About me
Blog archive
About me

Netflix might tell us why the election polls were wrong



by sjhoward

This is the 2,302nd post. It was published at 12:21 on Tuesday, 26th May 2015.

This post was filed under:
» Election 2015
» News and Comment
» Politics
»
»
»
»

Netflix headquarters in Los Gatos, California, Thursday, July 10, 2014 for Alibaba.  (Paul Sakuma Photography) www.paulsakuma.com

It’s said that so-called ‘shy Tories’ swung the General Election: those people who are unwilling to admit to pollsters that they vote Tory for fear of a negative response, but who put their cross next to a Conservative candidate’s name in the privacy of the polling booth.

‘Shy Tories’ is basically a shorthand for a particular form of social desirability bias. When asked questions by pollsters, we may be unwilling to admit views that we think might upset pollsters. After all, most people like to please people.

Online polls, so the theory goes, should be less susceptible to social desirability bias because they remove the imposing pollster from the equation. Peter Kellner, President of YouGov suggests:

One of virtues of online research is, or should be, that it allows respondents to submit their views with complete anonymity, as there is no stranger watching over them or listening to their answer.

Or should be. An important caveat given that YouGov, online pollsters, were as wrong in their General Election predictions as telephone pollsters.

london-by-night-735085_1920

So perhaps social desirability bias exists online just as much as elsewhere. Perhaps. And there’s an intriguing piece of evidence from, of all places, Netflix, which described this week putting much greater emphasis on recommending shows based on what people actually watch, as opposed to how they rate individual shows. Todd Yellin, Netflix‘s VP of Product Innovation reports:

Most of our personalization right now is based on what they actually watch, and not what they say they like, because you can give five stars to An Inconvenient Truth because it’s changing the world, but you might watch Paul Blart: Mall Cop 2, three times in a few years … so what you actually want and what actually say that you want are very different.

Is rating something highly because it’s changing the world rather than because you actually enjoyed it a form of social desirability bias? I reckon it probably is. I reckon it’s an attempt to make ourselves feel more educated and ‘worthy’.

And if we fake our Netflix ratings, conning only ourselves, why wouldn’t we equally fake our responses to online polls?


This post was filed under: Election 2015, News and Comment, Politics, , , ,

James is wrong about giving 16- and 17-year-olds the vote



by sjhoward

This is the 2,301st post. It was published at 17:38 on Sunday, 3rd May 2015.

As a bit of an experiment, you can access an audio version of this post here.

The images in this post are all from Flickr, and are used under their Creative Commons licence. In order of appearance, they were uploaded by Eric Hossinger, AdamKR, The Fixed Factor, and James West.

This post was filed under:
» Politics
»

My friend James O’Malley argues on his blog that 16- and 17-year-olds should be given the vote as it will help to tackle the seemingly undue attention given to old people by political parties:

By increasing the potential pool of voters at the bottom, it affects the potential electoral mathematics that the parties have to do to maximise their votes. If there are more young people who can vote, it tips the scales back towards the young. Pensioners might be reliable voters, but if there were an extra 1.6 million (ish) young people on the electoral register might be a greater motive for refocusing policies and priorities.

I’m undecided on whether 16- and 17-year olds should be allowed to vote―I see good arguments on both sides―but I think that James’s specific argument is wrong for reasons mathematic and democratic.

5976378132_3506c0094b_o

First, the maths. For the purposes of these back-of-the-envelope calculations, I’m using ONS projections for England (as that’s what I have at hand), defining the “youth” vote as 29-and-under,1 and defining the “grey” vote as sixty-and-over.

Currently, the “youth” vote accounts for roughly 20% of eligible voters, versus roughly 29% for the “grey vote”. Giving the vote to 16- and 17-year olds would move these percentages to about 22% and 29%: that is to say, it wouldn’t make much difference. And the difference is lessened further by the fact that most people agree that the “youth” vote is less likely to turn out than the “grey” vote.

But, of course, the real imbalance in the “youth” versus “grey” votes isn’t in 2015. The population is ageing: the scale of the imbalance today is nothing to what the scale of the imbalance will be in the future.

If we fast-forward a couple of decades to the 2035 election, ONS projections suggest that the “youth” vote (as currently defined) would make up roughly 18% of the electorate, versus 37% of the electorate being “grey” voters. Giving 16- and 17- year-olds the vote rebalances this a titchy bit (to 20% and 36% respectively), but this difference is really so little as to be meaningless―the imbalance remains far greater than it is today.

SONY DSC

My second problem with James’s argument is democratic. He reckons that the makeup of the electorate needs to be changed to better ‘balance’ it in age terms, because generational disputes cause problems in our country. As an example:

The old, who own property want the value of their homes to continue to increase, whilst it would be better for the young people who Ed Miliband calls “generation rent” if property prices were to fall, so that buying a house can become even a remote possibility.

But, surely, to suggest that’s a problem is profoundly undemocratic! We have decided that the best way to run our nation is by the majority electing representatives who they think will best serve their interests. The majority of the population is ageing. We shouldn’t go around thinking of ways to “fix” the result to better reflect youth interests because the youth is in the minority.

If we stick with our current form of representational democracy, then, for the foreseeable future, our politics will continue to be determined by the “grey” vote as it is the “grey” vote which makes up the largest part of the electorate. The different electoral turnouts between the generations certainly exacerbate the problem, but they are not the source of it.

5100190241_2c63f290f6_o

Having said all of that, there is a problem here. Actually, it probably is unreasonable for the electorate to become so imbalanced: not generationally imbalanced, but gratuitously imbalanced between net financial contributors to the state and net financial users of state services. It’s hard to see how a state can function when politicians essentially only have to appeal to those who use the state’s services (especially the elderly), and have to appeal less to those who (by and large) pay for it (largely the working aged). It becomes perfectly logical for politicians to whack up tax rates or borrow with little regard for the future.

Of course, this probably won’t actually happen. It’s more likely that the “grey” vote will be effectively capped at a certain size as people work longer, as neither the state nor individuals can afford to pay for pensions which increasingly approach or exceed the length of an individual’s working life. And, of course, outrageous levels of tax and spend would provide a good incentive to improve low turnout in the younger section of the electorate, which would provide a degree of rebalance in and of itself.

DSC_7000

On the other hand… things could get worse more quickly. We’re seeing national and international narrative opinion increasingly extending the length of childhood. We’ve already seen in the UK a major shift in legislation pushing the end of childhood (in terms of, for example, school leaving and consumption of cigarettes) from 16 to 18. There is increasing scientific evidence that key elements of development, particularly emotional development, continue until the early 20s. UNESCO considers our period of “youth” to continue until 25. The African Youth Charter considers it to continue until 35. In this context, it’s not inconceivable that a future government might choose to increase the voting age, not decrease it.

To summarise: give 16- and 17-year-olds the vote if you want. But do it for good reasons, not because you want to “fix” the outcome of elections in a way that will matter little and matter for a short time. And go and read James’s post, too.


  1. Woe is me, having just exited my own definition of “young”. 


This post was filed under: Politics,

As a bit of an experiment, you can access an audio version of this post here.

The images in this post are all from Flickr, and are used under their Creative Commons licence. In order of appearance, they were uploaded by Eric Hossinger, AdamKR, The Fixed Factor, and James West.

News organisations are wrong about A&E waiting times



by sjhoward

This is the 2,294th post. It was published at 21:01 on Friday, 23rd January 2015.

This post was filed under:
» Health
» News and Comment
» Politics

Hospital surgery corridor

Answer me this: what happened to waiting times in A&Es in England last week? To help you answer, here are some tweets published by reputable news organisations today:

You would be forgiven for thinking that waiting times had reduced. You would be… possibly right, possibly wrong. The correct answer is that we don’t know. Performance against the weekly A&E waiting time targets—which is what all of the above are actually reporting—tells us nothing about the waiting time in A&E.

As an aside, before we get into this properly, I should clarify that “waiting time” doesn’t mean what most people think it means. The “waiting time” referred to in these statistics is the total time a patient spends in A&E, from the moment they walk in the door, to the moment they walk out again (whether that is to go home, to go to a ward, to go to the pub, or wherever). That’s not what we think of as “waiting” in common parlance: while you’re with the doctor, you are—in statistical terms—still “waiting”.

The NHS doesn’t report on waiting times, only on the proportion of patients seen in less than four hours. When the reporters wrongly say that A&E waiting times have improved, what they actually mean is that a greater proportion of people entering A&E are leaving again in less than four hours. This tells nothing about the amount of time people wait on average.

Imagine an A&E department that sees only five patients: A and B have minor injuries, and are seen and treated within 30 minutes. C and D need a more complex set of investigations, so end up being in the A&E department for 3 hours. E needs a very full assessment and ultimately admission; as a result, E ends up being in the department for a total of 5 hours before a bed can be found. The average time these patients spend in A&E is 2 hours and 24 minutes; 80% of them were discharged in 4 hours.

Now let’s say that someone puts a laser-focus on that 80% and says it’s unacceptable: whatever the cost, it must be brought down. So the department tells the nurse that used to do the “see and treat” job (which served patients A and B so well) that she must help with only the most complex patients, because they are breaching the target.

The same five people with the same five injuries now come into the revamped A&E. A and B have minor injuries, but now must wait alongside everyone else. They hang around for 3 hours. C and D need complex investigations, but these are slower to start because of people with minor injuries clogging up the queue. They are discharged after 4 hours. The new complex patient team deals with patient E slightly faster, getting her up to the ward with seconds to spare before the four-hour deadline.

100% of patients were seen within 4 hours. The hospital’s management is overjoyed! The BBC tweets that A&E waiting times have decreased: 100% of patients are seen within four hours instead of 80%. Politicians become a little self-congratulatory.

Yet… what has actually happened? The average waiting time has increased from 2 hours and 24 minutes to 3 hours and 36 minutes. 80% of patients are waiting longer than they did before.

And that is why—whatever the news tells you—we have no idea what happened to A&E waiting times last week. The average time could have doubled; it could have halved; it could have stayed precisely the same. We simply do not know.


This post was filed under: Health, News and Comment, Politics

The content of this site is copyright protected by a Creative Commons License, with some rights reserved. All trademarks, images and logos remain the property of their respective owners. The accuracy of information on this site is in no way guaranteed. Opinions expressed are solely those of the author. No responsibility can be accepted for any loss or damage caused by reliance on the information provided by this site. This site uses cookies - click here for more information.