The biggest issue was the lack of a proper CSV formatted file. ONS were able to commission a programmer to output their data in a sensible format. Unfortunately GROS were using a different version of Scrol and so were unable to make use of this. The Scottish CSV file was provided with all the geographies mixed together and was a nightmare to sort out.

Next time it would be sensible for all three census agencies to fully test the outputs with their users to ensure that this does not happen again. This issue became apparent at a very late stage which mean that remedial action became rushed or in GROS case impossible.

This will not be the case in Northern Ireland as they will be having a totally new geography.

Extra information may be also required if a postcode \ residence has moved grid reference over time due to improved mapping rather than any physical change on the ground.

Was this summary information adequate?

Is there any additional information to aid comparison between the 2001 Census and the 2011 Census that you would like to see provided for use with the next Census?

Our aim is to provide metadata that is useful and appropriate to our users. To guide us it is helpful to know what changes/new information or products would be desirable to improve on the metadata provided for the previous Census. Your opinions matter.

As always with topics on this blog, please provide any examples of websites that provide information related to the topic of discussion that you found particularly useful.

Please let us know your thoughts about training needs and census user support in general.

Asking all and sundry to comment on this subject will lead to many varied opinions.
Personally I like to read articles that have been released to the newspapers. I realise all these will be edited to show facts that the Chief editor requires.

The Census bureau should always be releasing Media faqs everyday. Rather than only when asked by the media when preparing a current news item.

The varied forms of presentations can be a simple Pie chart, a pyramid, or even a clickable Flash page (swf).

The vast majority of people would prefer an Internet based database, which might be offered in a variety of formats for the same request.

Goodluck hope you surf the web and view all the other census sites, as I did, maybe your site will be easy to navigate.

ONS has already made some progress in re-thinking how statistics can be presented for different audiences. For example, the interactive, animated population pyramid has proven popular with a wide-ranging audience. A tactile, playful approach to data presentation certainly seems to have its place. However, we are also aware that - in addition to downloading data, people might also to access data directly using Web 2.0 technologies. Such a mechanism might be useful for integrating data with data from other locations.

We are interested to hear about people's expectations of how they will access and interpet Census data. Do users want tables of data and nothing else? Or do graphs, maps and other visual mechanisms have a role to play? Do people need guidance in interpreting Census data? What about places to share stories about census data? Do you already have plans for how you would like to engage with Census data? If so, please let us know.

New regulations will come into effect for the next Census that will require the National Statistics Institutes of member states to make their data available in data cubes as specified by Eurostat. Member states have contributed to the development of these specifications. Eurostat plans to build a web portal that can access data held by those member states who have web systems that are able to exchange data automatically with other websites, and are carrying out some pilot studies.

The UK Census Offices are at an early stage in planning outputs, and the systems that will deliver them, and it is too soon for them to be involved with the pilot work or commit to developing such a system. However, we are interested in hearing about similar projects, especially those involving web access to data cubes, and learning about your experiences using web portals that access multiple web sites.

(As a reminder the survey focussed on high level issues and asked firstly about previous experience and use of census data and secondly about high level requirements for 2011 census data including 2001/2011 comparisons, access and dissemination, metadata, SDC and output priorities).

The idea of access via a 'Data Lab' is a non-starter - inconvenient and slow, and completely absurd in the 21st Century!

If some are prepared for the blurring of data because of SDC at the lowest levels in order to get ease of access fine, whereas others may want the greater accuracy and are prepared to use restricted access, then also fine. I believe that some users will only be using the OA to aggregate to their own higher geographies and so are happy with the first option, whilst others will want to look at the detail and so the latter will satisfy them.

I don’t see that this would undermine SDC and would give users the choice to meet their needs.

If we are not careful there is a possibility with all of these consultations that we will be fixing the problems of the last census. The census agencies hear from those who were dissatisfied, but those who were happy will not be commenting. I would not be at all surprised if this was one of these situations.

We are aware that users were very unhappy about the effects of the late addition of Small Cell Adjustment in the 2001 Census, particularly to tables with low counts such as the OA level Origin/destination tables which resulted in users reporting that such outputs were “unfit for purpose”. Whichever SDC method is applied the data quality of sparse tables are going to be disproportionately affected. For these types of tables we are considering the option of applying lower levels of SDC and making the output available to users only under special licence or accessed in data labs with outputs checked prior to removal. For example, some outputs from the Origin/Destination tables could be made available publicly at Super Output Area (SOA)/Data Zone level, made available under licence at Output Area (OA) level, and allow users interested in specific journeys access to detailed micro-data in a safe setting.

What would you feel about access arrangements such as this?

This output consultation website will remain open. We are keen to engage as widely as possible with users on output issues so please continue to visit the site, see what other users are saying and give us your views.

The potential benefits to be gained from this area of continuing develop are obviously an important consideration for the presentation and web delivery of data from the 2011 Census. Decisions in this area should be strongly influenced by the depth and range of functionality that are either required, or that users feel are at least desirable.

What experience of using data presented in cubes have users had? What are some of the best examples, and some of the worst? What specific features or functions are considered to be crucial, and what would be a nice option?

Some things to consider are:

• being able to interrogate multiple cubes at the same time - is this essential, or should more consideration be given to providing single cube presentation, but with a better provision of a data source and API, web service delivery to data 'mashup' services provided by 3rd parties such as Swivel or Yahoo Pipes
• the range of online interrogation and manipulation tools provided - are advanced manipulation and visualization tools essential, or again are such needs better fulfilled by elementary functionality supplemented by enhanced service provision to 3rd party web 2.0 tools
• range us data formats that have been or will be used and appreciated - beyond simple csv and other elementary formats, what experience is their of using XML, SDMX, JSON, or KML
• metadata - depth and tightness of integration within and across cubes

We have had a good response from across the user communities, in particular from Local Authorities. The Census Output Consultation website will not be closing so you please continue to make your views known on output issues. Results from the survey will be reported on the blog as they become available

We produced vast amounts of metadata to support the 2001 Census outputs. We have it on the ONS Website/Nomis/NESS on the CDs and in the printed volumes. The most detailed metadata is available on the ONS website - but would you know where to find it?

I'm talking about footnotes, definitions, classifications, methodology, disclosure, geography and more but would be very interested to know:
• What 2001 Census Outputs metadata resource/s you used?
• Which was most useful?
• What metadata resources you would like retained for 2011 Census Outputs?
• What improvements you would like to see either to content or access?
• New ideas for metadata resources or methods of access?
We want your views both good and bad.

Have you used any websites that provided fabulous metadata for large datasets provided with easy accessibility, if yes - what were they and why were they good (please provide link if possible)?

Those of you that have completed our questionnaire have provided us with an insight to some areas of metadata that would be useful for the 2011 Census but now is your chance to elaborate on your comments.

I will be posting some metadata topics for discussion, some may be of interest to you and some not, but if you have any comments on any of the following please let us know:
• Metadata to aid comparison of 2001 and 2011 data
• Definitions/Glossary - is it essential to have a printed version as well as an electronic version on the website?
• Quality indicators (What type of indicators and at what level of detail)
• Information about Census methodology
• A user community developed Census folksonomy

Please let us know of any other metadata related topics you would like to introduce for discussion.

There are a lot of very good analysts working in the public sector, most of whom are not members of the GSS. It is not likely that their changing and increasingly voluminous needs will be able to be met by GSS members, Regional Statisticians or others with access to disclosive data. Meeting the Government requirements shown above will require them to have access to data in a flexible way. The suggestions around licencing data sound really good - perhaps ONS can consult on some of the detail behind those suggestions? NB disclosive data laboratories, often promoted by ONS are not the answer - the needs are greater than that allows. We need desktop access to data, to be able to import it into packages such as excel, access or GIS. That implies that such analysts have to conform to certain qualifications and standards, and inparticular that there is very clear guidance on the use of data, and disclosure control.

That kind of approach clearly presents challenges, but there is time for ONS to consult and come to conclusions well before census outputs need to be produced.

NB putting analytical and/or presentational products onto the ONS site might be a solution : NOMIS functionality is very highly valued by analysts, and is not adequately substituted by products such as Neighbourhood Statistics.

By the way - well done on this blogging approach. It enables comments to be made easily and quickly!

In reality I don't believe this will be the case and so they should concentrate on getting the data out. This will include allowing data to be aggregated across the UK, by using common broad bands. It is also important that bases are consistent as well as the counts. Last time this did not happen and caused many problems. This will obviously need to be considered in the question selection and routing. In 2001 for example there was a difference in occupation whereby ONS didn't ask for older people to answer, but GROS did.

The Registrar Generals’ Statement of Agreement on the Conduct of the 2011 includes an aim for consistency in UK output. The degree to which this is achievable is however largely dependent on the degree of consistency in questionnaire content of the UK Censuses, which is outside the scope of outputs and this consultation. However, within that constraint, there is still a decision to be made about the extent to which variables common to the UK Censuses are “joined-up”. How important is it to ensure that these variables have a common definition and either common categories or categories that can be combined to produce a (less detailed) common category at UK level? Would this satisfy user requirements or is it important for the UK Census Offices to actually produce UK level outputs? What are the arguments for “going this extra mile”?