Sunday, 18 November 2012

Reaction to first post and a common prisoner voting rights mistake

The reaction to my first post has been great. Thanks to everyone who read and shared, especially the community over at www.reddit.com/r/unitedkingdom. I hope it will only get better from here and would like to see more involvement in the comments and on Twitter. As a reminder, my Twitter is @mattjreason and you can suggest any stories that look fishy using the tag #datadredge.

Voting rights for prisoners

I was watching ITV news just now and they made the common mistake of utterly misreporting what the European Court of Human Rights ruling on prisoners voting rights actually means. This is also something that the politicians constantly get wrong.

Basically, the ruling is only against a blanket ban and does not actually prevent our government from banning some prisoners from voting. This is particularly relevant as their latest proposal to allow some short term prisoners and those awaiting trial to vote is most probably acceptable to the ECHR. So the attempts by the government to make it look like a middle finger to Europe is just pathetic spin.

This great blog from the Durham Pro Bono Society does a good job at explaining the ruling along with some well thought out opinion. This BBC article also covers how the UK can decide which prisoners can vote.

Thanks again for the reaction so far and I hope you keep coming back, there will be a new debunking by Friday.

Please share if you like and look at my previous article if you have not yet. You can tweet me 
@mattjreason with requests/abuse/praise etcetera (use the tag #datadredge).







Friday, 16 November 2012

A headline with sex in it, how depressing


The Daily Mail ran this headline on their glorious website the other day:

“Whatever happened to the joy of sex? One in three women feel depressed after making love”

So let’s just look at that headline. Well if you haven’t guessed it already, it is utter shit. For starters, it even misrepresents the survey they based it on.

The story explains: “More than 200 young women were quizzed on life between the sheets and 33 per cent reported feeling depressed after sex at some point in their lives.” The phrase ‘at some point’ is key. The headline clearly claims that a third of women are depressed after sex, not just at least once.  So if we rewrite the headline:

Not much has happened to the joy of sex. Only one in three women felt depressed at some point in their lives after sex”

The story just does not have that shock factor anymore, and this is just for starters. The article is based on a study conducted by the “Queensland Institute of Technology”, which does not exist. This is a pretty good early warning sign that the author of apparent ‘journalism’ is just a little bit crap at checking her work. As any quick Google will reveal, the research was actually conducted by the Queensland University of Technology, but no biggie. At least the study is real.

The study was conducted with a sample size of 222 female university students who responded to a questionnaire. It was designed as a starting point for further research into the rarely studied phenomena referred to as ‘postcoital blues’. The sample size is what Data Dredge will call just above the ‘L’Oreal level’ (95% of 183 women agreed that their hair looked revitalised…). So here is problem number one, the sample is relatively small.

The sample also has a demographic problem. The average age is just over 24, they are all from one current geographic location and their education is going to be of a higher standard than many in Australia. This means that blanket statements, like the Mail story, cannot actually be made on the basis of this one survey. The sample is simply not varied enough in every aspect to claim conclusive results.

Another problem is the sexual history of the sample. 36.8% reported experiencing some form of sexual abuse and 25.0% reported having been forced or frightened into an unwanted sexual act before the age of 16. Apart from the disturbing nature of these statistics they cause a real problem when it comes to the reliability of the survey. The Mail, if it had taken a cursory glance at the report instead copying and pasting a press release, would have by its own standards of proof been able to claim that 36.8% of women have been sexually abused. This, if true, would be incredibly scary. In fact the figure of child abuse, which the sample has at 25%, is estimated at around 12% of women in Australia. This shows the sample size to be a poor representation of Australian women and women as a whole.

This is not to attack the authors of the survey who by their own admissions understand that their approach of a self-assessed survey and limited sample size can distort the results. With self-assessment comes the conscious or sub-conscious desire to adjust some of the answers to appear to be within what is felt to be a social norm, even when the survey is anonymous.  

So, after all this, just how shit is the Daily Mail claim? Well, what we have is a prime example of perfectly reputable science being utterly misunderstood by a journalist. The Daily Mail’s desperate attempts to get their stories on Google by mentioning sex and making shocking sex claims has driven the quality of the journalism down. Mentioning sex over and over again in an article is just pathetic. Sex.

So what would the headline read as now?

“Something might have happened to the joy of sex. One in three of 222 female Australian University students with abnormal rates of sexual abuse report feeling depressed after sex”

Shocking read.

As for the actual number of women it is hard to say. As the report authors point out, no thorough research into the area of ‘postcoital blues’ has been carried out. How depressing.

If you like it then share it!

If you suspect a dodgy claim is being made by a member of our wonderful press then Tweet me @mattjreason with the tag #datadredge. I also accept abuse. If you think I have got something wrong then do the same. If you wish to sue or some other ghastly thing then fuck off.





Thursday, 15 November 2012

A quick welcome


This is the Data Dredge, a blog dedicated to debunking statistical inaccuracy and bogus claims in the press. It could be anything from the complete misreading of raw data to simply claiming something that is, in reality, utter bullshit.

The author is an unemployed journalist with too much time on his hands. He is a cynical and properly boring person; you really don’t want to be friends with him.

If you want to abuse, question or fall head over heels in hate with him then you can tweet him @mattjreason (please keep all abuse in uppercase).

The process of actually debunking some of the shit out there is actually hard, long and laborious so do not expect more than one story a week.