Our findings - Visibility

Children and young people are visible in key new outputs relating to the pandemic

A number of the key statistical outputs released during COVID-19 have made children and young people visible. The Office for National Statistics’ (ONS) COVID-19 Infection Survey includes children as young as 2-year olds and all ages are also covered by the various sources of administrative data that exists on hospitalisations, testing and deaths.

There are also good examples of separate analysis being conducted for children and young people in specific educational settings. The ONS’s School’s Infection Survey and Student Impacts Survey both acknowledge the unique circumstances of these groups, and endeavour to provide separate analysis in light of this.

NHS Digital and Department for Health and Social Care (DHSC) also commissioned a “Mental Health of Children and Young People Survey” in 2020 to understand the mental health impacts on children during the pandemic. This study was conducted by the ONS and NatCen Social Research and provides population level estimates of the pandemic’s effects on children and young people in England.

Age breakdowns are provided, but they can sometimes have a broad range

In the ONS’s COVID Infection Survey (CIS), a degree of granularity is given in the data, with three age category breakdowns made across the 2-24 age range. However, some of these age categories can be broad. For example, one of the age categories in the data is ‘year 12 – age 24’. This could encompass young people with extremely varied circumstances; including high school pupils, university students and individuals no longer in education.

In the ONS’s Opinions and Lifestyles Survey (OPN), ‘young people’ in the headline bulletins and underlying data tables refers to those between the ages of 16 and 29. The relatively small sample size for young people in this survey inhibits the ability for a more granular breakdown.

In many cases more granular age breakdowns would enhance the value of statistics; providing a better understanding of the impact of the pandemic on children and young people. We consider that producers should be alert to the usefulness of age category breakdowns in their data and endeavour to understand whether or not they are serving the needs of their users. Where small sample sizes limit the ability of the data to be broken down further, producers should ensure that this is made clear in statistical bulletins.

Small sample sizes in surveys can inhibit the granularity of analysis

As well as limiting the capacity of data to be broken down by age, small sample sizes can also inhibit more granular breakdowns by other demographic characteristics. For example, we found that small sample sizes in a number of the official statistics we looked at limited the ability for any meaningful breakdowns by factors such as sex, ethnicity, economic disadvantage or LGBTQ status.

Small sample sizes have also impacted the availability of reliable data on the impact of the pandemic on students. The ONS’s Student Insights Survey had extremely low response rates, with the most recent iteration receiving a response rate of just 1.1% (1110 responses). In addition, earlier releases of the survey identified that a large proportion of the respondents were older than 30, which is not what most people would expect when they think of a ‘student’. Although this caveat is included towards the end of the bulletin, there is a risk that people could be misled by the data.