Teacher Quality and Education Structures

The currently popular mantra (among politicians) that “an education system”, or even a “school”, “cannot be better than the teachers it employs” is vacuous. On the other hand it is not meaningless. Even rank nonsense, and this merits consideration for that title, can derive meaning from the way in which it is used.

If we don’t think about it, the mantra appeals to a common sense notion that the parts of a structure/system can impose a limit to its overall functioning. Famously, the US space shuttle The Challenger disintegrated 73 seconds after lift off. The whole complex system failed because of its O-ring seals which failed. QED? Not really.

In the early days of computing, when computers were so large that one could walk around inside them, Westinghouse would hand wire cutters to visitors on a tour with an invitation to snip some wires at random. The point of the demonstration was that the computer had such large in-built redundancy that when one circuit failed another would take over thus demonstrating that system redundancy can the compensate for parts failure.

The amplifying component is the main part of an electronic amplification system. However even when that component has far from the desired characteristics a system can be built which has them by using the appropriate feedback circuitry. So long as the gain of the component amplifier is much greater than the gain required of the system it is easy to show that the system gain is independent of the gain of the component amplifier. In other words the system connections determine the ultimate performance rather than that of its component parts.

None of this new. Writing of machines in general in his An Introduction to Cybernetics (1956), Ross Ashby wrote: “That a whole machine should be built on parts of a given behaviour is not sufficient to determine its behaviour as a whole: only when the details of couplings are added does the whole’s behaviour become determinate”. (Emphasis added)

Finally, system performance can be crucially dependent on its in-built error regulation mechanisms and this again can override weaknesses in component parts of the system.

If all this is true of relatively simply mechanical and electronic systems it is far more significant when we deal with more complex systems such as those of biology and society. The view therefore that the quality of a system cannot be greater than that of its component parts is, to put it bluntly, crass.

An education system is extremely complex and is embedded in an even more complex social system. It is therefore shocking that politicians talking about education should want to ignore the richness of the system connection and reduce pretend that overall quality is down to that of one component. All the same there is a clear meaning to such claims. They are all of a piece with Mrs Thatcher’s famous “There is no such thing as society”.

These reductionist views, which exclude the decisive impact that the richness of a system’s connections can have on its overall performance, are a key part of the neo-liberal approach to society. Neo-liberals favour ‘market democracy’ in which a summation of individual choices made in the market place leads (they claim) to an optimisation of the ultimate outcomes. Contrary to this is a view of democracy as being a matter of decisions made through due deliberation leading to a collective choice. The results produced by these two approaches can be very different and even diametrically opposed. The differences are as clear as they could be in education.

Neo-liberals tell us that only if parents have free choice in an educational market with diverse provision can standards be effectively “forced up” (as they like to say). Those who favour democratic collective decision making argue that setting schools and parents against each other in this way will ensure that there will always be failing schools. They argue that the market is the wrong mechanism and that only a collective and democratic approach can ensure that all schools are good (thereby dissipating the need for individual choice). Most parents want to send their children to the local school safe in the knowledge that it is a good one. Markets might make for better biscuits. We can throw away the ones we don’t like and buy different ones next time. Not so with children.

Schools and school systems are (or should be) connection rich systems. This means that they will include: (1)  a certain amount of redundancy so that when one part fails another can take over. This requires, of course, that teachers are not pushed to the limit and have some spare capacity; (2) appropriate error regulation in the form of picking up on early warning signs of things not working as planned through teachers working together and sharing problems as they arise; (3) a connection-rich system rather than the command and control system that is becoming such a feature of our schools, building instead a collegiate approach to allow for maximal feedback from teachers to management and vice versa; (4) an external local democratic system in which the school is embedded which will enhance the operation of the preceding mechanisms through openness and dialogue with the wider public.

None of this is to argue that we need anything other than well-qualified, high-quality teachers. The point is that this does not, by itself, guarantee system success. Making teachers into the be-all and end-all of the success of the system may seem at first to be due recognition of their importance. But by rejecting discussion of system structures both the reality and impact of the system is denied and teachers are lined up for the blame when things go wrong.


5 Comments on “Teacher Quality and Education Structures”

  1. David, thank you for an excellent post. You’ve explained very clearly why systems cannot be understood simply as the sum of their parts – they can have very different properties. Each level requires an appropriate level of study, so the effectiveness of an education system can’t simply be read across from the effectiveness of its component schools just as social trends aren’t simply explainable at individual level. Complex connections and interactions have unpredictable aggregate effects. Even David Starkey in his recent “it’s all about memorising facts” then had to add that the connections between facts and the new meanings they build are also important. It is blindingly obvious that in practice the market does not lead to better public services.
    Thanks for a rigorous contribution to an important debate.

    • David Pavett says:

      Thanks Eddie. Perhaps an easy way to sum it all up for politicians who clearly do not weigh their words, is to say “An education system which is entirely staffed by poor teachers could not be good, but everyone agrees that the majority of our teachers are good or better. A well designed system will help poorer teachers to improve and will enhance the performance of even the best teachers. A system staffed entirely by excellent teachers could still produce poor results if badly organised.” The constant reiteration that a school, or an education system “cannot be better than the quality of its teachers” is crassly misleading and one can only doubt the educational credentials of people who talk like that.

  2. Terry Loane says:

    I also want to say thank you, David, for writing so clearly about some of the implications of systems thinking and complexity science for the world of education. Your post was like a breath of fresh air in a stale room (the staleness being the overarching reductionist assumptions behind most contemporary educational discourse).

    I liked the way you emphasised the nested nature of the systems involved: the school within ‘the education system’ within society as a whole. And we can identify other nested systems involved in education and human learning: the individual human brain; a cohort of students; a community of practice. All are complex non-linear systems and so none can be understood by measurement or by analysis of its components. As you correctly point out, knowledge of complex systems enables us to see that it is nonsense to say that “a school cannot be better than the teachers it employs.” Other examples of reductionist nonsense that nevertheless dominate educational discourse are, I suggest:

    # Graded lesson observations – Since each lesson is a unique, complex, non-linear system it cannot be compared numerically with another

    # The requirement that teachers should produce lists of so-called learning objectives for each lesson – Learning is an emergent property of the system that is “unpredictable but retrospectively coherent” (Williams et al, 2011: http://www.irrodl.org/index.php/irrodl/rt/printerFriendly/883/1686) so we cannot specify a priori what learning will or will not happen. We can only examine the learning post hoc.

    # Using exam results and other numerical data as the basis for comparing individual students, individual teachers and individual schools. Again it follows from the complex, non-linear and unpredictable nature of the systems involved that this is nonsense. Actually it is a huge irony that those in power in the world of education are fetishizing measurement almost a century after scientists (e.g. Werner Heisenberg, Ludwig von Bertalanffy, Henry Landsberger) began to understand the limitations of measurement and the need to look beyond simplistic data.

    # The increasing prevalence of top-down, non-collegiate, ‘command and control’ management by objectives in education – I blogged about this a year ago here: http://terryloane.typepad.com/reallylearn/2013/05/command-control-and-the-real-world-of-education.html

    And of course the tragic result of relying on simplistic, reductionist approaches in education is that the educational experience itself becomes hollowed out. Teachers can focus either on their students as people, developing their learning and wellbeing, or they can focus on the numbers – grades, reductionist objectives and league tables. They cannot do both.

    Thank you again, David.

  3. […] and districts), which requires layoff decisions to be based primarily on seniority, as opposed to teacher quality or other factors. Echoing concerns voiced by a growing number of politicians and reformers in […]