LOB-vs
Download Lectures on Biostatistics (1971).
Corrected and searchable version of Google books edition

Download review of Lectures on Biostatistics (THES, 1973).

Latest Tweets
Categories
Archives

Shitij Kapur

Jump to follow-up

Stop press. Financial report casts doubt on Trainor’s claims

Science has a big problem. Most jobs are desperately insecure. It’s hard to do long term thorough work when you don’t know whether you’ll be able to pay your mortgage in a year’s time. The appalling career structure for young scientists has been the subject of much writing by the young (e.g. Jenny Rohn) and the old, e.g Bruce Alberts. Peter Lawrence (see also Real Lives and White Lies in the Funding of Scientific Research, and by me.

Until recently, this problem was largely restricted to post-doctoral fellows (postdocs). They already have PhDs and they are the people who do most of the experiments. Often large numbers of them work for a single principle investigator (PI). The PI spends most of his her time writing grant applications and traveling the world to hawk the wares of his lab. They also (to variable extents) teach students and deal with endless hassle from HR.

The salaries of most postdocs are paid from grants that last for three or sometimes five years. If that grant doesn’t get renewed. they are on the streets.

Universities have come to exploit their employees almost as badly as Amazon does.

The periodical research assessments not only waste large amounts of time and money, but they have distorted behaviour. In the hope of scoring highly, they recruit a lot of people before the submission, but as soon as that’s done with, they find that they can’t afford all of them, so some get cast aside like worn out old boots. Universities have allowed themselves to become dependent on "soft money" from grant-giving bodies. That strikes me as bad management.

The situation is even worse in the USA where most teaching staff rely on research grants to pay their salaries.

I have written three times about the insane methods that are being used to fire staff at Queen Mary College London (QMUL).
Is Queen Mary University of London trying to commit scientific suicide? (June 2012)
Queen Mary, University of London in The Times. Does Simon Gaskell care? (July 2012) and a version of it appeared th The Times (Thunderer column)
In which Simon Gaskell, of Queen Mary, University of London, makes a cock-up (August 2012)

The ostensible reason given there was to boost its ratings in university rankings. Their vice-chancellor, Simon Gaskell, seems to think that by firing people he can produce a university that’s full of Nobel prize-winners. The effect, of course, is just the opposite. Treating people like pawns in a game makes the good people leave and only those who can’t get a job with a better employer remain. That’s what I call bad management.

At QMUL people were chosen to be fired on the basis of a plain silly measure of their publication record, and by their grant income. That was combined with terrorisation of any staff who spoke out about the process (more on that coming soon).

Kings College London is now doing the same sort of thing. They have announced that they’ll fire 120 of the 777 staff in the schools of medicine and biomedical sciences, and the Institute of Psychiatry. These are humans, with children and mortgages to pay. One might ask why they were taken on the first place, if the university can’t afford them. That’s simply bad financial planning (or was it done in order to boost their Research Excellence submission?).

Surely it’s been obvious, at least since 2007, that hard financial times were coming, but that didn’t dent the hubris of the people who took an so many staff. HEFCE has failed to find a sensible way to fund universities. The attempt to separate the funding of teaching and research has just led to corruption.

The way in which people are to be chosen for the firing squad at Kings is crude in the extreme. If you are a professor at the Institute of Psychiatry then, unless you do a lot of teaching, you must have a grant income of at least £200,000 per year. You can read all the details in the Kings’ “Consultation document” that was sent to all employees. It’s headed "CONFIDENTIAL – Not for further circulation". Vice-chancellors still don’t seem to have realised that it’s no longer possible to keep things like this secret. In releasing it, I take ny cue from George Orwell.

"Journalism is printing what someone else does not want printed: everything else is public relations.”

There is no mention of the quality of your research, just income. Since in most sorts of research, the major cost is salaries, this rewards people who take on too many employees. Only too frequently, large groups are the ones in which students and research staff get the least supervision, and which bangs per buck are lowest. The university should be rewarding people who are deeply involved in research themselves -those with small groups. Instead, they are doing exactly the opposite.

Women are, I’d guess, less susceptible to the grandiosity of the enormous research group, so no doubt they will suffer disproportionately. PhD students will also suffer if their supervisor is fired while they are halfway through their projects.

An article in Times Higher Education pointed out

"According to the Royal Society’s 2010 report The Scientific Century: Securing our Future Prosperity, in the UK, 30 per cent of science PhD graduates go on to postdoctoral positions, but only around 4 per cent find permanent academic research posts. Less than half of 1 per cent of those with science doctorates end up as professors."

The panel that decides whether you’ll be fired consists of Professor Sir Robert Lechler, Professor Anne Greenough, Professor Simon Howell, Professor Shitij Kapur, Professor Karen O’Brien, Chris Mottershead, Rachel Parr & Carol Ford. If they had the slightest integrity, they’d refuse to implement such obviously silly criteria.

Universities in general. not only Kings and QMUL have become over-reliant on research funders to enhance their own reputations. PhD students and research staff are employed for the benefit of the university (and of the principle investigator), not for the benefit of the students or research staff, who are treated as expendable cost units, not as humans.

One thing that we expect of vice-chancellors is sensible financial planning. That seems to have failed at Kings. One would also hope that they would understand how to get good science. My only previous encounter with Kings’ vice chancellor, Rick Trainor, suggests that this is not where his talents lie. While he was president of the Universities UK (UUK), I suggested to him that degrees in homeopathy were not a good idea. His response was that of the true apparatchik.

“. . . degree courses change over time, are independently assessed for academic rigour and quality and provide a wider education than the simple description of the course might suggest”

That is hardly a response that suggests high academic integrity.

The students’ petition is on Change.org.

Follow-up

The problems that are faced in the UK are very similar to those in the USA. They have been described with superb clarity in “Rescuing US biomedical research from its systemic flaws“, This article, by Bruce Alberts, Marc W. Kirschner, Shirley Tilghman, and Harold Varmus, should be read by everyone. They observe that ” . . . little has been done to reform the system, primarily because it continues to benefit more established and hence more influential scientists”. I’d be more impressed by the senior people at Kings if they spent time trying to improve the system rather than firing people because their research is not sufficiently expensive.

10 June 2014

Progress on the cull, according to an anonymous correspondent

“The omnishambles that is KCL management

1) We were told we would receive our orange (at risk) or green letters (not at risk, this time) on Thursday PM 5th June as HR said that it’s not good to get bad news on a Friday!

2) We all got a letter on Friday that we would not be receiving our letters until Monday, so we all had a tense weekend

3) I finally got my letter on Monday, in my case it was “green” however a number of staff who work very hard at KCL doing teaching and research are “orange”, un bloody believable

As you can imagine the moral at King’s has dropped through the floor”

18 June 2014

Dorothy Bishop has written about the Trainor problem. Her post ends “One feels that if KCL were falling behind in a boat race, they’d respond by throwing out some of the rowers”.

The students’ petition can be found on the #KCLHealthSOS site. There is a reply to the petition, from Professor Sir Robert Lechler, and a rather better written response to it from students. Lechler’s response merely repeats the weasel words, and it attacks a few straw men without providing the slightest justification for the criteria that are being used to fire people. One can’t help noticing how often knighthoods go too the best apparatchiks rather than the best scientists.

14 July 2014

A 2013 report on Kings from Standard & Poor’s casts doubt on Trainor’s claims

Download the report from Standard and Poor’s Rating Service

A few things stand out.

  • KCL is in a strong financial position with lower debt than other similar Universities and cash reserves of £194 million.
  • The report says that KCL does carry some risk into the future especially that related to its large capital expansion program.
  • The report specifically warns KCL over the consequences of any staff cuts. Particularly relevant are the following quotations
  • Page p3 “Further staff-cost curtailment will be quite difficult …pressure to maintain its academic and non-academic service standards will weigh on its ability to cut costs further.”
  • page 4 The report goes on to say (see the section headed outlook, especially the final paragraph) that any decrease in KCL’s academic reputation (e.g. consequent on staff cuts) would be likely to impair its ability to attract overseas students and therefore adversely affect its financial position.
  • page 10 makes clear that KCL managers are privately aiming at 10% surplus, above the 6% operating surplus they talk about with us. However, S&P considers that ‘ambitious’. In other words KCL are shooting for double what a credit rating agency considers realistic.

One can infer from this that

  1. what staff have been told about the cuts being an immediate necessity is absolute nonsense
  2. KCL was warned against staff cuts by a credit agency
  3. the main problem KCL has is its overambitious building policy
  4. KCL is implementing a policy (staff cuts) which S & P warned against as they predict it may result in diminishing income.

What on earth is going on?

16 July 2014

I’ve been sent yet another damning document. The BMA’s response to Kings contains some numbers that seem to have escaped the attention of managers at Kings.

10 April 2015

King’s draft performance management plan for 2015

This document has just come to light (the highlighting is mine).

It’s labelled as "released for internal consultation". It seems that managers are slow to realise that it’s futile to try to keep secrets.

The document applies only to Institute of Psychiatry, Psychology and Neuroscience at King’s College London: "one of the global leaders in the fields" -the usual tedious blah that prefaces every document from every university.

It’s fascinating to me that the most cruel treatment of staff so often seems to arise in medical-related areas. I thought psychiatrists, of all people, were meant to understand people, not to kill them.

This document is not quite as crude as Imperial’s assessment, but it’s quite bad enough. Like other such documents, it pretends that it’s for the benefit of its victims. In fact it’s for the benefit of willy-waving managers who are obsessed by silly rankings.

Here are some of the sillier bits.

"The Head of Department is also responsible for ensuring that aspects of reward/recognition and additional support that are identified are appropriately followed through"

And, presumably, for firing people, but let’s not mention that.

"Academics are expected to produce original scientific publications of the highest quality that will significantly advance their field."

That’s what everyone has always tried to do. It can’t be compelled by performance managers. A large element of success is pure luck. That’s why they’re called experiments.

" However, it may take publications 12-18 months to reach a stable trajectory of citations, therefore, the quality of a journal (impact factor) and the judgment of knowledgeable peers can be alternative indicators of excellence."

It can also take 40 years for work to be cited. And there is little reason to believe that citations, especially those within 12-18 months, measure quality. And it is known for sure that "the quality of a journal (impact factor)" does not correlate with quality (or indeed with citations).

Later we read

"H Index and Citation Impact: These are good objective measures of the scientific impact of
publications"

NO, they are simply not a measure of quality (though this time they say “impact” rather than “excellence”).

The people who wrote that seem to be unaware of the most basic facts about science.

Then

"Carrying out high quality scientific work requires research teams"

Sometimes it does, sometimes it doesn’t. In the past the best work has been done by one or two people. In my field, think of Hodgkin & Huxley, Katz & Miledi or Neher & Sakmann. All got Nobel prizes. All did the work themselves. Performance managers might well have fired them before they got started.

By specifying minimum acceptable group sizes, King’s are really specifying minimum acceptable grant income, just like Imperial and Warwick. Nobody will be taken in by the thin attempt to disguise it.

The specification that a professor should have "Primary supervision of three or more PhD students, with additional secondary supervision." is particularly iniquitous. Everyone knows that far too many PhDs are being produced for the number of jobs that are available. This stipulation is not for the benefit of the young. It’s to ensure a supply of cheap labour to churn out more papers and help to lift the university’s ranking.

The document is not signed, but the document properties name its author. But she’s not a scientist and is presumably acting under orders, so please don’t blame her for this dire document. Blame the vice-chancellor.

Performance management is a direct incentive to do shoddy short-cut science.

No wonder that The Economist says "scientists are doing too much trusting and not enough verifying—to the detriment of the whole of science, and of humanity".

Feel ashamed.