Sustrans has described as “bitterly disappointing” statistics suggesting there has been no growth in levels of cycling in England in the year to October 2012, but British Cycling has previously hailed separate figures covering the same period, from the same source – Sport England’s Active People Survey – as reflecting a boom in cycling. So which is correct? The short answer is neither.
As we reported in December, British Cycling said that data from the Active People Survey for the year to mid-October 2012 showed that 200,000 more people rode a bike once a week compared to the previous year and provided proof of a successful legacy from the Olympic and Paralympic Games.
However, the figures seized upon by Sustrans today, contained in a release from the Office for National Statistics that focuses specifically on cycling, show that the percentage of respondents who say they ride a bike once a month is unchanged from the year to October 2011 at 15 per cent.
“It is bitterly disappointing to see that cycling levels have not increased over the last year, especially considering the surge in enthusiasm for cycling,” said Sustrans policy director, Jason Torrance.
“Governments must prioritise making our roads safe and appealing if we are to increase everyday cycling numbers.
“It’s time we put an end to the postcode lottery that means some areas aren’t as safe as others, preventing many people from cycling regularly for everyday journeys to work, school or the shops.
“We need safer streets and more dedicated cycling routes so that everyone can feel safe on a bike, only then will we achieve reduced congestion and a healthier, more active population,” he added.
The figures British Cycling was looking at, however, were rounded to two decimal places, while those in the Office for National Statistics release are whole numbers. Given that each 0.1 of a percentage point equates to around 40,000 adults in England, clearly there is potential for some big discrepancies.
The Active People Survey has a sample size of 160,000, meaning that results should be fairly robust, at least at national level, so it's unclear why the Office for National Statistics has worked on round numbers instead.
There’s also another vital difference between the figures – the ones cited by British Cycling only cover cycling for 30 minutes each time for health, fitness and competition, and exclude utility cycling such as riding to work or the shops, as well as shorter rides; the data cited by Sustrans, while coming from the same source, relate to cycling for any purpose, irrespective of time spent riding.
As we pointed out in our coverage of British Cycling’s response to those figures back in December, the data are subject to some big variations year on year, so while there is a clear upward trend observed in the percentage of people riding for 30 minutes for non-utility purposes at least once a week in recent years, care should be exercised when comparing one 12-month period with the one preceding it.
The Sport England figures based on that narrower definition of cycling also showed a drop over several years in the proportion of people cycling less regularly than once a week but at least once a month. In numerical terms, that decrease was a lot less than the growth in those cycling once a week or more; in other words, not only are more people cycling, they are doing so more regularly.
While that relates to non-utility cycling, recently published Census figures show big growth in the number of people in England and Wales over the last decade who are riding to work, which could indicate that people already cycling occasionally are doing so more regularly, and for reasons other than just to keep fit; again, it’s impossible to tell, particularly since the figures compare two years a decade apart.
So what’s the true picture behind those figures that have given Sustrans so much cause for concern? It’s impossible to say, and we can’t trace data going beyond that rounded 15 per cent.
As we highlighted above though, it’s dangerous to read too much into a comparison between two consecutive years, moreover one in which there could be a variation of the equivalent of around 400,000 people, depending whether it’s 14.51 per cent, or 15.49 per cent.
But looking at the bigger picture, the evidence form a variety of sources suggests that over the past few years, more people are cycling, and once they start, there’s a progression towards doing so more regularly, as well as a crossover between riding for fitness and for utility purposes, and vice versa.
Add new comment
15 comments
Sustrans has described as “bitterly disappointing” statistics. Well perhaps if Sustrans infrastructure wasn't so bitterly dissappointing, more people might cycle. As a group that supposedly represents cyclists, Sustrans is a joke.
The way in which raw data is interpreted is subject to statistical techniques employed. ONS at first glance may seem to be reliable, but when scaling up samples to large numbers and stating trends, a measure as crude as "nearest number" is essentially going to produce inaccurate results. Going to two decimal places on a sample and then scaling will be far more accurate, so this may appear to indicate that Sport England's figures are actually more valid, whilst those from ONS (unless the sample is huge, which is unlikely) may not be entirely based on completely reliable source data in the first place. GIGO (garbage in, garbage out) always applies when considering statistics and any conclusions which may be drawn. Sustrans comments, as noted, appear to be more than a little silly, the use of language is clearly intended to promote their cause, which is fine, but a more skilful use of words would be helpful.
Here's a link to the 2010/11 survey so you can see exactly what was asked and how it was phrased. Cycling questions are on pp6 and 7.
http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CD...
As the article above says, the sample size is 160,000 - that's huge in market research terms, and certainly big enough to go to one or two decimals at national level with a fair degree of confidence.
ONS itself points out that at local authority level, sample sizes often mean that accuracy can't be guaranteed (even when just using the rounded numbers - that doesn't stop people trying though...)
The way these statistics have been gathered, and then interpreted, seems to indicate that they don't mean much at all. 'Bitterly disappointing' is a bit of a strong reaction. A more realistic reaction to these statistics might be 'so what?', since we don't really know much at all about the increase in cycling. Especially if we are discussing all kinds of cycling. Who was asked, when, exactly how were the questions phrased and how might they have been misunderstood, or people not bothered to answer accurately, what were the exclusions etc etc etc, all have a bearing on what kinds of results are received. I wouldn't trust any figures like this without a close examination of the design of the surveys.
Thanks for this piece as its a good balanced view on the statistics and spin.
My personal observations are that (outside of central London) utility cycling hasn't really increased at all though there is a huge increase in recreational road cycling by mamils at the weekend. Also more informal looking groups which I'd put down to the rise in charity rides/challenges. Worryingly I'd say this gives all of the pain (see disrupted sportives) with none of the gain (ie reduced car usage, better facilities etc).
Sorry, to clarify - perhaps 30-40% of the cyclists we saw in Yorkshire were female, we never see that many on our usual rides which start in South Mcr and take in Derbyshire etc.
I'd say Manchester is pretty suited to cycling as the centre is relatively flat, bar a few bumps for old railway lines etc.
Mind you, when MCC decide put in a cycle route that means you have to turn right across two lanes of traffic turning left in order to cross a mixed use path (Pavement with a bike scribbled on it) at a tight pinch point to then traverse over tram lines that are slippery as hell in the wet (yeah, in Manchester, who'da thunk it?) i think its not the geography that's the issue...
My own observations would suggest that there has been an increase, even just over the last 12 months or so.
Regarding the Sustrans statement of the "postcode lottery" of facilities/infrstructure etc, on a recent group ride in Yorkshire, there was a startling number of female cyclists compared to the small numbers we see around Manchester/Cheshire. Don't know why?
Hmmm did any of their data collection/statistics take account of the weather??? It hardly encourages new cyclists let alone us old ones!! "lies, dam lies & statistics". I suspect the trend is still on the way up but no real data to substanciate this other than the number of cyclists I saw out on Sunday!
What is it with bike hire schemes in cities and towns that are plainly unsuited? Talk of this in Manchester too, I just don't see the value (other than it's a bit sexier for a press launch than cutting the ribbon on a newly-painted strip of tarmac).
Sorry, a bit off topic there. But still, infrastructure is what we need to get people onto bikes. Proper demarcation/segregation and sort out the black spot junctions - leave the white elephants on the drawing board.
Well said Gazza D. As reported before Liverpool Council are trumpeting the installation of 1,000 hire bikes. They cost £2.5M but are fully funded so a nice easy press score for the Mayor Joe Anderson. We are all looking forward to his press launch as he makes Eric Pickles look like a whippet and may never have ridden a bike.
Meanwhile as cycle commuters brave the dual carriageway route in from North Liverpool the council is allowing Peel Holdings to carve up the whole water front with no real plans to include a cycle route connecting the north of the city to the centre.
It makes me weep.
That Chris Boardman is involved with trumpeting the useless bike hire scheme with so far no discussion on his part in cycle infrastructure and proper commuting in the city is also pretty poor. Not that they'd listen to him either...
The reason riding bikes hasn't increased and I don't believe it has except in select areas, is purely because riding bikes is not seen as safe or convenient for many.
From my perspective this is due to:
The unrelentingly hostile road conditions and attitudes of some drivers.
The lack of obvious, convenient and direct, and joined up cycle routes, and the unsuitability of a lot of them for kids trailer bikes, shopping trailers, tandems, trikes etc.
The media portrayal of cycling and cyclists as law breaking outlaws.
The complete apathy and obliviousness that a majority of councils and major companies and organisations have towards cycling and cyclists.
I imagine the increase is not statistically significant when rigorously tested. e.g. a 1% change in cycling figures isn't large enough to confirm that cycling has increased as it is below the threshold of what could be expected by random variation.
If you can't tell it apart from a random variation you can't really say cycling has increased.
If orgs can't agree whether it's good or not it would suggest that there isn't a clear or obvious trend.
You have repeated four paragraphs.
I would trust the ONS take on the figures. They probably don't believe that the survey results are accurate to four significant figures (two decimal places). You interview a sample of people, and then scale up that number to the whole population.
I would say that the survey results show that the rate of cycling hasn't changed much.