General enquiries :
+44 (0)20 7602 6000

The dark side of out of date demographics

Friday 14 December 2018 Data Insight & Analytics

1

Thomas Fletcher-Wilson's picture
By Thomas Fletcher-Wilson

Using out of date data

The world of data is evolving and developing at such a rate it is often hard to keep up with the latest innovations. It is becoming far more complex to regulate this data and make sure compliancy is consistent across the board. The acronym ‘GDPR’ has become almost mandatory in any conversation concerning data and its usage. Rightly there is a major focus on oversharing, collecting and data misuse. Should there also be a stronger focus on using inaccurate and out of date data? In light of this, it’s important to turn the spotlight not on the over use and sharing of data but in fact the dangers of underutilising data. 

There has recently been a major oversight in data usage – those using up to date demographics and those that don’t.

 

Census or senseless?

A form of the census has been in use since the “Domesday Book” was compiled in 1086 by William the Conqueror. The census has evolved from use for tax collection solely, to becoming a comprehensive source of socio-economic statistics such as population, race, employment and income etc.

There is always controversy surrounding the use of the census. Accuracy has always been a key example as has reliance on the data collected. Did you know 390,000 people in England and Wales described their religion as Jedi Knight in the 2001 census and in 2011, 177,000 people declared their religion was Jedi. Was ‘Return of the Sith’ as fictional as we thought, or did a certain number of the population just grow up? Although this example is comical it is an important point. 

The census is such a large-scale operation and unique in the data that it can collect I don’t want to draw away from its importance. The value of the census comes in many forms, such as:

  • High coverage due to the legal obligation to respond
  • The level of detail across wide geographical coverage
  • The advantage of comparisons over time
  • Being a source of data that is independent from official information gathered by local governments for their own purposes

 

Time's up on out of date data

As highlighted by the ‘Science and Technology Committee’ in the Third Report on ‘The Census and social science’, the timeliness of the census is the larger issue.

The problem with the census most frequently mentioned to us was that census data may quickly go out of date due to the gaps between census-taking, exacerbated by the lengthy data processing after the data-gathering exercise. The advantages of the census— size and comprehensiveness—are diminished when the data is not utilised until almost two years later

The census can start to lose its relevancy in the ten-year data blackout that occurs in-between the dates the census is conducted. This problem is even more evident as development, population growth and migration increase year on year as expressed by Professor Leslie Mayhew of City University, "This can be a significant issue in areas experiencing rapid population change, or when the importance of a particular socio-demographic topic suddenly changes in response to new or emerging Government policies and priorities."

East London has developed significantly since 2011. We have seen the success of the Olympic games and the re-generation that came after as a result of large developments on undeveloped and vacant sites, the impact of Westfield shopping centre, etc. These have all resulted in a population increase in the area. Below is a map that shows the Population Change by Postcode Sector – this shows the raw change in numbers of population since the 2011 Census (key in bottom right corner).

 

Cowboy consulting

Consultants and organisations around the UK are supplying reports and insight using data from 2011. You would not expect to gain accurate insight from a company's records from ten years prior, so why do we so openly accept doing the same with demographic data? 

Below I have included a map showing the percentage of unemployment in 2011. Dark blue being the higher levels of unemployment and lighter being lower (key included in bottom left for clarity).

 

Using our up to date demographics we can clearly see the percentage of unemployment levels in 2018.

 

Put your selves in the shoes of a local authority/developer/charity or any organisation who has a vested interest in data such as unemployment. The difference is alarming yet there are still organisations openly supplying results and insight off the back of the first map in 2018. 

There are several organisations that aren’t reliant on the 2011 Census any longer that can be found on the approved suppliers' scheme to ensure a quality control, CACI being one of them. CACI was the first commercial organisation to process and supply detailed census demographics. For four decades CACI have worked on the UK Census supplying information to both the public and private sectors.

 

Assumption vs Insight

Population projections are only the tip of the iceberg. Making assumptions about an area using out of date data can cause serious repercussions. For example, social renting.

 

Social renting can provide indicators of many factors; affluence, affordability in an area, reduction of homeownership and even a low supply of affordable housing. The map above shows social renting from both Housing Associations and Local Authorities as a percentage. Renting of social housing has increased overall. The natural assumption would be that there has been an increase in expected less affluent parts of London and the postcode sectors that have a large percentage of social renters already. If we compare 2011 to 2018 below some of the results may surprise you.

 

Although the changes are not as significant as population change. Largely due to developments taking several years to be built and onboard new tenants. There are changes in areas such as Shepherds Bush, West Brompton and West Kensington. A reduction in social housing can even be seen in Stratford and parts of East London, an area of lower affluence. 

On its own the maps will not paint a full picture but when cross referenced with other sources, the out of date data could make significant miscalculations.

 

Conclusion

The census and other historical data sources can provide great levels of insight in certain situations. This would mainly be in making comparisons over several years or where other large data sources are unavailable/don’t exist.

There needs to be a shift in due diligence when using reports completed either internally or by external consultants. Although the examples used above are specific to certain sectors it really does apply across the board Care, Retail, Housing to name but a few. 

As expressed above data innovations have surpassed using out of date demographics and in my personal opinion, we should really be thinking forward instead of using techniques and organisations who have not kept up with the times, it’s VHS versus Netflix. My hope is that this article can shed some light on sources that are available and to provoke organisations to challenge out of date data and report providers.

If you would like to know more about CACI's up-to-date data please get in touch.

 

Sources

publications.parliament.uk/pa/cm201213/cmselect/cmsctech/322/32206.htm 

ftalphaville.ft.com/2018/04/11/1523422800000/Will-Big-Data-kill-the-Census-/

www.telegraph.co.uk/news/politics/7882774/National-census-to-be-axed-after-200-years.html

www.ons.gov.uk/census/2011census 

www.caci.co.uk/location-planning/demographic-data 

www.acorn.caci.co.uk

There has recently been a major oversight in data usage – those using up to date demographics and those that don’t. Tom Fletcher-Wilson questions the reliance on Census data.

The dark side of out of date demographics