image area 5


The social sustainability framework that has been developed for Sutton council uses three different types of data, the scoring process combines these disperate sources. Wherever possible, data is scored by the small local neighbourhoods identified through the natural neighbourhoods mapping.

The three sources of data used are:

  • hard data: census data or data from other robust sources
  • predictive social sustainability indicators: data from national surveys benchmarked to small local areas, which can be compared to a residents’ survey
  • soft data: data that drawn from one to one interviews or subjective professional assessments.

Each data source is scored as robustly as possible, given the constraints of the data type and the need for proportionality. The intention behind scoring all the different types and sources of data is to identify where:

  • the data indicates stronger social sustainability within this measure
  • the data indicates expected social sustainability within this measure
  • the data indicates weaker social sustainability within this measure

Go to the Useful Information page to see the list of questions from national surveys used in benchmarking, and to see the benchmarking and other scoring templates.

4.1 Hard data

The framework uses data from the 2011 census, including numbers of residents falling into different categories, and combined indicators of deprivation, and a variety of other data sources.

Census data

a. Numbers of people experiencing different social needs:

  • The number of people in each category is expressed as a proportion of the total population of the area (using population figures for output areas)
  • This is then assessed against the overall borough or ward figure, and scored.

For Beddington, all census data fell in the middle range of overall Sutton scores, being around the national average. Scores for neighbourhoods were given for areas that were significantly higher than the ward average (red) or lower (green).

b. Deprivation indices:

  • Output areas are ranked within Sutton
  • Scoring over 90 – green
  • Scoring under 30 – red

Other hard data

Where a pattern can be found that identified an area had a stronger or weaker score than the average, these are scored.

The Hard data scoring template sets out how data can be scored, using the Beddington pilot data as an example.

4.2 Predictive, benchmarked data

Residents survey data that can be benchmarked against questions from national surveys has been analysed by OAC or IMD group (depending on the question).

  • The % score for the local area is compared with the national average
  • The percentage difference is then tested for statistical significance
  • This generates a green or red score

The test of statistical significance used is known as “the z test for column proportions”. It is used to test the significance of a difference between two proportions (or percentages) taking into account relative sample size. This takes account of the fact that a percentage of a smaller group is less reliable than a percentage of a larger sample size, and that the significance of a result varies the further away the figure is from the mid point of possible answers.

The z test for column proportions results in figure that indicates the confidence level (in the “z” column in the benchmarking template). For a 95% confidence level the figure will be is +/-1.96.

The Benchmark template shows how residents survey data is benchmarked against comparable areas, using the Beddington pilot data as an example.

Other residents survey questions

Questions which had fewer than 10 respondents were excluded.

These are scored on the basis of:

  • very satisfied = 1.5
  • satisfied = 1
  • neither satisfied or dissatisfied = 0
  • dissatisfied= -1
  • very dissatisfied= -1.5

The scores added up for each question and then divided by the number of possible answers.

  • If the result was 0.8 or more, this was scored green
  • If the result was 0.6 or more, this was scored red

The questions where answers could be “yes” or “no” were scored differently depending on the topic.

  • “does your child have somewhere safe to play outdoors” was scored red where more than 10% said no, green where over 90% said yes. This is a high threshold because of the topic, it should be expected that the great majority of children have access to outside play space.
  • the two questions relating to adaptability and resilience were scored red where there were more no answers than yes’, and green where the indicator score was over 0.5.

Where there is no residents survey, predications can be made about key social sustainability issues from Social Life’s analysis (see under “predictive data” in section 2), and corroborated against responses to qualitative indicators.

The Scoring non benchmarked questions template shows how to score these questions, using the Beddington pilot data as an example.

4.3 Soft data

This draws on the survey of the built environment and the audit of community amenities and facilities. Scoring is made after assessment of the available evidence.


4.4 Combining different data sources

  • To combine disparate data sources, each “red” is scored -1, and each “green” +1.
  • The average from each indicator is found by adding the reds and greens, and dividing by the number of questions.
  • The average of each dimension is found in the same way, by adding adding the reds and greens, and dividing by the number of questions.

The table below shows how scores for each indicator is combined into scores for the four dimensions, using the Beddington pilot data as an example.

scores graph for stage 4

Assessing the final score for each indicator involves making a judgement about how much deviation above or below zero is needed to score either stronger or weaker social sustainability.

For Beddington, the indicator scores were assessed using three scoring rationales. This exposes where the bigger – and therefore more significant – social sustainability strengths and weaknesses lie. This is shown in the table below.

The Overall scoring sheet template shows how different forms of data are combined, using the Beddington pilot data as an example.

4.2 Presenting the data

The overall data is presented in circular diagrams to try and make it easily understandable. The aim is to be accessible and to present the overall picture, rather than concentrating too much on individual indicators. There are many options for presenting the underlying data, depending on what data has been used, the level of detail required, and the audience.

wheel explained


This template shows how a circular scoring diagram can be constructed:  Circular bar chart template.

This report – the Beddington social sustainability report - shows how the full report of the Bedddington social sustainability assessment carried out in early 2014 was presented.

Stage 5 looks at how to use the toolkit in your local area.