Social contact survey - Results

Here we will show some of our intitial results. We hope to give you an idea of how the survey information is being used, and why.

Our principle aim is to measure the social contact network within the UK – how we all fit together into one giant network. Understanding what this looks like will be very helpful in working out how far and how fast close-contact infections, like influenza, can spread.

Web submissions – recently received and by location

This plot shows where we've received contact information from, from the web survey. The bigger and more yellow the dot, the more recent the submission. As we only ask you for the first part of your postcode, the location of the dots are not precisely accurate (it depends on how many people live in that postcode). Remember, we cannot identify you by the first part of your postcode: many other people and households have the same code too.

You may need to refresh your browser to see the latest image (F5 key for Internet Explorer users).

Does your home area have any dots?

If not, we really need to know about your day and the people around you. Without your help, your area will be poorly represented and inaccurate when we simulate infections spreading through-out the whole UK, using all the information gathered in the survey. Even if there is a dot near your home, the more information about different people we can get, the more accurate our predictions will be.

Web submissions – contacts reported by location

This plot shows an indication of how many contacts people from different regions are reporting, from the web survey. The bigger and more yellow the dot, the greater the number of contacts made. Do the most sociable people in the UK live in Yorkshire? Note, the number of contacts is on a log scale.

Web submissions – Number of daily contacts

Thanks to the people submitting social contact info on the web, we can show you some early results.

This plot shows the variation in the number of different people you've reported meeting in a day (Web submission only). The plot is a histogram, but with a log scale on the x-axis: this makes it easier to include both the low numbers and very high numbers reported.

The blue vertical line marks the median average (17 different people within a day), and the green vertical line is the mean average (29.2 different people a day).

So, most people seem to encounter between 8 and 33 people in a day, but some meet much more than this, several hundred in a few cases. This variation is very important to know about, as it can explain why we see some patterns of infections in certain parts of the community.

Some example personal social networks

Here are some examples of 'social contact snapshots' taken from a few of the postal surveys we've received back.

The participant is each orange circle in the middle of their network 'wheel'. The number of contacts they make in a day is the number of other cirlces, or nodes, they connect to. Colours refer to where they met:
red = home
blue = work / school
yellow = travel
green = leisure / other

The bigger a contact's circle the longer the duration of their encounter.
The length of the 'spoke' denotes how far from the participant's home the encounter took place.
Darker, thicker lines indicate a 'touch' contact.

(Click on the thumbnail above for the full size image)

With enough different samples from around the UK we'll be able to start piecing these small network fragments together, like a jigsaw, and infer the giant contact network of the UK. Once we've done this, it will be an incredibly useful tool for us and other researchers to predict the spread of infections, and to direct healthcare policy decisions on how best to control or mitigation epidemics.