Paying Teachers by Results

Paying Teachers by Results (Research in Public Policy, Autumn 2005)

Research concluding that the introduction of Threshold pay points for teachers improved pupil performance by on average half a grade for pupils of eligible teachers. This despite arguments beforehand that teachers were professionals with a service culture and would not respond to financial incentives. Counterpoint to pre-seen material argument that professional control is adequate.

Also makes the point that the threshold system is not that rigorous – most applicant were successful. But teachers beforehand believed that it would be challenging, making it real.

School Performance: How Headteachers Respond to Measurement

School Performance: How Headteachers Respond to Measurement (Centre for Market and Public Organisation Bulletin, Winter 2005)

Behavoural implications: Because of the complexity of the measurement system, parents fixate on %5A*-C. As a result head teachers fixate on this measure, in their roles as marketers of their schools. Resources are then targetted on borderline C-D students. They feel that more complex, value-added measures will not replace the simpler result measure in the minds of parents. Also worth noting that heads are not using the externally reported measures internally – bit of a disconnect.

Private schools say league tables create fear of failure

Private schools say league tables create fear of failure (PF 2/5/08)

Negative behavioural consequences of PM schemes. E.g. tendency to be risk averse. Some pupils were prevented from taking part in extra-curricular activities for fear of harming exam results. Eton College and St. Pauls have refused to submit their results in future.

The numbers game, by Tony Travers

The numbers game, by Tony Travers (PF 23/11/07)

Difficulties in measuring public sector output: e.g. putting a teaching assistant in a class causes a fall in productivity, spending more on health has led to a fall in NHS productivity. Are the techniques for measuring productivity really helpful or appropriate? Leads to temptation to make excuses by attacking the numbers.

GCSE results show upwards trend

GCSE results show upwards trend (PF 26/10/07)

PM more useful in cases like this: GCSE results have intrinsic value in themselves – they aren’t just a measure. PM based on result trends can therefore be useful.

Schools face struggle to meet new PSA targets

Schools face struggle to meet new PSA targets (PF 12/10/07)

Targets to raise standards and to close the attainment gap could work against each other as the former helps the brightest pupils improve, at the expense of struggling pupils.

UCU on RAE

http://www.ucu.org.uk/index.cfm?articleid=1442

University and College Union on the RAE.

“The RAE has had a disastrous impact on the UK higher education system, leading to the closure of departments with strong research profiles and healthy student recruitment. It has been responsible for job losses, discriminatory practices, widespread demoralisation of staff, the narrowing of research opportunities through the over-concentration of funding and the undermining of the relationship between teaching and research.”

RAE is extremely transactional.

League tables ‘encourage soft exam choices’

League tables ‘encourage soft exam choices’ (Times Online 25/7/07)

Schools encourage students to do ’softer’, easier subjects in order to inflate performance results and league table positions.

Failsafe futures, by Conor Ryan

Failsafe futures, by Conor Ryan (PF 18/4/08)

What do targets mean? Discusses 638 schools below government’s floor target of 30% of pupils getting at least 5 good GCSEs including Maths and English. However, if you look at ‘contextualised value added’ performance, many of these are performing above average. Some have ‘good’ Ofsted reports.

No end of a lesson, by Anthony Browne

No end of a lesson, by Anthony Browne (PF 8/6/07)

Need for markets in public sector. Education – need to loosen up supply side to let new schools be set up and good schools expand. Neither command/control nor targets really work.