Introduction

National Records of Scotland (NRS) held four stakeholder events in June 2019, discussing non-standard outputs from Scotland’s Census 2011, geographies for census outputs and plans for the 2021 Scotland’s Census website.

NRS invited users who had previously expressed an interest in outputs or geography. The event was also promoted through the Scotland’s Census newsletter, Twitter, Facebook, and on the Scotland’s Census website.

The event was split into a morning and afternoon session. The morning session covered: microdata, origin-destination data and census geographies. There was a short presentation on each section, followed by a discussion with stakeholders led by NRS statisticians. For each of these topics, written feedback was also gathered from stakeholders via the handouts included in the appendix to this document.

The afternoon session involved an interactive website workshop around the current Scotland’s Census website and the requirements stakeholders had for the 2021 census website. The afternoon session also covered some of the other non-standard outputs produced in 2011 and a discussion on UK data access.

In total 47 stakeholders attended over four sessions, representing a range of different areas including: local authorities, academics and members of the public. This document summarises the feedback received from stakeholders on each of the topics discussed during these events.

Microdata

Census microdata are the record level data of households and individuals. Three microdata products have been made available from Scotland’s Census 2011 which provide samples of detailed census data for whole households and individuals. However the microdata products do not contain personally identifiable information such as name, address and date of birth.

These three products are the microdata teaching file, the safeguarded microdata files and the secure microdata files. The microdata teaching file contains records on a limited set of variables for a random sample of 1% of people in Scotland. This file is available publicly and is intended to encourage wider use of census data, particularly in an educational context.

The safeguarded files are two random samples of 5% of people in Scotland and contain a higher level of detail than the teaching file. The secure files consist of two non-overlapping random samples of 10% of people and households in Scotland. The secure files contain the highest level of detail of all the microdata products.

Given the increased level of detail in the safeguarded and secure files, these are only accessible under certain restrictions and after approval by NRS staff. In particular, the secure files can only be accessed in an approved location under NRS supervision. More information on each of these products and how to access them is available on the microdata page on the Scotland’s Census website.

At this stage NRS are not planning major changes for the microdata products to be produced in 2021 but are exploring the feasibility of an additional safeguarded household microdata file if there is interest from Scottish data users. NRS will also be investigating whether access to the safeguarded and secure products can be streamlined by working with our colleagues at the other UK Census offices, the Office for National Statistics (ONS) and the Northern Ireland Statistics and Research Agency (NISRA).

During these events, stakeholders were asked whether they had made use of the 2011 microdata products and, if they had, whether or not their requirements had been met.

Only three stakeholders reported that they had used the 2011 microdata products. Feedback from these microdata users indicated that the microdata teaching file had been used in academic settings and some had made use of the secure and safeguarded files for their work. Microdata users expressed broad satisfaction with how the 2011 products were created although they agreed that streamlined access to the safeguarded and secure files would be a positive development.

During discussions with those stakeholders who had not used the 2011 microdata products, several key issues emerged. In particular, the relatively small sample sizes of the microdata products, even in the safeguarded and secure files, limits their usefulness for in-depth analysis. Stakeholders also highlighted that the lack of geographical information contained in the microdata files limits their usefulness, particularly for users performing region specific analysis.

A number of stakeholders commented that the severe restrictions on the secure files may prohibit users from accessing them, particularly given that users must travel to approved secure locations. It was pointed out that the restrictions are likely to be another contributing factor in why the microdata files have not been more widely used. Hence, stakeholders were happy with NRS plans to make access to safeguarded and secure files more straightforward and agreed that they would be more inclined to use microdata products in the future if this could be achieved.

It was also clear from stakeholder discussions that a general lack of awareness around these products is likely to have contributed to their limited usage. A large number of stakeholders commented that they did not know that the 2011 Microdata products were available.

During discussions on how awareness and usage of the microdata products could be increased for Scotland’s Census 2021, a number of stakeholders suggested that example case studies, summarising the projects and analyses that have been undertaken using the microdata, would be a useful addition to the microdata page of the Scotland’s Census website. These case studies would be helpful for users to decide whether or not the microdata products could have useful applications for their work.

Origin-destination

Origin-destination statistics provide detail on the movements (flows) of people. Two broad types of origin-destination statistics have been produced from 2011 census data. These are the migration statistics, detailing the movement of people from their address one year prior to census day, and travel to work statistics which measure the flow of individuals from their current address to their regular place of work/study.

Similar to the 2011 microdata products, these tables are available either publicly or under safeguarded or secure conditions where access is subject to the approval of NRS staff. Public tables are mostly available at the Council Area level whilst secure and safeguarded tables drilled down to lower geographies. More information on the origin-destination statistics produced from 2011 census data is available in the presentation slides for these events and on the Origin-Destination statistics page on the Scotland’s Census website.

Looking ahead to Scotland’s Census 2021, NRS are currently investigating the feasibility of making many more of the origin-destination tables available publicly rather than under safeguarded/secure conditions. This will be possible thanks to a new process for protecting the confidentiality of data subjects by making very small changes to tables to introduce a level of statistical uncertainty.

The 2011 origin-destination statistics had been used widely amongst stakeholders attending these events and for a broad range of purposes. Stakeholders from a number of Local Authorities and in other local government roles had used the travel to work tables to measure commuting flows, useful for informing transport and infrastructure planning. Several stakeholders had also made use of the migration tables and provided feedback that these are useful for household development and education planning.                                                                                                                                 

For the most part, stakeholders had made use of the publicly available tables. A number pointed out that they had not made use of the safeguarded and secure tables due to the access requirements and others were not aware that the secure/safeguarded tables are available.

Users of the 2011 origin-destination statistics were broadly content with the tables that are available. However a number of stakeholders highlighted that delays in the publication of 2011 tables meant that origin-destination data may not have been used to its full potential. Feedback also indicated that having data available at a lower geography would have increased the usefulness of the data. It was also commented that while the ability to visualise origin-destination data via the Datashine tool was helpful, it would have been more beneficial if there was a way to easily download the data from the Datashine tool.

Stakeholders were happy with our proposal to make as many as possible of the 2021 origin-destination tables available publicly. In particular users were keen to see publicly available tables at geographies below Council Area. Stakeholders also provided details for a number of variables that would be useful to prioritise for cross-tabulations in publicly available tables. These included demographic variables like age and ethnicity, labour market variables such as economic activity, transport variables such as mode of travel to work/study and car or van availability and household variables such as household tenure.

The consensus amongst stakeholders was that the small changes made to tables as a means of protecting the confidentiality of data subjects was acceptable given that this would allow for many more of the origin-destination tables to be made available publicly. Most agreed that analysis would not be negatively affected given that changes to tables will be small enough to leave large counts/flows unaffected.

Geography

Standard output tables from Scotland’s Census 2011 have been produced for a number of geographies. More information on census geographies is available in the presentation slides for these events and on the census geographies page on the Scotland’s Census website. For 2021, NRS plans to make outputs available via a flexible table builder tool which will allow users to create their own tables with only a small number of standard tables being created by NRS. It is therefore important that the most widely used or important geographies are included in the flexible table builder to ensure that the needs of users are met.

Standard Geographies

The following geographies will be available in the flexible table builder as they have a wide usage or particular importance identified from historical usage. As such, these geographies were not covered during discussions for these events:

• Scotland
• Local Authority/Council Area
• Electoral Ward
• Data Zones
• Output Area
• Civil Parish
• Scottish Parliamentary Constituencies
• Health Board Areas

Stakeholders were therefore asked to provide feedback on their use of a range of other 2011 census geographies for which standard outputs are available. These geographies either didn’t have a wide usage or had known quality issues in 2011. Stakeholders were also asked to give feedback on whether they thought each geography should be included in the flexible table builder, or whether separate standard tables or a separate report would meet user need.

LC and DC Postcode Sector

Usage of the Local Characteristic (LC) or Detailed Characteristic (DC) postcode sector geographies was limited. During discussions, it emerged that there is some confusion about what LC and DC postcode sectors actually represent and what the difference between them is. Several stakeholders noted that they had a preference for using other geographies such as Data Zone rather than these postcode sector geographies.

Given the limited usage and possible confusion around these geographies, NRS is proposing to drop these geographies for standard outputs and not include them in the 2021 flexible table builder.

Settlement and Localities

The use of settlements and localities varied amongst stakeholders depending on their location of work. A number indicated that their preference is to create their own settlements and locality areas for outputs because the boundaries of NRS settlements and localities do not always represent the area that they are actually interested in. However other stakeholders reported that they made regular use of the NRS settlements and localities, in particular a number of stakeholders working in Local Authorities with large rural populations.

Ahead of the 2021 census, NRS plans to engage with Local Authorities to gather information on the boundaries of settlement and locality geographies that they use. This feedback will be used to inform the development of future NRS boundaries for these geographies. Nevertheless, given that there are a number of stakeholders that regularly use settlements and/or localities, NRS is currently proposing to include them in the flexible table builder.

 United Kingdom Parliamentary Constituencies

The majority of stakeholders had not made use of UK Parliamentary Constituencies, however those that had made clear that this geography is very important for their work. Outputs using this geography have been used by stakeholders for answering requests from ministers and MPs. Furthermore, stakeholders pointed out that no suitable alternative geography that meets their requirements is currently available.  

Given the importance of this geography to those that make use of it, NRS plans to include it in the flexible table builder.

Inhabited Islands

The majority of stakeholders had not used the inhabited islands geography. Of those that had nearly all described this as being of medium to low importance. Given this feedback, NRS will consider whether to include this geography in the flexible table builder, or if a separate report or set of tables would meet the user need.

Scottish Parliamentary Regions

Feedback on this geography was similar to that provided for UK Parliamentary Constituencies. Again, although the majority of stakeholders hadn’t used this geography, those that did marked it as important as there is not an alternative geography that can be used. As with the Constituencies, made use of this geography for answering requests from ministers and MPs.

Based on this feedback, NRS plans to include Parliamentary Regions in the flexible table builder.

Community Health Partnerships

The majority of stakeholders had not used the Community Health Partnerships geography. However, a small number made regular use of this geography and highlighted that it was of medium or high importance to their work. For some stakeholders, this geography is occasionally useful for health planning and ministerial queries on the topic of health. NRS therefore plans to include this geography in the flexible table builder.

National Parks

Only a very small number of stakeholders had made use of the National Parks geography. However, those who had indicated that it was of critical importance to their work. In particular this geography is essential for analysis informing local development plans, community action plans, analysis on the demographics of populations in Scottish National Parks and more.

Given the importance of this geography to relevant stakeholders, NRS plans to include this geography in the flexible table builder.

Non-Standard Geographies

The next discussion covered geographies that were not included in standard outputs for 2011. Stakeholders were therefore asked whether they had an interest in these geographies and whether any should be considered for inclusion in the flexible table builder for outputs in 2021.

Intermediate Zones

Intermediate zones were not included in standard outputs in 2011, but some commissioned tables have been produced at this geography. A large number of stakeholders expressed that they had used this geography, particularly those working in Local Authorities. Tables at intermediate zones have been used by stakeholders for community planning, area profiling, demographic analysis and more.

A number of data users also pointed out that intermediate zones are particularly useful as this is a widely used geography for a range of other official statistics outside of census. Based on their extensive usage and their usefulness for comparisons between census data and other official statistics, NRS plans to include this geography in the flexible table builder.

Travel to Work Areas

No 2011 census outputs were produced for the Travel to Work Areas so stakeholders were asked whether they thought this would be a useful geography to have for 2021 census outputs.

Only a minority of stakeholders expressed a distinct interest in this geography, however a wide range of potential uses were identified, in particular analysis on labour market, transportation and housing.

On the basis of this feedback, NRS do not plan to include travel to work areas in the flexible table builder at this time. However, if sufficient demand exists it could be added to the flexible table builder in the future.

Workplace Zones

Only 8 tables have been produced for workplace zones in Scotland from 2011 census data. These tables were also released late on in the census release timeline, only being made available in 2018.

Although most stakeholders had not made use of the workplace zones, a small number made regular use and made clear that this geography was highly important to their work. This geography has been used by stakeholders for a range of analysis on topics like population and housing statistics, monitoring regional economic development and transportation planning.

A number of stakeholders pointed out that they were either not aware that the workplace zone tables existed or that the tables had been released so late in the census release timeline that they were no longer relevant for their work. Amongst these several commented that they would be interested in using this geography if it was made available more quickly after census day.

Given the feedback, NRS will consider how to produce the workplace zone data more quickly.

Grid Squares

Usage of the 2011 grid squares was limited. Several stakeholders expressed a lack of awareness of this geography, however there are also known quality issues with the 2011 grid squares. Both of these factors are likely to have contributed to the lack of grid square usage.

Although use of the 2011 grid squares was limited, several stakeholders commented that they would be interested in using grid squares for 2021 outputs if these could be developed to meet the requirements of data users. Specific suggestions for improving the grid squares included the inclusion of more variables for use with grid squares, the ability to create visualisations in apps such as R Shiny and the functionality to integrate with Geographic Information System (GIS) software.

Given this feedback, whether or not the grid squares are included in the flexible table builder is likely to be dependent on the functional requirements of grid square data users. Feedback from several stakeholders suggested that having the grid squares in a separate platform outside of the flexible table builder may be more suitable for their needs particularly if this allows for easier integration with other software used for analysis of grid square data.

 NUTS 2 and NUTS 3

A small number of stakeholders had made use of the NUTS2 and NUTS3 European statistical geographies. These geographies were described as being useful for making comparisons between regions in Scotland and other areas of Europe for the purposes of demographic and labour market analysis.

By and large however, most stakeholders reported that this geography was of limited use for their work and it was noted by many that Council Area is usually a suitable alternative. On the whole, stakeholders agreed that that NUTS2 and NUTS3 were not a high priority for the flexible table builder or for standard output tables. Based on this feedback, NRS is proposing to leave these geographies out of the flexible table builder.

UK Harmonised Output Areas

For the final discussion on census geographies, stakeholders were asked to provide feedback on whether or not they would make use of UK harmonised output areas.

Census output areas in Scotland are smaller than those throughout the rest of the UK. This could potentially cause issues for making UK comparisons when output area is the geography of interest. As such, stakeholders were asked if it would be useful for their work to have a second group of harmonised output areas. It is important to note that NRS is not proposing to replace Scottish output areas, which will still be produced and retain their smaller size. Rather these harmonised output areas would act as a geography that is the same size as output areas throughout the rest of the UK.

Most stakeholders had limited interest in UK harmonised output areas as many expressed a preference for the smaller, more detailed output areas available for Scottish data. Given that the creation of these harmonised output areas is likely to be time and resource intensive, NRS does not plan to make this geography available for standard outputs or for inclusion in the flexible table builder at this stage.

Website Requirements

The next part of the stakeholder event was an interactive website workshop discussion on the Scotland’s Census 2011 website and plans for the 2021 census website. Users were asked to provide feedback on how they used the 2011 website, whether or not their requirements were fully met and what functionality they required from the Scotland’s Census website in 2021.

On the whole stakeholders were happy with the content and functionality of the 2011 Scotland’s Census website. Nevertheless, stakeholders made clear that there are aspects that could be improved to better meet user need. In particular, a large number of stakeholders commented that the inability to create and upload their own bespoke geographies for census outputs was a significant drawback.

Website navigation also emerged from discussions as a key issue as several stakeholders reported that content was occasionally difficult to find. Stakeholders commented that navigation was particularly problematic when a user did not know the specific terminology for the content they were searching for.

Similarly some stakeholders felt that metadata was difficult to find. It was also noted that in some cases metadata was not comprehensive enough potentially leading to misinterpretation of data/variables etc.

Based on these discussions, a large number of stakeholders commented that functionality to upload bespoke geographies for census outputs would be a beneficial addition to the 2021 website.

Stakeholders suggested that an improved search function with more flexibility to search within specific areas of the website as well as cleaner pages and better signposting would remove some of the issues around website navigation. Furthermore expanded and clearly signposted metadata with links connecting relevant areas of the website would ensure that users could more easily find what they were looking for. What’s more, improved metadata would remove the issues around users not being certain of what a particular variable/output/etc. represented.

The website workshop discussion also provided stakeholders the opportunity to provide suggestions for the general layout of the 2021 website and how different sections should be organised and signposted. All of this feedback will be used to inform the ongoing development of the Scotland’s Census 2021 website.

Other Outputs

A range of outputs in addition to standard and commissioned tables were made available from Scotland’s Census 2011 data. Details on these other census outputs are available in the presentation slides and in the links contained in each section below. Stakeholders were asked to provide feedback via a handout which asked questions on which of these other outputs they were aware of and whether or not they had used them.

The feedback gathered during these discussions will be used by NRS to inform decisions on which, if any, of the other outputs should be prioritised based on how well used they are. Suggestions from stakeholders for improving these outputs will also help NRS to ensure that the needs of census data users are met in 2021.

Results At a Glance pages

Of the 23 respondents who responded to the question on usage of the results at a glance pages, 10 stakeholders had made use however only 3 of these reported that they used these pages often. During discussions, stakeholders expressed that these are useful for a quick overview of census results on a particular topic) but are, understandably, of less use for more in depth analysis. 5 stakeholders said that they may have made use of these pages but had not been aware that they were available whilst 10 said that they had no use for the results at a glance pages. A suggestion from stakeholders to make these outputs more helpful was to include relevant charts for each topic page.

Area Profiles

Similar to the results at a glance pages, of the 22 stakeholders who responded to the question on area profiles, only 4 used them regularly although 9 used them on a less regular basis. Users of the area profiles reported that they are a helpful tool for obtaining quick statistics at different geographies of interest and are also useful for making quick comparisons across geographies such as Council Area. 5 stakeholders indicated that they were not aware that the area profiles are available and may have made use if they had known however 4 had no use for the area profiles.

During discussions, several stakeholders suggested that the ability to view the actual numbers behind percentages in area profiles would be a helpful addition.

Maps and Charts

Of the 22 stakeholders who provided feedback on their use of the maps and charts that were pre-built by NRS, only 3 stakeholders were regular users. A further 8 were infrequent users. Users of the maps and charts were mostly happy with their format and content, however during discussions it was noted that would be useful to have these available for geographies lower than Council Area. A further 6 said that they would have used the maps and charts but did not know they were available while 5 individuals had no use for these outputs.

Analytical Reports

22 stakeholders responded to the question on usage of analytical reports, only 7 indicated that they made infrequent use of these outputs with none reporting regular usage. Nevertheless 7 provided feedback that they would have made use of these outputs but were not aware of their availability. 8 of the 22 respondents had no use for the reports.

Comparator Tool

20 stakeholders responded to the question on comparator tool usage. 11 made infrequent use but none indicated regular use. 7 respondents would have made use if they had known of its availability whilst a further 2 had no use for the comparator tool. During discussions stakeholders had several suggestions for improving the comparator tool. Notably, users were keen to have the ability to make comparisons between 3 geographies at a time, rather than just 2 and to have quick access to the raw data behind the comparator tool so that this could be easily downloaded.

Parliamentary Profiles

Usage of the Parliamentary profiles was very low amongst stakeholders. 22 responded to the question on usage of these outputs and 19 indicated that they had no use. Only 1 individual reported that they had made infrequent use. The remaining 2 stakeholders were not aware that this tool existed but may have made use of it had they been aware.

UK Outputs

During the final discussion of each session, stakeholders were asked about their usage of UK level census data via a handout. They were also asked to provide details on their requirements for UK data in 2021. Of the 23 stakeholders who responded to the question on usage of UK level data, 9 had made use.

UK data has been used by stakeholders to make a range of comparisons between Scotland and the rest of the UK, in particular comparisons between major cities like Glasgow and Edinburgh and a wide range of comparative analysis focused on the labour market in Scottish regions compared to the rest of the UK. UK level origin-destination data had also been used by some stakeholders for analysis on UK level commuting and migration flows.

In order to access 2011 census data at the UK level, users must visit the website of each of the three UK census offices separately. Links to census data for England and Wales (ONS) and Northern Ireland (NISRA) are provided on the NRS census website.

Stakeholders were therefore asked whether or not the current arrangement for accessing UK level data met their requirements or whether they would prefer to have access to UK data alongside Scottish data on the NRS website. During discussions, stakeholders were overwhelmingly happy with the current proposal of accessing UK data via links on the NRS site and were mindful of the technological issues that may arise from having UK level data alongside Scottish data on the same site.

Next Steps

NRS will use the feedback from these stakeholder engagement events to further develop plans for outputs from Scotland’s Census 2021. In particular, NRS plan to make the following changes on the basis of the feedback from these events:

1. Add case studies to the microdata pages to illustrate how the microdata files can be used
2. Continue to pursue a way to improve access to safe-guarded and secure micro data and origin destination data files in Scotland.
3. Include Intermediate zones in the geographies used for standard outputs and in the flexible table builder
4. Remove LC and DC Postcode Sector from the geographies used for standard outputs and in the flexible table builder
5. Look at ways to improve website design, navigation, search functionality and metadata information.
6. Consider how to enable the uploading of bespoke geographies for 2021 outputs.

In addition, the feedback and suggestions gathered during website workshop discussions will be used to inform the development of the Scotland’s Census 2021 website to ensure it has the content and functionality required by stakeholders.

We will continue to engage with stakeholders as our plans develop further to ensure that the needs of data users will be met in 2021. In the near future we hope to share with stakeholders illustrations of the new website design and the changes to the flexible table builder tool as a result of feedback from a wide range of users.

For more information about on plans for Scotland’s Census 2021 and for details on upcoming events, visit the Scotland’s Census website, subscribe to the Scotland’s Census newsletter and follow us on Twitter @Scotcensus2021.