Notes for dashboard users

We would like to make you aware of the following important information which you should bear in mind when interpreting and analysing the survey data. It’s easy to get caught up in the numbers and it’s only natural for the human mind to jump to conclusions and we want to try to help with your interpretation of the data by pointing out some of the pitfalls.

• Firstly, whilst we expect that there is very little bias in these results, we should be aware that the respondents who kindly took the survey were self-selected. This means that it isn’t a truly random sample of UK social housing residents and therefore may not accurately represent the views of the UK social housing population as a whole. That said, it’s a good representation and there are many meaningful conclusions to be drawn as long as these are done with care.
• The Total Responses block at the top left of this dashboard shows the total number of people who responded to the survey for the filter that you have selected. This may not always tally with the numbers in the other graphs because respondents had the option to skip questions if they did not want to answer them. However, most people answered all questions.
• The filters at the top right of the screen enable you to choose different categories of filtering. For example, you might want to look at the results for the over 55 year olds. In which case you should choose the two uppermost age categories and “Add filter”. Note that you can choose multiple categories.
• There is also a location category which you can use to view results from different regions of the UK.
• You can also filter the data by clicking on the age bar chart and/or on the map.
• IMPORTANT: When you do this filtering you inevitably reduce the number of respondents to the survey (the “sample size”) For example, if we view the results for the over 70 year olds in this survey, we only have 180 responses to the questions. We need to be very careful interpreting data when the sample size, (like the over 70s), is small. Generally speaking, the smaller the sample size the less confident you can be that the results you are viewing represent the views of all people in that subset.
• You can, however, confidently say things like “Of the 180 people who were over 70 years old who completed this survey 51% feel safe in their neighbourhood”. However, it would not be right to assume that if we asked every over 70s resident in UK social housing that we would get the same result.
• It’s better to filter the data less in order to get bigger sample sizes. This means you can be more confident of the conclusions that you draw.
• Our reports that we are preparing and will publish on this this website will put some numerical levels of confidence on the conclusions that we draw.
• If you need some assistance interpreting the graphs then hover over the title and you will see a small icon, hover over the icon below to get some help.