Popular Posts

Caveat Emptor

The opinions expressed on this page are mine alone. Any similarities to the views of my employer are completely coincidental.

Friday 30 May 2014

Piketty. Footnote on the footnote.

I think I'm beginning to change my opinion about Chris Giles. I really wanted to give him the benefit of the doubt, but now I'm not sure. He's  responded to Piketty's response with five short paragraphs (six if you count the final one liner). Only one of these deals with the actual issues at stake and as far as I can tell the content is just ridiculous ie he just repeats the original accusations whilst ignoring Picketty's explanations and then makes several assertions that look manifestly untrue to me: 

"...he [Piketty] explains transcription errors as deliberate adjustments to overcome discontinuities in data, but does not provide formulas or an explanation of why these undocumented adjustments should apply to only one data point in a time series; he does not explain why it is consistent to favour household surveys over estate tax records for the US but not the UK..." 

er, no, that's not the way I read it...

I really expected better than this. Any more of this and I'll be dusting off my old copies of Marx which I thought I had consigned for ever to a cardboard box in the junk room.  The ideas of the ruling class are in every epoch the ruling ideas...

Everybody's Talkin'

Apropos of nothing in particular here's a charming version of  Fred Neil's classic performed by Ryan Cook and Sunny Acres. And here's Fred himself singing another classic. What a voice.

Piketty footnote

Looks like I just squeezed in my rather thin thoughts about the Piketty affair yesterday  before he delivered  his 10 page reply to Chris Giles. The reply looks pretty cogent to me. 

The bottom line is that we are in the realm of estimation, ie trying to figure out what the whole looks like even though we can only see part of it. Inevitably this requires the exercise of judgment and reasonable people can give different weight to different parts of the judgment process. But that doesn't mean that we are free to believe whatever we like unconstrained by any evidence or that we shouldn't believe anything until we observe the whole. We are never going to have perfect data and if we have to suspend our beliefs about everything we aren't perfectly informed about then the world will be a very quiet place.

There seems to be a swell of opinion suggesting that Giles did something wrong. I don't see it quite like that myself. He was simply doing his job as an economic journalist. After all Capital in the Twenty First Century is a big news story and Piketty made his data available precisely to facilitate the sort of scrutiny that Giles subjected it to. It wasn't his fault that the FT bigged it up to a front page story (and a whole inside page). That was an editorial decision, as presumably was the ambuscade style way in which Piketty was alerted to the coming onslaught.

I must confess though to feeling a bit disappointed by the FT. By default it's become just about the only serious daily broadsheet in the UK that even pretends to hold a non-partisan position. When I want my  heart nudged back into the correct position I read the Guardian. When I want to find out what is going on in the world I read the FT. The editor made a bad call on this one. I think Giles acted in good faith, but the test will be whether he has the good manners, and the balls, to acknowledge that he got some things wrong.

Thursday 29 May 2014

Piketty

Given my interest in picking holes in other people's numbers you might have thought I would have something to say about the FT's hatchet job on Piketty. At the moment I'm loath to say very much for the simple reason that though the book has sat on my desk for the best part of a month it's unlikely that I am going to have sufficient time to read it with the attention it deserves before the Summer vacation begins. Commenting on things you haven't read is rarely a good idea. In any case a lot of heavy weights have already waded in,  with more knowledge and insight than I'll be able to muster.

I did read  Chris Giles' piece on Saturday and my overall impression was that though he may (or may not)  have uncovered a few data errors - let's face it all data contains errors  but unlike Piketty most people don't  bother to expose themselves to the risk that others will sift through their work to find the odd tasty morsel - he didn't even attempt to make a serious case that these errors - if errors they be - are sufficiently important to torpedo Picketty's argument.

Obviously Piketty will have to respond to Giles' claims and I've no doubt that he will. Science is all about conjecture and refutation, so let's see what happens.

There is one specific claim that Giles  makes  which I do have a hunch about. Let me be clear, it is nothing more than an informed guess and I'm perfectly happy for somebody to come along and demonstrate that I'm talking nonsense.

Giles claims that inexplicably Piketty ignores (or decided not to use) evidence for the UK from the ONS  Wealth and Assets Survey (WAS). As it happens I know a little about this source. When I sat on the National Equality Panel we were given access to preliminary results from the first wave of it. And jolly grateful we were to have them because good micro level data about wealth is like gold dust.

What struck me at the time, and after spending a few minutes yesterday having another look at the survey documentation I see no reason to change my view, is that WAS is probably not a good source for producing sensible estimates of the wealth held by the top 1% of the wealth distribution. The reason is very simple. Error prone household surveys tend to do a poor job of measurement when it comes to the tails of the distribution - either those with very little wealth or those with a lot of wealth. All sorts of errors have a habit of accumulating in the tails and the ratio of signal to noise in these regions is probably quite low. Add to that that WAS only samples about 20000 households and you can readily see that the amount of information about the tails of the distribution is pretty sparse and the confidence intervals must be rather wide. And all this is before we even get into sample selection bias (do the super rich agree to be interviewed?), selective item non-response and all the rest.

No doubt HMRC tax data is also flawed in all sorts of ways, but it is not obvious to me that - in the context of creating a time series - it is obviously worse then  the WAS data.

Wednesday 28 May 2014

Party Pooper

Knowing a little about numbers and having a modicum of common sense tends to make you a party pooper. I woke up this morning to the sound of  Penny Young chief executive of NatCen Social Research being interviewed on the Today programme. Actually I think she did  a good job and if you listened carefully you would have heard her explain that whatever change there has been in people's willingness to admit racial prejudice has been going on for a long time - probably since 2001. 

But that story isn't going to make the headlines - or make an "impact" case study. The framing of the interview suggested that there was some genuinely new finding here and that certainly seems to be how the Guardian is selling it today: "New data from NatCen’s authoritative British Social Attitudes (BSA) survey, obtained by the Guardian, shows that after years of increasing tolerance, the percentage of people who describe themselves as prejudiced against those of other races has risen overall since 2001." Well, no not really. What's new is one new data point for 2013 which adds nothing  much to a data series that has long been in the public domain, and a bit of puff for volume 31 of the British Social Attitudes series.

Let's take a closer look at the numbers behind the story. To their credit NatCen have put them online  in a nice pdf . I hope they don't mind me borrowing one of their graphs (all credit to them, I've just corrected a small typo in the title).

I'll leave it to others to quibble about the question that is posed to respondents and simply assume that it has some claim to validity (if it hasn't then the whole story collapses anyway).

 Let's look first at the mauve line which  joins up the year on year data points, except it doesn't because it is clear from the data table that NatCen helpfully provide that there are no data for 1988, 1991, 1993, 1995 and 1997. Full marks to the Guardian for making that clear in their graph where they represent the missing years with a * on the x axis. I'll return to this below. 

My first impression is that there is quite a lot of year on year variation within the range of 38% and 25% affirming that they are at least a little prejudiced. A 13 percentage point spread in a social attitude is actually a fairly narrow band of variation given the magnitude of non-sampling errors that these sorts of investigations are prone to. It is entirely possible that nothing very much has happened since 1983. Consider the change between 2011 and 2012 this was, roughly 12 percentage points in one year! My inclination is to believe that racial prejudice is a fairly stubborn cognitive disposition rather than a febrile attitude and that leads me to regard such apparent differences  as implausible estimates of real  population aggregate level attitudinal change.

Clearly the NatCen team have similar concerns, hence the inclusion of the blue line for the 5 year moving average (which the Guardian omits in its report). Some smoothing of the data is obviously desirable. But a few things puzzle me. Why doesn't the 5 year moving average start in 1985? And why does it continue to 2013 (obviously we can't calculate it for 2013  (or 2012) because we haven't yet got observations for 2014 and 2015)? I'm going to hazard a guess that NatCen have got the moving average plot wrong by plotting points  at the 5th year of the average rather than the central year. This would make sense of the dip at the end of the blue line which mainly reflects the large (and suspect) difference between 2011 and 2012.

But something else worries me. The "natural" year on year variation can be quite large, yet this is already somewhat smoothed out in the first part of the series because of the missing years in which no data were collected. If data had been collected we probably would have seen much more of a saw tooth pattern in the 1990s The downward trend from 1992 looks quite convincing but how convincing would it have looked if every other year that line was jumping up and down to the same degree that it is in the second half of the series where the year on year data are continuous?  This is the kind of situation in which a good (multiple) imputation model would come in handy to simulate data for years in which it is missing (remember the point of imputation is to preserve the pattern of variability in the data). A subsidiary quibble is that the 5 year moving averages in the first part of the data can't be five year averages because of the missing years. In brief, the first half of the series is made much smoother than the second half of the series and that's a bit undesirable for a fair over time comparison. It would have been more honest, though less aesthetically pleasing, to have plotted points with confidence intervals around them rather than joined up lines.

And then there is item non-response. We aren't told anything about this, but my guess is that a reasonable number of people won't feel too comfortable about answering this question. Is the number of non-respondents stable over time or has it changed in any particular direction? Again this may be a situation where a fair representation of our knowledge probably, somewhat paradoxically, requires some data imputation.

So, what's my best guess at what has actually happened? I'd venture that these data give us some, not particularly strong, evidence to say that  (at the aggregate level) willingness to admit to prejudiced attitudes probably declined a little bit between the 1980s and the end of the 1990s and since the early years of the new century there may have been a slight increase in the tendency to admit prejudice.  However these changes, if changes they be, are within a fairly narrow range. Nothing particularly noteworthy has happened in the last four or five years and the 2013 BSA data add nothing new to the story.

But you can't make a case for impact, or get on the Today programme by saying that.





Monday 26 May 2014

Of Gove and Men

It doesn't surprise me that a Conservative Education Secretary would want to remove certain books from the school examination syllabi. It's more frightening that he is able to do so. It's also not so surprising that such a man would strongly dislike a book with the themes of loneliness, friendship and dreams. 

Like everyone  I read Of Mice and Men at school, but not for an examination merely in a third year English class taught by the inspired Mr Ritchie. It was the first book with genuinely adult themes that I  engaged with and it made me realize that novels were more than just stories you read in order to find out what happened. 

Mr Ritchie was a man with a mission. No doubt Of Mice and Men was the prescribed reading for the year, but after dispensing with that he took  the  decidedly nonliterary 14 year-olds  of 3L on an intellectual adventure. He decided we were mature enough to tackle King Lear (which at the time was an A level text). Boring? No. Because he also showed us clips from Jonathan Miller's 1975 BBC production with Michael Hordern in the lead so that we could see it as a play not just a text. 

Warming to his theme he decided that we should learn about the conventions of the renaissance revenge tragedy.  I remember sitting in the "stock cupboard" with several mates during an English lesson, reading Hamlet's "What a piece of work is a  man" speech  and then listening to the Hair version on an ancient record player. Brilliant, Shakespeare lives! I think the stock cupboard was sufficiently soundproof that the Cambridge trained Head of the English department wouldn't hear what we were up to.

Next up were the conventions of the gothic novel and the 18th century picaresque. He handed out to the class his own copies of The Castle of Otranto, The Mysteries of Udolpho, Dracula, Frankenstein, Vathek, Tom Jones, Tristram Shandy, Roderick Random and set us the task of writing a report showing how some of the conventions employed in these tales were still alive and kicking. Dry? Not at all, because we also had lessons in which he showed us clips from cheesy Hammer horror movies and from Monty Python sketches. Only connect.

I was given Tom Jones (I wonder if Mr Gove would approve of that?). I don't think I'd ever held a book that contained so many pages. I remember being disappointed because I wanted to read one of the gothic novels, but, with some reluctance I started to read it and to my surprise I found that it was   understandable, bawdy and quite hilarious.  By the time I had to give it back I must have read about 300 pages of a book that I would never have spontaneously chosen to open.

And then we were on to modern literature. Again Mr Ritchie distributed his own books. One of my friends got The Prime of Miss Jean Brodie  which I gathered was fun and I landed The Power and the Glory which was not. I think I found Greene's slim volume more of a challenge than Fielding. Since then I've changed my mind and despite failing to empathize with his Catholic sense of sin,  I've  read almost all his novels with pleasure and grown to admire his writing.

I suppose inspirational teachers like Mr Ritchie would be banned from today's classroom for not sticking to the prescribed texts. And look out if you point out to the Secretary of State that the subject is English literature not British literature. Patrick White, Naipaul, Joyce, Yeats, Hemingway, Steinbeck, Faulkner, Mansfield, Beckett, Synge away with you!  Far better if you to settle down for the night with a nice Trollope. Ooh, aah missus...

Friday 23 May 2014

Version of Record

Don't get me wrong, I'm a great fan of "online first" academic publishing and I look forward to the day when the outmoded conventions of print journals: issue numbers, volumes, endless waits to see your work appear etc are consigned to the dust heap of history.

But, there is one convention that is worth keeping: the single dated version of record that is only amended by errata or retraction. Once you have read and corrected the proofs and the article is in the public domain that should be it. You shouldn't be permitted to tinker with it once it has gone live and there should not be multiple versions of record.

The reason for this is that once something is published online the academic conversation has begun and the object of discussion needs to be fixed. It is clearly undesirable if the target of a debate or discussion is a moving one. Authors should not be permitted to  add or delete in response to criticism or second thoughts. They can, of course, react to critics in replies and responses. They can also, if they discover consequential mistakes in their work, issue errata. All of these things give us a clear audit trail and they constitute an essential foundation of discipline on the urge to publish quick and dirty, ill considered, work. 

It seems to me that we already have a large body of people whose modus operandi is to submit poorly conceived and half-finished articles  at the point where they reckon the chances of getting an R&R are evens. I've definitely heard people say they encourage their doctoral students to  submit early and benefit from referees' comments. I think this fundamentally misunderstands the nature the referee's job. Referees should not be  unpaid auxiliary doctoral supervisors, hidden and largely unacknowledged co-authors, methods advisers, proof-readers or translators. If I have to write five pages to tell you why what you have written is nonsense, you probably shouldn't have submitted it in the first place and you probably should have shown it to a  supervisor or a few colleagues  and then rewritten it before you even started to think about sending it off.

These thoughts came to mind when I saw that the journal Sociology appears to allow multiple online versions of record (as well as a final print version of record) distinguished only by the date of publication. As far as I can discover there is no information as to how these versions differ and the reader is left to figure this out for themselves, which I assume few will have the time or patience to do. I've no idea whether all that is happening is that a few minor typos are being corrected or that more substantial changes are being made, but personally I'm against any unexplained changes being made.  It would be trivial to produce an errata list and then the reader would feel assured that no funny business was going on (I'm not claiming there is). Science is not just about intellectual integrity, it's also about the appearance of integrity.


Wednesday 21 May 2014

Social Class and Obesity

The Great British Class Survey (GBCS) has escaped from its sociological host and has established a foothold in epidemiology. Be afraid, be very afraid. 

You can read my sceptical thoughts  here on an article published at the weekend in BMJ Open  in which the author attempts to use GBCS data to illuminate the  link between social class and obesity.  I must say that their response policy is a model of how science should operate. My response was up and linked to the article I was commenting on within 24 hours of being submitted.

Friday 16 May 2014

Are you a millionaire?

The Telegraph is running a story today with the headline: 'One in five university graduates becomes a millionaire'. How intriguing. The sub is: "More than two million degree-holders have a net worth of £1 m or more as new statistics reveal the education gap between rich and poor". This apparently stunning factoid is based on a report by ONS which I've not read, so I can't venture any criticisms of the source material. 

But...hang on a moment read further in the Telegraph report, down to the line that says: "The figures, which do not account for household debts such as mortgages, also showed that...". Play that by me again...So in fact a large proportion of those two million degree holders with net worth of £1 m or more won't actually have a net worth of £1 m at all. Possessing assets worth a million is, er,  not the same as having a net worth of £1 m. You'd think the Telegraph would get that right wouldn't you?

And, then there are pensions. I've not read the original data source, but I'll make an educated guess that pension wealth has been imputed. This is not a criticism, it is very hard to get accurate individual level information about pension wealth. What your pension is actually worth to you right now depends on assumptions about the discount rate to apply, how long you are going to live and for an increasing proportion of people the state of the stock-market at the point they retire. I'm sure ONS have made sensible guesses about these things but they will just be guesses no matter how fancy the methodology.

Combine all this with the arbitrariness of the £1 m threshold and the whole story disappears in a puff of smoke.

Friday 9 May 2014

Chakraborrty and the big bad economist

I wonder if sometime in the past an academic economist did something bad to
 
 

Thursday 8 May 2014

Leslie Thomas RIP

Until a month ago I'd not knowingly read a word written by Leslie Thomas. Despite my total ignorance I had him confidently pigeon-holed as a writer of slightly cheesy middle-brow humour, a sort of racy James Herriot with a swinging sixties edge. Then, browsing the book shelves of a charity shop, I came across a paperback copy of his first volume of autobiography This Time Next Week. The book deals with the time he was growing up in a Bernado's home in Kingston-upon-Thames. It was in pristine condition, the shop only wanted a pound for it and to tell the truth it was the only thing worth buying. I'm vaguely interested in the history of London's South Western suburbs so I thought "why not?". Flicking through the first few pages when I got home I noticed the book had been signed  by the author with a dedication: "To_____ with many thanks from Leslie Thomas. Salisbury 19__". So that was a pleasant bonus.
 
With nothing better to do in the evening I sat down and began to read. I have to say the book was excellent. It conveyed the horror of being orphaned and handed over to an institution without being maudlin about the experience and it succeeded in capturing the matter of factness about the way a 12 year old boy might accept his new situation. Thomas is not bitter about the way he was treated, he wasn't actually treated badly, but he did  come up against the routine neglect that comes from institutionaling care: nobody takes a personal interest in the welfare of the child so nobody goes out of their way to do anything they don't have to. The "it's not my job" attitude is illustrated perfectly by the fact that he was separated from his younger brother for more than 18 months and during that time nobody bothered to tell the younger of the siblings that his mother was dead. This was left to the older Thomas, still himself a child, when they were eventually reunited. It's only in relating that story that a slight note of anger slips through. Otherwise his account is full of humour and, to my surprise, extremely well written. If this book is anything to go by, Leslie Thomas knew his craft and was a writer of some talent.
 
I doubt I'll be rushing to read his novels, which don't really appeal to me but it is good to be reminded from time to time that you shouldn't always judge a book by its cover.

Friday 2 May 2014

Simon Wren-Lewis on Post-Crash Economics

Simon Wren-Lewis has a nice piece on his blog about the current mood of student discontent with the content of the orthodox economics curriculum. As I've said before I think it is best if economists fight their own battles. They certainly need no assistance from me. But in as far as I can understand what is at stake I can still stand on the touchline and cheer for common sense and boo what seems like trendy  posturing.