How many children are in gangs? The data’s not good enough to know
28th Jun 2018
More than 30,000 children aged between 10 and 15 now say that they are in gangs.
We don’t know the exact number, and it’s difficult to get an accurate measure of this. In 2013/14, a survey found 0.9% of 10-15 year olds in England and Wales said they were in a street gang, which implies around 30,000 children in England, but this should not be taken as exact.
“More than 30,000 children aged between 10 and 15 now say that they are in gangs, according to research that will fuel concerns about the country’s violent crime epidemic.”
The Times, 25 June 2018
The 30,000 estimate has significant limitations, and cannot be fairly used to draw a link to a supposed rising crime epidemic.
The figure of 30,000 children comes from the Office of the Children’s Commissioner. It produced this estimate in a 2017 study, using data from 2013/14 from the Office for National Statistics (ONS), which reported that 0.9% of 10-15 year olds it surveyed in England and Wales described themselves as being in a street gang.
But this figure is subject to significant uncertainty, as The Office of the Children’s Commissioner noted at the time. The survey didn’t include children who are detained, “missing”, or living in health and care residential establishments, and the definition of a street gang is quite broad.
The figure of 30,000 could vary a lot if there’s just a small amount of error in the survey’s findings. The 0.9% figure equates to about 26 young people (in a survey of 3,000) reporting themselves as being in a gang. In a repeat survey for 2016/17, it found 0.7% did so, but that’s the equivalent of about five fewer young people.
Ultimately, the sample size, infrequency and wider data recording issues mean we can’t confidently say exactly how many children are in gangs, and whether this number has gone up or down in recent years. While this is the best available data we have on gangs, it cannot be taken as evidence that young people are fuelling a “violent crime epidemic”.
This article does not assess whether or not there is such an “epidemic”.
The 30,000 estimate comes from 2014
0.9% of 10-15 year olds in England in Wales reported that they were a member of a street gang, according to ONS data for 2013/14, sourced to the Crime Survey for England and Wales. This is the data which the Times report says the office of the Children’s Commissioner used.
In 2017, the Office of the Children’s Commissioner used the 0.9% figure to estimate that around 30,000 10-15 year olds were in street gangs in England, based on population figures for England (mid-2015). It also assumes that gang-membership levels are the same in both England and Wales.
The ONS published more recent data in May this year (which wouldn’t have been available to the Children’s Commissioner when their report was published). It finds that 0.7% of 10-15 year olds in England and Wales reported being in a gang in 2016/17. It also reports that 0.2% of 16 to 24 year olds are a member of a street gang, compared to 0.7% in the 2013/14 data.
The data can’t give us a very precise number
The sample sizes of the research mean it’s hard to precisely estimate the number of children in gangs, or change over time. Applying the reported percentage of 10-15 year olds in gangs to population statistics suggests that the number in England has fallen by 7,000 from 2013/14 to 2016/17. Yet this is the equivalent to about five fewer people (in a survey of around 3,000) reporting that they are in a gang in the 2016/17 survey. Given what we also know about young children responding inaccurately, or giving “funny” answers, in surveys, this data isn’t good enough to give a very precise sense of change over time.
Street gangs are also not necessarily committing criminal activities, according to the ONS definition. It told us its survey defined a street gang as:
“groups of young people who hang around together and
- Have a specific area or territory
- Have a name, a colour or something else to identify the group;
- possibly have rules or a leader; or
- who may commit crimes together.”
There are more reasons these numbers should be treated with caution
The 30,000 figure formed part of a report on vulnerable children, in which the Children’s Commissioner recognised that the data available to it on street gangs was limited and “should be treated with caution”. It added: “The information for people involved in gangs or have been victims of gangs is very limited and our research revealed an important gap in the data.”
It also highlighted that the ONS data doesn’t cover “people living in health and care residential establishments as well as people detained are not included in these estimates. Children included in the missing population may face an increased risk to be involved in gangs and thus the estimates may underreport the real numbers.”
The Office of the Children’s Commissioner told us that: “We don’t collect data but collate it from many sources and sometimes the starting point for an estimate may be, in our own view, statistically dated, or weak but nonetheless, within certain standards, the best available at the time. Where that is the case we do what we think we can, transparently, to model towards a reasonable estimate of what the number is at the time of publishing.”
This data cannot be treated as evidence for a supposed violent crime epidemic
The lack of consistently published data and limitations in recording practices mean it’s hard to get a perfect picture around the number of young people in gangs. It certainly seems a stretch to use the existing data as evidence of a phenomenon which is fuelling a supposed “violent crime epidemic”, given the limitations we highlight above.
The Office of the Children’s Commissioner told us that: “Ever since the Office of the Children’s Commissioner undertook to assess the levels of vulnerable children in England in 2016, with a view to publishing findings in 2017, there has been a dual purpose.
“Firstly to fill an existing statistical gap and find the best possible and reliable estimates we could for numbers that previously have not been calculated, and to beg the question “why hadn’t they been?” …
“Secondly to highlight how hard it can be to arrive at those numbers when data collection can be patchy, sometimes non-existent, subject to local variation, variations of interpretation of official terms (not least the word ‘vulnerable’) and in what form the data came.”
They also said that: “In looking to next year part of our call on Government and local authorities and other services will be to provide our calculations on Vulnerable Children with data that is fresher, more robust and comprehensive on the area of gang membership.”