How the ONS have operationalised data from 56m people
Last Sunday, 21 March was Census Day. This once-every-decade survey aims to provide a snapshot of life in the UK in 2021.
When it comes to statistical exercises, this is about as big and as complex as it gets. Touching every individual and household in the land, the census has been described as one of the largest peacetime operations carried out in the country.
The question that fascinates us here at Clekt is, what happens once that data has been submitted?
The output is vitally essential to the nation’s future planning, the vital services and infrastructure we need to run society. Government departments, NGOs, community groups and businesses will all make use of the Census data to help shape strategic decisions on policy and investment.
How do you make this colossal dataset available to the agencies who need it quickly and in a meaningful and secure way?
In short, how do you operationalise census data?
In the 2011 census, lessons were learned on this score: critically, asking agencies to put in a data report request is not good enough.
“In 2011 tables were made by analysts inside the ONS in response to requests from customers. [ ] This process is called statistical disclosure control and for 2011 this was a manual task for each user request.”
This year, the ONS has come up with a better way: Self-serve.
Users will have “the ability to analyse multivariate data for themselves. They’ll be able to make complex queries and generate tables instantaneously. They’ll also download the tables into popular spreadsheet tools.”
While it may seem like an obvious innovation, the technology that has gone into making analysis possible on a data set of this vast scale is quite breath-taking. It’s a huge move forward when it comes to operationalising the insights contained in the raw information.
And it got us thinking.
The level of innovation we have seen since the last census ten years ago has been phenomenal; how will we use data to drive our nation from the 2031 census, ten years from now?
We asked Clekt Data Scientist Simon Sleight, where we thought we could be in 2031.
“How we collect the data is going to change. Expect to see a rise in voice data entry, driven by intelligent, dynamic questions. IoT data will add another dimension. However, I know for sure that data on abstract human concepts, for example, beliefs or sentiment, can never be replaced by data from things.
When it comes to decoding census data, I predict much more sophisticated and nuanced analysis techniques, already being developed and deployed in the private sector today. In summary, 10 years from now our analysis of census data will have moved on again, to the point that the data will be coming up with answers to the questions we should be asking.”
In the meantime, if you would like to know more about how your business data can answer the important questions you have today, then get in touch with the team at Clekt and let’s talk.
