Abigail Linyard-Tough, Research &
Evaluation Officer
Hannah Pattinson, Strategic Lead –
Resident Insight
Rich Stockley, Head of Research
Key
points raised during the discussion:
The Strategic Lead
introduced the slideshow presentation. It was agreed that this
slideshow would be emailed to Members after the
meeting.
Will
Forster Warner arrived at 12:14pm.
The Head of Research
explained that the residents’ survey was answered by a
demographically representative sample of residents, which gave more
accurate results than self-selective residents’ surveys, as
were used by some other councils. It was conducted by an external
company, Swift Research.
The Research and Evaluation
Officerinformed Members that the
residents’ survey was conducted through telephone interviews,
which was deemed cost-effective and relatively anonymous, and in
partnership with Surrey Police.
The residents’
survey had been operating for 12 years in Surrey, meaning there was
a large well of long-term data to draw on.
Each month, the
survey was conducted with 550 residents. This data was then
compiled quarterly.
Residents’ surveys were conducted in varying
ways by Local Authorities (LAs) across the country as well as by
the Local Government Association (LGA). The LGA questions and
guidance had been used in designing the Surrey County Council
survey, and national data gathered by the LGA could be used as a
comparator.
The Task Group felt it would be useful to see a
breakdown between different districts and boroughs perhaps not
frequently but on the basis of sharing an overview. In the same
vein, data on comparison with other LAs would also be useful, as
would a country-wide perspective based on the LGA survey for the
country as a whole.
On discussion relating to the veracity of the survey
and the reliability of the data it generated, the Task Group was
interested in understanding how bias was taken into account in the
survey. The Head of Research said that there was always bias, and
the survey was designed taking this into account. Given that it was
not possible to eliminate bias completely, the Head of Research
said that in interpreting the results, a standard deviation method
(95% accuracy with 2% either way) could be applied to show how
probable it was that the survey was accurate.
A Member noted that
some people would not want to answer a call from a number they did
not recognise, which led to some inevitable self-selection
depending on which residents would choose to answer the call. The
Head of Research acknowledged that this could cause bias, and said
that he would send Members more information on this.
A Member said that it
would be useful to see the results of the residents’ survey
broken down by Districts and Boroughs. The Head of Research
informed Members that it would broadly be possible to obtain
figures on a District and Borough level and even on a division
level, perhaps not on a monthly basis but on an annual basis
instead. However, another Member was not convinced that divisional
research would be of value.