A culture of target-chasing has become all pervasive in our schools. Phil Armstrong looks at the issues
In education, the importance of targets has increased relentlessly in recent years. Schools are finding themselves suffocated by the need to meet or exceed targets. Failure to do so inevitably releases the forces of Armageddon in the form of a short but intensive Ofsted visit. The inspection itself is stressful and the judgements made are based on a highly questionable view of education both in terms of how statistical models should be used and, indeed, how students actually learn.
‘Winning isn’t everything, it’s the only thing’ is a well-known quote and refers to the need to be single-minded in the pursuit of victory. Whether this is the right approach to sport is questionable. Even more debatable is whether meeting results targets is the only thing that matters in education. The Ofsted model places such an overriding emphasis on targets that schools are forced to act as if results are the only thing that matters. Failure to reach their targets could trigger an Ofsted inspection based on a ‘negative hypothesis’ about the school, especially in terms of teaching and learning. A school has no choice other than to do what it can to improve results, ideally more quickly than its ‘competitors’. However, the pursuit of higher results for their own sake is ultimately damaging as all other aspects of education are rendered unimportant; results in education should not be the only thing.
The origins of the problem are traceable. Most of us accept that when schools perform well and provide a high quality education, examination results will tend to be good, at least in terms of ‘value-added’. If schools improve the quality of education, students will learn more effectively which, in turn, leads to better results. Higher results then, in a general sense, are a good measure of the quality of education. So far, so good. The problem comes when this relationship is relied on for policy. Goodhart’s Law suggests that once an established relationship is relied on for policy the relationship breaks down.
In simple terms, when a measure becomes a target it ceases to be a good measure! In the case of education, once results become targets, the relationship between quality of education and results breaks down. Schools will simply target result improvements directly using whatever techniques are deemed most effective for the purpose. A system that forces schools to put so much emphasis on results must inevitably lead to a reduction in educational quality elsewhere.
Since schools are in competition with each other they must use ever more sophisticated techniques aimed at the direct improvements of results. Simply improving results is not enough, they must improve faster than the average of their competitors. Schools must pay extremely close attention to the exact results-based criteria by which they will be assessed. For example, schools may become very effective at maximising their 5 A*-C grade percentage relative to entry data of students but may be ‘caught out’ by the move to using 5 A*-C percentage including English and mathematics. A ‘successful’ school must understand the results – based assessment criteria very well and design their strategy to improve results with the utmost care. The whole focus of the school must be on results, negatively affecting education as a whole. Students become ‘statistics’ and inevitably schools target borderline students to the detriment of others.
Schools will be forced to pressurise talented youngsters from an early age with constant target- setting and monitoring. For many young people the homework burden becomes too heavy. In many cases students become disenchanted and do not enjoy their education. In the same way, staff will become demotivated by a set of results ‘below expectations’. The process will also divide staffrooms as some staff receive praise for ‘success’ and others criticism for ‘failure’.
Schools may try to educate pupils in the best way they can rather than use ‘tricks’ such as finding ‘easy’ courses, pressurising individuals, providing selective attention or buying in ideas from examiners. However, the penalty for relatively slow result improvement set in an environment where schools in general are using ’tricks’ to improve grades quickly is heavy and schools are inevitably be forced to resort to their own tricks. The end result of this process can only mean grade inflation. Price inflation affects the distribution of wealth. Past savers’ sacrifice is undervalued as current incomes and prices rise.
In terms of grade inflation, previous students who have worked hard to achieve particular qualifications find the relative value of these qualifications reduced as each new set of students emerges into the market for jobs and higher education with inflated grades generated by schools desperately striving to raise grade outcomes. Grade inflation beyond a certain level will reduce the acceptability of the grades and confidence in the system. Ultimately, this will require a new grade ‘currency’ to be issued to restore confidence (rather like the German government issuing the Rentenmark to replace the Reichsbank mark after its period of hyperinflation in 1923). This appears to be happening at the moment with the arrival of ‘tougher’ GCSEs and A levels.
The situation is even worse than this since the models used to generate estimates which provide the targets are themselves flawed and also used wrongly on a systematic basis. A good example is the ‘Alps’ model. The model bands students according to GCSE results and generates frequency distributions for A level results in a base year. The results are then used to provide estimates of expected performance in future years i.e. current results can be compared to these estimates and assessed to arrive at a ‘value-added’ score. Scores range from one (‘outstanding’) to nine (‘poor’). Scores one to three are sought-after and are coloured red on the Alps ‘thermometer’ whereas scores seven to nine are to be avoided at all costs and are coloured blue. (Scores four to six are white on the thermometer). Unfortunately, students in any given band are not homogenous over time, reducing the statistical validity of the model.
It is reasonable to believe that such models, based on past frequency distributions, may give a useful guide to expected future attainment. They could be used with some degree of validity to assess progress if they were used in conjunction with teachers’ judgements and other information. They could be treated as ‘aspirational’ and used as a motivational tool. However, they are not robust enough to be used as the sole means of assessing the success of a department or a school. Sadly, in many cases, this is just the way they are being used.
Developing this theme, if a school or department were fortunate enough to record a red ‘Alps’ score this would mean that its results were as good as the top 25% of schools or departments in the base year data set. There is of course pressure to maintain performance. Clearly, a one-off ‘red’ score is not enough – the pressure is on to stay ‘red’. In statistics, the phenomenon of ‘regression to mean’ is important. This means that if a variable is extreme on first measurement it will tend to be closer to the mean on the next. If a school has an Alps score of one in a given year it is likely that the score will move back towards the mean (five) in the next. A score of, say, four in the next year should not be taken as an illustration that the school is in decline!
The ‘Ofsted’ model is also very prescriptive in terms of teaching and learning. It suggests that active learning and peer and group work is, always and in every case, superior to teacher-led lessons. It may be that the former approaches have much to commend them but, surely, they cannot always be better for all students. Teachers have different personalities, styles, strengths and weaknesses. It is the freedom to play to their strengths that allows teachers to teach effectively and students to learn well. It is the variety of teaching approaches that students experience that makes them enjoy school and want to learn.
Ofsted stresses the overriding importance of ‘progress’ schools must show is being made over all time periods. Even in individual lessons or parts of lessons progress must be made. However, young people are not machines; they do not make linear progress. Sometimes they progress faster than others. Sometimes, dare I say it, they stay on a plateau and consolidate. Sometimes a teacher must take a class backwards a little in order to take them forward a lot. Students may possess prior knowledge that brings comfort but is incomplete or even wrong! The teacher needs to point this out and then rebuild the knowledge and understanding on a firmer footing (or perhaps help the students do this for themselves- both are fine). So it is quite possible, indeed likely, students may make little or even no progress (in an orthodox sense) in some lessons or parts of lessons – that will come later.
I would argue that statistical analysis is a useful tool which helps teachers to assess their pupils’ progress. However, it should not be the main driver in the evaluative process and should complement the teacher’s own judgements. The quality of education will only improve in the long term within a flexible system where teachers are given freedom and are actively encouraged to teach to their strengths and, above all, to inspire. Teachers must teach beyond the syllabus to do this and not reference learning only to externally generated examination criteria. Education is partly about monitoring, evaluating and training for examinations but there are other vital elements to it. Concentration on these areas alone will suffocate the supply of spontaneity and inspiration in education. Attempts to drive the process in the Ofsted way will ultimately be self-defeating – examination results will improve but the underlying quality of education must and will decline.
Reflecting on my thirty years in teaching I would like to think that the system I work in fosters schools’ ability to improve children’s educational experience. Sadly, especially in recent years, this has not been the case. It is even less likely to be true in the near future unless there is a radical change in priorities. Raising standards is not about designing a system that generates enormous pressure on schools to relentlessly raise grades outcomes. It is really about providing young people with the type of education that enriches their lives. If our education system does not change to reflect this fundamental vision, in the final analysis, it is our children who will pay the price. If we tolerate this our children will be next.