§
The Secretary of State may publish such statistics as she deems appropriate to inform the general public of the progress made by any element or elements of the statutory system of education, and shall ensure that, as far as reasonably practicable, such statistics are provided on the same basis in every year and that when changes are made in the basis of calculation, restated statistics are provided for the previous three years."
The noble Lord said: As a user of statistics in this House to help to hold the Government to account, and as a republisher of many government statistics in the course of my business for the Good Schools Guide, I have a great interest in the quality of the statistics which come from government. The amendment is in large measure unnecessary because the Secretary of State already has the power to publish statistics. I am seeking to impose that, when the Secretary of State changes the basis of calculations, restated statistics for previous years are provided where it is sensible and possible to do so.
A cause of immediate irritation to me are the changes made last year to the ways in which A-level performance tables and figures for 16 to 18 year-olds 1329 were presented. When that was done in Scotland, a three-year history was produced to provide a time series. In England, we had a change of basis but no time series. Time series are particularly crucial in respect of performance tables. A single year's figures tell one a great deal less than a three or five-year series—which will give a picture of consistency or progress. At any rate, a series provides much more data than one can obtain from one year's statistics—which may be considerably biased in one direction or another in relation to a school's average performance.
It is important to my business that the figures are right and that every other user can see how the statistics vary from year to year. Government should care about producing statistics that are usable by the public. I am sure that the Minister knows of other examples where the basis of statistics has been changed and the figures have not been restated. That is a bad principle because statistics become useless. Sometimes changes are necessary for political purposes—both parties have done that with unemployment figures as a necessary way of disguising bad news. Performance statistics are not a political plaything and government should take responsibility for providing quality information. I beg to move.
§ 1.45 a.m.
§ Baroness BlatchHistorical information and the reporting of trends is important, so that parents can make judgments about their children, the classes and schools that their children attend and even educational progress within the area in which they live. The information in question is for parents with a lay understanding of educational jargon and could be improved.
§ Baroness Sharp of GuildfordI too support the amendment. Coherent sets of statistics running over time are extremely important.
§ Lord Davies of OldhamI am one of those people who wince at the word "statistics", not least because one of my notable failures was not completing a university course in statistics. I was overtaken by a statistic that counted—I got more votes than someone else in a general election, so I left the course. That left me bereft of the ability to analyse statistics, which I have never been able to make up. I will therefore confine my contribution to the argument made by the noble Lord, Lord Lucas, and supported by the noble Baronesses.
I accept entirely that public information belongs to the nation and that government must provide the best and clearest information that they can. The Secretary of State shares the desire to see an unbroken statistical time series on education, and endeavours, where possible, to provide consistent information. The new national statistics code of practice requires all national statistics products to produce bridging tables for breaks in time series and recasts of prior years' figures where possible. The intention is that the code will cover not just the Department for Education and Skills but all government statistics. We are at one in terms of objectives.
1330 The noble Lord is particularly interested in information published in our annual schools performance tables. I reassure him that we shall make every effort to keep changes in the calculation of performance indicators to an absolute minimum.
I hesitate, however, to accept the amendment. There is a necessity to reserve some flexibility to respond quickly to concerns about particular circumstances that might damage an institution's credibility in what are high-profile publications. Given that the tables are based on individual pupil information, a recast of earlier years' results will not always be a reasonable proposition.
When using new information, the collection of data from schools on pupils in the same circumstances over the previous three years would be hugely burdensome and impracticable, as the noble Lord will recognise. For example, in order to reflect the success of the Qualifying for Success reforms, last year's performance tables reported the advanced level results for students at the end of two years of post-16 education. To enable that, schools and colleges had to tell us who the students were. It would not have been reasonable to ask them to identify the equivalent group of students over the previous three years. The noble Lord will recognise that difficulty. However, wherever changes are made to the presentation of information, we will endeavour to provide recasts whenever it is practical to do so without overburdening schools.
I hope that, on the basis of those assurances, the noble Lord will feel able to withdraw the amendment.
§ Lord LucasI shall certainly withdraw the amendment. The department has all that pupil-level data in its computers. It would not be that difficult for it to regurgitate the information in a different form. It had the entire A-level and GCSE database by pupil many years ago. I remember having access to that and playing around with it on my computer to examine the way in which some exams were harder than others. Doubtless, Ministers play with it similarly in their many peaceful hours in their offices in the department.
It is not that difficult to produce back figures, if the information has been collected. The department has been assiduous over many years in collecting information. I can understand that, with the changes made last year, there may have been some difficulty in going back. However, for the benefit of new policy, I hope that it will be possible to have the quality of information to enable the time series to be recast.
I do not like the idea that statistics cannot be changed. One must keep changing the basis of statistics—first, because there are always better ways of doing things and, secondly, because, if a statistic is left static for too long, the whole school system will be biased towards that statistic. We must keep things moving and keep the basis on which we measure schools changing, so that we are measuring the real school and not just results produced for the purpose of the statistic. To do that and still produce useful time series, we must have enough data to recast at least the 1331 previous two years' information in the form in which it has been produced for the current year. That is a basic requirement for the collection of statistics, and I hope that we shall see that in future. I beg leave to withdraw the amendment.
Amendment, by leave, withdrawn.
§ Baroness Finlay of Llandaffmoved Amendment No. 363ZA:
After Clause 202, insert the following new clause—