• Welcome to DOSBODS

     

    DOSBODS is free of any advertising.

    Ads are annoying, and - increasingly - advertising companies limit free speech online. DOSBODS Forums are completely free to use. Please create a free account to be able to access all the features of the DOSBODS community. It only takes 20 seconds!

     

Sign in to follow this  
Frank Hovis

More or less - misuse of statistics

Recommended Posts

There is an excellent programme on Radio 4 - and yes, I know that these days that sounds like an oxymoron - called More or Less, on 4.30pm Fridays repeated Sundays 8pm.  The chap who runs and presents it, Tim Harford, isn't a BBC staffer but an economist and journalist who writes for the FT.  This explains its lack of political correctness and dumbing down.

It takes quoted statistics suggested by listeners from reports in the press and analyses them for source and accuracy.

A recent claim was that there are more statues of goats in the UK than of non-royal non-mythological women quoted on a feminist campaigning website.  This isn't remotely true and when the website was contacted for source they said that they heard it at a conference.

https://www.bbc.co.uk/programmes/b006qshd

Anyway in tribute to More or Less I thought this thread would be useful for examples of statistics that are either clealry wrong, ludicrous, or so mangled that they don't at all support the argument that they're being used to support.

 

Here's one from (and my sincere apologies for this) The Guardian.

I take it as read that the stories the usual Guardian drivel, and I'm not trying to discuss that, but what jumped out is this criminal misuse of survey data:

 

Quote

 

“More than half of those aged 13 to 21 have felt unsafe walking home alone, experienced harassment or know someone who has, and nearly half feel unsafe using public transport,” the survey states.

This is an astonishing statistic.

 

https://www.msn.com/en-gb/news/uknews/of-course-girls-feel-miserable-they-can’t-move-freely-in-the-world/ar-BBNzdcI?ocid=spartandhp

 

Right - now did anybody here, man or woman, during the ages of 13 and 21 not feel at least once "unsafe walking home alone" or "unsafe using public transport"?  I did.

And does anybody here not know somebody whne between teh ages of 13 and 21 who has "experienced harrassment". I did.

 

What is astonishing to me about that statistic is that it isn't 100% given quite how ludicrously wide the definition of "unsafe" has been drawn.

 

It is useless and you cannot conclude anything from it so throw it away rather than using it as if it were a meaningful statistic.

 

Share this post


Link to post
Share on other sites

 

Quote

“More than half of those aged 13 to 21 have felt unsafe walking home alone, experienced harassment or know someone who has

100% of readers of this thread now know someone who has ...

 

Share this post


Link to post
Share on other sites

Yes, I once felt  "unsafe walking home alone" . It was cold and dark, the pavement was covered in melted snow that had refrozen back into ice. It felt very unsafe, I could have easily suffered a serious injury.

Please let me know who I should report this non crime to.

Share this post


Link to post
Share on other sites
1 hour ago, Frank Hovis said:

A recent claim was that there are more statues of goats in the UK than of non-royal non-mythological women quoted on a feminist campaigning website.  This isn't remotely true and when the website was contacted for source they said that they heard it at a conference.

https://www.bbc.co.uk/programmes/b006qshd

 

 

 

Yep, that's on the list -

A repost of the how to spot a SJW list:

- Ignore difficult questions

- Make unfounded claims

- Claim any source of information that goes against their ideology is bias or fake news

- Will then use a well known bias source of information as if it is gospel

- Are bigoted

- Get shouty/agressive and make large numbers of posts (I have no idea how they find the time!)

- Use reversal, e.g. you expose their bigotry and then they claim you are the bigot. You prove them wrong but they claim the opposite.

- Will call people names and use derogatory terms (I don't have a problem with this but it shows the level they are at)

- Are aware of Godwins law so try to avoid it by using terms such as white supremist or fascist

- Are very intolerant and problably the closest to being fascist I have seen in recent times

- Treat their ideology more like a religion and it's almost like an atheist trying to argue with the Pope that God does not exist

- Make ad hominem attacks

- Present assumptions and/or personal views as fact

Share this post


Link to post
Share on other sites

The lie about goats is out there and has become the new truth, top two search results on duckduckgo:

https://www.globalcitizen.org/en/content/there-are-more-statues-of-goats-than-real-women-in/

https://www.reuters.com/article/us-britain-women-culture/outnumbered-by-men-and-goats-more-women-to-be-celebrated-in-uk-statues-idUSKBN1KK27Q

 

That's the issue with fake news that supports an agenda, once it is out there plenty of people are happy to accept it as true if it aligns with their views.

Share this post


Link to post
Share on other sites

must be the same feminists (stella creasy) who think there has never been a female face on a banknote.

 

Is it really worth arguing with feminists, though? There are lefty groups who have certain agendas, but IMO feminists just dont 'do' logic. If they feel a certain way, that is reality, regardless of what the actual reality is. 

Edited by PatronizingGit

Share this post


Link to post
Share on other sites
28 minutes ago, Frank Hovis said:

I"m thinking occasional thread; as a regular listener I can give loads of examples but threw in the goats as a particular silly one.

The local rag that boasts a readership of between 1 and 2 million.

Share this post


Link to post
Share on other sites

A classic misuse of statistics is the government trick in hiding the real unemployment figures.

In government speak JSA claimants = the actual number of people who are unemployed.

In an ideal world we would have government reports on the real number of unemployed people and their proposals to rectify the problem. xD. No chance of that happening!

 

 

Edited by Van Lady

Share this post


Link to post
Share on other sites
2 hours ago, Frank Hovis said:

A recent claim was that there are more statues of goats in the UK than of non-royal non-mythological women quoted on a feminist campaigning website.  This isn't remotely true and when the website was contacted for source they said that they heard it at a conference.

I am sure it has been said but it is worth noting that this could be a prediction as it may be likely that those in power in the UK in the future put up a statue to their favourite goat.

Share this post


Link to post
Share on other sites
2 hours ago, Frank Hovis said:

There is an excellent programme on Radio 4 - and yes, I know that these days that sounds like an oxymoron - called More or Less, on 4.30pm Fridays repeated Sundays 8pm.  The chap who runs and presents it, Tim Harford, isn't a BBC staffer but an economist and journalist who writes for the FT.  This explains its lack of political correctness and dumbing down.

I really like this show and that it doesn't have a BBC agenda; only one I make the effort to listen to on iplayer when I remember. I quite liked his book too, the undercover economist. The chapter on how the singaporean govt decided on their healthcare system shows how things could be done if only our politicians were capable of using evidence and thinking through consequences.

Share this post


Link to post
Share on other sites

Interesting example here where the Guardian comes out as less (Orwellian) biased in its use of statistics to plug an agenda than the So-Called BBC and SKY.

1612638491_Screenshot-2018-12-13GoogleNews.jpg.01596f9c9efb3ceb25e8de4d3ff2d5d2.jpg

With the So-Called BBC and SKY articles (especially the SKY one) you have to read past the headline and right to to the bottom to put the statistics into perspective.

Prevent scheme: Extreme right-wing referrals up by 36%

Total deaths and injuries for alleged right wing terrorists 1 + 12 versus (no figures quoted in any article but here they are) 4 + 49 (Westminster Bridge), 22 + 129 (Arianne Grande), 11 + 48 (London Bridge), 0 + 29 (Parsons Green) giving a grand total of 37 + 355 and an attrition rate of 3700%.

These kind of statistically twisted articles are written for pure propaganda purposes.

https://www.bbc.co.uk/news/uk-46556447

https://news.sky.com/story/number-of-people-referred-to-prevent-programme-for-far-right-extremism-soars-36-11579553

https://www.theguardian.com/uk-news/2018/dec/13/prevent-referrals-over-rightwing-extremism-rise-by-over-a-third

https://en.wikipedia.org/wiki/List_of_Islamist_terrorist_attacks

 

Share this post


Link to post
Share on other sites
51 minutes ago, Chewing Grass said:

Interesting example here where the Guardian comes out as less (Orwellian) biased in its use of statistics to plug an agenda than the So-Called BBC and SKY.

1612638491_Screenshot-2018-12-13GoogleNews.jpg.01596f9c9efb3ceb25e8de4d3ff2d5d2.jpg

With the So-Called BBC and SKY articles (especially the SKY one) you have to read past the headline and right to to the bottom to put the statistics into perspective.

Prevent scheme: Extreme right-wing referrals up by 36%

Total deaths and injuries for alleged right wing terrorists 1 + 12 versus (no figures quoted in any article but here they are) 4 + 49 (Westminster Bridge), 22 + 129 (Arianne Grande), 11 + 48 (London Bridge), 0 + 29 (Parsons Green) giving a grand total of 37 + 355 and an attrition rate of 3700%.

 These kind of statistically twisted articles are written for pure propaganda purposes.

https://www.bbc.co.uk/news/uk-46556447

https://news.sky.com/story/number-of-people-referred-to-prevent-programme-for-far-right-extremism-soars-36-11579553

https://www.theguardian.com/uk-news/2018/dec/13/prevent-referrals-over-rightwing-extremism-rise-by-over-a-third

https://en.wikipedia.org/wiki/List_of_Islamist_terrorist_attacks

 

Even then the stats are completely meaningless.  See, the Prevent scheme is entirely voluntary.  Now, the stats are for referrals (not people that actually go), but with things like this it isn't a simple suggestion that once made is made -- it'll be a conversation where a referral will be discussed -- it'll only go onto the stats if, at the end of that discussion, the individuals agrees to be referred.

So the stats could equally well be something like -- 'right wing extremists, after a bit of discussion and a nudge, often say they'd like to be referred.  Whereas other extremists are less likely to engage in the process right from the start and refuse to discuss referral'.  

Of course, I don't know if that is true.  But I also don't know if the article's interpretation (ie, more right wing, less Islamic, extremists) is true.  And no-one will actually tell me the situation, for some reason.  So I just end up knowing nothing*.

Well, other than people seem to continue getting killed by terrorists, sorry, mad people. 

[* Socrates would be proud]

Edited by dgul

Share this post


Link to post
Share on other sites

The stats sleight of hand I always watch for is 'average', which can be arithmentic mean, median or mode, with very different results.

Also, when comparing stats for, say, two countries, are they using the same criteria?

'Weighted' stats are always suspect, unless the methodology of the weighting is disclosed.

Choosing only datasets that support a particular argument is another old trick.

 

Share this post


Link to post
Share on other sites
Just now, Happy Renting said:

The stats sleight of hand I always watch for is 'average', which can be arithmentic mean, median or mode, with very different results.

Also, when comparing stats for, say, two countries, are they using the same criteria?

'Weighted' stats are always suspect, unless the methodology of the weighting is disclosed.

Choosing only datasets that support a particular argument is another old trick.

 

Fuck me, you are awake and not pissed up!:)

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.