Hearing Jim Al-Khalili being interviewed about his own life this morning reminded me to finish this. He began by talking about his upbringing in Iraq, with an English mother and Iraqi father, and speaking English at home, and then toning down his RP accent when he came to the UK in ’79, as a refugee from Saddam Hussein. Still, if it were now, he may well have been noted as EAL by the school…
 

Take my friends, J and G. Neither speaks English as a first language, one is from the Americas and the other is from the Middle East. Being bright and ambitious, they had the opportunity to pursue academic study in an English-speaking institution, married, became academics, moved to the UK and then had children. These children are trilingual, but you can probably guess which they speak most of, given that the parents also have English in common. Or another friend, D who is English, and her partner, who is from one of those ex-Empire places where a local language is present, but there’s some officialdom done in English, and elites that use English. Their children have also been in the UK since birth. Another family, arriving in the UK as students, learnt to operate in English in high level jobs, but still speaking Urdu at home, then had a child who went to the local nursery and then to primary school.

Each of these children is noted as an ‘English as Additional Language’ pupil, in school returns and assessments. None of them, however, have any trouble with their English. Like the category of Pupil Premium (or Free School Meals), the administrative category is a crude measure of the extra barriers a child may have, and consequently a crude measure of additional assistance they might need. None of the children above need extra help, whereas a child who arrives alone aged 14, seeking asylum and with little English, is going to need a whole lot of help (for more than one reason).
​
The problem here is that schools have a double incentive to make the category inclusive. First, there’s any funding associated with the EAL tag: it’s not a lot, but all money counts. More important, though, is the assessment of schools through various attainment or progress statistics. The aim, from government and via Ofsted, is to ensure children with EAL don’t fall behind and stay behind, and so schools get feedback on the difference between the EAL cohort and the others (presented in various aggregate formats). A school does well if the EAL cohort gets similar SATS grades to the other kids, and this is highly likely if a good proportion of those with an EAL tag are the children of professionals with better English (especially re. vocabulary) than many native speakers.

Indeed, we know that the vocabulary of the most deprived children is much lower than that of better off children, and that vocab aged 5 is the best predictor of whether deprived children can ‘buck the trend’ and eventually escape poverty (Blanden 2006). This suggests that some of those kids who don’t get a great deal of communication may well have similar problems, even if parents are native speakers, and this is why we have Pupil Premium. I think there is an argument for rolling the two together.

Like Pupil Premium, then, EAL statistics should be taken with a pinch of salt. Some schools think it gives others an unfair advantage in the data, as some pupils are noted as unlikely to do well and then their (actually likely) success bumps them up. But there will be other schools, for one reason or another, who will get a cohort of EAL children with far bigger problems, and they will be below average on this particular measure. Few of these statistics come with the kind of comparison that really helps (see EEF for an exception): not only does the kind of EAL matter, but also any other barriers that are also present. I haven’t seen any research that matches each EAL child to an equivalent with the same other problems and attributes (e.g. social background, learning difficulties, school type) AND distinguishes between the level of the EAL barrier too. And, of course, what the statistics have never shown is what would happen if the EAL assistance wasn’t there.

That said, in this instance schools are likely to direct help where it is most needed. If a child tagged EAL doesn’t need help anymore (or never did) it is holding them back to insist on language lessons. Presumably, this is also the case for Pupil Premium, but because the funding is ring-fenced (see Allen 2018), it creates an incentive for doing extra work with children who don’t need it, while restricting how much can be done for those who do.

Discussions about the ‘white working class’ come and go, and being part of (or, at least from) that demographic I take an interest especially when they becomes culturalist and start to blame people for their own misfortune/failure. The debate about schooling, seemingly about whether ‘white working class boys’ (me again) are at the bottom of the pile, is ripe for some rethinking that takes in process and a critical realist take on causation. Counterintuitive it may at first appear, but it is to be expected that if:

• the employment system is more racist than the education system, and
• parental income decides free school meal (FSM) eligibility, but other parental attributes influence school success, then
• white FSM kids (me again, again) will do less well as a cohort than the other FSM kids.

 
I recently had the pleasure of hearing David Gillborn speak, and was reminded once again of the slippage in a lot of talk of the white working class and academic underachievement. Two parts are important here. First, we should note how some of this debate posits ‘white working class’ as an ethnicity, which could usefully be compared to all black Britons, as opposed to ‘black working class’. Regardless of how we want to think the status of ethnicity and culture – and I find these concepts problematic for such work because of their fuzziness and their not being a straightforward causal independent variable – this approach doesn’t help us untangle the various constituent parts. Second, we should also note that FSM status is not the same as ‘working class’, but is an administrative status with specific causes.

This, however, made me think about the problems of comparing even the free school meal cohorts that are noted as, on the one hand ‘white’, and on the other ‘BME’ or any subgroups. After all, for any robust comparison we should be looking at samples that differ only according by white v BME, or only FSM v non-FSM, and that’s something we can’t be sure of. Indeed, there are good reasons for believing that they are not. In this particular case, it only takes a racist employment environment to make these FSM groups not properly comparable. There are complications we could add to this – thinking about who migrates, for instance – but these are unnecessary. What we are studying, in these comparisons, is groups of pupils that are sorted by a prior process, that is the way that some pupils get FSM and some do not.

To get free school meals, a family has to have a low income. Indeed, under the Universal Credit system it will be less than £7400 per year after tax, before benefits. In essence, you can get FSM if income is so low that the state is making a benefit payment due to having a lower income than it believes people can live on. Now remember that unemployment rates are higher for BME people, even if they are graduates. Sadly, this data lumps all BME people together, while in reality there are differences across the groups, but this is enough for this argument.

This racism in the employment system means (all other things being equal) that the parental background of BME free school meals will be skewed towards a slightly better educated cohort. A higher proportion of those in a ‘BME unemployed’ group will be graduates than is the case in the white unemployed group. Therefore, if the education of parents has any influence (through whatever mechanism) on how well kids do at school, then the overrepresentation of graduates in the unemployed BME parents means that it is to be expected that their cohort of kids will have better results than that of the unemployed white parents. Indeed, given that one named route to FSM is that of asylum seeker benefits (support under Part VI of the Immigration and Asylum Act 1999) and many seeking asylum are highly educated but not allowed to work, then this relationship will be even stronger.

​ 
Of course, what this approach also suggests is more thinking not only about the process of who is in the free school meals cohort, but also what might cause them to do worse at school, and so what can be done about it. This can go hand-in-hand with thinking about why other groups do worse at school too: it’s pointed out in the official statistics that ‘the Gypsy/Roma ethnic group had the lowest percentage of pupils achieving A* to C in English and Maths GCSE (10% did so)’ but looking at the tables you find that FSM/non-FSM makes little difference here. Apart from in that case, FSM seems to always make a big difference in a negative direction, and makes most difference for White Irish pupils. Apart from Irish travellers/Roma/Gypsy, it’s the Black Caribbean cohort that has the lowest percentage.

What I find most disheartening about the public debate on this is that these ‘social facts’ are then taken as read, with the combination of ethnic identity, alongside FSM or class, given a more or less causal status that can be responded to as is. As in much work on extremism, the area I normally work on, this approach sometimes end up saying it’s about ‘their’ culture. Most obviously, if it is just about culture then why do many pupils still succeed despite being members of these cohorts. Those gaps described above are cohort figures, not individuals. As Becky Allen writes about the Pupil Premium – the extra per-pupil funding schools get alongside the free school meals – the targeting is crude because it uses the demographic indicators as opposed to checking who has more barriers to education and what they are. There is plenty of research on this, and a long list of factors including: parents that are not engaged with education because of their own experience; drugs and alcohol issues; a lack of settling because of school moves; poverty that makes getting to school with the right kit harder; teachers with low expectations of particular pupils; other priorities; assumptions that education isn’t needed for the expected job; and so on and so on. Perhaps it is time to reread Paul Willis’ Learning to Labour.

None of these factors are about ‘whiteness’, ‘working classness’, ‘blackness’ per se. A large part of the real story is that the pupils whose parents (for whatever reason) don’t invest as much in education tend to do less well. This is something we can address by identifying this, and working on it… Becky Allen’s approach is even more specific:
‘Categorising students as a means of allocating resources in schools is very sensible, if done along educationally meaningful lines (e.g. the group who do not read at home with their parents; the group who cannot write fluently; the group who are frequently late to school)’

Whether ‘reciprocal radicalisation’, ‘cumulative extremism’ or ‘interactive escalation’, a key question for policy makers as well as researchers is how new people get drawn into conflict. After all, at any one moment the number of British citizens or residents actively involved in extremism (whether violent or non-violent) is a miniscule proportion of wider society.

In my own research, where I spent time doing ethnographic and interview research with radical Islamist (mostly al-Muhajiroun associated) and far-right (BNP, EDL, and others), there were around 20 or so committed activists in total, with a few dozen other less so, in a city of a quarter of a million. I used to think that the solution to the choreographed confrontations between the EDL and al-Muhajiroun, was to send the small number of people involved to an island to fight it out. This would leave the rest of us, probably 99.9% of the population, to get on with life, rubbing together in a messy multiculturalism. However, at face value it appears that theirconflict is connected in some way to wider social conflict: the question is how?

I have come to an answer to this by thinking about radicalisation journeys, and particularly by thinking about what I and Phil Edwards have described as ‘microradicalisations’. This terminology is used because state and society (and, of course, research) is not only concerned with the change in behaviour from being an ‘upstanding’, ‘normal’ and entirely peaceable member of society to a lawbreaking ‘extremist’, as though these were two mutually exclusive and homogenous categories. As the term escalation implies, the processes we are examining include small conflicts getting bigger and as such includes activity that is not subject to legal constraint or response. In much the same way, two children bickering or adults arguing in the street is not illegal, but it could escalate into something that is. Furthermore, given the role of state and society in reacting against extremism whether through action or rhetoric, where this takes the form of fear, anger and a need to ‘do something’, then this is itself an escalation of conflict, another microradicalisation.

This is not to say that such escalations are inevitable. Most low-level conflicts can stay small or blow over, with escalation and de-escalation continuing as an iterative process. As Marc Sageman argues, “do not overreact… [many will] move on with their lives”: even in the world of the far-right and radical Islamism, it is possible to ‘flirt’ with ideas and behaviours that the rest of us find distasteful and perhaps dangerous, and then leave them behind. Furthermore, we can also include here a far wider range of conflict behaviours, including those based in gangs and territories, anti-authority sentiment, and even generational and other family disputes.

Here, then, I argue that the overbearing nature of a narrative that connects small scale extremisms, travelling jihadis, a ‘clash of civilisations’ and communal division, means that every instance of any conflict can be interpreted as contributing to this particular conflict. This is ever more the case when individuals’ beliefs and activities are explained in this common sense narrative as being attributable to community membership. Thus, in this model, one person is homophobic because they are Muslim, another is racist because they are white and working class, whereas the middle-class racist or homophobe is this way for their own individual reasons. Further, society’s response, including academic analysis, can support this narrative and generate more conflict.

I interviewed K in 2010, after he had come to the attention of the police for radical Islamist activism, but before terrorist activity for which he received a lengthy prison sentence. He told me he hadn’t been particularly religious as a child, and it did not sound like his family was either. However, he was a naughty kid, often in trouble with teachers at the comprehensive that was very working class, with mainly white British and a sizable minority of pupils with Asian (mainly Pakistani) backgrounds. I saw the beginnings of his Islamist identity in a blow-up with teachers, after he told me:

“I was fasting that day. That is the only day I fast. Like I wasn‘t practicing but I, I really wanted to fast this day. I woke up in the morning to fast and I said I‘m never, not going to break my fast. No. I said to my teacher no I said to her I‘m fasting I don‘t want to… We were making cakes or something. She said you have to make it. I said I don‘t want to you know what I mean bro? I don‘t want to. Because I‘m fasting. She said go to your own country and fast. Go there and do the fasting. I got angry. I got angry. I was an angry kid them days. I started throwing stuff. I got very, very angry. Then the head teacher came and like the senior head came in and they all… I said she said go to my own country. And why didn‘t they say nothing to her because she was an authority.”

Here, K’s microradicalisation is not informed by a radical Islamist ideology, but is rooted in a fairly unformed Islamic identity mixed with the anti-school and anti-authority sentiment common in working-class boys, echoing Paul Willis’ Learning to Labour. Teachers, having a similarly unformed Islamophobia or national chauvinism responded in a way that made the Islam/Britain relationship into a dichotomy.
​
In each direction, then, behaviour and attitudes that are not necessarily components of extremism or communal conflict are interpreted as though they were. England flags could be a sign of the world cup, or because the residents inside want immigrants to go home. Hijab could be a fashion statement, or a statement of religious superiority. As the dominant explanation of extremisms is that they are outgrowths of particular ethno-religious, or other ‘communities’ – a problematic term that could mean anything from the set of people sharing one attribute, to a set of people sharing everything – then small and unproblematic cultural forms are then interpreted as the pre-cursors to the most serious problems. Our current obsession with risk (reduction) makes this ever more important. Even young Asian boys playing football in a war memorial park can be interpreted as part of the conflict, even if they themselves do not see it that way, as they are seen as exemplars of a culture (Islam) that does not respect another (Britishness).
​
Thus, those that are genuinely part of extremist subcultures are then seen as a part of a threat that is imagined as those that belong to those ‘communities’ they are stereotyped as being rooted in. The threat is therefore magnified from a few thousand to few million. This, I argue, is how the accumulation of conflict operates, with the white noise of a potentially infinite number of microradicalisations, some connected and many not, being crystallised into communal conflict, as they are all examined through the lens of a ‘clash of civilisations’, or its local equivalent in ‘community cohesion’.

In an idle moment at work the other day, I picked up a recent issue of a high-ranking journal. I knew a few of the contributors, so had a read, and found my own words reproduced. I’d been a research participant a few years ago, interviewed in the home, and long ago lost touch with the researcher. Now the article had finally been published, and I found that it misrepresented me in fairly serious ways, and if it hadn’t been anonymous I’d have been pretty cross. More important, though, this raises questions about how sociological research makes claims to knowledge, how participants’ words are interpreted by researchers, and how it is difficult to understand someone or something without a huge amount of contextual knowledge, something we can’t have now because we don’t have the time.

I am not going to quote from the article, or even provide a reference. First, this is in order to preserve the anonymity of participants. Second, because it would not be fair to the researcher, who only has the spoken words and his own theory of the world to understand it. As you’d expect, the work was on the things I have an interest in – why else get involved? – and covered aspects of locality, class and leisure. The misrepresentations were related to class and local knowledge. The first came in an argument about class that assumed I was a particular kind of middle-class person, and so disconnected from other currents of knowledge and culture. The second was related, in that it was arguing that I’d made a claim of knowledge about the local area that was not true. Because it’s about Stoke, this latter part is what felt most like a personal affront.

Indeed, this second argument seemed to have come from a basic misunderstanding of what I was talking about, which can be attributed to the researcher’s shorter-term engagement with the city than my own. So while I talked of the decline of a particular street in town as being a recent phenomenon, the researcher felt I was wrong because to him it had always been full of empty shops.* However, my use of the term recent meant the last two decades, whereas it seems he thought I meant in the last few years.

The first argument is much more deep, however, and comes from what is not said as well as what is said. As I’ve pointed out recently, I’m now firmly middle class but certainly wasn’t when I grew up in one of the most deprived neighbourhoods in a very deprived city. So while on the one hand I know have taken on some very middle-class ways, on the other I remain connected – in numerous ways – to the working-class environment I grew up in. Depending on the audience, I have often presented very different faces and in some circumstances I would not talk about any of this. I don’t remember the interview, but I doubt I will have talked about my upbringing or family. What I said in the interview was written up as a middle-class snob, being disparaging of a local working-class culture that I know nothing of, the very stereotype of the privileged not understanding what it is like not to be privileged. When I showed this to friends and family, the response was ‘that doesn’t sound like you at all’.

I presume the quotes attributed to me are accurate reflections of the words, although I know from experience that small errors made by transcribers (especially missing words) can make a great difference to the meaning. Even if accurate, though, the quotes are partial. An assessment made on other quotes from the interview would be different. On another day, I might have formulated my thoughts into words differently.

From some perspectives this does not matter. If the words were said they were said, and they are part of, and evidence for the way that ‘we’ represent what is going on ‘out there’. So whether the words express my beliefs or not, they can be seen as connected to a societal discourse on class. This, however, makes sociology merely journalism with theory. The research that is funded, the questions that are asked, and the use of parts of the interview transcript are chosen because they fit with sociology’s (and at least part of wider society’s) theories about the way the world works. I’m certain I’m also guilty of this practice too. The problem is, another theory could just as easily be supported by other words, interviews, or individuals. Just as computer algorithms stand accused of taking a problematic social reality and making it solid, so sociology runs the risk of reinforcing its own theories by only using data that supports them. Despite claims of radicalism, some sociology is then deeply conservative as the writing is more reflective of researcher bias than the mess of social reality, and this bias reflects the traditions and backgrounds of the discipline and researchers.

So what should be done? There is no view from nowhere, and my biases are surely similar. A plurality of voices may help, as would openness in methods, data and analysis. Mostly, though, we need to pay more attention to context and process, not imagining each individual person, interview or statement as a ‘data point’ (unless under experimental conditions). This, though, would require different methods, somehow holistically seeing the wider picture. Sometimes I think historians have got it right – with a method of read, think, talk, repeat.

​*Example is changed to preserve anonymity

Recent talk of extremism and universities, the Prevent duty and so on reminded me of something I wrote back in April 2001. It feels odd now, especially given that some of the explanation of the online was done because it was all so new (imagine that!).... this first appeared in a Goldsmiths publication:

In the week of the first UK Holocaust Memorial Day I watched as a Goldsmiths’ student attempted to purchase a Hitler Youth Armband. Regrettably I did not attempt to talk to her and so at this point I can only guess her motives. I cannot also be fully sure of the reasons behind my own actions and inaction. The situation throws up questions of ethics, authority and political action. However, without talking to the student these questions cannot be answered.

I’m not a particularly nosy person. When the student sat next to me in the first floor computer room to do her email I noticed her red hair, boots and fishnet tights my only thought was that there are always new punks to replace the retirees… much the same as the goths. I took more notice of her when she was browsing the online catalogue of Data records, a Coventry record shop specialising in old and new punk. Working in a record shop myself and for a long time being responsible for this shops online catalogue I am always curious about competitors. So as I thought about planning my essay I took occasional glances at the next screen.

The student’s record browsing told me that her main interest was Skinhead and Oi!. Although the skinhead movement is widely perceived as being white, working class and racist this image has been contested since its beginnings. Many bands resent being categorised as racist and describe themselves as ‘anti-racist Oi!’. Apart from the notorious Skrewdriver the overt stance of most bands is either anti-racist or ambiguous. In addition, this straight translation of political affiliation from band to fan is problematic even if one is in the audience. This is also more often the case with fandom at a geographical or temporal distance. As with the current goths, many punks in London are from other countries in Europe and so do not have the same background influences as the stereotypical white, English, working class punk or skin.

It was what happened next that surprised me. The student began browsing listings of Nazi memorabilia on Ebay (www. ebay.com), an online auctioneer. She spent between thirty minutes and an hour browsing through photographs and descriptions of medals, armbands and signed photographs amongst other souvenirs of the Third Reich. At first it looked like youthful curiosity but all the way through the browsing the student was using a calculator to convert currency (into UK pounds or her ‘home’ country’s currency?). And then it happened. She saw a Hitler Youth Armband at the right price and made a bid of 15.50 for it. I have tried to think of explanations for this behaviour. Research? But why actually buy an item. Fashion? Medals maybe, but in what circumstances would you wear the armband? And it’s hard to see how signed photos of prominent Nazis could be a fashion accessory. As a prop for a play? She would have been looking for a specific item and not comparing prices of medals AND armbands. As part of a historical collection? Surely she would have looked at memorabilia from other wars.

It was the requirement for authenticity that really closes off these possibilities. For fashion purposes the painting of a swastika onto a ripped T-shirt would be enough. It is hard to imagine a situation in which this level of authenticity would be a requirement. A trip to an army and navy store would usually suffice.

At this point I felt a need to speak to her. But I also knew that in the library I do not have the authority to tell others what they can or cannot do. I also knew that as soon as she knew I was watching her behaviour would change. A warning would also enable her to find an explanation for her potential purchase. However, I needed to know that I could find out who this student was before leaving the library. In the end I decided to speak to whoever was manning the computer helpdesk. I was informed that there is an ethical policy regarding the use of Goldsmiths’ computers. He was sure that pornography was banned but did not know whether the purchasing of Nazi memorabilia was permitted (the purchasing of Nazi memorabilia is legal in the UK but banned in France, Germany, Austria and Italy). He also assured me that the records of who is logged on where and any network traffic is stored. I knew at this point that any conversation or confrontation could wait.

I left the library a little confused. I felt that I needed to talk to someone and ask advice about what to do next. I also felt disappointed that I did not take the opportunity to talk to her and would probably not follow it up with more research. In some ways that is why this is being written. As it is now in the open it compels me to finish the story, to find the student and discover her motives.

​Technical Note: When using the Internet your activities are recorded in many places. Your computer keeps a copy of the material (a cache) in order that it does not need to be retrieved again if it has not changed. Your web server at your Internet Service Provider (e.g. Goldsmiths or Freeserve) can log any page requests and may also keep a cache of data to speed up access to commonly requested pages. In addition the server that hosts the website keeps records of which computers have requested the pages. If I was browsing the BNP website from the Goldsmiths’ library the creator of the website would have a record of goldsmiths.ac.uk requesting pages. In commercial circles this is used to see how popular the site is and from where people are looking. Of course the BNP webmaster has no way of telling whether the user was a supporter, an enemy or just indifferent but even by browsing the site you inform him that he has had an influence. 

​I enjoyed Catriona Paisey and Betty Wu's article in the Conversation, which argues that the gender pay gap is not the only discriminatory inequality that government should look to examine and perhaps address. They are right that unequal distributions of goods that correlate to attributes that should not affect worth (e.g. gender, ethnicity, originating social class) are evidence of some form of problem. It seems likely that, for whatever reasons, the economy does give some types of people better chances than others. That said, these approaches rarely open the can of worms which is what we think about inequality in general. 

Much of the politics in this area seems to be based on a form of luck egalitarianism. That is, because the individual can't be held responsible for whether they were born male or female, white or brown, and so on, then they shouldn't get any detriment. This is opposed to 'desert', where society accepts that someone who works hard should be rewarded more than someone who is lazy. So, the argument goes, we should try to do something to equalise distributions based on the former, but not on the latter.

However, what do we count as luck? Is being born clever luck? Is having parents who read to you at night luck? Is choosing an interest, aged 10, that happens to coincide with what the economy wants 20 years later luck? In the end, almost everything might have an element of luck. 

Hence my response to the article, below. I don't think we can iron out all effects of luck as it is everywhere. Therefore, it makes more sense to address inequality more generally, which will have the effect of reducing all the gaps we are currently concerned about, as well as ones we ought be concerned about...

The other problem with such analysis is that in analysing a gap between the average incomes of two definable groups, we then stop being interested in the incomes distributions within these groups or, by extension, within the group of ‘all humans’.
Indeed, such gaps appear in a myriad of ways. Average income is lower for short men, higher for women considered good looking, lower with increasing distance from capital cities and, of course, lower in the global South. Average income is higher for the better educated, lower for the young, and so on.
By looking at averages, we would find the gap changes by the substitution of as few as one person. On a global scale, if the 8 richest men were actually women, then the global gender pay gap would reverse as these 8 would have the income of 50% of the world population. In a company, the pay gap would disappear if one man and one woman at the top earned millions, and the rest earned an equal pittance. Would it really create more justice if a few women at the top meant that an average pay gap did not exist, even though the women at the bottom could still be earning less than the lowest paid man?
There is plenty of political philosophy work that addresses these issues, asking: how much luck or background should impact on livelihood and which reasons should be allowed to be relevant to livelihood; whether policy should prioritise raising the quality of life at the bottom or attempt to promote equality; and how equality of opportunity differs from equality and so on. I recommend starting with Anne Philips’ article from 2006 (http://onlinelibrary.wiley.com/doi/10.1111/j.1467-9760.2006.00241.x/full) in which she argues that:
‘Majority opinion currently favours something more than the minimal—anti-discrimination—interpretation of equality of opportunity, and looks to a more substantive neutralising of the effects of social background. Yet no government is going to introduce measures that would genuinely neutralise these effects… Equalising opportunities means equalising our chances of the good things in life, but almost by definition, leaves untouched the distribution of rewards between ‘good’ and ‘bad’… They therefore leave untouched the really big questions about inequality.’
While all the gaps between groups described here are problematic, they are all far smaller than the gap between the average of the bottom 50% and the average of the top 50%, whether assessed globally, nationally or, I would assume, within most organisations. And the great thing is, dramatically reducing overall inequality would reduce the gaps between groups, whereas the converse would not be true (see John Hills et al. https://eprints.lse.ac.uk/28344/1/CASEreport60.pdf).

Picture: Cynthia/ Flickr (CC BY-NC-ND 2.0​)

I love analysing social science statistics, especially when they’ve been in the hands of PR professionals and journalists. I think it’s probably because errors are easily disputed and easily traceable, so we can see how social knowledge is created and transmitted: tracking back sources is usually easier as one can search for specific ‘facts’, and where those ‘facts’ feel wrong one can search out the original data. Here I’ll discuss two of this weeks stories, one on burglary and one on university entrance.
The ‘burglary’ story caught my eye because the hotspots seemed unlikely (Chorlton-cum-Hardy, Blackheath) and there seemed to be some slippage in what the story was about. The story is originally from moneysupermarket.com(so it’s a PR story) who analysed their insurance inquiries. So the first important observation is that the data is from a) people using the internet, to b) renew their insurance, and c) doing some shopping around. This is a biased sample in lots of ways: obviously they are people who buy online and have insurance (around 10% of people aren’t covered, presumably the poorest). That they are shopping around suggests that they have made a claim recently too. It’s not hard to imagine that this biases the sample towards 25-34 year old professionals.
Furthermore, the original press release gives away the slippage. Although it has ‘home theft hotspot’ in the title, and spokesperson quotes like ”Home is where the heart is and there’s no denying that having it burgled is an emotional and frightening experience”, the data refers to ‘claim[s] for theft on home insurance’. So this includes muggings, thefts from the beach, bikes being stolen, lost wallets on the bus and so on. Again, this is probably biased to young professionals: the kind of people who get their phones nicked in the pub, or have their bike nicked from outside.
So to education.
‘Just 1% of poorest students go to Oxbridge’ cries the Guardian. ‘In contrast, 10,827 students attending Liverpool John Moores University and the University of East London claimed a full bursary – 4.7% of the total for the whole country’
This is statistical nonsense. Without knowing how many students in total go to each university we can’t know if these figures are better or worse than average, or bang on. Perhaps 10% of all students go to Liverpool John Moores or UEL, so they aren’t attracting as many poor students as they should. We also don’t know if there are other of the poorest students who are attending and not counted as they get other forms of support or none at all, but I’ll let that go.
The original report has the more important figure, ‘the proportion of full fee-paying students this number [those receiving bursaries in the lowest income group] represents’. This in itself is problematic as bursaries are only for UK students, and I don’t know if the ‘full fee-paying students’ include EU and overseas students: I suspect not, but if so then those with more non-UK students will have a lower percentage.
Anyhow, the figure for Cambridge of full-bursary over students is 11.1%. This is low, but some are lower: Grennwich – 9.1%, Guildhall – 11%, Leeds Met 0.4% (very strange, I need to look into this). UEL is high at 61% but LJM isn’t particularly high… it’s just a big university.

​Railway Suicides
As a railway commuter, I have experienced multiple delays due to suicides, and enough to get some sense of their volume and patterns. The one time I was on the train that hit the person, there was an amazing flourishing of a sense of community, with people hugging the driver and conductor when we got off at Manchester Piccadilly. I talked to the conductor a little, while we were waiting, with discussion of the time patterns (Mondays, the week after Valentine’s day, the end of January) and the potential causes for this, which include the arrival of bank statements and credit card bills, broken relationships, jobs and so on.
While thinking about this, I did hunt out some statistics, only to find more of the Guardian’s failures to really grapple with numerical data. A datablog on railway suicide has some representations of data with arguments about what is going on, almost all of them wrong.
The first is a chart showing an increase in railway suicides over a 10-year period, alongside data on passenger journeys. I’m not even sure any increase isn’t just random variation: because the numbers are so small, what looks like a big percentage change is actually what we’d expect annual variation to look like. Saying that there’s been a 17% jump from 2010 to 2011 ignores the fall of a similar size from 2009 to 2010. And I’m not sure what the passenger numbers have to do with suicides: people of all sorts can go to a station or more usually a bend where the driver won’t see them to commit suicide, it’s not just a subset of passengers.
The data does show ‘Fewer suicides on the underground’, and this is argued to be ‘more indicative of lower passenger numbers than station safety’. However, passenger numbers on the underground are at the same order of magnitude as on the national rail (1.4bn v 1.6bn) so this would not explain suicide on the tube being a tenth of that on overground rail. Again, I’m not sure what passenger numbers have to do with it: suicides come from the population of humans, not from the population of train commuters or tube riders (although there could be a correlation due to confounding variables). More likely is the design of the tube system. On the national rail system, it is possible to climb a fence and get onto the track, somewhere where a train cannot stop, or to step off the end of a platform where a train is passing through. All tube trains stop at all stations, and most of the time we can’t get to the track in between stations, so attempting to commit suicide on the tube is far more difficult.
The final chart titled ‘One of the highest rates in Europe’ does not show this at all. Yes, we are near the top of the chart, with more railway suicides than all EU nations bar Germany and France. But this is because we are one of the most populous nations. ‘The UK's rail network accounted for around 7% of all suicides that took place on EU rail’ but this needs to be put into the context of having just under 9% of the EU population. A rate needs to be worked out as suicides per population. We’re mid-table here, with the Catholic countries lower than us, but many others higher.

I have long considered writing some autobiographical notes on social mobility, as my own path and position is relatively unusual and demonstrates some seldom made points on gender, the multi-factoral nature of the process, and generational effects. However, this has been prompted by a holiday in Italy, in which it seems I have definitely become middle class, a conversation there with a Somalia-born British journalist (Ismail Einashe), and then finding the BBC radio documentary on this topic by Hashi Mohamed, who coincidentally is another Somalia-rooted Londoner who has made it into a very middle-class profession.

Even on the journey to Italy, I had been thinking about how this was a very middle-class trip. It wasn’t that I was flying Business class, or even with a non-budget airline, but what I was to do once I was there. Not only was I going to be in Naples, with the obligatory visits to the archaeological museum, Pompeii and Herculaneum, and which of course were great: I was also dropping into an academic conference and then hiring a car to get to small town where I could meet a friend who grew up there. Thus, this holiday was also generating the cultural and social capital that we know helps ‘middle class kids get middle class jobs’.  I even got to chat to a prominent British sociologist who happened to be at the conference, and who has edited two books on migration with people I know from my time in London. I don’t think this is the end of any journey, but I started life very working class and this change has been a gradual accretion of various attributes.

I was talking to Ismail about this, and he mentioned his piece on Britishness for the Guardian, with its focus on belonging and second-class citizenship, the latter being a trope I’ve used in academic work in the past. Indeed, in other work I have written about the process of becoming British that does take in more than the moment of citizenship. What caught my eye in this, though, was the mention of the passport as the end of the process and his recollections of schooldays:

As the end of term approached, my classmates would ask where I was going on holiday. I would tell them, “Nowhere”, adding, “I don’t have a passport”

At sixth form, doing a History A-level, and armed with citizenship and so passport, he was then able to go on the school trip to Versailles. This, I replied to Ismail, made him a few years younger than me when I got my passport and first went out of the UK.

So why was I 20 when I got my first passport? I had the right status as a citizen, but in my family and wider circle there was no history of going abroad: I didn’t think it was something one did, I hadn’t the confidence, I didn’t know how to. This is another attribute that can be a marker of class: like the kids I remember seeing in a documentary about Newcastle-upon-Tyne, I’d barely been out of Stoke-on-Trent for more than a day or so in my life up to the age of 17, other than for Stoke City away games, and had never been to London. The documentary by Hashi Mohamed, now a barrister, mentioned other things, including experience of enjoying and attending classical music and sport events, drinking fine wine and more, that are part of how people ‘fit in’ and so reproduce the same types of people. He also mentioned how particular opportunities for work experience would be available for the children of barristers’ friends. I recommend this documentary as an introduction to the ‘unwritten rules you must learn to get a top job’, but with the caveat that this is true of far more opportunities, with the ‘top jobs’ being some of the most desirable. Me, I spent most of my childhood in a street found to be in the top 1/3 of 1% in the index of multiple deprivation (i.e. 299/300 other areas are less deprived). I’m the only person in my family not to leave school at 16, and all my grandparents and both parents worked in the pottery industry. I also had my education disrupted by family breakup, and was eventually in a household that on today’s standards would be a Troubled Family due to crime, long-term ‘out of work benefits’ and more. I’ve seen alcoholism and violence at first hand. Just as important, though, was the fact that our social networks did not include anyone outside Stoke, and I knew no one that had been away to university, nor anyone who had a job that was not ‘skilled manual’ (except for one I’ll come on to). I didn’t know anything of classical music, or great literature. I didn’t know of the unwritten rules, or even when they might come into play.

Avoiding ‘certain fates’ is not just about talent and hard work. Ismail’s turning point was the move to a high performing sixth form, with a diverse range of students. Hashi Mohamed’s turning points came courtesy of Newsnight and a letter asking for work experience, and then a subsequent personal introduction to someone who could help arrange the pupillage. My turning points came both earlier and later, but also involved my mum, who’s a great example of how social mobility isn’t just about improving education and getting a middle-class job. Furthermore, this complicates my story, because the disadvantages described above were coterminous with a couple of serious advantages. The headline complication is that, while my childhood home and where my mum still lives, is one of the most deprived neighbourhoods in the land, when she moved there my mum had just gained a PhD in mathematics.

So, yes, my mum grew up in a council house, with parents who both worked in and were killed by the pottery industry, and she did work in the pots too. She got into a grammar school at 11, left at 16 and worked in industry, married, had kids and divorced. But while me and my siblings were small, she did a maths degree with the Open University, and then the doctorate at Keele. After divorce, she became a maths teacher at an ex-grammar that had changed to fee paying, but this was after I’d passed the 11-plus and been offered a place at the same school under the 1980s ‘assisted places scheme’, which enabled kids with poor parents to get fees paid by the government. So what I did gain was educational capital of straight As in science/ maths A-levels and then a Cambridge degree in Natural Sciences, and this paralleled my mum’s educational capital. However, neither of us had the kinds of social or cultural capital, or the being comfortable in middle-class environs, that would often go along with this. My mum still lives on a council estate, just a mile or so from the one she grew up on. By some measures, she has not been socially mobile at all, despite the PhD and teaching job. Perhaps it was a gender issue for a particular generation, certainly a parent/ family issue, but the moving out and moving on just didn’t happen.

On the other hand, I stayed at school post-16 and went away to university at 18. At that point I didn’t have much of the transferable cultural and social capital, being still a working class kid into house music and football, and knowing little of the world. But I’ve been able to build on that, and much of it because of the need to make new friends when leaving a home town, and the diversity that comes with doing this in Cambridge and then London (itself geographic capital, perhaps). In the end, after not totally fitting in at school, and moving into another world at university, I’ve become comfortable in middle-class worlds. And as a consequence, I’m doing the next stage of the social mobility, where I move in very different circles to the rest of my family and do the omnivorous cultural consumption of the group I’m now in - ‘Established middle class’ - according to the Great British Class Survey. Knowing this, any denial that I’m middle class would be absurd.

While I’m all in favour of cleaner air, and wish I hadn’t got a diesel for a second vehicle, I’m not sure that the Labour party’s claim that ‘nearly 40 million people in the UK are living in areas where illegal levels of air pollution from diesel vehicles risk damaging their health’ (Guardian) is particularly meaningful. This is both on the grounds of how the data they have is constructed, and also on how pollution affects real, mobile human beings.

The first part is partially down to this concept of ‘area’. As a counterpoint, I’ve seen it written that ‘99% of the UK meets European air quality standards’, and we know that this is because it is for geographical area as opposed to population. The 1% that does not meet the standard is where lots of people live. However, it seems that the Labour figures are merely a repetition of a parliamentary question from October. This produced a list of ‘Local Authorities with exceedances of NO2 annual mean limit value’ based on a 2015 exercise*. This exercise appears to have itself (at least where I live) on data from 2014, which comes from numerous sensors across the area.

However, in the original (LA level) report for my local authority we find that the NO2 average is exceeded at five locations in two smaller geographic areas inside the borough, with six sites in three locations coming close. This is out of 38 monitoring locations across the whole borough. Thus, for each area, the Labour analysis only tells us that at least one location went above the limit, but it could be just one or it could be all of them. I, therefore, count as one of the people living in an ‘area where illegal levels of air pollution from diesel vehicles risk damaging their health’, but I live in a relatively rural location quite a way from the nearest sensor that detected a breach. Indeed, given that sensors are placed next to main roads, and not by a random sample, the numbers of people living within a specified distance of a breach is also a function of the geography of the area. Indeed, what counts for health is not the indicator of the monitoring station, but how much pollution is present in each person’s air. Presumably there is some sort of diffusion gradient for the pollutants, and the breach figure somehow relates to the potential for danger to someone near to it, but also to the danger for everyone else living further away. Indeed, the breach figure for the sensor is 40µg/m3, but research shows that risk comes from breathing in lower concentrations: I guess the assumption is that no one is living on top of the sensor, and that people are living a distance from those highest polluted spots. Better research would assess the level of pollution in people’s homes.

There is, of course, some kind of diffusion gradient for the humans too. We don’t sit in our houses breathing the same body of air all the time, but we move about. I spend some time in places much nearer to the monitors with breaches, as that’s where the shops are, but not much. I also go to other cities and travel through more polluted areas. However, the relationship between the population’s exposure to pollution and the exceedance of the annual NO2 average is not clear: if the breach is in the town centre it is shoppers that will experience it, and if the breach is on a dual carriageway then it is drivers. Once again, the connection is down to the spatial configuration of the place.

​
What is likely, however, is a relationship between income/ wealth and pollution, which goes back a very long way. The configurations of some cities came about because of the way the wind blows: the least desirable places to live were those where the factory smoke blew. At that time a big house next to a main road was desirable, however. Now the pollution mainly comes from vehicles, a house next to the main road is far less desirable, and those who can afford it maintain a healthy distance from those sites with bad air.

Suffice it to say, I’m not expecting the plan that the government has been forced to publish to clear any of this up. As ever, one set of politicians want the problem to ‘just go away’, ideally for no cost, and the others want to use the data as the stick to beat the former.


*We could also ask why NO2 and not particulate pollution has become the focus of the debate. In the borough I discuss PM10 (particulates <=10microns) is measured at only one site, and annual average limits are not exceeded, which I presume gives a different story for politicians and activists. If it’s not measured, it doesn’t exist!