good is that the business of political polling is in large part a way to funnel campaign money to the right people, and serving the public true information is less important.
Then you have situations where someone who isn't even a pollster takes a bunch of other pollsters outputs as a black box and puts it into his own black box and pretends that this is useful somehow at the level that it can calculate the probability of winning the election at exactly 54% (eyeroll).
As a crosscheck of the current polls, based on the voting demographics of recent election cycles, Harris needs to be up several points nationally to be ahead in the battleground states. Either the national polls are wrong by several points, or she is well behind in the EC. I guess we'll find out soon enough.
Ann Zelzer in IA not looking as art anymore either :/
How can a poll be anything other than disinformation? It's an attempt to influence an election based on highly problematic data, which is sanitized and made "legitimate" by numeric manipulation. There's a good reason why they speak of Democratic and Republican pollsters.
We need an updated confidence interval blog in this new era of negative probabilities
Makes sense, but do you think it'd be more defensible that changes in polling numbers, if using the same sampling strategy and the same post-corrections, contain some useful signal?
And to state the obvious, assuming the poll responses are unbiased and it's a representative sample, a forecast where the Expected Prob (X beating Y) = 80% reflects a strong likelihood but not certainty. If Y wins, this outcome does not necessarily discredit the poll, as events with a 20% probability sometimes occur. However, since elections are one-time, non-repeatable events, the poll is a snapshot in time, and its accuracy depends on factors such as sampling error, late shifts in opinion, or methodological biases. While probabilistic models account for the occasional occurrence of lower-probability outcomes, the unique nature of each election means such an outcome provides only limited insight into the overall reliability of the polling methods.
There is no punishment in getting election predictions wrong. There is no punishment when Goldman predicted we will have a 100% chance of falling into a severe recession in the next 6 months (this was made last year and we have not had a severe recession yet i don’t think or nowhere near 2008 bad). Nate Silver has Trump - Harris at 60%-40%, 538 has Trump - Harris at 40%-60%, the betting market has Harris at a slightly higher percentage than Trump. And for all we know, what do we learn from it? Tbh, I’m not sure lol. Just let the experts do their thing and we do our part by voting. I do find polling results entertaining to read about though.
Arguably the trust is established through incentives: pollsters and media companies will get skewered online if their polls turn out to be inaccurate. But this incentive exists mainly for exit polls, and to a lesser degree for polls immediately preceding an election. There is no incentive for truth-telling in September, because there is no ground truth in September. There is only an incentive to obtain "publishable" numbers.