Consumer Data Research

198 Pages·2018·41.183 MB·English

by  James Cheshire, Paul Longley, Alex Singleton

Big Data collected by customer-facing organisations “An insightful, state-of-the-art guide into the social
– such as smartphone logs, store loyalty card and commercial value of applying geographical
transactions, smart travel tickets, social media posts, thinking to the study of consumer data.”
or smart energy meter readings – account for most of Professor Richard Harris, University of Bristol
the data collected about citizens today. As a result,
they are transforming the practice of social science. “An excellent guide to leveraging the value of
Consumer
Consumer Big Data are distinct from conventional academic research on valid data. Partnerships
social science data not only in their volume, variety based around consumer data should be encouraged
and velocity, but also in terms of their provenance and supported by all and their outputs used to
and fitness for ever more research purposes. The better the way we manage the world we live in.”
contributors to this book, all from the Consumer Bill Grimsey, retailer and author of
Data Research Centre, provide a first consolidated The Vanishing Highstreet
statement of the enormous potential of consumer data
Data
research in the academic, commercial and government “The use of data from everyday consumer
sectors – and a timely appraisal of the ways in which transactions is a potential game-changer for
consumer data challenge scientific orthodoxies. understanding economic and social patterns and
trends. This is an excellent overview of the field.”
Dr.Tom Smith, Managing Director, Office for
National Statistics Data Science Campus
Research
Paul Longley, James Cheshire
and Alex Singleton
UCL_DATA_BOOK_cover_AW.indd 1 27/03/2018 17:21

Consumer
Data
Research
Paul Longley, James Cheshire
and Alex Singleton

Acknowledgements
The editors are grateful to the Economic
and Social Research Council for funding
and supporting the work of the Consumer
Data Research Centre (CDRC), an ESRC Data
Investment, grant ES/L011840/1 and all the
research featured in this book.
Sarah Sheppard (CDRC Project Manager)
has been particularly instrumental in the
success of CDRC and, by extension, this
book. Her efforts to coordinate researchers
as well as maintain close working
relationships with data providers are
greatly appreciated! Thanks also to Patrick
Morrissey (Unlimited) for his excellent
work designing and typesetting the book.
The authors and the CDRC would also like
to thank our Data Partners for making the
data available for the research featured and
for their continued support.
Consumer
Data An ESRC Data
Research Investment
Centre
Contents
8 INTRODUCTION
Consumer Data Research – An Overview
Paul Longley, James Cheshire and
Alex Singleton
—
PART ONE
PROVENANCE AND CONSUMER
DATA INFRASTRUCTURE
15 1. Consumer Registers as Spatial Data
Infrastructure and their Use in Migration
and Residential Mobility Research
Guy Lansley and Wen Li
29 2. The Provenance of Customer Loyalty
Card Data
Alyson Lloyd, James Cheshire and
Martin Squires
41 3. Retail Areas and their Catchments
Michalis Pavlis and Alex Singleton
53 4. Given and Family Names as
Global Spatial Data Infrastructure
Oliver O’Brien and Paul Longley
PART TWO PART THREE
DYNAMICS AND CONSUMER NEW APPLICATIONS
DATA INFRASTRUCTURES AND DATA LINKAGE
71 5. Ethnicity and Residential Segregation 141 10. Geovisualisation of Consumer Data
Tian Lan, Jens Kandt and Paul Longley Oliver O’Brien and James Cheshire
85 6. Movements in Cities: Footfall and its 153 11. Geotemporal Twitter Demographics
Spatio-Temporal Distribution Alistair Leak and Guy Lansley
Roberto Murcio, Balamurugan Soundararaj
and Karlo Lugomer 167 12. Developing Indicators for
Measuring Health-Related
97 7. The Geography of Online Retail Features of Neighbourhoods
Behaviour Konstantinos Daras, Alec Davies,
Alexandros Alexiou, Dean Riddlesden Mark A Green and Alex Singleton
and Alex Singleton
179 13. Consumers in their Built Environment
111 8. Smart Card Data and Human Mobility Context
Nilufer Sari Aslam and Tao Cheng Alexandros Alexiou and Alex Singleton
121 9. Interpreting Smart Meter Data of UK —
Domestic Energy Consumers
Anastasia Ushakova and Roberto Murcio 190 EPILOGUE
Researching Consumer Data
Paul Longley, James Cheshire and Alex
Singleton
8 CONSUMER DATA RESEARCH
Consumer Data Research –
An Overview
Paul Longley, James Cheshire and Alex Singleton
It has become a cliché to observe that new of the analyst. Second, different individuals
sources of Big Data are becoming available have different wants, needs and spending
in ever greater variety, in unprecedented power, and so some individuals in the
volumes and with ever more frequent population at large will be represented
temporal updating (velocity). This book more prominently than others – and at the
is about ‘consumer data’ that arise out other extreme, those that consume nothing
of every-day transactions for goods and from a particular retailer / service provider
services, carried out between individuals will not be represented at all. A related
and organisations. Such data account point is that few consumer organisations
for an increasing real share of all of the have a monopoly of their markets, and
characteristics and activities of active many focus upon particular market niches.
citizens today, and offer the prospect Taken together, this means that there
of better understanding the nature and is bias in the content and coverage of
functioning of society. consumer data sources, and that the source
and operation of bias cannot be ascertained
Consumer data are not created for the without reference to external sources.
edification of researchers and analysts. In many ways these issues are akin to
Instead, they are a by-product of the those that characterise volunteered or
myriad consumer transactions that created crowd sourced data – in that individuals
them. This has important implications for need to feel motivated in order to
the data’s content and coverage when they contribute data, and the distinctive
are reused for research purposes. First, the characteristics of those that feel motivated
traces of (some kinds of) transactions or may affect the content and coverage of the
those people conducting them may be more resulting dataset (Haklay, 2010).
evident or detailed than others, and this
outcome is usually well beyond the control
Introduction 9
This situation contrasts sharply with the The research reported in this book has
design of conventional social surveys, developed using the Consumer Data
where the principles of scientific sampling Research Centre’s (CDRC) ‘ladder of
are used to ensure complete coverage of the engagement’, whereby initial collaborations
relevant population of interest at the design with consumer organisations are focused
stage. Nevertheless the quality of social upon specific small MSc projects. A number
surveys is diminished where acceptable of these have developed into co-sponsored
response rates are not achieved, or there is PhD projects, or shared projects staffed by
bias in the relevant characteristics of those CDRC Data Scientists. Some data providers
that respond to the surveys and those that then progress to providing data for wider
do not. In this context, it is important to use by the academic community, under
recognise that recent years have seen agreed terms set out in data licensing
cumulative declines in response rates agreements. Finally, it is also possible to
throughout the developed world (e.g. Sax engage data providers in the co-production
et al 2003) and that in important respects of data with the CDRC itself. Good examples
social surveys are no longer a panacea for are provided by our engagement with
social science research. More generally, players in the domestic energy provision
there is also no guarantee that we will be and retail sector who have participated
able to rely on the long-term availability of in the Master’s Research Dissertation
those traditional sources of data such as a Programme before going on to co-sponsor
Census of the Population, as within many PhD research. This latter development
countries these expensive and time- in turn led to providing CDRC with a
consuming surveys have come under nationwide dataset; which is available
increasing threat in line with fiscal to access by other researchers through
constraint (Singleton et al, 2017). the CDRC service. The collaboration with
the Local Data Company (LDC) reported in
Many of the chapters in this book arise this book represents the highest rung of
out of shared challenges that are faced this ‘ladder of engagement’ and follows
by academics and the organisations that, successful collaboration on MSc and PhD
to differing degrees, create consumer data. projects as well as the co-production of
There are, of course, differences too: the nationwide data with CDRC for further
timescales that characterise academic research and development.
research offer horizon scanning that
business organisations are less likely to Many consumer-facing organisations are
have resource to facilitate; usually focused highly sensitised to the risks of disclosure,
upon more operational matters, such as although these risks are absolutely
optimising the next set of sales figures. minimal where data are anonymized
There may be tensions too, in that prior to transfer, and appropriate resources
consumer data providers may safeguard to access them are put in place. To this
their competitive position, while end, CDRC uses a number of secure data
contributing to research that ultimately facilities (one of which is accredited by
increases the competitiveness of their the London Metropolitan Police), and
industrial sector as a whole. There are CDRC researchers are familiar with using
also differences of emphasis in method, novel data access technologies such as
technique and application that have secure links to sensitive data-sets held
evolved in different ways between the by different organisations.
academic and business sectors. But it is
also possible that there is shared interest The approaches to consumer data research
in better understanding the form and that are reported in this book come at an
functioning of social systems. interesting time in the evolution of data
landscapes in advanced economies. There
10 CONSUMER DATA RESEARCH
is emerging consensus that data are ‘passporting’ of data originally acquired
the world’s most valuable resource for government statistical purposes to
(The Economist, 2017). To the behemoths researchers. Such arrangements would
of the Internet age – Alphabet, Amazon, also have favourable implications for the
Apple, Facebook, Microsoft – data are a preservation and curation of many sources
strategic resource, largely to be acquired of consumer data under the provisions for
and siloed within corporate organisations. research exemptions of the General Data
From the broader public good perspective, Protection Regulation (GDPR).
data provide infrastructure for individual
and societal decision-making. For example, This vision begs a number of important
there is abundant evidence that Open strategic questions concerning the form
Data platforms and open Application and detail of the emerging data landscape:
Programming Interfaces (APIs) lead to
wide economic and social benefits, with 1) Are Big Data to be thought of as a rival
the data feeds from Transport for London or non-rival resource? The siloed
(TfL) providing one of the most well- approach of large corporations
known exemplars. Such initiatives can lead suggests that data are a valuable
to the creation and successive updating of commodity and strategic resource,
new data infrastructures, although in many the potency of which is diluted if data
cases this process is impeded by difficulties are shared with competitor ‘rivals’.
in apportioning the cost of infrastructure Seen from this perspective, they are
creation and maintenance. Whilst there not to be traded or otherwise shared.
has been significant progress, the freer Yet data sharing has been shown to
movement of data within and between leverage wide benefits, particularly
jurisdictions and industrial sectors if data platforms can be made open
still presents daunting challenges for to the widest constituency of users.
government, not least because there 2) Does GDPR present a threat to the
exists no open market for many sources creation and maintenance of datasets
and forms of data. for research purposes, or an
opportunity for researchers to create,
Without a strong precedent, the work of maintain and preserve data-rich
CDRC relies heavily upon the attitudes to representations of social systems?
data licencing of a wide range of industrial 3) How can the Big Data ‘exhaust’ of
partners with their own policies and consumer transactions and
procedures (over 20 data licensing interactions be reused in
agreements have been signed to date). representations of social systems
These partners provide their data for the that are genuinely inclusive? How can
public good and pursue research questions scientific methods be repurposed to
that contribute to a more competitive analyse data that are created and
economy and fairer society. Some of these possibly assembled without any
shared objectives were integral to the 2017 scientific research design?
Digital Economies Act, which includes 4) How can public trust and
provisions to require business to assist understanding of science be developed
in the compilation of national statistics. and maintained in support of research
The spirit of the approach underpinning that realises more of the potential of
the chapters of this book is to go beyond consumer data?
narrow official requirements and engage
in truly collaborative inter-sector research CDRC’s mission includes the creation and
of common concern. It is our hope that maintenance of new measures of the ways
these arrangements might flourish further in which ‘smart’ urban systems function,
in the future, for example through the for example with respect to pedestrian