“all models are wrong…some are useful”

george box
Think outside the box…

This day one year ago (March 28th 2013) saw the death of one the country’s most quietly influential exports, Professor George Box. Born just after the first world war in Gravesend, Kent and attending University in London for both his batchelor’s degree and his PhD, he ended up spending most of his life in the USA.

This modest and witty Kentish man, who stumbled into the world of statistical analysis is now considered amongst the top ten statisticians of all time.

I recall seeing a George Box presentation at the 1995 First World Congress on Total Quality Management in Sheffield; he described how industrial progress has been based on increasing knowledge and systematic design (he used a schematic of historic developments in shipbuilding to which I have often referred people to illustrate this point).

As a statistician made famous for the development of models to better describe  phenomena, Box’s most memorable quote was:

“essentially, all models are wrong, but some are useful”

If you come across a model, remember it is just that – something to map onto your mind to help make sense of things. If the model helps, make it something useful to you – by using it. But don’t waste time picking small holes in things – ‘worry selectively’ as Box would say. There is no ‘best way’ to see anything, but if there are fundamental flaws in a model, or it is useless, then drop it. As much as not worrying about everything, there is also little time to keep flogging dead horses!

Reading:

Box G.E.P. (1976) Science and Statistics. Journal of the American Statistical Association, 71 (356): 791-799

Box G.E.P. (2013) An Accidental Statistician, Wiley-Blackwell

Jones B. (2013) George Box: a remembrance. http://blogs.sas.com/content/jmp/2013/03/29/george-box-a-remembrance/

 

 

 

Culture “Change”: a new frontier or more disruption and waste?

The idea of ‘culture change’ has been around at least since the 1970s.

Company culture was flagged as the new route  to progress and competitiveness. A good company culture was seen as the antidote to inefficiency, obsolescence and lethargy. The old ways were habits to throw away, to be ashamed of, to turn our back on. People who don’t adapt are seen as dinosaurs or stuck in the dark ages.

This trend in thinking gave rise to a plethora of ‘culture change’ programmes, usually involving energetic efforts to describe company values, visions, extended programmes of training, sometimes introspection (on the part of managers), browbeating and exhortation (of employees), ‘communication cascades’, ‘town-hall meeting’ and suchlike. To support this, various four-box models, multi-ring schematics, life-cycles and illustrations sprung into life to describe this intangible ‘thing’ of culture. But is ‘culture’ a cause or an effect of what happens in organisations?

Peter Drucker stated: “company cultures are like country cultures. Never try to change one. Try instead to work with what you’ve got.” This is pragmatic thinking – and in many senses he is right – but not wholly so. A different perspective is needed, since sometimes the pervading culture can be damaging, counterproductive or simply unfair or unethical.

The culture in a company, or department, or team, or any type of organisation CAN actually be changed, but it is not achieved by trying to change the culture itself.  Seddon suggests that we should never make efforts to change a culture by ‘doing it to them’ (Seddon 2005). People will resent it  – and also people tend to detect any manipulation or ‘brainwashing’ a mile off. This increases resistance, undermines trust, garners cynicism and is generally unhelpful – the opposite of what you intend.

Don’t try to change people by attempting to change people, instead influence them to change themselves. The same is true of organisations. We can avoid a great waste of time, energy and resources if we skip this approach and instead work on things which really matter to people – and matter to our organisation. Just like forcefield analysis, it is better to identify and remove the forces that are driving the negative culture, rather than push at the positives.

The most effective approach is to intervene at the point of work.  Deal with the issues which people already find difficult or frustrating. Remove the conditions which impose upon them the negative behaviours which we want to eliminate. Give them a sense of purpose to fix their ideas upon – how things could change for the better and what THEY can do about it.

Reading:

Drucker PF ( 1993) Managing for the Future: the 1990s and beyond New York: NY, Dutton.

Jacobs, C.J. (2009) Management Rewired: Why Feedback Doesn’t Work and Other Surprising Lessons from the Latest Brain Science, Penguin Group Portfolio, NY

Seddon, J. (2005) Freedom from Command and Control, Vanguard Press, Buckingham, UK.