The Scottish Government

« Previous | Contents | Next »

Listen

PRE-RELEASE ACCESS

Proposal:

29. The consultation paper "Improving Statistical Services in Scotland" proposed to draw up detailed plans for pre-release access to Scottish devolved statistics that will reflect the different types of government statistics. It asked:

• What do you consider to be the different types of government statistics?
• Should different pre-release access rules apply to different types of government statistics?

Findings:

Consultation Paper

30. None of the respondents to the consultation paper objected to the principle of pre-release access. Nor was the timing of the pre-release access period the key issue for respondents. Transparency and clarity of the rules were seen to be paramount.

• 'The comments for and against the pre-release of publications are both convincing however, I agree with the conclusion that pre-release access should be retained and detailed plans for pre-release access to Scottish devolved statistics.' (Clackmannanshire Council)
• 'Other than for market-sensitive statistics, there appears little benefit (and much scope for confusion) in drawing distinctions between different types of Government statistics.' (Dr David S Gordon)
• 'Pre-release of statistics to ministers and policy officials is accepted practice in the UK. As long as this is well known and the time allowed for access is agreed, then we believe that this can continue but should be limited to the shortest possible time.' (Falkirk Council)
• 'Highlands and Islands Enterprise agrees with the statement in the consultation paper that pre-release access should be retained but that some restriction is desirable, to provide assurances that there will be no trailing of statistics and selective early release; this would fully meet international standards' (Highlands and Islands Enterprise)
• 'The guidance covering pre-release access in the consultation paper should result in a transparent decision making process.' (Scottish Museums Council)
• '..public confidence will best be enhanced if the rules for pre-release access are determined by the Board as part of the Code of Practice, not - as the Bill proposes - determined separately by Ministers.' (Statistics Commission)
• 'We think that the onus, in relation to any particular set of statistics, should rest on demonstrating why pre-release access is required, and why each person granted access should have it, rather than on defining large sets of statistical series that would all be subject to the same inequality of access that pre-release arrangements afford.' (Statistics Commission)
• '..the rules for pre-release should be accessible, transparent and adhered to. There should only be a small set of rules to avoid over-complicating the types of pre-release permitted.' (Scottish Water)
• 'It would be inappropriate for the Executive to apply the same pre-release access rules to market sensitive statistics as to, say, secondary release statistics produced from data already published.' (Stow College)

Finance Committee Scrutiny

31. In its final report, the Finance Committee flagged up the strength of opinion on the pre-release access issue and suggested that consideration should be given to these concerns in developing proposals for pre-release access arrangements, including considering the possibility of transferring responsibility for pre-release access rules to the Authority.

Parliamentary Debates

32. There was no mention of pre-release access arrangements when the legislative consent motion to extend the Act to Scotland was taken in the Scottish Parliament, which may signify the relatively low profile that this issue has in Scotland.

33. In contrast, during the Westminster debates on the Act, pre-release access was one of the most controversial issues. At Westminster, agreement to Ministers setting the rules for pre-release access was secured by the addition of a new provision to the Act to ensure that the UK Statistics Authority would be consulted.

34. The Westminster debates focussed on the negative aspects of pre-release access (for example, risk of selective leaks and preferential access for government). The debate therefore concentrated on limiting the period of pre-release access, with little consideration of its contribution to good governance and the integrity of the statistics.

35. The UK Government have since proposed, in the publication "Limiting pre-release access to statistics: a consultation document", to reduce pre-release access to 24 hours for all their statistics ¹¹.

Scottish Administration Discussions

36. Changes in pre-release access arrangements have the biggest impact on those that produce the statistics and those that are expected to comment when the statistics are published. Thus it was very important for us to consult those within government on the pre-release access rules also.

37. Although timing of pre-release access was not a big issue for respondents to the consultation paper, it became apparent that statisticians like to work to a timing norm and that adequate time is needed for statisticians to manage the briefing and influence the accuracy of what is said publicly.

Scottish Government Response on Pre-release Access:

38. The Act does not allow for the pre-release access rules to be set by the UK Statistics Authority in its Code of Practice for Statistics. Scottish Ministers are to set out the rules and principles in an Order to be considered by the Scottish Parliament.

39. The pre-release access arrangements apply to access to Official Statistics in their final form prior to publication. This does not interrupt arrangements for quality assurance, allowing statisticians to enlist the help of experts on the policy or geographical area that the statistics are concerned with. Quality assurance is a vital step in the production of statistics.

40. The principle of pre-release access to final statistics publications, to allow Ministers and others to prepare for the public release, is accepted - because it is understood that Ministers and others are expected to comment at the time of release by the public, Parliament and media.

41. In the light of the consultation comments, we have prepared a draft pre-release access Order for Scottish devolved statistics, which is presented at Annex C. Comments on the draft Order are invited. Please let us know your views via the ScotStat website atwww.scotland.gov.uk/scotstatup until 1 August 2008. This is shorter than the usual 3 months consultation as we have already consulted on pre-release via the "Improving Statistical Services in Scotland" paper. This is a final chance for users to express their views on the draft Order specifically.

42. Our approach to pre-release access in Scotland recognises the benefits of pre-release access and the role of the professionally independent statistical staff in applying the rules. In reaction to responses to the consultation paper, the draft Order sets out a transparent and clear approach to pre-release access in Scotland.

43. We recognise that pre-release access is not a privilege reserved for Ministers for political purposes. It is needed for good governance and to ensure the integrity of the statistics. Hence it should be controlled by the professional statistical staff. This desire to give the decision-making powers to statistical staff, independent from Ministers, as a matter of law is not incorporated in the draft Order itself because the enabling Act does not give Chief Statisticians or Statistical Heads of Profession a legal status. However, the intention is that the decisions on how pre-release access works in practice - including determining which statistics are eligible for pre-release, who gets pre-release access and ensuring awareness of the rules - will be taken by the Scottish Statistical Heads of Profession¹². The Scottish Ministers are therefore proposing to accept these officials' professional advice in all cases (subject to their legal duties to properly exercise their own functions).

44. The draft Order sets out explicitly the legitimate uses of pre-release access:

• to ensure that Ministers, public bodies and others understand the statistics and are able to comment on them properly at the time of release without misinterpretation;
• to ensure that where Ministers already have access to data for management purposes, this is not used in a way which would damage public trust or the business of government. For example, where statisticians are preparing official statistics publications they might identify errors in other related figures that are available directly from the source of the figures e.g. health boards or local authorities. Statisticians should be able to make Ministers aware of such issues - especially where these figures are intended to be used in decision making or in public statements;
• to ensure that publications or other material being prepared for release at the same time as the statistics (or shortly after) are properly informed by the correct figures; and
• to allow access where necessary in exceptional cases provided this would not damage public trust in official statistics.

45. The draft Order sets a workable time norm to allow all these legitimate uses to be carried out efficiently - a normal maximum of 5 working days but a maximum of 1 working day for market sensitive statistics to reduce the risk of leaks. This is in line with current practice, which works well. The statistical Heads of Profession can decide to vary these periods if, in their opinion, more or less time is required to fulfil the purpose of pre-release access. Where a statistical Head of Profession increases the maximum he/she must publish that fact.

46. The draft Order contains important conditions to prevent Ministers, officials and others from commenting on the statistics before the public release.

47. When considering the approach to pre-release access in Scotland, we struggled to find information on the pre-release access regimes followed by government departments that produce official statistics around the world. This is because internationally there tends to be a focus on the pre-release access arrangements and rules followed by National Statistics Institutes ( NSIs) which do not produce the full range of government statistics and where Ministers have a reduced role and therefore have a reduced need to be well briefed on their activities. Hence the pre-release access regimes followed by NSIs are not necessarily appropriate for government.

48. However, we are keen to learn from other countries and to adopt best international practice, which is appropriate to our circumstances. Therefore we will review the pre-release access arrangements 12 months after implementation. This review will include:

• an appraisal of how other countries manage pre-release access to statistics, especially those produced by government departments from management information, such as hospital waiting times and school examination results.
• how the rules in the Pre-release Access order for Scotland, and the arrangements in the rest of the UK, have been implemented.

« Previous | Contents | Next »