Review of pre-release access in Scotland

Evaluating current processes around pre-release access to Official Statistics in Scotland

This section draws mainly on data from interviews and focus groups with stakeholders who work with PRA to explore the ways that PRA is currently used, and the benefits and challenges associated with it, participants’ experiences and views of changes to the PRA period for economic and market sensitive statistics and of the differences in PRA between Scotland and other parts of the UK. Reflections from the discussion which took place with the public engagement group are also included where relevant.

Purpose and current use of PRA

Generally, the ways that stakeholders used PRA across the different professional groups aligned with what they perceived to be the purpose of PRA, and with its stated purpose in the Pre-release Access to Official Statistics (Scotland) Order 2008. That purpose is, broadly, to ensure that Ministers and officials have an accurate understanding of the statistics and their implications so that any comments made about the statistics at, or shortly after, publication can be properly informed. The main purposes of PRA identified by participants were:

• To give relevant actors who have not been involved in the production of statistics time to understand, interrogate and query the statistics and how they have been produced and, particularly in cases where statistics are published annually or infrequently, to understand statistics in the context of general trends and previous years.
• To enable the effective briefing of Ministers so that they understand the statistics and are prepared to communicate them clearly and accurately and answer questions on them from the media, opposition and in Parliament.
• To allow the preparation of communications lines and materials and any quotes or Ministerial statements prior to publication, and to have these lines reviewed and signed off by the various levels of approval.
• To allow those who provide statistics or data at a local level (such as health boards or local authorities) to prepare for their release.

…ensure that Scottish Government officials, policy and communications officials, and also ministers can be involved in ensuring that at the point the stats are released that a minister can speak on an informed basis about them to make sure that any proactive statements that the government makes are correct, properly informed by the stats and if we need to make any reactive lines, they're prepared and everything, they’re done properly, just making sure that ministers know what's coming out, really, that's the main reason that we do it. (Scottish Government statistics producer)

In addition, participants used PRA for activities which they did not consider to be the main purpose of PRA. These were:

• To enable the Scottish Government to develop its own publications which combine official statistics collected by external statistics providers (e.g. National Records of Scotland, Public Health Scotland) with internally collected statistics and policy content so that these could be published to coincide with the release of the statistics.
• For quality assurance of statistics, correcting errors or inaccuracies and proof-reading of publications.

It was noted by participants that quality assurance was not the purpose of PRA, but that this time did allow any errors or inconsistencies in the statistics to be queried and rectified.

While the 5 day period of PRA was generally standard, participants noted that they did not always receive the statistics a full 5 working days in advance of publication and that PRA was sometimes delayed. There were some exceptions to the 5 working days where shorter periods of PRA were used, such as in the case of statistics which are published weekly or very regularly where PRA was only given 1-2 days prior to publication due to the short turn-around period for this data.

In general, participants described the process of receiving PRA as robust and that the restrictions around how the data could be used and who it could be shared with were well communicated. In instances where those receiving statistics under PRA did inadvertently share these with people who were not on the list to receive PRA, this was described as being promptly dealt with, and several participants reflected that they could not recall any recent incidences of this.

Benefits and challenges

Participants identified a number of benefits and challenges associated with the current arrangements for PRA. The stated benefits generally aligned with views on the purpose of PRA.

It was seen as beneficial for officials to have to time to distil, understand and query statistics. Stakeholders spoke about the complexity of some publications, some of which contain “a panoply of information” and stated that it would not be possible to read and understand these in a short space of time. It was also beneficial for stakeholders to have time to understand the methods by which statistics had been collected and collated, and any changes to the methodology. Where there had been large or unexpected changes in statistics, participants found it helpful to have time to investigate what may be driving these changes.

This benefit was particularly marked for the publication of health statistics, due to the number of publications which are released on the same day and the complexity of these statistics. Participants gave examples of up to 15 publications being released on the same day and valued the 5 day PRA period for allowing them to navigate and engage with these multiple publications.

Having time to brief Ministers in advance of publication of statistics was seen as beneficial in supporting the Minister to respond in a way which is clear, accurate and does not misrepresent the statistics during media interviews or in Parliament or Cabinet meetings. It also allows the Minister to respond to questions from an informed position and to speak about relevant policies, and to give a response which reflects the Scottish Government’s overall policy approach. In addition, PRA allows officials to prepare to brief other Ministers who may be asked about the statistics but who do not receive them under PRA as soon as the statistics are published.

In terms of media and communications, PRA was seen as beneficial because it allows the government to respond to media enquiries without delay, and to effectively counter any inaccurate reporting on the day of publication, rather than dealing with queries from internal stakeholders. There was a perception that opposition parties are able to mobilise a large number of staff to scrutinise a publication as soon as it is released and may put out press releases without a full understanding of the statistics and how they have been produced. For these instances, PRA was perceived as allowing the government to counter with an informed response. PRA also allows all parties who are involved in developing a response to be involved and for the response to go through the various levels of approval, including review by statisticians, to ensure that any communications messages do not misrepresent the statistics. It also enables communications teams to develop infographics and accessible ways of sharing key messages which were seen to increase engagement with the publications, even if the statistics were not picked up by the media. In cases where a centralised communications function is provided for other organisations, PRA allows communications message to be developed with these organisations and for alignment in the messages put out by both organisations. Prior sight of the statistics freed Ministers up to be able to attend visits to publicise the release, which was seen to enhance the public profile of the statistics and increase engagement with them.

…when the stats come up to have our position and if there are things out there that are being inaccurately reported, we can just rebut them straight away rather than having to go through the kind of slightly insane Scottish Government clearance processes that can take five or six hours to get a line out there, and by that point this kind of thing has been spreading like wildfire. (Communications stakeholder)

The additional time to check and quality assure statistics was also highly valued. While participants noted that this was a rare occurrence, several examples of incorrect statistics or issues with analysis which had been identified during the PRA period and rectified were given.

While some participants noted that there could be the perception that some officials having prior sight of official statistics could be detrimental to public trust in statistics, these stated benefits of PRA - Ministers being well briefed and able to respond to questions, clear and effective communications messaging and identifying and correcting any errors in the statistics - were thought to increase the accuracy and clarity of communication of official statistics to the public and, in turn, to increase public trust in both official statistics and the government.

I think having access to pre-release statistics vastly improves the government's ability to respond and to be able to tell the wider picture of what's going on. I think that enables the public to have more confidence in government's handling and also competency. (Special Adviser)

And also just making sure that those who are actually speaking publicly about the statistics are using them in the correct way because if they didn't use them in the correct way, either misunderstood the statistics and started speaking about them, then that would reduce the trust in publics. (External statistics producer)

In addition, stakeholders working in policy felt that PRA allowed them to add value to statistics publications by highlighting any relevant changes in policy. The example was given of policy teams adding information on changes to eligibility criteria for certain benefits to publications to improve accurate communication and engagement with these benefits.

Finally, PRA was thought to be beneficial in workload planning for departments, as it allows the workload to be spread across team members, particularly for those who work part time or compressed hours, or in cases of holidays or staff sickness.

…the great benefit of a working week, it seems to me, is as a team leader, we're, you know, if that is a release that's happening during the summer holidays or at Christmas when people are off on leave or you know otherwise coming and going or somebody is sick, who's very critical to it, it gives us more resilience. (Policy staff member)

The challenges which were associated with PRA tended to be more operational than ideological. The challenge which was most commonly identified was the difficulty associated with the limits placed on the number of people on PRA lists or with swapping someone on the list who would be unavailable during the PRA period. Participants found it difficult to know who was on the list and shared instances where they had been missed off the list. For those who granted PRA, there were difficulties in judging whether someone needed to be added to the list as they often did not receive enough detail as to why people needed to receive PRA.

Less commonly, participants mentioned instances where statistics had been inadvertently shared with individuals who were not on the approved list to receive PRA by being copied into emails but felt that this happened infrequently and was addressed appropriately. One participant highlighted that PRA did give the opportunity for statistics to be leaked but noted that they couldn’t remember any instances of this happening within the Scottish Government. One participant felt that there was an inherent unfairness in that the party in government was able to see the statistics in advance, whereas PRA was not available to those in opposition and others interested in the statistics who would like to be able to comment on the statistics as soon as they were released. However, another participant felt that this view belied “a lack of understanding of the responsibilities and of what has to happen within a government around the statistics from those who are not accountable for them” (Special adviser).

So where the imbalance is most marked is the political part of it, because you've got, you know, if you happen to be in government, you get it a week in advance. If you're in the opposition benches, you're not, you're not getting that access. (External statistics producer)

When asked about their views on PRA being limited to the political party in government, some public participants thought this to be unfair, although this view was partly based on a misconception that all elected representatives of that party received PRA rather than a limited list of individuals. However, others felt that not receiving PRA was disadvantageous to opposition parties in that they were less able to engage with the data and question the government on it.

It's not really fair because it doesn't allow the opposition parties to come up with their own understanding and good answers to the right… you know, doesn't let them come up with good criticisms of the data or whatever. (Public participant)

Differences in PRA across UK

Levels of awareness of the differences in PRA arrangements across the UK varied among participants. Those who worked with UK departments, particularly external statistics producers who had a remit for the whole of the UK, tended to be more aware of differences and, in some cases, either worked with the different arrangements for the different devolved governments or had taken a decision at an organisational level to implement 24 hours PRA across all their statistics. Those within the Scottish Government who had experience of working with statistics which were produced by other organisations with no PRA or 24 hours PRA described difficulties with producing Scotland specific reports to coincide with the publication of the official statistics which combine Scottish statistics with data sets from across the UK when the UK data was not available in advance.

And it is manifestly unreasonable that that kind of sledgehammer approach is applicable to Scottish Government when NRS are looking to take that ONS data and publish it in reference to Scotland, because that same sensitivity around leakage does not exist in the Scottish context. (Policy professional)

When asked their opinion on Scotland having a different PRA arrangement to other parts of the UK, most participants stated that they did not see a problem with this. For some, differences across the different parts of the UK are part of devolution and they highlighted what they felt to be much more significant differences in policies between the UK and Scottish governments. For others, the differing lengths of PRA were justified by the difference in the sizes of the government and the number of staff available to work on the statistics. However, it was noted that the differences between the parts of the UK had the potential to become a political issue. Several participants felt that the differences between the PRA arrangements reflected a culture of leaking in the UK government which was not present in the Scottish Government.

But I fundamentally don't understand what problem is posed by other bits of the UK having different rules. I mean, devolution implies difference. So I sometimes think people think equity means the same everywhere. It doesn't. (Special adviser)

The capacity, there is less capacity, I would say. So I think that kind of reflects the size of the different governments. (Special adviser)

A small number of participants felt that different PRA arrangements across the UK was problematic and supported the principle of having consistent arrangements across the UK. Another view was expressed that, in principle, differences across the UK can be acceptable if these are well evidenced and articulated, but it was felt that this was not currently the case and that the differences are a legacy of different structures and decision-making processes rather than reflective of any well-evidenced reasoning.

There was a perception among stakeholders that the public is generally not aware of the differences between different parts of the UK and that differences would be acceptable to the public as long as the reasons behind this were clearly explained. When asked about their views on the differences in PRA in the different parts of the UK, participants in the public group were broadly of the opinion that PRA should be consistent across the UK.

I feel like if the if England and Northern Ireland didn't find 24 hours good enough, I think they would have made the moves to change it. So I think I agree with that should be enough. (Public participant)