To ask Her Majesty's Government further to the Written Answer by Baroness Morgan of Drefelin on 11 December (WA 167), whether they will provide an example to justify the statement that the release of the data requested might lead to individual children being identified. [HL812]

To ask Her Majesty's Government further to the Written Answer by Baroness Morgan of Drefelin on 11 December (WA 168–9), whether they will provide an example to justify the statement that the release of the data requested might lead to individual children being identified. [HL813]

Some years ago a journalist located two children convicted of murdering another child using statistics provided through the Department of Heath secure units bulletin. Since then the Department for Children, Schools and Families has adopted protocols that prevent individual children from being identified through published statistical information.

The department has closely examined and reviewed its statistical disclosure controls when considering the release of data that could lead to individuals being self-identified or identified by others. Where there is a risk that individual children could be identified, the data are released by aggregation with other data, or released in a form where small numbers are rounded or suppressed.

We have applied this policy to the information collected recently on home education. We released much of the raw data in the form of graphs, which allowed users to see the distribution of data but which did not identify individual local authorities. This is because where numbers are low it would be possible for journalists or other researchers to combine press reports or other information collected formally or informally with published statistical information to identify individual children subject to child protection plans, or young people not in education, employment and training, for example.

This could lead to home-educating families being pursued for further information about their children—for example, the sort of information set out in the Answers given on 11 December (Official Report cols. WA 167 and WA 169) and 15 December (Official Report cols. WA 206 and WA 207-WA 209).

We are considering how we can release more of the statistical data collected in a format that protects individuals from being identified but which is more informative than the data already released.