Tuesday, May 03, 2011

Ammo: Applied churnalism.com - the BBC and the Environment Agency


(For an explanation of 'Ammo' - see here).

In a previous post I highlighted the fantastic new facility, Churnalism.com. I used the example of a recent idiotic climate change story to highlight how useful the site was.

However, aside from individual queries, it is possible to use the site to process more substantial amounts of data that can potentially reveal trends or biases, and that is what I have done today.

My interest piqued by the utterly moronic 'Llama' story (eviscerated here) I decided to carry out a thorough survey of exactly how many press releases are cut and pasted (and how much is copied) by our media class in the UK.

My findings shocked even cynical old me.

I took all of the press releases from May and June 2010 available on the Environment Agency site. (Oddly, as I was working through the data the public link to the May 2010 data simply disappeared...however this isn't a problem as I stored the URLs of every single press release so others can verify this for themselves).

The research covered a total of 166 press releases. I will work out the statistical significance later, however I believe the sample is representative as it covers two out of twelve months of data. The data needs to be arranged in a tidier format; when this is done I will add an update to this post with the data in an excel spreadsheet so others can replicate my results.

My methodology was simple, every press release was copied into churnalism.com in its entirety. Some had a lot of data that could have been regarded as extraneous, however I didn't want to start editorialising and putting all of the data in is more favourable to the media organisations. I have carried out some analysis of the data below; I'm sure there are more interesting links and patterns to be found - you are welcome to see what you can do with the data yourself when I upload it. Additionally the figure I used as a percentage was the percentage of the article that is made up of pasted material. Every entry can be verified on Churnalism.com

Summary by Media organisation:

The BBC:
The BBC cut and pasted significant content (at least 20%) from 78 articles. Out of those, 39 had 50% or more of their content cut and pasted from the press release. In a number of cases the BBC also generated 2-4 articles from the same press release. In percentage terms, and regarding the sample as representative I conclude that the BBC cuts and pastes content from 45% of Environment Agency press releases. Out of all press releases, 23% are partial (20-49%) copies and 23% are majority copies (50%+) of the Press Release content.

Can you see the political bias yet?

I probably don't have to spell out the implications of this - any level of churnalism is obviously undesirable, yet such an enormous amount carried out by the BBC, uncritically carried on behalf of the Environment Agency, is absolutely not acceptable and completely shocking. BBC 'impartiality is in our genes', or BBC 'the worlds greatest media organisation' my arse. Not to mention the fact that it has billions at its disposal, is publicly funded (by compulsory tax) and is the nation's media monopoly organisation.

None of the other media organisations came even close:

The Independent:
14 Articles cut and pasted. Of those 12 were less than 50% pasted content and 2 were more.

The Telegraph:
10 articles cut and pasted. Of those 7 were less than 50% pasted content and 3 were more.

The Daily Mirror:
7 articles cut and pasted, all of which were lesss than 50% of content pasted.

The Guardian:
7 articles cut and pasted. All were less than 50% of content pasted.

The Express:
6 articles cut and pasted, all of which were lesss than 50% of content pasted.

The Daily Mail:
6 articles cut and pasted, all of which were lesss than 50% of content pasted.

Financial Times:
1 single article, less than 50% pasted.

The Times:
1 single article, less than 50% pasted.

Caveats:
The data should not be trusted for statistical analysis on the basis of dates as I believe the environment agency may have put the wrong dates on their some of their Press Releases - sometimes up to a week ahead of articles that have copied them.

There may be some crossover either in articles cut and pasting material from more than one article or from parts of the Environment Agency's press releases being repeated from one press release to another. Further work is required to weed these out (which could either strengthen or weaken the churnalism respectively).

Finally there are likely some minor errors in calculations; I've gone through the data twice so far but will be applying a finer level of attention later.

**Data to be uploaded later in the week**

I'd like to invite others to replicate this work - potentially crowdsourcing such efforts could allow us to check the entire content for the Environment Agency's press releases and churnalist enablers such as the BBC so there would be no ambiguity whatsoever, especially if the data is kept public.

There are countless other comparisons that could be made using churnalism.com between particular organisations, individuals and media organisations to reveal what was previously hidden.

Fight the endarkenment!

9 comments:

banned said...

I can only comment about The Telegraph but I am surprised that their record is not worse than your research finds.
Pages 4-5 often consist of nothing more than regurgitated press releases with no analysis and only rarely comment.

Katabasis said...

I agree Banned. Bear in mind this is only for the Environment Agency so far though. There will be press releases from other orgs regularly regurgitated by the likes of Louise Gray at the Telegraph.

This is why I'm keen to inspire other people to get involved - who knows what other patterns are there waiting to be discovered.

Frankly I'm still blown away by how much more churnalism the BBC engages in on this topic than any of the others.

WitteringsfromWitney said...

K: Great work and have retweeted! Tad busy at present otherwise would get involved - maybe later.......

Anonymous said...

I'd love to see what the churnalism rate for the likes of ASH and Alcohol Concern's press releases are in our national media. Extremely high, I suspect.

Katabasis said...

Anon - this is why I'm keen to encourage people to get involved and share their results.

Crowdsourcing this kind of thing could lead to an enormous amount of useful data and we will have statistical proof of bias and shoddy journalism.

James Higham said...

Interesting breakdown of the gazettes there.

Pointman said...

A great piece of work Katabasis.

Pointman

James Higham said...

I decided to carry out a thorough survey of exactly how many press releases are cut and pasted (and how much is copied) by our media class in the UK.

Don't give yourself the small tasks, do you, KB?

Katabasis said...

Oops - I probably should have worded that better; I was referring in this context to the Envrionment Agency.

I'll be sampling other similar organisations next, probably the Met Office.

Churnalism.com offers an API so I'm currently putting some time aside to work on a generic web spider I can make to automate submission of press releases of said organisations to the churnalism engine.

This is why I'm regularly promoting the churnalism site - if we can pick up a crowdsourcing effort on this we could between us cover large areas of coverage between us.