To ask the Secretary of State for Innovation, Universities and Skills

(1) if he will recalculate the higher education initial participation rate for each year between 1999-2000 and 2005-06 using the same methodology as that used to produce the figures for 2007-08 issued on 31 March 2009;

(2) what discussions his Department had with (a) higher education institutions, (b) bodies representing higher education institutions and (c) the UK Statistics Authority prior to the methodological change made to the calculation of the higher education initial participation rate for 2007-08.

The Statistical First Release published on 31 March estimated the Higher Education Initial Participation rate using a new methodology. The new method checks that an entrant aged 18 to 30 has not had more than six months' previous experience of higher education using data matched back over 12 years (10 years for FE colleges). As the SFR explained, 2007/08 is the first year that this has been possible using 12 years of matched data. However, in order to provide some time series comparison, and to demonstrate the impact of this change, the 2006/07 figure was also re-calculated using one fewer year of matching, as it was estimated that the missing twelfth year would make little difference to the rounded figure.

The figures published in the SFR provide a robust basis for monitoring progress in future years. Earlier years' figures based on the old methodology were also given. Estimates for earlier years using the new methodology would be progressively less accurate, since fewer years' data are available to check for previous experience of higher education, and so they were not included in the SFR. However, it is possible to calculate such estimates with an assessment of the possible scale of the bias. Additional estimates for earlier years, alongside information on how to interpret them and compare them to the existing published trend, will be added as an appendix to the published SFR on 15 May.

Following National Statistics procedures, the SFR release is timed to allow a minimum of time between production and release of the HEIPR, and prior access to the unpublished statistics is restricted. The change to the methodology was developed at a late stage in the publication process to resolve emerging concerns about the quality of the data. DIUS analysts consulted with HEFCE statisticians prior to release to confirm the robustness of the old and new methodologies. Given the fact that a robust alternative methodology was implementable in the time available, a decision was made by analysts to publish within the pre-announced timescale. In the interest of transparency, clear descriptions for the new methodology were shown within the SFR and figures for 2006/07 based on both the old and new methodologies were also provided.