These small area model-based estimates of subjective loneliness have been produced by Age UK, for people aged 65 and over at Borough level and middle and lower super output areas (MSOA/LSOA) in London.
They are calculated using a model based method using a combination of data from the English Longitudinal Survey of Ageing (ELSA) from 2010-12 and small area Census data from 2011. Further details of the methodology can be found at Age UK.
The different measures included in the data are
1. The ‘log odds’ of the risk of loneliness among older people
2. The estimated prevalence of loneliness among older people
3. The area ranking for London of the relative risk of loneliness among older people
4. The area ranking for England of the relative risk of loneliness among older people
1. Prevalence estimates: these estimates of prevalence are based on characteristics significantly associated with loneliness which could be estimated at small area level from 2011 Census data. Not all characteristics could be measured from the Census and so these estimates underestimate the true level of loneliness based on national survey and regular polling data. They are designed to be used to examine variation between areas rather than as a true estimate of the prevalence of loneliness in a given area
2. LSOA: Care needs to be taken when using estimates based on LSOA estimates due to other potential confounding factors such as communal establishment within the area such as care homes, hospitals, hospices etc. The risk factors were based on a survey of domestic rather than communal establishments only
3. The main aim of the modelling was to produce small area estimates which are not available for other sources. Borough level estimates are included for context, but alternative estimates at this level may give a more accurate assessment of prevalence
A map of the MSOA data can be viewed on the Datastore.
Source: Office for National Statistics