FOI reference: FOI-2024-2422

You asked

Could I please ask for the following information:

  • The count of unique ONS employees who have swiped into ONS sites (any) 40% or higher of the total working days in a month for each month from October 2023 to November 2024

  • Attending any ONS site would contribute to the 40% or greater calculation e.g. Newport on Monday, Titchfield on Tuesday, London on Wednesday would be counted as three days in the period

The data would look something like this:

  • October 2023 - 300 unique ONS employees attended ONS sites 40% (or higher) of the total working days in the month

  • November 2024 - 132 unique ONS employees attended ONS sites 40% (or higher) of the total working days in the month

We said

Thank you for your request.

We anticipate that the cost limit threshold would be met in order to fully answer your request at present. This is due to the following manual processing that would be required.

To answer this, due to the size of the dataset, multiple files need to be exported and merged externally before being able to aggregate the required data, which must be filtered by person, date, and site, to then allow for the removal of duplicate instances so that the final result includes occurrences at each site for each person.

Following this, the data then needs to be disaggregated into percentages created to show the requested 40% office attendance. Then a calculation has to be made to work out the total amount of people attending the NPT and TCH at 40% or over. Due to the volume of data that we'd need to work with spanning 13 months, the system being used also has reliability issues which could delay the process even further.

We have estimated this to take at least 27 hours to complete. The cost limit for replying to a Freedom of Information request is £600 or 24 hours work, and in our view we would need to exceed this to fully action your request. Therefore, Section 12 of the Freedom of Information Act 2000 is engaged, whereby a public authority is not obliged to answer a request if the cost of complying would exceed the appropriate limit.

We would recommend limiting the scope of your request so that we may answer your request within the cost limit. Please see the following guidance regarding this.

There are many caveats that come with the data that could be supplied and it is important to note that this will not provide a reliable measure of employees who have complied with the policy requesting colleagues to spend 40% of their working time in the 4-week monitoring period or higher.

Firstly, we only hold representative data from Newport and Titchfield colleagues. The data held for other sites is all depersonalised and the individually identifiable information is managed by a separate controller. As an side, any swipe data from London in particular could not be reliably regarded as complete, as a number of colleagues attending the London site on a regular basis do not have swipe card access and are admitted by security following a manual pass-check. This is because there are a certain number of colleagues who regularly work from the London site but are contractually based at another site.

Secondly, the policy allows for instances or hours to be used to calculate whether 40% attendance in a 4-week monitoring period has been achieved. Therefore, whether or not someone has fulfilled the requirement to spend 40% of their working time in the office measured in hours, as opposed to instances, would not be accurately reflected in the data.

Thirdly, it is also important to note that swipe data will not be representative of part time workers, those who have taken holiday or are on sick leave, and those who have reduced targets due to reasonable adjustments.

Therefore, we would advise resubmitting your request with a limited scope, to include only Newport and Titchfield ONS sites and limit the time frame to 4-6 months-worth of data. With these variables we would be able to provide the number of instances staff have attended the office 40% of that time period, subject to the above caveats.