Introduction to Profiling Attendance
Sex, age, socio‐economic group, disability levels, ethnicity, educational attainment, employment status, home ownership and car ownership are some of the more commonly asked aspects of people's personal profile. The possible list of questions is almost exhaustive but needs to be balanced against considerations such as:
- Why do you want to know?
- How will the data be used?
- What impacts will collecting the data have on the time and cost of data collection?
Profile questions should be asked in a comparable way to national surveys so that valid questions are asked and where appropriate comparisons can be made with wider populations. If event organisers adopt practices such as those recommended, then it also becomes possible to make comparisons between events.
In an online survey of 2,250 London Freewheel participants, the following profile information was identified. For an example of how profiling questions might translate into a surveying tool, please see the IRB questionnaire in the resources section below.
Sex: 1,400 Male (62%); 853 Female (38%)
Age: Mean Age 42 years. Actual age was requested and the frequency distribution can be grouped according to national categories such as Census 16‐17, 18‐19, 20‐24, 25‐29, 30‐44, 45‐59, 60‐64, 65‐74, 75+ or Active People
Disability: 116 (5%) considered that they have a long standing illness or disability which limits their daily activities.
Employment: Work FT (30+ hrs/wk) 73%, Work PT (<30 hrs/wk) 9%, Self‐employed 8%, Student 3%, Retired 5%, Not in paid work 8%
Ethnicity: White British 73%, White Other 15%, Mixed Race 2%, Black/Black British 2%, Asian/Asian British 2%, Other 5%. This is a truncated version of the groupings from the Census which may be too complex for the average survey.
Socio‐economic: In terms of the socio‐economic groups to which respondents might belong, the National Statistics Socio‐ economic Classification (NS‐SEC) is now used. The questions necessary to derive NS‐SEC are numerous and difficult to code, which may be more than is required by the average event evaluation. A pragmatic approach maybe to ask for the occupation of the main earner and the annual household income. Should event organisers wish to examine NS‐SEC in more detail use the downloadable manual from ONS. Taking Part is another reference option when looking in more detail at young people.
Profiling Attendance by Residence
Profiling by residence may be required to identify other impacts (economic), or establish performance against targeted demographics (e.g. % of local residents or socio-economic categories engaged by the event).
Geographic analysis of survey results is made possible by the inclusion of a question asking for people’s postcode. Although actual postcodes do not have legally binding boundaries, their positions can be plotted using a Geographic Information System (GIS) by making reference to a directory file, such as the National Statistics Postcode Directory. The advantage of this system is that the majority of people know their postcode, and are able to supply it when responding to a survey. The level of detail on any subsequent map is determined by the amount of postcode data recorded. Thus, a full (or unit) postcode can be plotted at street level, while the first two letters of a postcode are sufficient to identify the postal area (of which there are 124 in the UK). It is therefore possible to produce maps at a range of geographic scales with a suitable map 'theme' to show the results of the survey. The examples below demonstrate the difference between mapping survey data at postcode area and postcode district level (the first half of a postcode, sufficient to identify a post town).
Unit postcodes are sufficiently accurate to be mapped to within 100 properties (with an average of 15 per postcode). Once plotted, this point data can be aggregated up to any other level of geography within the national statistics hierarchy (e.g. super output area, ward, local authority, county), enabling comparison with other geographic datasets, such as population counts or market segmentation directories. Other spatial analysis techniques can be used on postcode level data, such as hotspot mapping and travel distance modelling (see examples below). It is not possible however, to reverse-engineer the postcode from postcode area to unit postcode level. It is therefore preferable to collect full postcode data from survey respondents.
Local authorities and central government departments/agencies have access to spatial data sets under the Mapping Services Agreement (MSA) with central government. The MSA provides a wide range of data from a number of suppliers including Ordnance Survey, including road network, address and postcode data, boundary datasets and other topological data. Under the MSA, authorities are able to supply data to consultants under a ‘contractor sub-licensing agreement’, which is a standard contract template identifying the datasets to be used. The use of such data is tightly controlled by the terms of the contract, which usually requires that data is only used for the purpose for which it is supplied, and cannot be retained beyond the period of the contract. Acknowledgement of copyright is required at all times, and must refer to the relevant licence number.