Conversation with Edmund Chattoe-Brown about agent based models

A little while ago I linked to an interesting article by Edmund Chattoe-Brown about  agent based models. It stimulated me to think about the conditions under which new methodological tools are adopted. Edmund got in touch and posed a few tough questions. I think we both found the dialogue useful and enlightening. We also thought that there might be at least a few people who would be interested in our conversation. So here it is, my part is in italics:

Dear Colin,

 Thanks very much for your kind mention of my article in your blog. I think I pretty much agree with you (which is why I have been trying to publish more in "subject" journals than simulation ones.) However, I have a couple of thoughts/comments: 

1)       What would you say, in your area (however you define it) that a reasonable number of sociologists agree can't be done but needs to be?

That is one hell of a question! I'm not sure it's possible for me to imagine anything but a tiny group of sociologists agreeing about anything. If I had to point to an area where it seems to me that there is already an active interest in the sorts of things that simulation can do, I would point to the interface between family sociology and social demography. I'm thinking principally of the work of people like Rob Mare and our own Francesco Billari. The kinds of things they are interested in tend to be about how observed macro-level demographic patterns can emerge from multiple micro-level processes with lots of endogeneity (ie "mediating variables"). My guess is that there are probably lots of connections here with social stratification, mobility, homogamy etc. The basic problem with applications to the latter is that we are still struggling to accurately describe what the basic patterns are.

2)       Although it is difficult to generalize reliably from one’s own experience, at least a couple of papers I have published might fall under: "publication in mainstream journals of a few articles reporting realistic applications to substantive problems that enough sociologists care about that tell us something believable and important that we didn’t know already." My BJS article "overturned" a result from previous analysis which in turn was based on an extensive empirical literature going back to the early seventies. (Are strict churches strong? Not in a properly dynamic environment rather than a simplified partial equilibrium model.) My BJC article showed that, for poor response rates (as you would expect in criminal networks for example), unreliable qualitative "third party" data might outperform quantitative data in reconstructing social networks. (Something that both the rather formal SNA and "real" users of network ideas, police forces, might want to know.) Now, of course, one can always argue that the problems one tackles could be "more substantive" or "more popular" than they are (and these articles were considered good enough to go in reasonable journals at least) but I'm not sure the response these papers have received is really in proportion to their "substantiveness".

It seems to me you are doing the right thing and one has to live in hope that if the message is read and received by the right people and they realize that it will meet their needs then they will take it up. Build it and they will come, if they have any need for it. My guess is that road to Damascus conversions are rather rare. The basic ideas about log-linear modelling had been knocking about in the early 1960s, but it wasn't really until the late 1960s early 1970s that it got taken up in the bio-medical field and it was only in the late 1970s that it really was first introduced into the sociological mainstream. Undoubtedly a big impetus came from the dissemination of Goodman's (relatively) user friendly ECTA program. What is clear is that publication in methodological ghettos, SM, SM&R etc doesn't necessarily reach the right audience. Also one shouldn't underrate the role of arbitrage. There are some people (I won't name them!) who specialize in "translating" the technical innovations from one field into another. They are often very good at picking examples that will sell.

3) Ages ago, Robert Andersen asked me why simulation and statistical data had so much trouble "getting on". Like all good questions I have been thinking about it on and off ever since. The other day, when trying to calibrate a simulation of attitude change, I had a sudden (minor) epiphany. I needed to know (very roughly) how often people discuss political matters. One large reputable survey asks people to report ("never", "rarely", "sometimes", "often") and another, equally large and reputable reports ("daily", "several times a week", "weekly", "monthly".) Both are perfectly OK if you want to look at statistical association but one is completely useless if you want to model an underlying process. I knew I wasn't imagining that there is more to these issues that "just" data!

I agree and would go even further. I'm not even sure the use of vague quantifiers is that enlightening in  social survey applications without some serious attempt being made to understand how sub-populations understand the category labels. There are ways of modeling this, for example Gary King has been a pioneer,  but there are powerful vested interests in the data collection world that sometimes inhibit sensible innovation. And if a behaviour is well defined and the unit of time is sensibly chosen, then I can't see why  we shouldn't attempt to get frequencies. Of course, it isn't always that straightforward. What is a political discussion? Is it a discrete countable event  with somewhat obvious boundaries like say visiting your GP's surgery? Then there are questions of time units, seasonality etc. And of course well known memory effects like telescoping.

4) You talk about simulation models that should be "reporting realistic applications". Statistics has many virtues but can its uses be assessed as realistic (or otherwise?)  To take another example I've just read. Simple regression makes normality assumptions on data. Sometimes one can "fix" data that fails to be suitably normal by logging the variable. For sure that solves the technical problem but what do we conclude about an association between something and "log age" (or age squared come to that). Is there a danger that every method seems "realistic" to its advocates and what we need are standards of realism that don't presume the virtues of a particular method?

I was thinking of "realistic" here as meaning something like "a realistic degree of complexity". I think - but this is just the impression of a possibly naive but sympathetic observer - that "toy applications" don't do much to persuade enthusiastic take-up. Another dimension of this problem would be to say that demonstrating that a given outcome could be produced in a particular (simplified) way, is not the same as demonstrating that it has in fact been produced in this way. Of course this is not a problem that is any way unique to simulation. If I think about the aggregate distribution of votes between parties at an election what tends to impress me is that the reality is that this is the result of lots of different decision processes that are going on in different sub-populations. Therefore to propose a single "theory of voting" is absurd. Some people are cogitating about the relative advantages to them of voting one way or another, others are just doing what they've always done or what their parents have done, some are voting strategically and so on. All these processes are going on at the same time to produce the aggregate outcome. In some sense an adequate model would try to capture this (and perhaps produce as a by-product some sort of estimate of the proportions involved).

In terms of statistics, the way I look at it is that statistical models are just smoothing devices that permit the estimation of some quantities that you happen to be interested in. How you go on to explain whatever patterns are revealed is quite another matter (and simulation has a big role to play here). This is, of course, not the standard econometric justification - structural parameters and all that.  If economists have believable models and sensible identification techniques then they should estimate their structural parameters. My feeling is that in sociology we usually are a long way from this position, not because our statistics are
no good, but more usually because either we don't have a good (precise enough) theory and/or because we don't have data that permit identification of what we are really interested in. Problems of, course remain even when we have identification - consider the classic experimental design. We have a well defined target and we have identification via randomization. But unless we have a good theory we also have a black box! Well, that discussion will take us off in another direction...

To which Edmund replied in a follow up email:

On 1: _That's_ why we simulators find it hard to build "generally appealing" models ... :) Interestingly, I certainly have demography on my list, particularly having read with interest this http://ideas.repec.org/a/bla/popdev/v37y2011i1p89-123.html) and seeing how statisticians, modellers (and even qualis) could have a debate around what each method can contribute to this specific problem and why each thinks that the other "hasn't got it". (One needs to convene a small "fair/broad minded" group to discuss.)

On 3: This shades into "hard core" simulation methodology (which even some simulators conveniently neglect). The biggest critique I have tried to make (still unpublished interestingly) is that the average simulation paper still has nothing to do with data (even when it is virtually free) and the "field" has forgotten several old papers (particularly Hagerstrand 1965 on spatial diffusion) which have higher standards. (Bad news for "science".) To do something useful with statistical data, a simulation doesn't necessarily have to have it exactly right (because it doesn't feed straight into parameter estimation). Obviously a model that is based on the idea that people talk about politics once a year not once a day almost certainly won't produce good data but with all the other social processes represented it may be that 1.5 times per week versus 1.8 times per week won't alter the "basic qualitative behaviour"of the system (like turning points versus trend). And sensitivity analysis can tell you roughly where it is most important your data be accurate even before you "get at" any real data. 

On 4: There's a lot in this response.

1) "Realistic degree of complexity" is hard to nail down. Qualis usually say that simulations are excessively formalistic and simplistic. Quants (particularly economists) say they are needlessly complex and ad hoc. This makes me laugh in seminars.) One argument I am trying out is that we need to distinguish clearly between complexity we think exists and complexity we can show _matters_. (Again, we need less "method embedded" ways to justify the claim that a model is too simple or not simple enough: "Too complicated for my taste" is not at all the same as "Too complicated".) For sure, ethnographers can tell us all sorts of things about, for example, family size aspirations (paper above) but how do we tell that these "add up to" a particular pattern of family size (or that  one couldn't do just as well with four key variables). Conversely, quantitative researchers can't usually show that process x (differing socially reproduced family norms for family size?) _doesn't_ affect outcomes because if they don't already have the data it is a huge faff to collect and, in any event, some reasonable causes just aren't very quantifiable. As far as I know, ABM is the only way that you can say "OK, we are now going to give our agents brains - or social networks or whatever - and see how much difference it makes". This is the "simulations as thought experiments" idea. (One thing I find thought provoking is that I think that Social Network Analysis fairly convincing that "networks matter" and yet social statistics - which almost never includes network variables - also seems to achieve sensible things. So is one approach "wrong" about the fact that "networks matter" or is it that the methods just aren't geared up to adjudicate on this? Perhaps networks _do_ matter but social statistics can lose these effects in lower R2 and error terms in ways which don't "ring any alarm bells" with practitioners.) On that score, what do you think of the statistics in: http://cumc.columbia.edu/dept/healthandsociety/events/documents/Haynie.pdf (network effects and delinquency.) 

2) Mixtures of agents with different decision making processes (including "no decision") are exactly something that ABM is good at. (But you have to watch to make sure you don't get good fit by just adjusting the fractions till the graphs match!) 

All the best,

Edmund

RIP standards in British empirical sociology

Hat tip to Mike Savage at Stratification and Culture Research Network for his rapid  and candid response to our questions about the Great British Class Survey data.

I'm rendered, temporarily, speechless by the implications of the answer to our second question.

Music, Culture and Distinction

As you can probably guess I've been spending time  reading the literature on cultural consumption. It's entirely possible that I have not been reading the right bits of it, but one of the things I'm missing in what I have been reading is any sense of process. To explain, forgive me if I lapse into the autobiographical again. 

For anyone born after, let's say 1900, there is one central fact about our cultural lives, the influence of  popular commercial culture delivered by the mass media: radio, cinema, TV, tabloid newspapers. This is part of everyone's cultural diet. It is the  given for everybody on  top of which other cultural tastes and preferences get added. You don't have to choose commercial culture, it is just there in the background all the time and there is bound to be some of it that you like at some point in your life. You simply can't avoid consuming it without going to enormous lengths (a couple I knew used to keep the TV in a cupboard so that they would have to make a positive choice to watch something, but they were in many respects quite exceptional). The question then becomes, how are the cultural omnivores created?

I don't think the answer is very deep: some are born to it and others have to seek it out. If there are alternatives to pop culture available in the family home then you have a wider range of things to choose from and some of them will stick. If, as in my case, there were few non commercial cultural resources then the problem becomes, how to find out what there is out there in the big wide world? You can only express preferences among the set of things you know about.

I learned nothing about music in school. For three years we had something called music lessons, but they were taught by a  borderline psychopath.  Mostly we sat in silence for fifty minutes because somebody had peeped their recorder at the wrong time causing the "teacher" to fall into an uncontrolled rage and the suspension of the lesson. In a rational system this individual would have been fired (or sectioned), but he saw his time out to retirement and strangled the musical experience of generations of pupils.

Another possible source of musical insight could have been church. If you are an Anglican then you have exposure to a rich fund of liturgical music. We though were non-conformists and although there was something called the church choir - a handful of warbling middle-aged ladies  - it didn't inspire me. I think there might once have been a performance by a visiting choir of that mainstay of middle-brow church music, Stainer's Crucifixion but that was about it.

As I've mentioned before Alan Freeman's Saturday afternoon radio show gave me an entre into a certain type of popular music that one wouldn't hear on the regular Radio 1 shows or more to the point Housewives' Choice which was what tended to be on in the background in our home. It seems to me that this and analogous shows functioned  like Alice's rabbit hole. You could fall down it and be taken to another world. I find this a persuasive argument for the retention of  nationwide public service broadcasting. It may be true that in our new digital media world there is so much more choice and opportunity, but existence of the material is not the same as having someone regularly pointing you in the right direction.

Also important for me were two  record stores within a 10 minute walk from my home. They were both essentially little corner shops with racks of vinyl, but two things were crucial. Firstly you could spend a long time  browsing and secondly they played music. I can still remember standing in one of them in 1977 and hearing the Clash's Career Opportunities. It was like nothing I'd ever heard before and I knew a new age had dawned! These guys were singing about me and kid's like me living in the grimy streets of English towns not in sunny California.

The local Carnegie library was another possible source of musical intelligence. They had just started to lend vinyl, though there was, at least for my pocket, a fairly hefty charge. I think I must have borrowed a few albums, but what I mostly remember is that every time I looked in the LP rack there was a copy of John and Beverley Martyn's Stormbringer that nobody ever seemed to borrow (perhaps somebody on the library staff thought they should support local girl Beverley Kutner!). I had no idea who John Martyn was or what kind of music he made. From the fact that nobody borrowed it I inferred that it must be rubbish! Without any sense of direction taste formation is just a process of serendipity.

By the time I got to university I thought I had popular music sorted. I also had a guitar and that focused my attention on a particular sort of acoustic music with the illusion that if I listened hard enough and practiced long enough I might be able to play like John Martyn. In fact my talent barely stretched to a ham fisted imitation of Loudon Wainwright III, but that is another story.

Classical music was still terra incognita. I literally hadn't a clue, but decided I wanted to find out, simply because, like Everest, it was there. So in the vacation between my second and third year I again went to our local library and borrowed vinyl more or less at random (I was now a bit richer as in those halcyon days students were entitled to social security during the Summer vacation and the idea of finding a job in Coventry in 1981 was, to put it mildly, ludicrous). This was an awfully big adventure, but if you don't know where you are going it's best just to set off and see where you end up. I still have a few of those recordings, copied (illicitly) to  cassette tape and now digitized. Some Mozart piano concertos, Purcell songs, Bach Violin Concertos and a brand new boxed set of Haydn string quartets. Somewhat bizarrely Haydn featured again that year in my musical education. The local Anglican church (a  fine 14th Century edifice) hosted an amateur chamber orchestra performance of Haydn's Seven Last Words and together with a couple of friends who were equally keen on self-improvement I went to my first classical concert.

And now another factor became important. Girls. In my social circle there were girls who either were studying at the music schools, were keen amateur performers or just knew about 'proper' music. This  introduced me to a social life in which going to concerts was just the natural thing to do. For a while I had a girlfriend who was a very good piano player and I have a fond recollection of going with her to a Murray Perahia concert at the newly opened Barbican. I was, of course, playing well out of my league, but it was fun while it lasted, my musical fumbling comically mirroring my sexual fumbling.

My interest in Jazz really came much later and was facilitated by one thing, having money in my pocket. Once I  had a job I had money to spare. I was renting a friend's apartment in a West London suburb right on the edge of civilization. It was a lonely place and a lonely time, but again I was lucky. The 'village' around which the suburb had grown had a record shop and a bookstore. The bookstore had a few books on jazz and the record store had a few recordings. Every Saturday, after doing my Waitrose shopping I had a punt and spent a few pounds on Miles Davis, Dexter Gordon and  Oscar Peterson.  I just bought on spec and found out by trial and error what I liked and didn't like. It made the week-end interesting. You could say it was the transformation of one sort of capital into another, but that would just be trite.

So I wasn't born a cultural omnivore. I had to choose to become one and I had to have lucky breaks and the financial resources to take chances. In retrospect what I was doing was simply trying to make my life more interesting, to add savour and variety to what I had inherited - the commercial popular culture that everyone gets as standard. It was all about having life and having it more abundantly. What it wasn't about was making cultural or social distinctions. I doubt if any of my friends, colleagues or acquaintances had the faintest idea of my musical tastes and I certainly knew little in detail of  what they listened to. As I by then was moving in a  middle-class environment where some kind of generalized omnivore taste was standard it is difficult to see what kind of status advantage I could possibly have gained from advertising my musical preferences. It would have been as odd as proclaiming to all and sundry a love of HP sauce. Which is not to say that sometimes you aren't startled by the cultural preferences of people you think you know, like the eminent and highly cultured professor of sociology who once told me that he couldn't see the point of poetry. I suppose it's mildly shocking, in the same way that I'd find it odd if somebody told me that they didn't like chocolate. But I didn't think any the worse of him.

Asking a few questions...

Me and some mates have been asking a couple of questions about the GBCS data over at Stratification and Culture Research Network. I'll let you know when we get the answers.

Musical consumption again

Desert Island Discs is, I suppose, one of the great guilty pleasures of the British middle-classes. We know  the format is contrived and the guest is putting on a face, but  it is still fascinating to be on the receiving end of this aural presentation of self. I've spent many a happy hour listening to programmes from the archive while doing various bits of house renovation and repair. And of course I've engaged in the fantasy of what I would  choose if I could take only eight discs to an island paradise. In fact though, my fantasy has mostly been stymied by my inability to discipline my choices. My problem is that I like too many different kinds of things, but my musical preferences depend on what I am doing at the time, which is, I suppose, the curse of using music as a wall-paper or soundtrack to everyday life. With this sort of problem, evading the issue by having multiple choices of eight differentiated by genre doesn't help. I can't just have eight hit lists and choose number one from each, because I don't know what I will actually be doing on my desert island, and for my manner of musical consumption that is important.

If I'm doing housework, hoovering or cleaning, it has to be, in the broadest sense, popular music  - rock, pop, MOR, perhaps soul or reggae. Anything that is lively, upbeat and  happy will do to get me through the drudgery. The same would have been true when I exercised regularly, but since I cut that out, the problem of what to listen to while jogging has been removed. The same kind of taste prevails when I'm driving and have got bored with or irritated by Radio 4. Rock, 1970s disco, Madonna, Prince... anything as long as it isn't soporific. This may be tempered by who else is in the car with me. If it is my daughter then her tastes will prevail.

At home I might  have music on at low volume in the background if I'm reading. String quartets, Byrd, Gibbon, Palestrina, lieder (as long as they aren't in English) or Webern, Schoenberg, Boulez, Wagner, things that are either soothing or so abstract they become part of the general background hum. Miles Davis, Duke Ellington, Art Pepper, Lester Young, Keith Jarrett, Pat Metheny, Benny Goodman, Brad Mehldau all work in a domestic setting. Armstrong, Beiderbecke, Coleman Hawkins, Charlie Parker and Bebop in general  demand too much careful attention to be doing something else at the same time, as does all symphonic music. I have lots of symphonic recordings, but I rarely listen to them. To appreciate symphonic music you have to listen to it at the right volume and that isn't appropriate in a small terraced house. If you put the headphones on you just have to sit and listen, which is great, but when I'm at home I usually have too many other things to do to devote time to just listening. This is, of course, a great reason to  make yourself go to symphonic concerts where you can't do anything else but concentrate on the music.

About the only time I have for really listening is when I am walking from one place to another and then I tend to listen to folk or singer songwriter type things, usually people singing and playing a guitar. Nic Jones, Dick Gaughan, Richard Thompson, Gillian Welch, Show of Hands. The interest is partly in trying to figure out how they are playing what I can hear, how they achieve the effect.
So, if somebody asks me whether I like or dislike a certain sort of music, my answer is: it depends.

Sociological Science is Live

Sociological Science is open for business and I for one wish them every success. What they are doing, or something like it,  is probably going to be the future of academic journal publishing. Their submission and publication fee schedule looks reasonable to me and is tailored to career stage which is as good a rule to go with as any. The only down side I can see is the requirement to prepare an accepted article in LaTex. I guess its fine if you've already made the up front investment  to learn it, but a bit off putting for people who have to start from scratch. I don't though have any better ideas about how to produce low cost uniform copy that doesn't depend on proprietary products.

Promises promises....

Interesting developments over at Stratification & Culture Research Network. Mike Savage has posted a statement outlining the future research agenda of the Great British Class Survey research group. I'm looking forward to the article he promises that will deal with (refute?) the arguments of the GBCS's critics. In fact I'm wondering how they are going to manage it within the constraints of the 8000 word limit that Sociology insists on. There is just so much to say in order to do justice to the points the critics raise. Still, I've made it easy for him to address my particular criticisms by listing them in the form of questions at the end of my critique and it will be easy for the interested reader to keep a tally of how many are actually seriously addressed.

Also of interest is more information about the archiving of the GBCS survey data. This will eventually be deposited at the UK Data Archive though we are asked to be patient because of the large amount to work that will have to be put into cleaning it. Fair enough, my guess is that there are lots of issues to do with data format and coding and half a million cases is quite a lot to chew on even if there aren't that many variables.

What is less understandable is the delay in archiving the GfK survey. This has only 1026 cases and (I assume, though we are not told this) about the same number of variables as the GBCS. There surely can't be any complex issues involved in archiving a data-set of this size and no reason to delay depositing it. I think I know a little about archiving data from a depositor point of view and quite a lot from a user point of view. To clean, document and archive a data-set of this size, even if it just consists of columns of asci  number, is about the work of a weekend for someone that knows what they are doing: let's be generous and say a week to allow for the unforeseen.

And the point is this. The class schema generated by the GBCS team is entirely dependent on the data from the GfK survey, not the data in the GBCS survey which plays no effective role in defining the GBCS class categories. The GfK survey is the foundation of the whole enterprise. Why delay in disseminating these data? The GBCS team may prefer us to believe that it is the GBCS data that people should be interested in, but these are, as I pointed out elsewhere, just the mountain of bad data sitting on the molehill of (relatively) good GfK data.

I can see the slightly ludicrous prospect of the GBCS team getting into print to "refute" the claims of the as yet unpublished critics who will have had no chance to actually examine the data on which the original set of claims was made. It seems to me that there is a serious disproportion between the amount of effort put into puffing the GBCS and the amount of effort put into facilitating the assessment of the science. I find it difficult to think of any reasons why the GfK data should not already be in the public domain.

Origins of musical taste

I've been thinking  about how I came to like the music I do. Despite being classified by the GBCS as a cultural univore, my tastes are, I think, quite omnivorous. I even like country and western, well, at least Hank Williams and Johnny Cash. Of course I know nothing whatsoever about the music that  young people, say under the age of  30, listen to, but then again life is short and choices have to be made. I guess my knowledge will improve a bit in 6 or 7 years time when my daughter becomes a teenager and realizes that a great way to annoy her parents is to play whatever she happens to be into at ear splitting, fight provoking, levels.

The musical consumption technology in my family home was, by the standard of the time, quite good. Round about 1972 my dad bought a hi-fi: Keltron speakers, Amstrad IC 2000 amplifier and Garrad SP25 Mk. IV deck. What was played on it was not high culture, it wasn't even contemporary popular culture - as I've mentioned before, my parents' tastes were really formed before the the sixties pop revolution. So we had a diet of maudlin ballads about old shep, bagpipe bands marching through the heather to scare the shit out of Johnny English (though as Billy Connolly once observed, you can't march through heather without falling on your arse), Mantovani, James Last, Val Doonican, Jim Reeves (my mother's favourite), Andy Stewart and non-cast recordings of family musicals like My Fair Lady and The Sound of Music picked up from the Woolworths bargain bin. I think we had two LPs of classical music. One was a Classics for Pleasure 1812/ William Tell overture combo. The other was Grieg's piano concerto. Why those? I have no idea. 

As a child I  had an overdose of Grieg. At primary school an ancient crackly recording of the Peer Gynt Suite was played every morning  as we filed into  assembly. I guess educational theorists recommended it as soothing. It is in fact one of the few things I can remember about my primary education, the others being  taking a penny for the "Boot Fund" box so that poor children could have shoes (amazingly this still exists) and my dad racing up to the school in his lunch hour to give the headmistress a row because another child had scribbled in my writing book and I had got the blame. After that the teacher hated me and shortly afterwards I was moved to another class.

I didn't pay much attention to popular music until I was 12 or 13. By that time I had my own transistor radio and could lie in bed at night listening to Radio Luxembourg. This was not terribly conducive to the development of any sort of musical appreciation. The reception was awful,  the earphone hurt, the music was crap and anyway I was more interested in listening to football commentary than pop music. I wasn't even interested in Top of the Pops, let alone the Old Grey Whistle Test which I don't think I saw until around 1976. 

My epiphany was listening to Alan "Fluff" Freeman's Saturday afternoon Radio 1 show while I was supposed to be doing my homework.  In fact I was mostly recording the music on a reel-to-reel tape-recorder my dad had bought off a work-mate for a couple of quid. That's where I first heard Free, Derek and the Dominoes, Led Zeppelin, Cream, Bee Bop Deluxe, Barclay James Harvest, Pink Floyd, Renaissance, Tom Petty and the Heart Breakers, Black Sabbath, Focus and so forth. That's what I liked until I was 16 or so. All pretty standard white-boy prog rock. 

Was there any social basis to this taste? To some degree there was, but it was very heavily restricted by  social homogeneity of my peer group.  There simply weren't many middle-class kids to interact with. At my school there may have been the odd son or daughter of a professional say a secondary teacher, a dentist or a solicitor, but I can't actually remember any. The axes of distinction were more along the lines of rough/respectable working class, grammar school stream/others, boys/girls, age-group and to a limited extent white/ethnic (though actually there were very few ethnic minorities unless you counted the Scots, Irish and Welsh). 

If you were a lad in a grammar school stream it was social death not to like some form of progressive rock. For reasons that were and are still quite obscure to me it was not possible to like both Deep Purple and Led Zeppelin, you had to choose. Pink Floyd was compulsory whatever your other preferences. Genesis and Yes were permissible, but considered a bit wet. Liking Faust, Hawkwind or Barclay James Harvest put you amongst the cultural avant garde.  Clearly this was all about tribal identity and your membership of the tribe was announced by reproducing in acrylic on the back of your army surplus haversack an album cover of one of the clan totems. I don't remember which particular groups you professed to like was particularly important for friendship choices as long as they fell within the approved set. Though I claimed to like Led Zeppelin, most of my friends belonged to the Deep Purple faction and one even liked The Who.

Certain preferences were marginal but accepted because of the charisma of the individuals holding them. Elton John and Sparks were regarded as potentially a bit poofy but OK because one of the year's  football stars liked them. Similarly Queen was within the pale because they were the favourites of the year's outstanding rugby player (I think camp was not a word in his or our vocabulary). Other preferences were tolerated without being endorsed. There was a group of "Soul Boys" in the non grammar school stream, but we wouldn't have anything to do with them. There was also a small group of reggae/funk fans who were into Bob Marley, Steel Pulse, War and that kind of thing.They were tolerated on a live and let live basis mostly because they tended to be tough black kids who commanded a certain amount of wary respect. Popular amongst some of the  art-crowd, a little older than us, were The Faces, Roxy Music and David Bowie.

What was on the other side of the cultural divide was commercial pop, the sort of thing that  girls liked: The Bay City Rollers, The Osmonds, disco, Sheena Easton, the stuff you could see on Top of the Pops. To say you liked that was social death.

What strikes me now about all this is its arbitrariness. For all I know a mile down the road at a neighbouring school the lines of division could have been  and probably were quite different. There might have been some generic similarity but the details, and it was the details that were important for us, were purely local. I doubt if there was anyone of my age in my social milieu that liked or knew anything about classical music, jazz, folk or world music. I learned about them much later. Over the past week I've relaxed in the evening to Bix Beiderbecke, Wayne Shorter, Oscar Peterson, Benjamin Britten, Arnold Bax and Gil Scott-Heron. Cultural omnivore or cultural univore? High on cultural capital or low on cultural capital?

Into the mystic - Bourdieu and MCA

What is it with Bourdieusians and multiple correspondence analysis? Why do they place  so much faith in one modest little data reduction tool. Why don't they love other ways of projecting points into  low dimensional sub-spaces - PCA, factor analysis and MDS  - quite as much? And why do they feel the need to denigrate other ways of summarizing  and smoothing data - let's not even get on to causality -  often with "arguments" (I use the term loosely) that are so spurious and/or muddled one wonders whether the purveyor actually understands what they are writing about? 

You can find a new example of this woeful tendency here complete with the usual approving nod towards the great Chicago General Linear Reality Dragon Slayer and obeisance to Wuggenig. Pity the author didn't care to cite Chan and Goldthorpe's utter demolition of Wuggenig's folly - you can read it here. Really a bit naughty to ignore it. Must have slipped by on the referees' blind side.

The only explanation I can come up with is a primitive psychological need to see magical  connections between substantive objects of interest - cultural tastes, practices, consumption and the scientific tools we use to learn about them. Which is a good excuse to listen to Van Morrison getting into the mystic.

What's the point of Critical Realism?

Just in case you missed it, over at orgtheory.net there is a great calling out by Kieran Healey of the  sociological adherents of Critical Realism. Predictably the CR true believers and fellow-travellers resort to the tactic of the spectacular dive, rolling over clutching their ankles crying foul beloved of certain Southern European football teams in the 1970s before they decided to actually play football and thus wipe the floor with the rest of us. My gut instinct is that the latter outcome is not one that the CR groupies are likely to realize.

 I should say that my own engagement with CR has been limited to  asking myself whether there was anything as a sociologist that I would do differently if I took  CR seriously. Being unable to find a single thing I decided that I could safely ignore Bhaskar, Harre & Co and carry on as normal. I haven't discovered anything in the storm of words provoked by Healey's intervention that makes me think I was mistaken.

Also worth an honourable mention amongst the supporting gallery of commentators is Cornell's Stephen Morgan who has some nice remarks on the brain-dead parrots who repeat stuff about "general linear-reality" "variable sociology and its ontological assumptions" etc. We need more people in the discipline like Professors Healey  and Morgan who are prepared to stand up and call it the way that they see it. There is far too much seriously intellectually weak drivel in sociology that only survives because the purveyors of it won't or are unable to survive outside of their climate controlled ecological niche. They need to be smoked out and their arguments exposed to the cold light of reason and evaluated empirically (if there are any actual empirical claims). If this leads to frayed tempers, bruised egos and tears before bed-time so be it. The argument, which I've heard a lot recently, that sociologists shouldn't air their dirty linen in public is so much hog-wash. Unaired linen rots and produces rank odours. How are you going to build a discipline with that kind of foundation?

How do you get people to use a new method?

The always stimulating Edmund Chattoe-Brown has a new article out at Sociological Review Online called: Why Sociology Should Use Agent Based Modelling. Well worth reading (if you have institutional access). A minor quibble (wouldn't a better title substitute Sociologists for Sociology?) set me thinking. 

Proselytizing is fine up to a point, but what is it that really makes a methodological innovation take off? After all ABM has been around for quite a while - I remember attending Nigel Gilbert's inaugural in 1992(?) which was an advertisement for the promise of ABM (does anyone else remember Ig and Ug?) - yet within sociology, even amongst open minded and broadly sympathetic people like myself, it still seems to be lumbering down the runway without achieving lift-off.

If I think about methodological innovations that - for good or ill - changed the things that people routinely did in their research it seems to me that a common circumstance is that they appeared to provide an answer to, a clarification of, or a way of dealing with, a problem that was generally regarded as of substantive importance but which heretofore had no satisfactory solution. In other words there was already a conversation about subject matter substance that  a critical mass of people cared about.

Think about log-linear models. People had been struggling with how to make sense of social mobility data in circumstances where you wanted to compare 2 or more populations in which the observed marginal distributions of the populations were known to differ and in which the detail of the pattern of association was important (ie a single correlation coefficient was for various reasons inadequate). Cometh the hour, cometh the method. Regarding this as a problem in the modelling of odds-ratios revolutionized the way people thought about a substantive issue (sweeping away dead end ideas about "structural" and "exchange" mobility).

The word appeared, is important here: I'm not claiming that succesful innovations  actually did what people thought they did or would do (good examples of  cases where the over-enthusiastic adoption completely obscured the very careful and cautious claims of the pioneers are structural equation modelling and multi-level modelling) but simply that they seemed to help answer questions that enough people at the time found important for substantive reasons.

The lesson I take from this is that what will determine whether ABM ever takes off in sociology will be the publication in mainstream journals of a few articles reporting  realistic applications  to substantive problems that enough sociologists care about that tell us something believable and important that we didn't know already. When methods give us, or appear to give us, the insights we crave then they will be taken up.

We shouldn't forget though that sometimes it takes a long period of promise before delivery can happen. I first learned something about network analysis in 1989 by going to a course given by Martin Everett (and very good it was too). Sociology is only now, at least in the UK, getting beyond the "social network as metaphor" phase, but progress is being made. For a long time, it seems to me, methods of network analysis spiraled upwards in sophistication and complexity mostly within a substantive vacuum. The object of inquiry was network analysis itself. Now things are changing, largely because of the greater availability of data. We simply have more data about more or less complete networks relating to things that we care about for reasons that are central to the substance of the discipline. When network tools are seen to give answers to sociological questions, then they get taken up. I think there is a lesson here for ABM.

Bayes, Religion and academic politics

Over at Error Statistics there is an interesting post  reblogging a piece by Normal Deviate on the question of whether Bayesian inference is a religion? This is something that I  have no opinion about whatsoever, even when taken within the rather measured and circumscribed terms in which Normal Deviate frames it. There is a link though to a fascinating Youtube interview with Dennis Lindley that gives you a lot of  insight into one particular Bayesian mind. 

Starting around 10:28 Lindley, talking of the time when he was HoD at UCL makes the following extraordinary admission: "...there were two members of the department who weren't playing the Bayesian line at all and I didn't think they were very good, so I went to the administration of University College and asked if there was any chance of being able to sack them?" 

Well, I suppose that plotting to get rid of people that don't agree with you - let's leave aside the question of whether that is the reason you judge them to be no good - is perfectly coherent... Being quite untroubled in confessing it is, to my mind, quite chilling. Tenure, which we gave away without much of a fight, was invented for a reason.