Tony Dent to Ed Humpherson: ONS Covid Infection Survey and Code of Practice for Statistics

Dear Mr. Humpherson,

Thank you for your letter dated 28th April 2022.

We unreservedly welcome the opportunity to discuss the Covid Infection Survey (CIS) with you and your team, whenever it is convenient to do so. As background, to the discussion, there has been extensive correspondence between Better Statistics and the CIS team, including recent correspondence with Pete Benton, and we would be pleased to provide you with copies of that correspondence prior to the meeting, if that would be of any value to you. From my personal perspective that correspondence provides evidence of the requirement for stronger guidance on a number of aspects of the code and the need for the OSR to have some real powers to enforce changes, when necessary. I return to this point below.

Meanwhile, with regard to our wider concerns, I should explain that we started out as the Campaign for Better Business Statistics because of our concern that the ONS were failing to accurately reflect the value of smaller business and of the gig economy. An opinion that was somewhat reflected in the Bean report of 2016. However, our original ambitions were derailed by the Covid pandemic and we began to take an interest in other areas of our national statistics – please note the use of n, rather than N, so we are including opinion polls. We therefore formed the Community Interest Company early last year specifically to encourage greater public awareness of and interest in the production of accurate and relevant statistics, including those produced by Market Research Companies and organisations such as Google or Facebook. Indeed, we would like to see the code of practice more widely adopted than presently, with closer integration with the MRS code.

It is partly this latter consideration that encourages us to consider strengthening the code in a number of areas because, to be of real value, it needs to have genuine ‘teeth’. It is my opinion that the present references to Value for Money in the code are inadequate in a number of ways but, for now, I will suggest only that the code should enforce the principle of data access and that every project or other data activity within the GSS should have a specific plan to ensure that the resulting data is more widely accessible for further analysis, to ensure that maximum potential value is attained. This is certainly one area where the CIS has failed, although there is nothing specific in the code to pinpoint that fact.

Finally, may I ask if you are content that we should publish this correspondence on our website? With all good wishes,

Tony Dent
Director, Better Statistics

Tony Dent to Ed Humpherson: ONS Covid Infection Survey and Code of Practice for Statistics

Related links