Age, Gender and Sport in England

I apologise for the change from the advertised topic for this post. I was planning to talk about the relationship between childhood experiences and involvement in sport, but have been sidetracked by writing up my PhD thesis which is now taking up most of my time. Part of this process has been to crunch some numbers from Sport England’s Active People Survey, a massive survey of over 150,000 people across the country, which tracks sports participation changes from year to year. I’m using the data to contextualise running within the full range of sporting and active leisure activities popular in the UK today. In particular I’ve been looking at how running compares to other sports in terms of participant demographics.

I have posted a simple ranking of sports in terms of the occupational class of their participants before, but now that I have analysed more up-to-date data that includes extra sports as well as taking into account gender and age, I thought I’d come back to it again in a bit more detail.

First of all, here’s a chart showing the relative gender balance (x-axis) and mean age (y-axis) of participants in a number of popular sports. The dark red lines indicate the mean age and gender for the sample. The fact that the mean gender is just over 0.4 indicates that more women than men responded to the survey. Sports to the left of the dark red line are more popular with women, sports to the right more popular with men. The further they are from the red line the stronger this bias is. The mean age of the sports (on the y-axis) runs from 22 (basketball) to 60 (golf).

Sports participation by gender and age

The next two charts show occupational category data sorted in two ways. Each shows the percentage of participants in a selection of popular sports that have professional or managerial jobs (NS SEC 1-2, shown in blue) and the percentage that have traditionally working class jobs or are currently unemployed (NS SEC 5-8, shown in orange). The first chart shows the sports sorted by the percentage of ‘higher status’* occupations, the second is sorted by ‘lower status’* occupations.

I’m not going to comment on these findings at this stage, other than to say that they reinforce a strange fact about running. On the one hand it is a highly accessible sport with low barriers to entry that attracts men and women in almost equal quantities; on the other it attracts a disproportionately high level of middle class participants. In fact running sits above golf, tennis and mountaineering in terms of its proportion of higher status occupation participants. This is a paradox my research seeks to shed light on, and something I will come back to in future posts.


* The use of ‘higher’ and ‘lower’ here is related to the typical level of income and education of holders of these jobs – i.e. the socio-ecomomic status. They are not meant o imply that managerial jobs are inherently superior to working class ones.


Active People Survey:

Sport England. (2016). Active People Survey, 2014-2015. [data collection]. UK Data Service. SN: 8038,


Age and Gender Maps of Running

In a previous post I showed how different forms of running (road runners, fell-runners, obstacle course racers, ultra marathoners, track athletes) attracted different socioeconomic groups. We visualised the way in which participation was structured by plotting the mean income and education ranks for participation in each sport onto a chart with axes of education (x) and income (y). This was the result:

A quick look at the relative positions of the different forms of running suggests that the longer the distance of the event, the higher the average income level of those attracted to participate, and that education level is somehow connected to the type of environment people like to run in; lower education levels are associated with highly constructed, artificial spaces like the obstacle course and running track, and higher education levels are associated with participating in unstructured, natural environments and the wild. Perhaps there’s also a suggestion of a link between education level and preferences for communal or solitary running experiences.

But of course education and income can only explain so much. In fact there are other, more powerful drivers behind the choice of running form that need to be looked at.

In terms of their influence over our choices around sport, two of the most important social variables of all are gender and age. And of course these two factors are strongly linked to income (men and older people tend to earn more), so perhaps some of the effect we can see in the above chart can be explained simply by the age and gender of those taking part.

Below I have plotted the same five forms of running onto a similar chart, but this time with axes of gender and age. The age axis is self-explanatory, it’s simply the mean age of participants for each point plotted. The gender axis shows the relative proportion of male and female participants in the sport. The pink line marks the sample mean gender balance, so points to the right of this have more male participants than average, points to the left have more female. The further from the central line the more lopsided the gender balance gets.

I’ve also add some extra plots for key motivations (red) and included two extra running forms: Jogging [Jog] (non-competitive runners), and Orienteering [Ori] (for which I have just collected a booster sample of 300 respondents).

Key to motivations (red points): ‘Looks’ = strongly agree with ‘I run to improve my appearance’; Weight = ‘to lose or maintain weight’; ‘Social’ = ‘to socialise with friends”; Explore = ‘to explore the outside environment’; ‘Races’ = ‘to do well in races’.

Interestingly the locations of the five forms we saw in the first chart are broadly similar in this one, even though we’re ostensibly measuring different things. This suggests that there may well be a relationship between gender/age and income/education.

We can see that the motivations associated more strongly with women are those around managing weight and improving their appearances, as well as social motivations. I should say that this is absolutely not to say that these are priorities for all female runners, the positions on the chart represent averages from my sample of almost 3,000 runners surveyed. Men are more likely to be motivated by competition (races) and exploring the outside environment.

To an extent these motivational tendencies are reflected in the forms that men and women participate in. Men are more likely to appear at a fell race or ultra marathon, both of which would often involve both competition and training in remote outdoor environments. Orienteering is the most male dominated of all the forms on the chart. Again we have a strong element of exploring the outside environment.

Women are more likely to go for OCRs (obstacle course races), which foster team spirit and camaraderie (i.e. feeding social motivations), and jogging, which is often practised to lose weight and involves no athletic competition. The location of sprinting is gendered female is harder to explain, but this may be because the sample of sprinters is relatively small, so the data may not be so reliable.

In terms of age, sprinting (unsurprisingly) attracts the youngest participants, with OCRs the second youngest group. Orienteering again stands out as by far the oldest of the forms.

Almost in the middle of it all sits the half-marathon. I think this reflects the open, easy access nature of this event. For many people it may be their entry point into running, and can be run as a motivation to stick to a weight loss plan or as a highly competitive race. It appears to be the one-size-fits-all event of running, attracting a gender and age balanced participant base.

Behind the Structure

How can we explain why different sports attract different social groups? This is a difficult question, and is key to anyone interested in promoting particular forms running and broadening their appeal. There are a number of possibilities:

  1. People prefer to join up with sports that are already populated by ‘people like them’, be that class, gender or age. This is certainly true, and would help explain why social differences harden, but not how they formed in the first place. This may require a historical account of how each form of running came into being.
  2. Some groups have greater physical access to the right infrastructure for a particular form of running, or have higher practical barriers to overcome to take part. For instance perhaps parents don’t have as much free time, so can’t fit in the training for doing ultra marathons, or self-employed people have more flexibility to fit in the large volumes of training for such events.
  3. Some forms of running cultivate an image that puts off or favours certain groups. Forms such as Ironman triathlon and races such as the Man versus Mountain appear to be masculinised through their names. Could this make them less appealing to women? Are they positioned as symbols of masculinity?
  4. People choose a form of running that suits their particular motivations, so competitive people choose more competitive forms, and people concerned about body image choose forms that they think will best address this. Again, this is certainly a factor, but sociologically speaking it’s quite a superficial answer. It fails to address why certain groups are more likely to have particular motivations than others. What is it about being a woman that makes you more likely to be motivated by losing weight; why are more educated people more interested in exploring?
  5. People’s life experiences, linked to their age, gender, social class and other factors lead them to develop particular tastes and identities that make some forms especially appealing and some less so. This is an important perspective. It pulls together ideas about motivation and taste and the demographic variables we’ve been looking at. By understanding how people’s class, gender or age result in different life experiences, and how these experiences lead to particular preferences or tastes, we might be able to get at the underlying cultural reasons for the structuring of the forms of running.

Picking up on the last point, in my next article I’ll be looking at some real examples of the ways in which running participation is influenced by people’s life experiences. In particular I want to look at how different experiences of sport and running in childhood are influenced by the kind of school people go to (state/private, rural/urban) and their gender. These differences, which we can explore using the interview data I’ve collected over the last year, show how variations in the opportunities and experiences of childhood can lead to quite different orientations to running in later life.

As always, any comments, ideas or suggestions would be very welcome! Just use the box below.

Running Riot: Victorian Track Hooligans

The Forgotten Story of Lillie Bridge

Hooliganism is not something we associate with running today. The crowds at track meets and road races are some of the friendliest and best natured of any sport. Yet at the dawn of the era of modern athletics running had a serious problem with crowd violence. In fact, the emergence of the modern sport was driven in part by the desire to rid running of the unsavoury elements it seemed to attract.

One of the most vivid examples of the violence connected with running in the nineteenth century is the riot at Lillie Bridge, in 1887. Lillie Bridge was the headquarters of the Amateur Athletics Club (forerunner of today’s Amateur Athletics Association), and the venue for the national championships. With a capacity of over 12,000 set around a third of a mile cinder track, it was state of the art when it was built in 1866.

Battle of the Bridges

At the end of the 1870s the pre-eminence of both Lillie Bridge and the Amateur Athletics Club was under threat from the powerful London Athletics Club, which opened its own stadium close by at Stamford Bridge. In a bid to take control of the sport the London AC also inaugurated a rival national championship hosted at the new track. The result was that two competing national championships took place in 1879.

In order to ensure its survival in this competitive environment the Amateur Athletics Club needed to maximise the income it received from Lillie Bridge. So the track was rented out to organisers of professional running events, which attracted large crowds and meant high gate receipts, but also brought the gambling and rowdy working-class crowds that many of the privileged founders of amateur athletics wanted to exclude.

The burning of the Bridge

In 1887 a professional race between Harry Gent and Harry Hutchens attracted large crowds to Lillie Bridge, and big stakes were bet on both runners. Things started to unravel shortly before the start when a rumour swept through the stadium that the race had been fixed. One of the runners was not fit, and it seemed gangsters had pressed for the contest to go ahead anyway so that they could make a killing betting against him. The bookies called for the race to be cancelled, and angry spectators demanded their ticket money back. And when the managers of Lillie Bridge failed to provide refunds, the crowd erupted into violence.

The rampaging mob pulled down the wooden structure of the stadium and surrounding buildings, smashed seating and set the wreckage ablaze. The police that were present were beaten up after trying to resist them, and one man died in the chaos. Further police arrived on the scene and eventually managed to restore order, but not before Lillie Bridge was put beyond repair. The venue was closed the following year, never to host a running event again.

The destruction of Lillie Bridge seemed to confirm the belief already held by many, that professional running was corrupted by criminality and cheating, and that only by adhering to principles of strict amateurism could spectators’ confidence be restored. In the decades that followed the administrators of amateur athletics worked tirelessly to marginalise professionalism, ensuring that prestigious events like the national championships and the Olympics were open to amateurs only. With money out of the equation there was little incentive for athletes to throw races or for criminals to get involved. As a result, amateur running quickly achieved a reputation for honesty and gentility that helped it become the definitive form of the sport for much of the following century.


This article was originally published in Running magazine, Winter 2016.

Running and Class

One of the anomalies about running is that despite the fact that it’s one of most accessible and cheap sports to be involved in, it’s disproportionately dominated by members of the middle-class. There’s plenty of evidence from all kinds of sources, including the massive Active People Survey by Sport England, to show that runners tend to have higher incomes and levels of education than participants in almost any other mainstream sport.

A few months ago I put together this table to show where running ranks amongst a range of sports in terms of its relative popularity with people in ‘high’ and ‘low’ status occupations (as per the National Socio-Economic Classification model).

To explain the figures: If a sport gets a score of 2 on the ranking that would mean it is twice as popular with the high status group as it is with the low status group. Or, if a sport gets 0.5 then the likelihood of a high status person participating in the sport is half that of a low status person. A sport scoring 1 is equally attractive to people of both groups.



Participation Rate Ratio

1 Tennis 3.89
2 Squash 3.00
3 Keep-fit classes 2.43
4 Golf 2.42
5 Mountaineering 2.40
6 Running 2.28
7 Road Cycling 2.25
8 Swimming – outdoor 2.09
9 Athletics – Track & Field 2.08
10 Aerobics 2.05
11 Badminton – indoor 1.87
12 Hockey 1.67
13 Swimming – indoor 1.60
14 Netball 1.53
15 Fitness & conditioning 1.50
16 Gym 1.49
17 Table tennis 1.29
18 Boxing 1.20
19 Karate 1.06
20 Equestrian 1.05
21 Bowls 1.03
22 Shooting 1.00
23 Cricket 0.97
24 Football 0.94
25 Rugby union – 15-a-side 0.73
26 Tenpin Bowling 0.71
27 Basketball 0.63
28 Snooker 0.60
29 Pool 0.56
30 Angling 0.54
31 Darts 0.40

[EDIT: A reader asked me what skiing would score in this table. Looking at people who ski at all (rather than once per week participation, which is the criteria for inclusion above) it would score 4.8, topping the table.]

Running, then, occupies a place just behind mountaineering and golf on the social scale, two sports which require significant financial outlays and are traditionally regarded very much as preserves of the middle-class. We can also see that running is the leading member of a cluster of individual outdoor racing and fitness sports – the others being cycling and wild swimming – that sit as a group on our socioeconomic hierarchy.

In this article I’m going to unpick this a little further, both by looking at the differing influences of income and education (two key factors underpinning understandings of social class) on running participation, and by unpacking that broad category of ‘Running’ into its different forms. As we will see, a similar hierarchy to the one shown above between sports also exists within the sport of running itself, with different forms attracting participants of significantly different backgrounds.

Unpacking Running

Anyone with more than a passing interest in running will know that the catch-all term ‘running’ covers a wide range of forms with significant differences in practices and appeal.

Track athletics, for instance, is a very different sport to fell-running. Ultra-marathon is a world apart from mud races. Multi-day adventure races have little in common with parkrun – save for running itself.

So perhaps, like the sports in the table above, different forms of running also attract different participant bases. Perhaps there’s a social hierarchy within running. In order to explore this we need first to unpack running into several different sub-sports we can compare.

The data from the Big Running Survey allows us to do this in a systematic way. Because each respondent listed the forms of racing they were involved in, we can look at groups of runners participating in different forms of the sport and see how they compare in terms of the socioeconomic factors of education level and income.

In this article we will look at how participants in the following forms of running vary in terms of their socioeconomic status:

  • Road racing (up to half marathon)
  • Fell-racing
  • Track racing
  • Ultra marathon
  • Obstacle course racing (OCR)


The chart below shows the mean personal income rank for each type of runner.

But first, a caveat: Because most of the runners in the data I have collected participate in more than one form of running (for example they have competed in both fell and road races in the last year) we have to bear in mind that the distinctions between different forms of running we observe may be watered down somewhat.The same runners appearing across more than one category will naturally make those categories more similar than they would be if we were looking at runners who only participate in one form of the sport.

So we need to interpret the figures carefully, and bear in mind that the differences we see below are probably smaller than what they might be if we compared ‘purists’ from each version of each sport.

Mean income rank by forms of running

Red line: Mean personal income in the UK

We can see that for all of the forms the mean income rank is between 2 and 3, which equates to between about £18,000 and £25,000. This makes sense in that this range includes the mean salary for the UK. Within this range though, there is a significant variation between forms, even without accounting for the number of people participating in multiple forms of running.

Ultra distance and fell-running lead the way, with road runners not far behind. There’s a significant gap back to OCR participants, and, on the lowest average income, track athletes.

However, these differences can be in part explained when we look back at the demographics of these different sports. In an earlier post we saw that different forms of running attracted men and women in different proportions. This is also true when it comes to age.

Ultra and fell running both attract more than their fair share of men, and of older runners. Both of these variables are associated with higher incomes. OCRs are associated with younger runners and are more popular with women. Track distances are very much the domain of the young. So perhaps this explains these differences. Let’s see what happens when we take gender out of the mix. We’ll just look at male runners in the next chart:


Surprisingly, looking only at male runners appears to have increased the variation. Another change is that fell-runners have dropped below road runners, with OCR runners snapping at their heels.

The change in ordering is probably because there were few women ‘dragging down’ the mean income rank of the fell-runners group compared to, say, road running, which has a much higher level of participation from women.

Let’s look at the same chart, but for female runners:


The most striking thing here is how similar the income scores are for all kinds of female runner, aside from track athletes. There’s almost no variation – so little that it may just come down to chance.

This is interesting. There appears to be a distinct economic hierarchy amongst male runners, but not amongst females. Let’s see if the same is true in terms of their education levels.


These are the mean ‘highest level of education achieved’ ranks for all runners:


There is significant variation between the forms of running in terms of the average education level of their participants, with ultra runners tending to be the highest educated and OCR competitors the least. Here there is less of an issue with gender skewing the results, as the men and women in the survey had similar average education levels. But that’s not to say splitting this down by gender won’t reveal differences in the distribution of education ranks. Let’s see.

Male runners:


Clearly if we look at male runners in isolation they present a different distribution to the mixed gender chart. Track athletes overtake road runners in mean education level, and fell-runners catch up with ultras.

And for female runners:


This more closely matches the overall chart, with the exception of the especially low level of mean education for female track runners in this sample. The similarity between the female chart and the overall chart is explained by the fact that there were more women in the survey sample than men by about a 3:2 ratio.

We can see that the amount of variation between the means is not significantly different for men and women. There’s about 0.5 rank points between the lowest and highest educated male groups, and for women the spread is about 0.4.

Map of the Social Space of Running

It’s worth remembering that at the start of this post we saw that running in general is very much a middle-class sport in terms of its participant base. So the variations within it can be seen as variations within middle-class taste, with different sub-groups preferring different forms of running.

To visualise this we can create a social map of running tastes with the data we examined above. The map is structured by income and education.

the social space of running

The orange lines indicate the sample means for income and education rank.

The chat shows us that within the (rather middle-class) population of runners there are some distinctions in terms of the relationship between different forms of the sport and the social status of those participating in them.

Road races up to half marathon sit at the centre, probably because this kind of racing attracts the widest range – and largest number – of participants. As a result participating in road running is the least socially distinctive form of the sport.

Fell and ultra racing share a quadrant of the map, attracting people of higher economic and educational status than average. However, we’ve seen that for men the mean income for fell-runners is significantly lower than sample average.

OCR and track racers occupy the low status quadrant, with lower than average income and education (compared to other runners, not necessarily to society in general). For track racers this can be explained in part by their relative youth.

Why the differences?

That’s it as far as this whistle-stop description of the social space of running is concerned, but we have yet to address the question of why different forms of running attract different social groups. This is a difficult question that means looking at the history and culture of the different forms of the sport, as well as issues such as access, geography and the demands of the different types of running.

In the next post I will start to unpick these distinctions, looking at each form of running in turn, attempting to explain how they have come about.

Gate-crashers and Ghost Runners

John Tarrant - the Ghost Runner
John Tarrant c. 1957. (copyright unknown)1957

Today, running is one of the most inclusive of sports. Athletics clubs are open to all, and mass races such as the London marathon offer anyone with the bottle the chance to test themselves in one of the world’s great sport events. But running hasn’t always been so democratic. Victorian hang-ups about women and professionals (the latter a thinly veiled euphemism for working-class athletes) lingered well into the twentieth century. Women were barred from long-distance races for decades, and anyone who had ever accepted payment for participating in any sport was banned for life. This left the way open for a clique of self-funding male athletes to dominate the sport for years.

As the twentieth century wore on however, social attitudes were changing, and running’s draconian rules began to look old fashioned and unfair. But any change was resisted by the privileged elite of former athletes who controlled the sport. Frustrations built, and some excluded runners, feeling like they had exhausted all routes to participation, decided to take direct action.

The Interlopers

Two key figures in this struggle were John Tarrant and Kathrine Switzer. Tarrant was a tough and talented, working-class runner from Derbyshire. He was refused entry to official races in the 1950s because as an 18 year old boxer he had received £17 in expenses, inadvertently stigmatising himself as a ‘professional’ for life. Switzer was an aspiring American marathoner who was barred from the 1967 Boston Marathon simply because she was a woman. Both decided to fight their exclusion in a similar way. They would gate-crash a race incognito, simultaneously achieving a sporting ambition and drawing attention to their plight.

Tarrant became a regular sight at road races in England, running without a number, hiding in the crowds at the start so as not to draw attention. Once the race was underway he was too fast for the officials to stop, and with the crowds roaring him on, he became more famous than many of his officially sanctioned rivals. In fact, so popular was he that race organisers started asking him to gate-crash their races, advertising the presence of the fabled ‘Ghost Runner’ in advance to boost crowd numbers!

Switzer entered the Boston Marathon as ‘K. Switzer’ dressed in an inconspicuous grey tracksuit. Although she couldn’t rely on her speed to escape angry officials she did have the advantage of being accompanied by her burly, hammer-throwing boyfriend, Thomas Miller. When one of the race’s organisers leapt onto the course to physically eject her, Miller brushed him aside. Switzer ran on to complete the race that was supposedly ‘too tough’ for women.

The Court of Public Opinion

The media played an important role in both stories. Tarrant was an exceptional athlete whose David versus Goliath story was mana from heaven to the newspapers. He built up a powerful group of supporters, including many fellow runners, and eventually the pressure told – he was cleared to compete nationally. Unfortunately his international ban remained, barring him from the Olympics, his greatest goal. However, he went on to win many official races and broke two world records at ultra-distance before his tragically early death at just 42.

Switzer wasn’t the only female runner to gate-crash the 1967 Boston Marathon. She wasn’t even the fastest (that was Bobbi Gibb, 3 hours 27 mins). Her good fortune was that the incident when the race official tried to grab her was caught on camera. The picture created a media sensation and generated the sympathy and support that eventually led to the opening up of the Boston Marathon to female runners in 1972. Switzer went on to win the New York Marathon in 1974.

Thankfully for runners today, the old barriers have been broken down. The accessibility and diversity of mass participation running have become some of the sport’s most attractive features. Trailblazers like Tarrant and Switzer played a small but important role in helping this happen.


This article was originally published in Running magazine, September 2016.

You can read the full story of John Tarrant in Bill Jones’ excellent book: The Ghost Runner: The Tragedy of the Man they Couldn’t Stop.