Sport England’s unconvincing claims regarding sports participation

1474

Sport England recently released an update to their sports participation statistics. The source of the estimates is the Active People Survey, an ongoing telephone survey of sports participation in England. The key measure reported is participation in 1 session a week of sport, with the condition that participation has involved “at least 4 sessions of at least moderate intensity for at least 30 minutes in the previous 28 days”.

If you were to simply read the press release (‘More young people playing sport’ 13/06/2013), it would look as if a positive picture of sizeable and consistent growth in sports participation was being found. Phrases such as: “most of the increase seen following the Games in 2012 has been retained” and “750,000 more people playing sport than the previous year”, suggest a pretty impressive impact from the Olympics and the various policies, initiatives and funding put into sport by the government and sports organisations.

The most impressive statistic relates to how 1.4 million more people are participating than when London won the right, in 2005, to host the 2012 Olympics. But if you reframe this impressive figure as a proportion of the population, it equates to an increase of around only 1% (the 1.4 million figure is influenced by population growth). When you plot the data, an even less impressive picture emerges. The 1.4 million growth relates to the difference between the first and last data points shown in the graph below (the vertical bars at the data points represent 95% confidence intervals around the point estimates)…

National sports participation
National sports participation
(*last data point is a rolling estimate including half of the previous year's data)

The last four data points show that the figures are somewhat unstable. So what would the graph look like if you disregarded the first point as being unusually low? I personally doubt anyone would be convinced the figures were indicative of growth. As with all statistical plots, it’s always worth scrutinising the presentation too; my graph has a narrow range y-axis. What does the chart look like with an axis that incorporates the entire 100% of the population? Answer: pretty flat…

National sports participation v2
National sports participation with a 100% y-axis
(*last data point is a rolling estimate including half of the previous year's data)

Another focus of all the current ‘Olympic legacy’ interest is participation by young people. In their press release, Sport England write: “The number of young people aged between 16 and 25 playing sport regularly has reached 3.86 million. This is an increase of nearly 63,000 on the previous 12 months”. But again, if you plot the data, it looks a little less impressive over time. Taking into account the confidence intervals, there is little evidence of any change in recent years…

16 to 25 sport participation
16 to 25 sport participation
(*last data point is a rolling estimate including half of the previous year's data)

The story painted by the data isn’t all bad, however. The trend in participation by disabled people is encouraging, showing a consistent rise over time, although participation rates are still low compared with the rest of the population…

Disabled sports participation
Disabled sports participation
(*last data point is a rolling estimate including half of the previous year's data)

Overall, I’m not convinced sports participation is really increasing, except perhaps amongst the disabled. Sport England has funded and carried out a lot of very high quality research into sports participation in the past, but press releases like this, which do not seem to honestly represent the data they are describing, only serve to undermine confidence in Official Statistics in the long run. I think Sport England is letting itself down by allowing its press office to release this kind of spin. The reasons, however, are apparent. Sport is a highly politicised topic at the moment, what with the ‘Olympic legacy’, the Select Committee on school sport, and all the rest of it; budgets are being cut, and organisations are battling to make the case for their funding.

But perhaps this is all academic anyway. Sports participation does not provide the whole story on levels of physical activity. For example, the Sport England measure does not include all recreational walking and cycling, as they are not defined as ‘sport’. But routine physical activity, not sport, is of more concern for the health of the population as a whole. Also, you have to wonder how many of those people who are included in the Sport England measure are meeting the recommended levels of physical activity (DoH, 2011):

  1. Adults should aim to be active daily. Over a week, activity should add up to at least 150 minutes (2½ hours) of moderate intensity activity in bouts of 10 minutes or more – one way to approach this is to do 30 minutes on at least 5 days a week.
  2. Alternatively, comparable beneits can be achieved through 75 minutes of vigorous intensity activity spread across the week or combinations of moderate and vigorous intensity activity.
  3. Adults should also undertake physical activity to improve muscle strength on at least two days a week.
  4. All adults should minimise the amount of time spent being sedentary (sitting) for extended periods.

The Health Survey for England has generally found that about a third of the adult population reach healthy levels of activity. An accelerometry sub-study found that only around 5% of the adult population were sufficiently active. In 2002, the New Labour government released Game Plan (DCMS & Strategy Unit, 2002), which included targets to get 50% of the population reaching healthy levels of activity by 2010 and 70% by 2020; the first target has been missed (even according to self-reported activity) and the second must now be seen as a pipe dream. If the current government is serious about raising activity levels in the population, they have their work cut out and desperately need to come up with some better ideas. I doubt that ‘sport’ as defined by Sport England can solve this problem on its own.


UPDATE: DCMS claim more women are playing sport since the Olympic bid was won
The DCMS are at it again. On the 5th July 2013 they tweeted this:
DCMS tweet
In case you are interested what this looks like with their data, the graph is presented below. I would argue, this isn’t too convincing either…
Female participation in sport
Women's sports participation
(*last data point is a rolling estimate including half of the previous year's data)