Analysing attendance data for Durham Area Youth

 
Durham Area Youth is a Charitable Incorporated Organisation (CIO) that provides youth provision in several villages on the outskirts of Durham. They run drop-in and outreach facilities for young people aged between 8 -18 years old.

The request

This was to examine data concerning the attendance of Young Persons (YPs) at the various Youth Centres (YCs) located within the locality of East Durham. This is a quote from the original specification document:

“Through analysing the data of attendees, we will be able to see how many young people we are reaching in areas (including deprived areas and families and the hard-to-reach families). This data will also identify areas where we can expand.  It will also identify trends and enable us to forecast for the future”

The approach

The supplied data consisted of initials of individual YPs (thus preserving anonymity – an important consideration in work of this kind), their home postcodes (and an associated level of deprivation for that postcode), their age, gender and whether or not they had siblings, and then a record of attendance for each calendar month at YCs.

The statistician working on the project was Nick Wray, and Durham Youth Services’ representative was Sarah Robinson. Because of Covid all contact was via Zoom, mobile phone and email. Data was sent via Dropbox and as email attachments.

The deliverables asked for by the client were
  1. graphical displays of attendance patterns
  2. forecasts of future attendance
  3. identifying of the YPs in the most deprived areas to enable better targeting of services.
The first of these was straightforward. The second was not really doable, because of the limited timewise distribution of the data (11 consecutive calendar months).

The third involved some regression techniques which required a certain degree of explanation to the client. We were relating deprivation indices to attendance levels and wanted to see whether there might be a relationship between these two variables. I did not want to pull results “out of the hat”, so added some notes to the report which explain in outline how least-squares regression and confidence limits work. The client found these useful.

An unforeseen benefit was that the analysis led to using hitherto unfamiliar R programming techniques in importing maps from Google Maps, and using these as a graphical display to show home locations in the Durham report. This was tricky and took some time to do, but the code for doing this, once worked out, was a useful additional to the toolkit.
 
The result

The analyses showed, amongst other things, that attendance at YCs was independent of deprivation, and also home distance from the centres, a result which suggests that Durham Area Youth is reaching all youngsters across the deprivation and geographic range.
 
Impact and benefits

Sarah Robinson, business development, HR & office manager at Durham Area Youth commented:
The information provided to us has enabled us to look at trends, length of journeys that a young person is willing to make to attend the sessions and the break-down of age bands and genders to see if one is more popular than the other.  We were able to identify that we are not specific to any areas of deprivation and that we do not have any areas to target.
This information is only the beginning of the process, as we will collate further data to help us build the youth service and expand community centres and areas that we deliver in.
 
The volunteer and Durham Area Youth have agreed to make this an ongoing project – they will supply the volunteer with more data as they become available, so they can begin to do through-time analysis in coming years.