12.9.14

Order Order

So in the Houses of Parliament, it's Prime Ministers Questions that gets most of the news. But a lot of the deeper and more searching work is conducted in the committee rooms with the various Select Committees.   They typically cut to the quick, and are wonderful combination of  authority, focus and personality.

No one is out of scope for the select committees.   Ministers, Chief Constables and Chief Executives get "invited" to attend to "give evidence" and be "witnesses".  And there are often the charismatic and forthright committee chairs, that take no prisoners.  So there's a couple of examples below.

And with recent cases of witnesses correcting themselves some time after giving evidence, some committees are now seeing fit for some witnesses to take the oath, with hand on the bible, "to tell the truth, the whole truth and nothing but the truth".  That makes it purgery for anything that not the truth there and then.   It's a powerful turn of phrase to set the scene for the witness..... "The Clerk will now administer the oath to you"

Not only is there strong directness but the role of evidence and data seems to be getting more profile.  Here's a couple of examples of the staggeringly direct and demanding challenge from the committee chairs.

Public accounts Committee on Out of hours GP services in England

Monday 1st September 2014.

The audit commission reported on this topic...  Out-of-hours GP services cost an estimated £400 million in 2013-14 and an estimated 90 per cent of GP practices have opted-out, and delegated the service.

Again here's the committee purpose...."This inquiry will challenge the Department and NHS England on their oversight and assurance of out-of-hours GP services, on the performance of the system and on the level of public awareness of out-of-hours GP services."

Oversight is such a great term with its different definitional options. It's both "overview" and "overlooked".  Can be great to describe "overview" but infer over "looked".  That feels like the case here.

So Margaret Hodge is in the chair, questioning the Chief Exec of NHS England, and the Director of Acute Care


"You should know it. With the greatest respect"
  • "You out to have the basic data rather than just give an example"

  • "You see, you don't have data on anything"

"You don't know it's value for money. You don't know because you don't understand your cost base.
"You're creating an absurdity of it"
"I don't know what that means. It's a non answer".


Hodge referring to whistleblower evidence..
"The data that was being provided was falsified....there was no monitoring through the system that enabled that to emerge"
Then some super lines from Richard Bacon...
"The chairs question was about cost not quality" 
"That was not my question neither was it Mr Pugh's"
"That was the question to which we were looking for an answer...and it was striking to me that you didn't know that."
"You need to understand your cost base and I don't think you do.  You have to understand your underlying cost base and you don't".
And some back and forth from John Pugh creating a turn around answer.....
Pugh... "Have you done any drilling down into the figures?" 
Willet... "We haven't got a breakdown and we don't know the individual breakdowns and the individual consultations and ratios" 
Pugh..."But I'm surprised at that as that's quite a key factor to know on terms of integration of services" 
Willet..."Those data better understood locally.  There's going to be a variation from locality to locality" 
Pugh..."But a variation is a concern". 
Willet... "And that's were we have to get to measure the all fragments in the same way.  Look across the system to see how it's performing"
And then the build up to a strong challenge....
Willet... The "local health economy working with the public will need to decide how best they want to provide and offer the service and what is the best arrangement.   Walk in centres remarkably different, across the country where they were what they did, whom they were supported by".
Pugh... "You didn't have information centrality on the extent to which GPs were sub-contracting".
Willet ..."We have a complete inventory of the whole country and we can tell you exactly what the position is"
And then comes the simply sublime intervention from Sir Aymas Morse - Comptroller and Auditor General of the National Audit Office - responding to a real sense of flippancy from Willet in that last response.
"That's really great.  So you're collecting more information.  All we're trying to push you towards is to make sure the information is sufficiently coherent that you know whether the out if hours service is value for money or not.  We're not telling you to collect all the info that could possibly be asked for everywhere  in the world or things that are burdensome. Just asking you take a logical approach to putting something together that allows you to make good judgements at the centre. Is that fair?"
 "And we agree with that" Stevens


Home Affairs Select Committee on Child sexual exploitation in Rotherham .

Tuesday 9 September 2014

Chaired by Keith Vaz and "taking evidence" from the previous Chief Constable.

There's a clue of what's to come in the pre-committee published purpose from Keith Vaz.  "The Committee will be questioning those in positions of authority at the time that these offences were being committed on how the sexual exploitation of children on a horrifying scale was allowed to go unchallenged for so long."

Here's some Vaz statements to the previous Chief Constable.


"That's an interesting point. It's not born out by the facts" 
"I don't think your treating these proceedings with the respect they deserve". 
"How would you know that Mr Hughes, when your telling this committee that you don't know about anything else and can't remember anything else" 
"We find your evidence totally unconvincing"
And some reference to the data from the previous Chief Constable.
"Improvement"..."That's not some sort of numerical idea".  "You look at recorded crime figures you look at the data to establish the problems". There were "good relationships with Local Authorities and others who would bring data to me". "I had no knowledge of this". 


What Gets Measured Gets Done...But....

....What gets pressured gets spun.



Targets. Targets. Targets.  A curious thing.  They include not only a prediction of the future, but also a prediction of influence on that future, and specific measurement of all that too.   Seeking that special sweet spot between being both challenging and achievable at the same time.

Of course in some cases it might well be visionary and aspirational, as in the old Eastern proverb ..."Aim for the top of the tree and you'll never leave the ground.  Aim for the moon and you might at least get to the top of the tree".  But more often than not it's about performance, and about success or failure.  Binary rather than multidimensional.

So big headlines around falsification of the measurement of cancer treatment waiting times to meet government targets.  The Care Quality Commission (CQC), the independent regulator of heath and social care in England found that measurements in one hospital, nearly a third of the patient records examined (22 of 66) had been altered to hide "extensive" delays for treatment.

What is measured and targeted sure does get attention and focus.  But once the that pressure tips the balance from being a priority, to being the pure judgement of success or failure, then unhelpful creativity and system redesign sets in.

That might just be managing the system.   There's the example from transport logistics moving electronically tagged parcels.  The sorting centre will bring the parcels inside only once they are ready to process them. Simply bringing them over the thresholds starts the electronic clock.   So best they wait outside until we can start the clock when it suits us.. That may well minimise the amount of time in the sorting centre, but overlooks the bigger picture of overall faster delivery.

It's also possible to entirely design out measurement of failure.  There were maximum waiting times from booking to appointment to see your local GP.  The solution.....take away the advanced booking system so you can only book on the day for the day.  No luck today, then it all starts again tomorrow.

What's disappointing is that the target it set at the minimum acceptable level.  So take the original four hour target for waiting times for Accident and Emergency departments.   So the margins can be managed, for example ambulances can be left queuing before patients are accepted which starts the hospital acceptance clock, and patients can also be transferred to "assessment units" which stops the clock.

That was strongly reinforced, hearing super presentation at the recent Royal Statistical Society Annual Conference (Sept 2014) from the joint academic and medic research team at the University of Sheffield. There's a fantastically clear spike where patients were admitted in the moments before they exceed the 4 hour waiting time.  The target has gone now but by all accounts the 4 hours is still a strong corporate driver.



What's even more staggering is some of the background measurement activities.  Different hospitals measure these things in different ways.  In fact it's possible to stay two nights and still not be classed as an admission.

So rather than the one dimensional focus on four hours, what about a four dimensional focus.....on the number of people waiting for 1hr, 2hrs, 3hs and then 4hr.   The elapsed time is being measured anyway so there's no additional measurement overhead.  That's a much more refined perspective.

Post four hours might well be failure, but how about some success focus too, at the 1hr and 2hr end of the spectrum.  Sure there might then be micro management at the hour margins, but at least it's a fuller and more multidimensional picture. Two A&E departments with identical performance at the 4hr threshold, might well look different when it's unpacked into the hour performance slots.  Even mean waiting times might give a more insightful view.



5.7.14

Some Psychological Context for Evidence

Evidence based decisions.  A mantra that gets used widely.  But there is of course a balance in decisions between evidence and judgement.  Decisions are easy, good decisions are more difficult, as they rely on getting that balance right between evidence and judgement.  But it does not stop there.

The reality of the world is that the evidence is set in the context of individual and social behaviours, which it can be easy to overlook.  The psychologists and sociologists have articulated and categorised plentiful behaviours that influence us both personally and collectively.  That adds another layer of challenge to power of evidence.

Here's just a few of those psychological contexts...

Cognitive Dissonance.  This is all about self perpetuating perspectives.  Where something is counter to a prevailing view (cognitive consistency), there are tendencies to find ways to assimilate those contradictions.  (Leon Festinger 1919-1989).  So the remedial action can typically include decreasing the perceived importance of the dissonance.  Dissonance can also be designed out....only taking information or news from specific sources for example.

Groupthink.  Really just cognitive dissonance for groups.  Coined in 1971, this describes the mutual reinforcement of group behaviour, often demonstrated as a general collective over confidence.   

False Memory.   Memories may not always be real.  They can be a distorted recollection or entirely imaginary.  So where memory is incomplete, the gaps can be filled with other information from other circumstances.  Entire false memory can be quite easily induced in experiments, often through mis-information.

Invisible Gorilla.  This about missing something obvious by being so focussed on something else.  In an experiment, a person in a gorilla suit (or dressed as a ghost or on a unicylce) walks through as baseball game and goes unnoticed by half of the spectators who have been briefed to count the number of passes.   More interestingly perhaps, even expecting the unexpected doesn't help you see it.

Halo Effect. Where one positive trait or outcome can lead to the assumption that other traits or outcomes are positive.    It works the other way around too as the "Horns Effect".  Its a long standing cognitive short cut for dealing lots of complex or missing data.

Muller-Lyer Illusion.  This is about the influence of visual context, especially important with the ever increasing infograhics and visualisations.  This is where the same two lines will seem to be of different lengths depending on their immediate visual context (typcially inward or outward facing arrows at each end).  The lines with the inward facing arrows will seem shorter, due to visual conditioning around perspective. 

Barnham Effect.  Named after the US Showman P.T. Barnum (1810-91), this describes statements which on the surface appear detailed or specific, but in fact will be vague or ambiguous, and even self contradictory and hence widely applicable as it covers all angles....Barnam described it as "having a little something for everyone".




Sources: Freudian Slips. Joel Levy. 2013




1.2.14

Serial Thinking

There's an emerging mainstream "thinking about thinking"  Given that "Learning about Learning" has become more common place and even expected - the explicit focus on Study Skills for example - so the same interest is applying to thinking.    In short there's increasing recognition about how our brains can work better if we're more concious about it.  Welcome to one such example....Serial Thinking.

So move over multi-tasking and welcome to serial tasking.  So if multi-tasking is parallel working, then serial tasking is linear working.  Of course multi-taking is really just lots of micro serial tasking with lots of switching.


There seems to be some increasingly recognition that the busyness of constantly switching from one area of thinking to another has more explicit downsides.   So natural multi linguists who start counting in one language then switch to another will stumble on the switch.  Its the overhead of switching tasks.  The more the switching the more the overhead, and hence becoming less productive, and often when we need to be more productive.  I'm not sure we should be surprised at this.  It would appear that the word "priority" only took a plural addition "priorities" in the 1970's.

So it's about having the mental capacity, space and energy to both focus on the specific and to have the capacity to think more broadly too, and without carrying in the baggage of previous tasks or the anticipation of the next one. "It's not just about time but about bandwidth...when we are most busy we think we are becoming better users of time when we are really becoming worse managers of our bandwidth" (Sendhil Mullainathan, Prof of Economics at Harvard).  As ever it's a matter of balance in these things.


That can translate beyond the personal to the more corporate and cultural.   Easy to be busy on the wrong things or in the wrong way (wrong busy), rather than making the space to make sure that it's doing the right things in the right way (right and less busy).



Sources: Wired October 2013

11.1.14

Rush to Wrong

In the world of strategy, planning, analysis and performance, there's a real balance to be struck between planning and action.   Sometimes this can be simple enough, especially in short term and reactive situations.

My favourite example here is generating parliamentary answers to parliamentary questions - with the red flag clipped to the wallet "Ministerial - Immediate at all stages".  But as scale increases, and where there is more choice, then that balance needs to be more sensitively managed.



There are two ends of that action-planning spectrum.....

Rush to Wrong.  Without sufficient planning then action can be wasted, even contradictory or counter-productive.  In a desire to do something and get busy, that might end up being "wrong busy".  That may take us further from our objective and makes things more difficult or longer to subsequently achieve.  If we're looking for the buried treasure, it's digging in the wrong place (ineffective) or with the wrong tools (inefficient)

Paralysis by Analysis.  With to much planning there is a risk of over-analysis.  Elapsed time can take a real toll.   There's the now immortal quote...."Move fast and break things. Unless you are breaking stuff, you are not moving fast enough" (Mark Zuckerberg).  That would rather assume that it's right things that are being broken.

So the challenge is to get the balance between those two extremes just right for the context.


30.11.13

News Numbers - Helipcopters

With plenty of numbers being used in the news, it's often very tempting to overplay the effect of the numbers, often quoting without some context.

Hence having personally coined the term "Numberator"....the use of a number without some necessary contextual denominator here's another example....Another of those news number naughties.... or news numberation.

So the UK is very helpfully sending aid to the Philippines to support victims of typhoon Haiyan.  HMS Darling is helping out but will be replaced by HMS Illustrious announces the Prime Minister...

"And I can announce today that once Daring has started its work, we are actually going to be able to replace in time HMS Daring with HMS Illustrious which is of course a carrier with helicopters, seven times as many helicopters as on HMS Daring and with the key ability to process fresh water. So we will be giving further assistance in the best way we can". 

So "seven times" as many helicopters, that's quite some emphasis.  That's also the edited sound bite that made the main news headlines.  So how many would that be in practice?

Actually it's seven helicopters in total.  HMS Darling had one and HMS Illustrious has seven.  So yes technically seven more.

"Seven times more helicopters" does seem to have a more powerful emphasis than "Seven helicopters". Seven times is a relative statement that can apply in multiple situations for higher numbers.  So if there were 2 helicopters on HMS Darling there could be 14 on HMS Illustrious, if 3 helicopters on Darling then 21 on Illustrious.    So while the relative statement can be applied for larger (and even much larger) scenarios, here it's applied in the lowest possible extreme scenario, but at the same time can, and arguably does, imply larger scenarios.  In the larger scenarios the "7 times" is helpful shorthand.  Rather than saying something in increased from 123 to 861, the "7 times" gives a better quicker sense.   So when we hear "7 times more" were more naturally drawn to assume it's shorthand for bigger numbers than simply 1 to 7.  

An there's another twist.  Of course while HMS Illustrious does have seven times more helicopters than the one on HMS Darling which is already deployed. The net increase in the Philliplines is of course only 6 extra helicopters.

One simple calculation step away from the raw data, but looses so much meaningful context in the presentation.  It's also a comparative measure rather than an absolute one.

This is a great simple demonstration of the powerful message around exploratory data analysis  The further we are from the underlying data, the greater the chances of misunderstanding and misrepresentation. Add to that the messages from the Statistical Process Control movement,  "No data has meaning apart from it's context".

28.11.13

Statistics Views - Summit

For a couple of days, the great and good of planet earth's statistics academics and closely associated players
came together to take stock on the past, present and future.  All as part of the International Year of Statistics, and hosted by our own Royal Statistical Society (11-12 November 2013).

So a mixture of presentations, workshops and discussion.  Much like a conference, but very efficiently high level.  The subject diversity was huge and very technical too.  Here's my key generic messages from all of that.

Collaboration is key. A much stronger emerging collaboration between disciplines, and even between disciplines.  Add a call to "divide and conquer" as approaches to big problems need breaking down for the solutions, then building up the solutions.

Big data.  It's not just big it's granular and partial ("missingness").  To that I would add "now" and "open".  So big, open, now, granular and partial.  Potentially a new scientific paradigm.

Numeracy Paradox.  Recognising different reading ages and languages, maybe we need to explicitly target different numeracy levels.

Transparent Representation.  A clear desire for informed and impartial.  Prof. David Spiegelhalter's always so very engaging and eloquent here.  The example of the high profile unemployment figure which increased by 34k, with the appendix small print which gives an error rate of +/-87k, which means it could in fact have gone down.

Graphics over numbers.  Some pioneering representation of statistical risk, especially around breast cancer, which both avoids probability and an overall judgement, just presents both pro's and con's graphically - icon arrays and frequency trees.  Breaking new ground here, and can be slow going.  It's make your own mind up based on your own circumstances and  what's important to you.

Now over future.  Communicating the risk to life is evolving.  Rather than describing the risk of various heath issues as effective length of life, implicitly focussed on lost time at the end of life, the ideas now is to measure the current health age of body components.  So you might well be 40, but as a smoker your lungs have are already 50. It's about the power of now and accelerating through life, ageing faster.... that cigarette ages your lungs and extra 15 minutes right now....

Anecdotal Reasoning.  Broad aspiration to reduce that all round.

And the prize for greatest technical phrase goes to super polynomial hyperbolic relaxations, closely followed by  The Bag of little bootstraps".

A scientific paper will come in due course, and designed as a lobbying tool for more stats skills too.