In search of the data revolution: Has the official statistics paradigm shifted?

Abstract

What distinguishes revolution from evolution? What events or movements in the world of official statistics have been sufficiently disruptive or transformational to deserve being called revolutionary? Using the definitions of data revolution put forward by the Independent Expert Advisory Group on a Data Revolution for Sustainable Development in their report A World that Counts to identify potential data revolutions and then deriving a framework to evaluate those definitions from Thomas Kuhn's work The Structure of Scientific Revolutions, this paper investigates, through the lens of official statistics, whether there has been a data revolution or not.

Full text

Statistical Journal of the IAOS 36 (2020) 1075–1094 1075
DOI 10.3233/SJI-200662
IOS Press
In search of the data revolution: Has the
official statistics paradigm shifted?
Steve MacFeely
Centre for Policy Studies, University College Cork, United Nations Conference on Trade and Development
E-mail: steve.macfeely@un.org
Abstract.
What distinguishes revolution from evolution? What events or movements in the world of official statistics have been
sufficiently disruptive or transformational to deserve being called revolutionary? Using the definitions of data revolution put
forward by the Independent Expert Advisory Group on a Data Revolution for Sustainable Development in their report A World
that Counts to identify potential data revolutions and then deriving a framework to evaluate those definitions from Thomas Kuhn’s
work The Structure of Scientific Revolutions, this paper investigates, through the lens of official statistics, whether there has been a
data revolution or not.
Keywords: Open data, public good, datacratic, ethical frontier, New Global Data Deal
‘You are the sum of your data’
Don DeLillo, White Noise
1. Introduction
In 2013, the High-Level Panel of Eminent Persons on
the Post-2015 Development Agenda called for a Data
Revolution [1] to exploit the opportunities presented by
the new data landscape. No sooner had the words been
uttered but term was immediately adopted in public
discourse and diplomatic declarations. Statements, such
as: ‘The term “Data Revolution” is not an exaggeration
when one considers the drastic shifts in the digital land-
scape caused by the data avalanche’ [2, p. 18]; ‘Har-
nessing the data revolution would help the United Na-
tions system to deliver better results at a lower cost’ [3];
‘We are living in a worldwide data revolution’ [4]; ‘tak-
ing advantage of the many opportunities offered by the
data revolution’ [5]; or ‘the data revolution is having
an enormous impact’ [6]; ‘we are now in the midst of a
data revolution’ [7, p. 929]; or ‘the era of the data revo-
lution has started’ [8] have become commonplace. The
High-Level Group for Partnership, Coordination and
Capacity-Building for Statistics for the 2030 Agenda
for Sustainable Development (HLG-PCCB) have even
conducted a ‘Review of Implementation of Data Revo-
lution’ in 2020 [9].
The data revolution, if indeed there has been such a
revolution, is a curious one. Notwithstanding the chatter
about big data and data analytics, there has been no
obvious coup d’État, no shouting or marches on the
streets, no Viva la Revolución, no data riots. Although
the report A World that counts [10] could conceivably
be considered a manifesto for a data revolution. But that
aside, like many scientific revolutions before it, the data
revolution has been decidedly silent. So much so, it is
appropriate to question whether there has been one or
not. But sometimes history doesn’t need to be loud to be
decisive. Occasionally the most profound changes slip
past silent and unnoticed; sometimes the significance
of events are evident only with the passage of time.
This paper sets out to explore this question from the
perspective of official statistics; has there been a data
revolution or simply a phase of rapid evolution?
It is not immediately clear how to answer this ques-
tion. What distinguishes revolution from evolution? For
an event or movement to be revolutionary, it must pre-
sumably be in some way disruptive or transformational.
What might those disruptions or transformations be?
Is the emergence of big data sufficiently transforma-
tional to qualify as a revolution? Or what about the
1874-7655/20/$35.00 c
2020 – IOS Press and the authors. All rights reserved

1076 S. MacFeely / In search of the data revolution
Open Data movement – is that disruptive enough? Is the
data revolution about new partnerships between NSOs
and civil society or citizen science, where official statis-
tics moves from a purely production based model to
a mixed business model combining the manufacture
of official statistics with franchising under license? Or
is something more profound like a deepening of hu-
man capital and statistical literacy required to justify
the term? Deriving a framework or criterion to assess
potential data revolutions from Thomas Kuhn’s work
The Structure of Scientific Revolutions [11], this paper
investigates whether there has been a data revolution
or not, but from the perspective of official statistics.
Consequently, many dimensions of the data revolution,
in its broadest sense, are not discussed.
This paper is organised in nine sections. Section 2
outlines the etymology of the term and concepts of data
revolution. Sections 3 and 4 provide definitions and
criteria for identifying data revolutions. Sections 5–7
identify and analyse a selection of supply and demand
side data revolutions. Section 8 explores some potential
crises and provides a very preliminary analysis of the
impacts of COVID-19 for official statistics. The paper
is concluded in Section 9.
2. Etymology
Described as the ‘the idea of the moment’ [12] or
dismissed as ‘hype’ [13] – one way or another it seems
everyone is talking about the ‘Data Revolution’. Hype
it may be, but the term ‘Data Revolution’ has caught
the imagination of politicians, diplomats, statisticians
and data scientists. The idea of a data revolution seems
to have captured the zeitgeist perfectly. In retrospect, it
seems as if the term ‘data revolution’ was just waiting
for someone to say it aloud. But where did the phrase
come from and what does it mean?
As the proverb says, ‘success has many fathers’ and
so it appears when you try to trace the ancestry of
‘Data Revolution’. Although credit is often given to
the High-Level Panel of Eminent Persons on the post-
2015 Development Agenda for having coined the term
in 2013 [1], there are other contenders for the title of
parent. Andreas Weigand, former Chief Scientist at
Amazon is sometimes cited as the originator after he
published an article The Social Data Revolution in the
Harvard Business Review in 2009 [14]. In this paper
Weigand argued that two distinct data revolutions were
underway. The first revolution began with e-commerce,
as internet companies began to appreciate the value of,
and store, the massive volumes of data being gener-
ated indirectly from internet search patterns. The sec-
ond data revolution, he said, came about as people ac-
tively shared data directly on various social networks
and posted product reviews which has led to the evolu-
tion of the wiki approach (a neutral space where peo-
ple share information) and massive volumes of data. In
other words, the data revolution was the confluence of
digital and cultural revolutions. Others have credited
Nandini Ooman of the Center for Global Development
(and presumably a Beatles fan) with giving birth to the
term, in a 2010 blog entitled You say you want a data
Revolution! where she discussed the need for system-
atically collected data to improve research on health
outcomes [15].
But in fact, ‘data revolution’ has a much longer lin-
eage. A Google NGRAM search for uses of the term
‘data revolution’ reveals that the term pre-dates Ooman
and Weigand by several decades. The term data revo-
lution, fittingly, can be traced back to the 1960’s. The
earliest reference identified is Scotten in 1961 [16]. But
some other early references are noteworthy. Allardt and
Littunen [17], discussing data revolution, note the con-
nection between changing social conditions, cultures
and political processes, and the ability to assemble and
collate data for evaluation and comparison purposes.
Halina [18], discussing data revolution in 1966, is no-
table as he ponders whether there has been a data rev-
olution or just another phase in the evolution of data.
Mackenzie [19, p. 30] argues ‘there has been since
1945 a “data revolution”, in the sense that very large
quantities of data can be stored and analysed electroni-
cally’. His work also discusses the challenges of large
datasets, electronic storage and the problems of relia-
bility and completeness. Deutsch [20] identified four
separate data revolutions: the first being comprised of
disorganized data collections in the 17
th
and 18
th
cen-
turies for administrative, tax and military purposes; the
second revolution was the use of historical data by so-
cial scientists; the third was largely statistical, with the
development of new quantitative techniques for orga-
nizing and interpreting data; the fourth data revolution
arose from technological advances in computing which
have allowed greater storage, complex databases and
analytical capacity facilitating ‘all-to-all’ comparisons.
This idea of four data revolutions continues to enjoy
some favour today [21,22]. In summary, not only is the
term ‘data revolution’ not new, but the meaning of term
hasn’t evolved that much if at all. On the contrary, it is
remarkable how little has changed.
Although Data Revolution is not a new term, it is
probably fair to say that the 2013 report of the High-

S. MacFeely / In search of the data revolution 1077
Level Panel of Eminent Persons A new Global Part-
nership [1] is responsible for its current prominence.
This report outlined what international development
might look like in a post Millennium Development Goal
(MDG) world. The authors called for a ‘Data Revo-
lution’, and this call attracted a lot of attention. In a
data driven world, a world witnessing an exponential
growth in data, the call for a data revolution seemed
to make perfect sense, even if no one knew what that
meant exactly. While the panel did not offer a precise
explanation, they stated ‘A true data revolution would
draw on existing and new sources of data to fully in-
tegrate statistics into decision making, promote open
access to, and use of, data and ensure increased support
for statistical systems’ [1, p. 24]. The authors also made
clear that the call was made in the context of sustainable
development with the aim of improving the quality of
statistics and information available to citizens in order
to strengthen accountability and decision-making and
ensure government decisions are evidence based. While
they refrained from mentioning big data, they identified
information and mobile technology, improved connec-
tivity and crowdsourcing as all having the potential to
facilitate real-time monitoring. They also lamented how
these developments had remained ‘largely disconnected
from the traditional statistics community’ [1, p. 23].
Thus, the current call for a data revolution was framed
in the context of delivering improved official statistics.
3. Defining the ‘Data Revolution’
Following the publication of A New Global Partner-
ship the then Secretary-General of the United Nations,
Ban-Ki Moon, established an Independent Expert Advi-
sory Group on a Data Revolution for Sustainable Devel-
opment, asking them to translate the broad-brush con-
cept of a data revolution into something more concrete.
In their subsequent 2014 report A World that Counts the
expert group (or data revolutionaries) concluded that
a data revolution is ‘an inspiring vision of a world of
fast-flowing data deployed for the public good, and of
citizens and governments excited and empowered by
the possibilities this creates’ [10, p. 20]. They saw the
data revolution as a dazzling vision of how data could
support human development.1
1
Dazzling perhaps, but also terrifying. The idea that ‘No one should
be invisible’ is not necessarily a vision that everyone finds attractive
and arguably runs counter to the European General Data Protection
Regulation which explicitly safeguards the ‘right to be forgotten’
The expert group defined ‘data revolution’ in two
parts: the first part provided a generic, supply side def-
inition which broadly speaking focused on new and
emerging data sources. The second part was aspira-
tional, defining the data revolution from the demand
side, focusing on the uses to which data could be used
for sustainable development. The two sides of the defi-
nition are:
Part 1: A supply side definition of data revolution
‘An explosion in the volume of data, the speed with
which data are produced, the number of producers
of data, the dissemination of data, and the range
of things on which there is data, coming from new
technologies such as mobile phones and the “inter-
net of things”, and from other sources, such as qual-
itative data, citizen-generated data and perceptions
data; A growing demand for data from all parts of
society’.
Part 2: A demand side definition of data revolution
‘The integration of these new data with traditional
data to produce high-quality information that is
more detailed, timely and relevant for many pur-
poses and users, especially to foster and monitor
sustainable development; The increase in the useful-
ness of data through a much greater degree of open-
ness and transparency, avoiding invasion of privacy
and abuse of human rights from misuse of data on
individuals and groups, and minimizing inequality
in production, access to and use of data; Ultimately,
more empowered people, better policies, better de-
cisions and greater participation and accountabil-
ity, leading to better outcomes for people and the
planet’ [10, p. 6].
The first definition aligns quite well with the concept
of a revolution in ‘data’ as it focuses on the explosion of
new data and new data sources arising from the digital
(Art.17 – Right to Erasure). Clearly the intention of this report is well
intentioned and altruistic – their focus is on measurement to assist
development. The idea that ‘No one should be invisible’ pre-empted
the SDG call that ‘No one is left behind’ and which was translated for
statisticians by Mogens Lykketoft, President of the seventieth session
of the UN General Assembly, as ‘leaving no one uncounted’ [23]. But
such an inspiring vision borders on totalitarianism – imagine Stalin,
Hitler or several more contemporary leaders (or Silicon Valley CEOs
for that matter) uttering ‘No one should be invisible’ or ‘Leaving no
one uncounted’ and the aspiration takes a somewhat more sinister
meaning. “We are now reaching a point when new technologies – the
combination of artificial intelligence and biotechnology, biometric
sensors, face recognition, voice recognition – make it possible for
the first time in history for a dictatorial government to follow all the
citizens all the time,” Harari [24] has warned.

1078 S. MacFeely / In search of the data revolution
revolution. So, to a large degree this first definition is
consistent with what Weigand [14] had proposed as
it focuses on an aspect of data that arguably has been
revolutionary – the increase in volume.
The focus of the second definition is very different,
as the emphasis is on statistics and their use rather than
on the narrower concept of data. While it should be ob-
vious, it is worth noting for the sake of clarity – data and
statistics are not the same thing, although frequently,
but incorrectly, used interchangeably. Data are basic
elements or single pieces of information. Statistics are
numerical data that have been organized or summa-
rized (i.e. data in-formation) through mathematical op-
erations in line with conceptual frameworks [25]. The
second definition outlines a revolution in ‘statistics’.
In presenting two definitions together, the ‘data revo-
lutionaries’ underlined two important points. Firstly, by
presenting supply and demand side perspectives, they
illustrated that there are two perspectives, which are
different, and not necessarily in equilibrium. Secondly,
they also underscored how data and statistics are inter-
twined and inseparable, yet distinct. This latter point
is of importance to this investigation; while the focus
is on official statistics, it is not always easy to make a
clear distinction from data.
Other definitions are available. The MacMillan
crowdsourced open dictionary [26] defines a data revo-
lution as ‘a movement that focuses on producing, cap-
turing and developing data to improve the way it is used
to facilitate change’. This definition is quite similar to
the demand side definition provided by the Independent
Expert Advisory Group. Thus, the crowd’s definition of
the data revolution is actually a statistics revolution with
an emphasis on impact. PARIS21 provides a comple-
mentary definition ‘delivering the right data to the right
people in the right format at the right time’ [27], thus
emphasising the use and impact of data for outcomes.
4. Criteria for assessing a data revolution
In today’s parlance, revolution is an overused word.
We have had industrial, sexual, political, religious and
scientific revolutions. More recently we have had tech-
nological, digital and information revolutions. To this
catalog we must now add data revolution. But as Hack-
ing [28, p. xv] points out, the meaning of the word ‘rev-
olution’ has not only changed but also become some-
what meaningless, ‘Nowadays revolution is pretty much
a praise word. Every new refrigerator, every daring new
movie, is announced as revolutionary. It is hard to re-
member that the word was once used sparingly’. So,
what do we mean by the term? The Lexico Oxford on-
line dictionary
2
provides a range of definitions for the
word ‘revolution’, ranging from the movement of ce-
lestial bodies to the overthrow of governments. But of
more relevance to our study, it also defines a revolution
as ‘a dramatic and wide-reaching change in conditions,
attitudes, or operation’.
The challenge for this investigation is to decide
whether the changes occurring within the world of data
and official statistics are simply the next stage in an
evolutionary process or whether the changes are of suf-
ficient magnitude (i.e. disruptive or transformational
enough) to merit being described revolutionary. In other
words, how big a change is required to justify using the
term revolution and against what criteria?
Thomas Kuhn, in his 1962 book The Structure of
Scientific Revolutions [11] decomposed scientific revo-
lutions into their defining characteristics. Kuhn argued
that the development of science is not linear, but is
marked by shocks or revolutions, that introduce step
changes or paradigm shifts that drive the development
of science. Adapting Kuhn’s approach, a set of crite-
ria for identifying a data revolution are derived. The
criteria are:
1.
Paradigm shift – this means the rejection of one
paradigm (a time honoured theory and practice)
and the acceptance of another.
2.
Paradigm crisis – severe and prolonged anomalies
or controversies are a precondition or prerequisite
to revolution.
3.
Paradigm vocabulary – the emergence of a new
paradigm may result in the construction of new
equipment, classifications and vocabularies.
4.
Paradigm speciation – this is not a necessary con-
dition per se, but with a paradigm shift, an ex-
isting paradigm may split in two i.e. more than
one paradigm may emerge. A new paradigm may
spawn a new profession or discipline.
Thus, we see that a shift in paradigm, involves a si-
multaneous Schumpeterian type paradigm creation and
destruction, often in reaction to crisis during the pre-
revolutionary phase. Kuhn [11, p. 48] notes that the
pre-paradigm period is often ‘marked by frequent and
deep debates over legitimate methods, problems, and
standards of solution’ and these debates do not neces-
sarily vanish with the appearance of a new paradigm.
In some cases, this may lead to splits within existing,
2https://www.lexico.com/en/definition/revolution.

S. MacFeely / In search of the data revolution 1079
or the creation of new, professions or disciplines, with
the corresponding development of new terminologies
and vocabularies. He argued that ‘professionalization
leads . . . to . . . considerable resistance to paradigm
change’ [11, p. 64] and also, quoting Max Planck, noted
that ‘new scientific truth does not triumph by convinc-
ing its opponents and making them see the light, but
rather because its opponents eventually die, and a new
generation grows up that is familiar with it’ [11, p. 150].
5. Identifying data revolutions
The concept of data revolution can be assessed from
both supply and demand side perspectives. This sug-
gests we must consider multiple data revolutions, rather
than just one. Indeed, a quick search online, suggests
there are many: data privacy revolution; open data revo-
lution; big data revolution; and a social data revolution,
to name a few. As noted above, Weigand [14] identi-
fied two, while Deutsch [20] identified four. Raderma-
cher [8] identified 3 revolutions: (1) Zettabytes and yot-
tabytes; (2) Evidence and Decisions; and (3) Facts and
alternatives. The HLG-PCCB interpreted the data rev-
olution to have four components: developing a global
consensus on principles and standards; sharing tech-
nology and innovation; developing new resources for
capacity development; and providing leadership for co-
ordination and mobilization [9]. For the purposes of
this paper, four potential data revolutions are identified
and examined, two supply side and two demand side.
These ‘revolutions’ are in keeping with the spirit of
data revolution as described by the High-Level Panel of
Eminent Persons.
On the supply side, the two potential revolutions
investigated are:
–Data revolution #1 – Broadening the definition of
data
–Data revolution #2 – Use of secondary data
On the demand side, the two potential revolutions
investigated are:
–
Data revolution #3 – Recognition of data and
statistics as public goods
–
Data revolution #4 – Data and statistics to inform
decisions
Each of these ‘revolutions’ are discussed and investi-
gated using the criteria detailed in Section 4.
6. Supply side data revolutions
This section presents the two potential supply side
revolutions that have been selected for assessment.
6.1. Data revolution #1 – Broadening the definition of
data
The emblem of the data revolution must surely be the
byte
3
as it is inextricably linked with the information
and digital revolutions that preceded it. The link is the
vast amounts of digital data created by our daily activi-
ties. Digitisation has also changed the very concept of
data itself. Now data includes text, sound and images,
not just numbers. ‘The days of structured, clean, simple,
survey-based data are over. In this new age, the messy
traces we leave as we go through life are becoming the
primary source of data’ [29, 97]. Consequently, just
about everything we do, from taking pictures, book-
ing flights, searching online, watching movies, or just
carrying a phone, is creating data. Every transaction or
activity leaves a ‘digital footprint’. Today everything
we do is data or can be converted into data. If Descartes
were alive today, he might well summarize existence as
‘I am counted; therefore I am’. Perhaps the old adage
‘data doesn’t grow on trees’ is still true, but it appears
data grow on just about everything else. This torrent of
by-product data being generated by our daily lives, plus
a vast amount of data being produced by satellite obser-
vations, utility meter sensors, surveillance cameras and
other elements of the internet-of-things, all combined
with cheap storage, has resulted in a ‘data deluge’. This
deluge is also the result of an important behavioural
or cultural change, where people now record and load
content for free. Weigand [14] has described this phe-
nomenon as a social data revolution (this development
arguably qualifies as a paradigm shift in and of itself).
Definitional differences make quantifying the data
deluge difficult. Consequently, there are various es-
timates. Waterford Technologies [30] estimated that
2.7 zettabytes of digital data existed in 2017. Surpris-
ingly, no equivalent of Moore’s law
4
has yet been postu-
lated for the data world, although Goodbody [31] states
that 16 zettabytes of data are produced globally every
year and that by 2025 he predicts this will rise to 160
zettabytes annually. IBM [32] reckon we create an ad-
ditional 2.5 quintillion bytes
5
of data every day. Irre-
spective of which definition one choses, the underlying
message is clear, a massive volume of digital data now
exists.
3
A byte is a unit of measurement used to measure data. One byte
contains eight binary bits, or a series of eight zeros and ones.
4
Moore’s Law predicts that the number of transistors on a mi-
crochip will double every two years.
5A quintillion bytes is the equivalent of 1018 bytes or 1 exabyte.

1080 S. MacFeely / In search of the data revolution
Thus, we see this supply side data revolution, has
its antecedents in the digital revolution. Digitalisation
has utterly changed the concept of data, from a nar-
row numeric viewpoint to a much broader concept that
also now comprises audio, visual and text information.
This rescoping has led to an explosion of data. This
can legitimately be described as a paradigm shift. This
shift was not driven by crisis or anomaly. The change
in concept and the eruption in volume is an inadvertent
by-product. No one set out to deliberately increase the
volume of data, it just happened. Without question this
shift has led to a new vocabulary.
6
It has also created
a speciation of sorts, as now data (and statistics) are
being produced by a wide variety of actors, ranging
from traditional producers, such as, national statistical
offices, but now also by the private sector (most notably
IT platforms), citizen science, civil society, and criti-
cally by everyone who is digitally connected when they
tweet, share photographs, or rate restaurants. The emer-
gence of data science, which is establishing itself as a
new profession, distinct from statistics also qualifies as
speciation. We have also witnessed the hegemonic rise
of the word data, which has steadily usurped statistics.
Summary analysis:
1. Paradigm shift? Yes
2. Paradigm crisis? No
3. Paradigm vocabulary? Yes
4. Paradigm speciation? Yes
Data Revolution? Yes
Although no crisis precipitated this supply side revo-
lution, Data revolution #1 arguably still qualifies as a
revolution. There has been a paradigm shift, and this
shift has been accompanied by new vocabularies and
speciation.
6.2. Data revolution #2 – Use of secondary data
One of the most important changes within the world
of official statistics, both national and international, and
in industry, over the past 20 years has been the grow-
ing use of secondary data to compile statistics. Sec-
ondary data can be defined as any data holdings con-
taining information which were not primarily collected
for statistical purposes. Thus, secondary data includes
everything from national and local administrative or
6
For example: algorithms, data analytics, predictive analytics,
cloud, dark data, data lakes, data mining, data scientist, unstructured
and semi structured data, biometrics, clickbait, clickstream analytics,
zettabytes to list a few.
public-sector data (e.g. tax, social security, or education
records, public registers and smart meter utility data, to
name a few), private and commercial data (e.g. credit
rating, utilities data or store loyalty card information)
7
and big and GIS data (e.g. credit/debit card purchase
transactions, mobile phone CDR, satellite imagery, or
ASI ship identification records).8
Although, accessing and using administrative data
has probably been to date the richest and most pro-
ductive source of secondary data for compiling offi-
cial statistics, big data has sparked the most discus-
sion. Although an inherently vague concept, there have
been various attempts to define big data. But over the
past 15 or so years, the concept of big data has quickly
evolved, and so too have the definitions. Perhaps, the
most well-known definition, the 3 V’s (high-volume,
high-velocity, and/or high-variety) proposed in 2001
has today been replaced by the more general definition
where big data are described as large data sets that are
almost impossible to manage and process using tra-
ditional data management tools due to their size and
complexity [35].
Using secondary data brings both opportunities and
risks for National Statistical Offices (NSOs). Opportu-
nities on offer include, potential cost savings, improved
timeliness, burden reduction, greater granularity, the
ability to link, match or geotag microdata (provided a
functioning data infrastructure is in place), greater ac-
curacy, improved international comparability, the possi-
bility to compile greater variety of social, economic and
environmental indicators, including dynamic indicators.
Global big data, such as, satellite imagery also offer
the possibility of truly comparable international data
(and more controversially, the possibility of globally
rather than nationally produced statistics). Big data may
also offer solutions to data deficits in the developing
world where traditional approaches have so far not been
especially successful.
On the flip side, using secondary data presents some
real challenges for NSOs. There will always be uncer-
tainty regarding the quality or veracity of secondary
data as it lies outside the control of the statistical com-
piler. There may also be concerns regarding the stability
and consistency of secondary data, most particularly,
with big data. Secondary data, like any data source, will
7
See UNECE [33] Using Administrative and Secondary Sources:
A Handbook of Principles and Practices. For a fuller discussion and
list of what compromises administrative and secondary data.
8
For a fuller discussion on the definitions of and distinctions be-
tween administrative data and big data see MacFeely [34].

S. MacFeely / In search of the data revolution 1081
have their own set of biases that must be addressed.
Access to commercial data is typically still outside the
remit of statistical legislation, and thus access often
triggers data protection and confidentiality concerns.
Uncertainty over access or stability of the underlying
data, introduces a significant risk for NSOs as it means
surrendering control of the data source and in many
cases an inability to influence the shape and quality of
these data.
Data revolution #2 qualifies as a revolution. NSOs
are increasingly accessing and using a greater vari-
ety of secondary data to compile official statistics.
Sources that Stephens-Davidowitz [29] has described
as ‘paradigm destroying phenomena’. Beginning in the
1960s, Nordic countries began using administrative
data. Resisted by many for a long time, the practice
spread south, and now most NSOs understand the ex-
istential necessity of using secondary data to produce
the types of statistical products expected of a modern
statistical service. It is almost impossible to measure
trade flows, migration flows, life events (births, deaths,
marriages and divorces) or crime and recidivism with-
out recourse to secondary data [36]. Without doubt this
qualifies as a paradigm shift, albeit a slow and gradual
one. This shift was precipitated by a crisis – two in fact.
The first, budgetary constraint, and the realization that
primary data were prohibitively expensive and unsus-
tainable. The second, in the face of sustained criticism
from policy makers, was the realization that without
harnessing secondary data, official statistics could not
meet the demands being placed upon it. As seen in Data
revolution #1 secondary data brought a new vocabulary,
most particularly with regard to big data and geo-spatial
data. It has led too, to a speciation of sorts whereby
some statistical offices have been quicker to embrace
the use of administrative data than others. Today we
see a similar pattern whereby some offices are more
resistant than others in embracing the new possibilities
of big data and geospatial data.
Summary analysis:
5. Paradigm shift? Yes
6. Paradigm crisis? Yes
7. Paradigm vocabulary? Yes
8. Paradigm speciation? Yes
Data Revolution? Yes
In conclusion, on the supply side, there has been a
data revolution. Data revolution #1 meets 3 of the four
criteria and most importantly experienced a paradigm
shift. Data revolution #2 meets all four criteria.
7. Demand side data revolutions
This section presents two potential revolutions, se-
lected to assess whether there have been demand side
data revolutions. As noted above, using the definition
of a data revolution supplied by the Independent Ex-
pert Advisory Group on a Data Revolution for Sustain-
able Development, these would be better described as
statistical revolutions.
7.1. Data revolution #3 – Recognition of data and
statistics as public goods
Hervé Conan and Isabelle Valot of the Economic and
Financial Governance Department of Expertise France,
in their article, ‘Official statistics are a public good for
all’ [37] opined that ‘Quality statistics [are] a prerequi-
site for democratic dialogue’. Today, this seems noth-
ing more than common sense, but it wasn’t always so.
The purpose of official statistics has evolved. Originally
statistics were the exclusive preserve of the state, col-
lected to help rulers levy armies or tax populations (as
described by Deutsch’s first data revolution [20]). While
this is still true today, statistics now enjoy a broader
mandate; now they are expected to serve the state and
society. Official statistics are now recognized as a pillar
of democracy and accountability.
This progressive view is quite a recent development.
As recently as the 1980’s, the Thatcher administration
in the United Kingdom, viewed government statistics
as serving only the needs of central government [38]. In
the United States too, both Reagan and G. H. W. Bush
administrations viewed data and statistics as an eco-
nomic resource rather than a public good [39]. Today,
this view is roundly rejected, and expressions like ‘Data
belongs to the citizen, not to the government’ [40] are
more typical of the norm than the exception. The ac-
ceptance of official statistics as a public good has gone
hand in hand with the notion of freedom of agency,
of democratic and participatory government, political,
economic and personal liberty.
Reaching this enlightened position has taken a slow
silent revolution to which there have been a variety of
contributing factors. As the culture of scientific man-
agement or evidence-based decision-making spread to
public policy, the demand for rigour, for hard evidence,
for transparency and objectivity in decision making
contributed to the demand for official statistics. Within
academia there has been a move towards more open
access, as represented by the FAIR principles for sci-

1082 S. MacFeely / In search of the data revolution
entific data management and stewardship [41],
9
on the
grounds that closed access is detrimental to the distri-
bution of knowledge. There has been a similar push
within the creative arts as expressed by the Libre Cul-
ture Manifesto [42]. Without doubt, the communication
and information revolutions have also contributed. Like
all media industries (print, music etc.), official statistics
has had to grapple with the transition from hard copy to
online dissemination and in doing so has been forced to
decide between open and closed, between free and sub-
scription models. Official statistics have been carried
along by these swirling tides.
A public or collective good can be defined as a good
that is produced for the benefit or well-being of the
public
10
– a good that is consumed ‘together rather than
separately’ [43, p. 128]. In other words, for official
statistics to be a public good they should be produced
for the benefit of all. The idea of official statistics as a
public good was first formalised in Europe, where, at
the 1992 Economic Commission for Europe Ministe-
rial conference
11
the Fundamental Principles of Official
Statistics were adopted. These principles were subse-
quently adopted by the United Nations Statistical Com-
mission in 1994 to become the United Nations Funda-
mental Principles of Official Statistics. In 2014, these
principles were endorsed by the United nations General
Assembly.
12
The notion of official statistics as a public
good is set out principle 1 of the United Nations Funda-
mental Principles of Official Statistics [44] which states
‘Official statistics provide an indispensable element in
the information system of a democratic society, serving
the Government, the economy and the public with data
about the economic, demographic, social and environ-
mental situation. .. are to be compiled and made avail-
able on an impartial basis by official statistical agencies
to honour citizens’ entitlement to public information’.
Thus, when the United Nations General Assembly en-
dorsed these principles, heads of state from around the
world were explicitly saying that official statistics were
a public good.
Stiglitz [46] has argued that data are public goods,
which in economics literature, means they are both
non-rival and non-excludable [45]. Non-rival means
the use or possession of data by one person does not
9
https://www.go-fair.org/fair-principles/. The 4 underlying prin-
ciples of FAIR are: Findability, Accessibility, Interoperability, and
Reusability.
10https://www.lexico.com/definition/public_good.
11Decision c42.
12Resolution 68/261.
exclude simultaneous and full possession by another. In
other words, non-rival goods can be copied, shared and
used by many people at the same time. Furthermore,
data are also characterized by positive externalities [47]
meaning that data will almost certainly be useful to
others, and vice-versa.
Warsh [48] notes that many non-rival goods are made
at least partially excludable, by controlling or restricting
access. Jordan [49] explains that data can be converted
into a rival or excludable good by artificial means. The
distinction between data and official statistics in this
context is important, as for official statistics only aggre-
gates are made accessible – microdata must remain con-
fidential (excludable). But many private sector data are
also excludable owing to their proprietary nature [34].
Other data are in effect excludable owing to their am-
biguous licensing which prevents safe use from a le-
gal perspective. The notion of data and statistics as
non-excludable can also be challenged on the grounds
that much of it today requires significant expertise and
contextual knowledge to use these data properly. Thus,
asymmetries accessing data and the increased levels of
skills and general statistical literacy required to process
and understand statistical information have arguably
introduced frictions into the information market. In
other words, the argument that data and statistics are
public goods presupposes a certain level of statistical
literacy.
Rolland [50]
13
proposes that official statistics might
also be considered a ‘merit good’, meaning they are so
meritorious or fundamental that their production must
be publicly funded. Radermacher [8, p. 526] presum-
ably subscribes to this view, as he states ‘official statis-
tics belong to the core of inalienable public services.’
The notion of official statistics as a public or merit good
has now spilled over to data more generally, with calls
that the ‘data revolution must benefit all’ [51]. But a
counter argument, that data are a ‘club good’ has also
been put forward [52], meaning that some data are only
available to those in the club.
The issue of excludability is of growing importance.
In recent decades it has led to the Open Data movement.
According to Chignard [53] the term ‘open data’ was
first used in 1995, but Fretwell [54], identifies 2007
as the genesis of Open Data as we understand it to-
13
Readers of Rolland’s article will be rewarded by a fascinating
discussion on the pros and cons of whether official statistics are public
goods or not. He also explores whether in fact they are goods at all,
and not services.

S. MacFeely / In search of the data revolution 1083
day, when the 8 principles
14
were drafted in Califor-
nia. Less than two years later, on his first day in office,
President Obama, recognizing government information
as a public good, signed the ‘Memorandum on Trans-
parency and Open Government’. In 2013, the European
Union [55] passed legislation promoting the re-use of
public sector data. Gray [56] argues that the history of
Open Data and the motives behind it are more complex
and less enlightened that the standard narrative given
by open data advocates. He traces the origins back to
the 1990’s and a Randian, libertarian gospel of indi-
vidualism, small government, and market fundamental-
ism. In recent years open data has been recast as inno-
vative, civic minded and revolutionary and has down-
played the ideological and political connotations [57].
Whatever the motive, open data is the paradigm du
jour, enjoying a meteoric ascent, with governments
all around the world opening their data vaults. It even
has its own international day – Open Data Day, held
in early March every year. Some examples of coun-
tries adopting open data include: United Kingdom [58];
Australia [59]; Philippines [60]. The OECD too has an
Open Government Data initiative [61] that promotes
transparency, accountability and value creation by mak-
ing government data available to all. It publishes an
Open-Useful-Reusable Government Data (OURdata)
index that ranks countries progress towards data open-
ness. But as UNCTAD [62] has noted, many ‘open data’
initiatives are drives to open government data only and
this may inadvertently contribute to a growth in data
inequality by adding to a growing asymmetry in open-
ness between private and public sector data, where pub-
lic data are classified as a public good but corporate
data are classified as a marketable asset. The argument
that public sector data should be open, within sensible
limits, is relatively straight forward; open, transparent
government promotes accountability, and in any event,
taxpayers should have access to data as they paid for
them with their taxes.
But what about corporate data? Corporations now
perceive data as prized economic assets [63]. But who
owns those assets? In an argument reminiscent of Marx-
ist theory where capitalism exploits labour by appro-
priating their surplus value, or what Zuboff [64] calls
behavioural surplus, the argument has been made that
people also own much of the data held by search
engines, payments systems and telecommunications
14
Data should be: (1) complete; (2) primary; (3) timely; (4) acces-
sible; (5) machine-processable; (6) nondiscriminatory; (7) nonpropri-
etary; and (8) license-free.
providers [65] as they produced them with their labour.
Taylor [66, p. 1] notes, unsurprisingly, ‘this movement
is not gaining traction with those who hold the highest-
value data’. The debate on data ownership seems to
reflect wider economic trends where capital trumps
labour. The contention that data paid for with taxes
(capital) seems to be accepted, whereas the arguments
that data ‘paid for’ by labour is contested. But the de-
bate over data ownership is hugely complex, involving
property rights, copyright and IP, with implications for
contract and competition law [67]. Jordan [49] notes,
this already complex debate is further muddied by re-
cursion, the production of data from data, that is driving
the data deluge. Complex or not, Deichmann et al. [68]
have already proposed a market mechanism where ‘data
suppliers authorize the marketplace to license their in-
formation’.
The fact remains that for the moment the focus is
on opening public or government data only. Prender-
gast [69] contends that a new cold war has begun – a
war between individuals, corporations and States for
control of the ‘currency of the modern age’ – our per-
sonal data. At stake in this war is individual privacy;
sovereignty of data ownership; and weaponization of
data. A compounding problem, Coyle [47] notes, is
that many governments don’t understand the value of
data and continue to underinvest in administrative data
and official statistics. Davis [70], pondering the future,
describes a ‘potentially frightening proposition’ where
official statistics might be privatized. Citro [71] too
has sounded alarm bells, noting that federal statisti-
cal agencies in the United States are struggling to sur-
vive. In November 2019, the Federal Reserve in the
United States testified before the Congress Joint Eco-
nomic Committee that they are increasingly relying on
private sector statistics, rather than official statistics,
to understand the state of the economy [72]. Given the
systematic underinvestment in official statistics, this
shouldn’t come as any surprise. Even in China, Baidu
(China’s largest search engine) now compiles national
employment and consumption indices [73].
From an official statistics perspective, an important
development was the emergence of Open Data Watch in
2013, with a mission of ‘unlocking the data revolution’.
They apply 10 criteria to assess the openness of offi-
cial statistics for 178 countries. Indeed, open data has
quickly become the accepted standard for disseminat-
ing official statistics. Nevertheless, in their 2018/2019
report Open Data Watch noted countries with a high
openness index remain rare – only 7% of high-income
countries were awarded a score above 80 [74]. No low-

1084 S. MacFeely / In search of the data revolution
income countries were awarded a score above 60; 71%
of low-income countries were awarded a rank between
21 and 40. At international level, both the Data Strategy
of the Secretary-General for Action by Everyone, Ev-
erywhere: With Insight, Impact and Integrity [75] and
the UN Data and Statistics Roadmap [76] adopted by
the United Nations in 2020 include an explicit objective
that all UN agencies should adopt open data standards.
The report A World that Counts [10, p. 22] too argued
that data are a public good. The report outlined nine
key principles for harnessing the data revolution for
sustainable and inclusive development,
15
including data
transparency and openness, where the authors proposed
‘All data on public matters and/or funded by public
funds, including those data produced by the private sec-
tor, should be made public and “open by default”’. They
also argued that growing inequalities in access to data
and information was creating a new inequality frontier
between the data haves and have-nots, between those
who know, and those who do not. The principle that
data and statistics are public goods is also evident in the
2030 Agenda [77]. The importance of having access
to information is explicitly emphasized. For example,
Targets 12.8
16
and 16.10
17
set out the aspiration that
people should have access to information to help live a
sustainable life and to protect fundamental freedoms.
But implicitly, open data are even more central to 2030
Agenda than is immediately obvious. For example, Tar-
get 1.4
18
aims to give men and women equal rights to
economic resources and access to basic services. In a
data driven world, data must surely be considered as
both an economic resource and a basic service.
Assessing whether there has been a paradigm shift
is not straight forward; arguably there have been two,
but moving in opposite directions, and to some extent
cancelling each other out. On the one hand it is clear
that official statistics (and other public sector data) are
15
(1) Data quality and integrity; data disaggregation; data time-
liness; data transparency and openness; data usability and curation;
data protection and privacy; data governance and independence; data
resources and capacity; and data rights.
16
Target 12.8: by 2030 ensure that people everywhere have the
relevant information and awareness for sustainable development and
lifestyles in harmony with nature.
17
Target 16.10: Ensure public access to information and protect
fundamental freedoms, in accordance with national legislation and
international agreements.
18
Target 1.4: by 2030 ensure that all men and women, particularly
the poor and the vulnerable, have equal rights to economic resources,
as well as access to basic services, ownership, and control over land
and other forms of property, inheritance, natural resources, appropri-
ate new technology, and financial services including microfinance.
recognized as public goods, in rhetoric at least. On
the other hand, private sector data and statistics, now
understood as valuable assets are clearly not. In time
this may block or impede the ability of NSOs to fulfill
their mandate.
Developments in official statistics have been
prompted by global developments, such as, the emerg-
ing culture of scientific management; the opening-up of
science, academia and the creative arts; and the commu-
nication and information revolutions. Official statistics
were to some extent forced to evolve in this changing
atmosphere. But the development of the Fundamental
Principles of Official Statistics were also prompted by
crisis; the collapse of the Soviet Empire urgently re-
quired some sort of principles to help statistical offices
in former Eastern Bloc countries transition from the
closed and secretive soviet regime to more open, west-
ern liberal democracies where official statistics were
impartial and independent. The changes noted above
have led to new languages and understanding with re-
gard to the importance of statistics as an important pil-
lar of democracy and the openness of data. While there
have been multiple new bodies created, it’s not clear
this could be described as speciation.
Summary analysis:
9. Paradigm shift? Yes (but official statistics only)
10. Paradigm crisis? Yes
11. Paradigm vocabulary? Yes
12. Paradigm speciation? Unclear
Data Revolution? Yes
As the focus of this study is on official statistics,
Data revolution #3 can be described as a revolution. All
criteria except speciation have been met and it seems
impossible to argue that the transition of official statis-
tics from a restricted state asset to a public good is not a
paradigm shift deserving of the adjective revolutionary.
7.2.
Data revolution #4 – Data and statistics to inform
decisions
The idea that statistics can used to inform decisions
rests on the Churchillian ideal that figures can be the
neutral arbiters in political debate. According to Boaz et
al. [78, p. 234] the term ‘Evidence Based Policy Mak-
ing’ emerged during the run up to Tony Blair’s gov-
ernment being elected in the United Kingdom, while
noting this as a new label for a longstanding concern,
going back to the late 19
th
century. In fact, the existence
of Minoan linear B tables, census of population in Bib-
lical times or the Doomsday Book in Norman England,
all suggest that statistics were used to inform decisions

S. MacFeely / In search of the data revolution 1085
well before ‘Cool Britannia’. As Engel put it, statecraft
‘is a mere sham without a statistical foundation’ [79].
The emergence of evidence informed or evidence-
based decision making, and the role of official statis-
tics in this history, is a fascinating one. A wide range
of phenomena have led to the increased demand for
statistics. For example, Bernstein [80] argues that the
concept of risk emerged with the steady secularization
of society, as people began to believe they were free
agents, and that they had to take responsibility for their
own lives and decisions. Rostow [81] has made similar
arguments. Agar [82, p. 14] identifies ‘two great histor-
ical tidal waves: industrialisation and empire’ as having
had a profound impact on the need for information. The
industrial revolution, globalization and empire all drove
trade, which required information on markets, banking,
financial systems, credit and debt, prices and other as-
pects of commerce to help quantify risk. These global-
izing phenomena also led to wars and the development
of transport and logistical systems, and cyphers and
cryptanalysis – all of which required data and statistics.
Two events in the early 20
th
century had a profound
impact on official statistics – the great depression and
World War 2. As Keynes noted [83, p. 13] ‘every gov-
ernment since the last war (i.e. WW1) has been un-
scientific and obscurantist, and has regarded the col-
lection of essential facts as a waste of money’. Prior
to the great depression, much of the demand for data,
as noted above, were produced largely to meet the de-
mands of commercial interests. But the 20
th
century
saw the emergence of new concepts like the ‘economy’,
‘society’ and ‘unemployment’ which required state in-
tervention. FDR’s egalitarian response to the depres-
sion ravaging the USA, the ‘New Deal’, required a level
of statistical sophistication that hadn’t existed prior to
that point. Suddenly policy makers needed to under-
stand not only the size and structure of the economy
but also the distribution of real incomes and wealth,
and the likely impacts of redistribution. This was clear
in Roosevelt’s 1936 budget speech – ‘statistical infor-
mation concerning business activities and markets, em-
ployment and unemployment, incomes, expenditures,
and savings is urgently needed as a guide for economic
policies’ [84, p. 86]. The emergence of interventionist,
countercyclical, social and economic policy, such as,
Keynesianism, brought about a need for evidence to in-
form policy. As Fogel et al. [85] note, laissez-faire eco-
nomics was rooted in small business and didn’t require
any information on the national economy.
While the great depression had a profound impact
on official statistics, it was the world wars, and WW2
in particular, that really provided the impetus for the
development of modern official statistics and the con-
cept of evidence informing policy. WW1 ‘led to the
use of statistics, not only as a record of what had hap-
pened, but as a vital factor in planning what should
be done’ [86, p. 87]. The interwar years witnessed a
growing recognition or acceptance that statistics might
also have peace time applications for policy. During this
period work began on some of the most important and
enduring statistical concepts and indicators, most no-
tably the development of national income, labour force
and trade statistics [86,87]. The emergence and domi-
nance of these key indicators was in and of itself revo-
lutionary. Prior to WW2 economic planning was largely
descriptive with very little quantitative input [88]. But
the war and subsequent reconstruction efforts led to sea
change in the demand for quantitative evidence to in-
form decisions. So much so, Patinkin [89, p. 1110] de-
scribed the war as a ‘statistical revolution’. The global
scale and massive mobilization of WW2 required more
planning than previous engagements, involving secur-
ing and distributing materials and commodities, orga-
nizing labour, obtaining shipping, and imposing price
controls. It also involved balancing the requirements
of domestic economies with war economies, including
limiting non-essential imports. Underscoring the im-
portance of statistics, almost immediately after the war,
the UN Statistical Commission was established in 1947,
one of the very first commissions of the United Nations.
Taylorism, or scientific management, that drove de-
mand for data in industry in the early twentieth cen-
tury, later morphed into New Public Management in the
1980’s and spread this culture and demand for statistics
to the public sector. ‘Accountability’ and ‘performance
measurement and metrics’ became buzzwords among
politicians and policymakers as they were seen as a
guarantee of objectivity [90]. The demand for rigour,
for hard evidence, for transparency and impartiality also
contributed to the demand for official statistics as the
ideas of scientific management or evidence-based de-
cision making spread to public policy. Desrosières, in
what he terms the ‘five forms of interaction between
state, market and statistics’ [91], notes that this neo-
liberal model led to an ‘indicator culture’, that objec-
tified and massively proliferated statistics. This he as-
serts, paradoxically, led to statisticians losing control of
production, increasing the risks of indicators being con-
structed by vested interests and loss of general coher-
ence and reliability. He has also noted that the develop-
ment of statistics is typically presented as unconnected
or entirely autonomous of economic doctrine and prac-

1086 S. MacFeely / In search of the data revolution
tice. In other words, ‘these two histories, that of politi-
cal economy and that of statistics, are rarely presented,
much less problematized together’ [92, p. 553]. But in
setting norms and standards for the practical manage-
ment of the state, he argues that statistics are both a
tool of proof and governance, as they simultaneously
measure and shape reality, and thus play a dual consti-
tutive co-production role.
19
More broadly, the spread of
democracy has also been central to this story, as citizens
need information to fully participate and hold their gov-
ernments to account. Not unrelated, the development
of 24 hour newsfeeds, with their insatiable appetite for
content, has also generated a huge demand for data.
With the avalanche of big data, facilitating algorith-
mic based decisions, we are seeing an important shift
from evidence-informed decisions to data-driven deci-
sions. The former acknowledges and makes transparent
the judgements and trade-offs involved in democratic
decision making, the later implicitly adopts a datacratic
approach. This important transition is at the heart of
concerns around the use of artificial intelligence or AI
in public policy. Algorithms, or what Tufekci has called
‘semi-savant mini-Frankensteins’ [94] are playing an
increasingly greater role in our lives. But algorithms
like any computationally reducible model, are just sim-
plifications of reality. They are manmade and are not
immune from poor or biased choices. They may uti-
lize the wrong data or inadvertently measure the wrong
thing. We have seen examples of this throughout his-
tory. Famously, McNamara the Secretary of State in
the USA, inappropriately used ‘body count’ as a met-
ric of success for prosecuting the war in Viet Nam.
Lewis [95] details the fascinating story of how in the
sport of baseball, coaches and selectors used inappro-
priate metrics to judge and value players. But more pro-
foundly, what does data driven decision making mean
for legal systems and for human rights? The ‘end of
theory’ approach, where petabytes of data have allowed
us to replace the traditional hypothesis approach to sci-
ence with a ‘correlation is enough’ methodology is a
further complication. Lohr [96, p. 96] has described this
change as a ‘Copernican shift in discovery and decision
making’. It poses interesting questions for what ‘knowl-
edge’ will mean in this new environment – what does
it mean to ‘know’ something without understanding
19
This is not unconnected to what physicists describe as the ob-
server effect, or what George Soros has termed reflexivity [93] i.e.
the act of measuring economies or societies, leads to changes in ideas
or behaviours which feedback and change the economies or societies
being observed.
causation? T.S. Eliot, in his poem Choruses from the
Rock, presciently asked ‘Where is the knowledge we
have lost in information?’
The adoption of data and statistics to support decision
making was a paradigm shift, but one that took place
gradually over the past century. More recently, we have
witnessed a new iteration of this paradigm shift, from
evidence informed to data driven decision making. The
first paradigm shift was the result of crises – the great
depression and world war. The more recent shift has not
been the result of crisis but may well provoke one. Al-
gorithmic based decision making and proponents of the
‘end of theory’, have brought new paradigm vocabulary
and speciation (not least the separation of statistics and
data science).
Analysis:
13. Paradigm shift? Yes
14. Paradigm crisis? Yes
15. Paradigm vocabulary? Yes
16. Paradigm speciation? Yes
Data Revolution? Yes
Thus, we can conclude that Data revolution #4 is a
revolution as all criteria have been met.
8. Coming crises?
We are still in the “Dawn of the ‘Data Revolu-
tion”’ [2, p. 10]. As the digital revolution unfolds, so
too will the next stages of the data revolution. Almost
certainly, there will be crises ahead, some foreseeable,
some unexpected. In keeping with Kuhn’s theory, some
of these crises may well spark new data revolutions, but
others will not. As noted above, one can anticipate any
number of crises: the loss of national data sovereignty –
where data holders and the data themselves are beyond
the reach of national legal systems, meaning govern-
ments cannot enforce national laws or protect their citi-
zens; and data ownership – clarity regarding ownership
will be important, not only to safeguard personal pri-
vacy, but also commercial and national security. “Data
is the lifeblood of international trade” [97, p. 8] and will
be central to who controls any future global digital or
crypto currency [98]. UNCTAD [99, p. xix] has argued
that “key policy questions include how to assign owner-
ship and control over data; how to build consumer trust
and protect data privacy, how to regulate cross-border
data flows, and how to build relevant skills and capabil-
ities for harnessing digital data for development”. For
this discussion, two potential crises have been selected
that, should they arise, could have a profound impact
on the governance and compilation of official statistics.

S. MacFeely / In search of the data revolution 1087
8.1. Coming crisis #1 – Competition leading to
irrelevance
Charles Dickens novel A Tale of two Cities opens
with the immortal line “It was the best of times, it was
the worst of times, it was the age of wisdom, it was the
age of foolishness, it was the epoch of belief, it was the
epoch of incredulity, it was the season of Light, it was
the season of Darkness, it was the spring of hope, it was
the winter of despair .. .” Although written in 1859,
the comparisons drawn in these opening lines seem
extraordinarily apropos the situation official statistics
find themselves in today. Never has the worth of of-
ficial data and statistics enjoyed such a dichotomous
existence. Statistics have never been more recognized
as a critically important tool for supporting decision
making and providing public accountability, and yet
simultaneously statistics have never been so irrelevant
to public discourse. Today, in the midst of a data deluge,
official statistics are competing with myriad informa-
tional sources, all claiming to be authoritative. It seems
Huxley’s prophesy that truth would drown in a sea of
irrelevance hit the nail on the head [100].
The existential question for official statistics is
whether the good times or the bad times will reign
supreme? Fukuyama [101] warns ‘in a world without
gatekeepers, there is no reason to think that good in-
formation will win out over bad’. The noise in public
debate is increasing and public animosity to evidence-
based policy making is rising. Stiglitz [102] makes the
argument that after 40 years of evidence based, ne-
oliberal policy, which has brought only slow growth,
rising inequality, stagnating wages and massive asset
price inflation, has left ordinary citizens feeling conned.
He argues that we are experiencing the political con-
sequences of a grand deception: distrust of elites, of
economic ‘science’ on which neoliberalism is based,
and of a money-corrupted political system that made it
all possible. This neo-liberal doctrine has not only un-
dermined public services in general, but as noted above,
has resulted in continued underinvestment in official
statistics and national data infrastructures. Small won-
der then that Davies [103] has concluded that official
statistics is losing the battle for relevance, declaring
‘the declining authority of statistics is at the heart of the
crisis that has become known as “post-truth” politics.’
Novelist John le Carre [104] succinctly captured the
issue when he wrote ‘one day somebody will explain
to me why it is that, at a time when science has never
been wiser, or the truth more stark, or human knowl-
edge more available, populists and liars are in such
pressing demand’. This may change. If the outbreak of
COVID-19 has illustrated one thing, it is the need for
credible data and scientific expertise. But COVID-19,
climate change, and rising inequalities (to name just a
few global challenges), have all signaled too the need
for more integrated, holistic statistics, not simply longer
lists of indicators.
Engineer and management guru W. E. Deming
opined ‘Without data you’re just another person with an
opinion’. This of course, was only his opinion. Deming
made the common mistake of treating data and statis-
tics as synonymous, which they aren’t. Having data
doesn’t really add anything to a discussion, unless it
has been organized and aggregated so that knowledge
can be gleaned from it. In other words, he should have
said ‘without statistics. . . ’ But what makes a statis-
tic a statistic? Today, official statistics are competing
with a barrage of ‘statistics’ masquerading as science.
But many of these so called statistics are little better
than opinions as they are unaccompanied by paradata,
metadata or any transparent methodologies. They are
nothing more than mysterious black box numbers and
their real quality and utility is unclear. No doubt some
are of underlying good quality, but others are deliber-
ately misleading. Critically, these ‘statistics’ compete
with official statistics in the evidence and knowledge
markets.
The debate surrounding the quality of COVID-19
statistics has brought many of these issues to a head.
Arguably, COVID-19 has also sparked a greater ap-
preciation of the need for public services and public
infrastructure. This reevaluation may provide an oppor-
tunity to make the case for greater investment in public
statistics. The emerging debacle over the veracity of
statistics relating to COVID-19 cases and mortality in
some countries, also highlights a perennial dilemma
for international official statistics. Should international
organisations (IOs) always trust the official statistics
reported by their member states? MacFeely and Nas-
tav [25] have suggested previously, that there might be
good reasons to be circumspect about some national of-
ficial statistics. How statistical offices, both national and
international, should respond to falsified statistical re-
ports and fake news is not always clear. It is not always
easy to verify the veracity of data. It is often hard, if not
impossible, to do so without the NSO or IO becoming
politicised in the process. In an age where quantification
is cult, in a world of ‘alternative facts’, where data and
statistics are being weaponized [105,106] how should
official statistics (both the profession and the product)
distinguish themselves from other statistics and protect
their legitimate role as custodians of knowledge and
protectors of deliberative public spaces?

1088 S. MacFeely / In search of the data revolution
8.2. Coming crisis #2 – Confidentiality
For official statistics, safeguarding the confidential-
ity of individual data is sacrosanct. The importance
of confidentiality is enshrined in Principle 6 of the
United Nations Fundamental Principles of Official
Statistics [44], the ‘basic law’ or ‘constitution’ of offi-
cial statistics [107]. Principle 6 states ‘Individual data
collected by statistical agencies for statistical compi-
lation, whether they refer to natural or legal persons,
are to be strictly confidential and used exclusively for
statistical purposes.’ For a national statistical system to
function, the confidentiality of the persons and entities
for which it holds individual data must be protected
i.e. a guarantee to protect the identities and information
supplied by all persons, enterprises or other entities. In
short, everyone and anything who supplies data for sta-
tistical purposes does so with the reasonable presump-
tion that their confidentiality will be respected and pro-
tected. In most countries, safeguarding confidentiality
is enshrined in national statistical legislation.
Meanwhile, in the private sector, the captains of
industry 4.0, such as, Mark Zuckerberg (Facebook),
Scott McNealy (Sun Microsystems) and John McAfee
(McAfee Associates) have all in recent years argued
that the concept of privacy is extinct [108
–
110]. Many
disagree and have voiced concerns over the loss of pri-
vacy [61,111,112]. New data protection legislation in
Europe (General Data Protection Regulation) and in
California (California Consumer Privacy Act) suggests
that at least some legislators still see a value in privacy.
Nevertheless, it is difficult to see how the concept of
privacy can survive unscathed with the relentless drive
towards the Internet of Things – smart phones, smart
TVs, smart cars, smart homes, smart cities. Soon it
seems everything we do will be monitored. One can’t
help but wonder whether privacy as an ‘ideal’ might
still be alive and well, but privacy in ‘practice’ is on
life-support; day after day, we read about enterprises
and institutions failing to protect personal records.
But what if Zuckerberg & Co. are correct? Ta-
plin [113, p. 157] ponders this, musing ‘It very well
may be that privacy is a hopelessly outdated notion and
that Mark Zuckerberg’s belief that privacy is no longer
a social norm has won the day.’ What then are the im-
plications for official statistics, and the requirement to
guarantee confidentiality? If other statistical providers,
not governed by the UN Fundamental Principles, take a
looser approach to confidentiality and privacy, or ignore
it completely, this will leave official statistics in a rela-
tively disadvantaged position vis-à-vis other statistical
compilers.
In addition to the broader crisis outlined above, NSOs
will surely need to deal with two specific issues. The
first centres on differential confidentiality and whether
it is sustainable to offer the same level of confidentiality
to persons and enterprises; in particular, multinational
enterprises. Arguably, a distinction can legitimately be
made, most especially, when the masking of multina-
tional enterprise activity inhibits the publication of other
statistics that are of vital national interest and impor-
tance. The second is how to respond to emergencies and
crises. In a time of crisis, populations expect their Gov-
ernments and public services (including their statistical
systems) to step up and provide services (and infor-
mation) more quickly than usual. Furthermore, ethical
frontiers and social license are elastic – contracting in
good times but loosening in bad. For example, following
9/11 many legal barriers to data sharing were quickly
swept aside [114] as the political focus shifted from
privacy to security. The COVID-19 pandemic may be
the justification for our next Faustian bargain. In March
2020, 19 countries reported accessing citizen data to
track the virus [115] including Austria, Germany, Italy,
the UK and the USA, while Liechtenstein was planning
to electronically tag and monitor its citizens [116].
20
Yale’s professor Sudhir neatly sums up the situation;
‘privacy concerns are on the back burner during this
emergency’ [117]. But what happens after crises? It is
not obvious that publics elasticity is transitive; nor is it
clear that Governments, having discovered the power of
tracing technology, will be willing to surrender it. Will
we look back ruefully on 2020 someday, the year of the
pandemic, with a deeper understanding of what 2020
hindsight really means?
This circles back to the earlier discussion, regarding
public-private asymmetries, and competition. Another
important question emerging from the pandemic is why
Google were allowed to publish detailed statistics, har-
vested from their applications and platforms, on pop-
ulation movements [118,119], but NSOs are routinely
prevented by Data Protection Commissioners from ac-
cessing and using ‘big data’ to provide similar types
of useful information? On March 27 the UK Informa-
tion Commissioner’s Office noted that “data protection
law enables the data sharing in the public interest and
provides the safeguards for data that the public would
expect” [120]. How do we reconcile this with official
statistics being a public good?
20
Outside the scope of this paper; but contrasting developments in
Liechtenstein with the protests against ‘excessive’ quarantine restric-
tions in some US states, and the attacks against the WHO, Covid-19
has unwittingly exposed the battle between individual and community
rights, and between nationalism and multilateralism.

S. MacFeely / In search of the data revolution 1089
8.3. COVID-19: The crisis that came
Just as this paper was being finished, the fourth
horseman of the apocalypse once again rode out of the
shadows, this time calling himself COVID-19, and un-
leashed chaos. NSOs and IOs around the world were
forced to quickly adapt to the challenges of contain-
ment. An interesting summary of their innovative re-
sponses is presented in the joint United Nations – World
Bank report Monitoring the state of statistical opera-
tions under the COVID-19 Pandemic [121]. For the pur-
poses of this paper however, contributions to the special
IAOS discussion on Official statistics in the context of
the COVID-19 crisis
21
provide more reflective insights
on the implications of, and opportunities arising from
COVID-19 for official statistics. The debate is too rich
to summarise here, but immediately relevant are the dis-
cussions on whether, and how, official statistics might
strategically reposition for the future, and the greater
levels of innovation and openness to solutions that are
required. Pedro Diaz Muñoz [122] caught the spirit of
the discussion well, noting that ‘most industrial and
institutional sectors will have to launch gigantic trans-
formations in order to survive; we cannot lag behind’.
But what might those transformations be? Writing from
within the eye of the storm, it is not easy to identify,
let alone comprehend, the countervailing forces of his-
tory. Who can say what the long-term repercussions of
COVID-19 will be for official statistics?
The pandemic has already had a profound impact
on NSOs; in short order they have had to grapple with
a whole series of knotty conceptual, methodological,
logistical and technological issues [123]. But it is not
clear, that any fundamental transformation or paradigm
shift is, as yet, underway within the statistics commu-
nity. There can be no doubt however that the tectonic
plates of the global data ecosystem are shifting; as noted
above, key users are already diversifying their informa-
tional sources. The question is, will COVID-19 stem
this tide, bringing a renewed appreciation of official
statistics or will it accelerate the process, driving an-
other nail in the coffin? It seems hard to imagine that
the technological tracking solutions, discussed in Sec-
tion 8.2 will not have profound long-term implications
for statistics, and perhaps for freedom and democracy
too. Cheung [124], rather pessimistically, has argued
that the designation of NSOs as a ‘non-essential ac-
tivity’ in many countries, just when official statistics
21
https://officialstatistics.com/news-blog/crises-politics-and-
statistics.
should be in peak demand, is ‘a severe indictment of
NSO’s relevance and usefulness’. Di Gennaro Splen-
dore [125, p. 234] offers a more optimistic view, sug-
gesting that COVID-19 presents NSOs with ‘extraor-
dinary opportunities to redesign their roles and their
tools’ but also warns they ‘need to learn from the crisis’.
In the wider world, the pandemic has highlighted the
risks of steadily eroding public services and infrastruc-
ture. In doing so, it has provoked debate on whether
societies and economies should return to the ‘old nor-
mal’ or dare to imagine a better ‘new normal’ – one
that might also address the climate crisis or persistent
inequalities. The official statistics community might
usefully engage in a similar contemplation. What, as
a profession would we like a ‘new normal’ to be? For
those in need of inspiration, Walter Radermacher’s arti-
cle How Statistics Can Help – Going Beyond COVID-
19 [126] is not a bad place to start.
9. Conclusion
Data – the single word that defines our age; data have
assumed a new importance for today’s economies, soci-
eties and politics. Data are the glue that binds and drives
the digital economy, the cloud, blockchain, the internet-
of-things, and increasingly our politics too. They offer
promise but also peril – they are a tool for liberation,
but also potentially a weapon for exploitation. Data
and statistics have both a knowledge and a governance
effect.
The argument put forward in this paper, is that there
has not been a single data revolution, but many. The
Data Revolution is in fact a series of revolutions. Those
revolutions are a function or consequence of other rev-
olutions; digital, informational, cultural and social. Us-
ing definitions of data revolution supplied in A World
That Counts, and criteria derived from Kuhn’s semi-
nal work on scientific revolutions, their existence have
been investigated, through the prism of official statis-
tics. Two supply and two demand data revolutions were
identified. Broadening the definition of data and use
of secondary data on the supply side, and statistics as
a public good and to inform decisions on the demand
side. The conclusion is that there has been both supply
and demand side data revolutions. However, on the de-
mand side that revolution is more complex and more
fragile. The paper has also speculated that future crises
might trigger new data revolutions, not least, compe-
tition leading to irrelevance and changes provoked by
privacy and confidentiality concerns.

1090 S. MacFeely / In search of the data revolution
The data revolution(s) as we now understand the term
is inextricably linked to the SDGs. It began as an aspira-
tion, a plea for better data but quickly transformed into
a fact. Diplomats make reference to it, not as a future
state, but as the solution. But as Froman and Shah [51]
warn ‘although data are ubiquitous, and the opportuni-
ties offered by the data revolution are even larger, not
everyone is poised to gain from it equally’. Perhaps
ubiquity has led to a paradox of plenty, a resource curse,
a Dutch data disease? One can also question whether
in fact data are ubiquitous? Without question the digi-
tal revolution has led to a data deluge but as discussed
above, many data are excludable. So much so, there is
in fact not one data economy but several, each with its
own ideology and membership [52] and all pretty much
unregulated.
There needs to be a New Global Data Deal for safely
accessing and using data (and by extension statistics)
while protecting the rights of citizens. Such a deal or
convention will need to be global in order to address
ethical and sovereignty issues. It must reestablish some
sort of social contract that strikes a balance between
community dataveillance and individual and human
rights, between security and privacy, between com-
merce and public good, between asymmetries in private
and public openness
22
and funding, between data own-
ership and reward. As data and statistics are joined at
the hip, official statisticians must be part of this discus-
sion to ensure their perspectives are properly reflected
and their mandate refreshed. In an era of faltering mul-
tilateralism, it may be convenient to turn a blind eye,
but given the importance of data to all of our futures,
the United Nations cannot ignore this challenge. Gov-
ernments cannot abdicate their responsibility either. A
real challenge in this respect is that many governments
don’t understand the value of data.
23
Coyle [47] notes
that many governments are ‘mismanaging what will
become one of the key assets of modern economies’,
signing away the rights to national (public) data assets
as they outsource services to the private sector [129].
Perhaps an important conclusion is that there is a data
revolution underway in the private sector, after all, they
hold and have access to our digital exhaust. The public
sector has been slower to react, for a variety of reasons,
cultural, lack of expertise and resources, historic un-
22
Perhaps news [127] that Microsoft have embraced open data
offers a glimmer of hope?
23
Of course, others understand the power of data all too well – as
Larsen [128] points out in her article ‘Who needs democracy when
you have data?’
derinvestment in data and IT, more legal restrictions
prohibiting access and use.
We are at a fascinating moment in the history of data
and statistics. The landscape is changing before our
eyes. But irrespective of the criteria discussed above, the
overarching question, is whether the changes we have
witnessed have been transformative or not. Overall, it
is hard to argue that they have not been. Yet we see
the benefits are unevenly distributed, and in the context
of the 2030 Agenda this is important. Are the benefits
of the data revolutions being shared equally across the
world? Clearly not – the digital divide has created a data
divide [34]. Runde [130] correctly notes that without
skills and infrastructure, for many countries the data
revolution will remain an aspirational notion. Without
funding too. This is perhaps the key conclusion – data
revolutions are not free!
References
[1]
United Nations (2013). ‘A New Global Partnership: Eradicate
poverty and transform economies through sustainable devel-
opment – The Report of the High-Level Panel of Eminent
Persons on the Post-2015 Development Agenda’. Available
at: http://www.post2015hlp.org/wp-content/uploads/2013/
05/UN-Report.pdf [last accessed: 20.03.2017].
[2]
Association of Chartered Certified Accountants/Institute of
Management Accountants (2015). The Data Revolution.
Available at: https://www.imanet.org/-/media/3cc5e721552
9457a8dc3816c81452d95.ashx [Last accessed 01.11.2019].
[3]
United Nations System (2015). Conclusions of the High-level
Committee on Management at its twenty-ninth session
(UNESCO headquarters, Paris, 19 and 20 March 2015). Chief
Executives Board for Coordination CEB/2015/3. Available at:
https://www.unsceb.org/CEBPublicFiles/CEB_2015_3%20%
28HLCM%2029%29.pdf [last accessed: 01.04.2020].
[4]
PARIS21 (2017). Informing a Data Revolution – video. Avail-
able at: https://paris21.org/informing-data-revolution-idr [last
accessed: 30.10.2019].
[5]
United Nations (2018). Statistical Commission Report on the
forty-ninth session (6–9 March 2018). Economic and Social
Council, Official Records, 2018. Supplement No. 4. Doc.
E/2018/24-E/CN.3/2018/37. Available at: https://unstats.un.
org/unsd/statcom/49th-session/documents/Report-on-the-
49th-session-E.pdf [last accessed: 30.10.2019].
[6]
Mohammed, A.J. (2018). Opening Remarks – World Data
Forum, Dubai. 22 October, 2018. UN Web TV. Available at:
http://webtv.un.org/search/deputy-secretary-general-opens-
the-world-data-forum/5852039006001/?term=Data%20Revo
lution&page=11 [last accessed: 30.10.2019].
[7]
Fritz, S., See, L., Carlson, T., Haklay, M., Oliver, J.L., Fraisl,
D., Mondardini, R., Brocklehurst, M., Shanley, L.A., Schade,
S., When, U., Abrate, T., Anstee, J., Arnold, S., Billot, M.,
Campbell, J., Espey, J., Gold, M., Hager, G., He, S., Hepburn,
L., Hsu, A., Long, D., Masó, J., McCallum, I., Muniafu,
M., Moorthy, I., Obersteiner, M., Parker, A.J., Weissplug,
M., West, S. (2019). Citizen science and the united nations
sustainable development goals. Nature Sustainability, 2, 922–
930.

S. MacFeely / In search of the data revolution 1091
[8]
Radermacher, W.J. (2019). Governing by the Numbers: Re-
flections on the future of official statistics in a digital and
globalised society. Statistical Journal of the International
Association of Official Statistics, 35(4), 519–537.
[9]
United Nations Statistics Commission (2020). Review of
Implementation of Data Revolution. Prepared by the United
Nations Statistics Division for the High-Level Group for
Partnership, Coordination and Capacity-Building for statis-
tics for the 2030 Agenda for Sustainable Development (HLG-
PCCB). Background document Item 3 (a), Statistical Com-
mission, fifty-first session, 3–6 March 2020. Available at:
https://unstats.un.org/unsd/statcom/51st-session/documents/
BG-Item3a_Review-of-Implementation-of-Data-Revolution
-E.pdf [Last accessed: 24.02.2020].
[10]
Independent Expert Advisory Group on a Data Revolution
for Sustainable Development (2014). A World That Counts:
Mobilising the Data Revolution for Sustainable Develop-
ment. Report prepared at the request of the United Nations
Secretary-General. Available at: http://www.undatarevoluti
on.org/report/ [last accessed: 20.03.2017].
[11]
Kuhn, T. (1962). Structure of Scientific Revolutions. 50
th
Anniversary Edition (2012). University of Chicago. Chicago.
[12]
Melamed, C. (2014). The data revolution is coming and it
will unlock the corridors of power. The Guardian, 1 October,
2014. Available at: https://www.theguardian.com/global-
development/poverty-matters/2014/oct/01/data-revolution-
development-united-nations [last accessed: 24.04.2017].
[13]
Jerven, M. (2014). ‘The problem with the data revolution in
four Venn diagrams’. The Guradian, 17 December, 2014.
Available at: https://www.theguardian.com/global-develop
ment-professionals-network/2014/dec/17/data-revolution-
limitations-in-images [last accessed: 24.04.2017].
[14]
Weigand, A. (2009). ‘The Social Data Revolution(s)’.
Harvard Business Review, May 20, 2009. Available at:
https:// hbr.org/2009/05/the-social-data-revolution.html [last
accessed: 24.04. 2017].
[15]
Ooman, N. (2010). You Say You Want a Data Revolution!
Center for Global Development – Global Health Policy Blog.
Available at: https://www.cgdev.org/blog/you-say-you-want-
data-revolution [last accessed: 24.05.2017].
[16]
Scotten, D.W. (1961). Marketing adjustment to the environ-
ment. Bureau of Business Management, College of Com-
merce and Business Administration of the University of Illi-
nois.
[17]
Allardt, E., Littunen, Y. (Ed) (1964). Cleavages, ideologies
and party systems; contributions to comparative political so-
ciology. Transactions of the Westermarck Society, Volume
10. University of Michigan, USA.
[18]
Halina, J.W. (1966). Data transmission – current trends and
future prospects. Electrical Communication – The Techni-
cal Journal of the International Telephone and Telegraph
Corporation, 41(2), 177–195.
[19]
Mackenzie, W.J.M. (1971). The Study of Political Science
Today: Studies in Comparative Politics. The Macmillan Press
Ltd, London and Basingstoke.
[20]
Deutsch, K.W. (1970). The Impact of Complex Data Bases on
the Social Sciences. In Bisco, R.L. (Ed) Data Bases, Comput-
ers, and the Social Sciences, pp. 19–41. Wiley-Interscience,
New York.
[21]
Heim, K.M. (1982). Data libraries for the social sciences –
introduction. Library Trends, 30(3), 321–325. Winter 1982.
University of Illinois.
[22]
Russet, B. (2005). Four Methods and Five Revolutions, in
Mintz, A. and Russett, B. (Eds.). New Directions for In-
ternational Relations: Confronting the Method of Analysis
Problem. Lexington Books, Oxford.
[23]
Lebada, A.M. (2016). Member states, statisticians address
SDG monitoring requirements; available at: http://sdg.iisd.
org/news/member-statesstatisticians-address-sdg-monitorin
g-requirements/?rdr=sd.iisd.org [last accessed: 07.04.2016].
[24]
Harari, Y.N. (2019). INTERVIEW/Yuval Noah Harari:
Mankind still has time to stop rule by computer algorithms.
The Asahi Shimbun. October 4, 2019. Available at: http://
www.asahi.com/ajw/articles/AJ201910040015.html [last ac-
cessed: 04.04.2020].
[25]
MacFeely, S., Nastav, B. (2019). You say you want a (data)
revolution? Statistical Journal of the International Associa-
tion of Official Statistics, 35(3), 309–327.
[26]
MacMillan (2017). Data Revolution – Definition. MacMillan
Crowdsourced Open Dictionary. Available at: http://www.
macmillandictionary.com/dictionary/british/data-revolution
[last accessed: 27.09.2017].
[27]
PARIS21 (2015). A Road Map for a Country-led Data Revo-
lution. PARIS21Secretariat. Available at: https://www1.oecd.
org/publications/a-road-map-for-a-country-led-data-revolut
ion-9789264234703-en.htm [last accessed: 30.10.2019].
[28]
Hacking, I. (2012). Introductory Essay’ in Kuhn, T.S. (2012).
The Structure of Scientific Revolutions – 50
th
Anniversary
Edition. The University of Chicago Press, Chicago.
[29]
Stephens-Davidowitz, S. (2017). Everybody lies – What the
internet can tell us about who we really are. Bloomsbury,
London, UK.
[30]
Waterford Technologies. (2017). Big Data Statistics & Facts
for 2017. Available at: https://www.waterfordtechnologies.
com/big-data-interesting-facts/ [last accessed: 20.03.2020].
[31]
Goodbody, W. (2018). Waterford researchers develop new
method to store data in DNA. RTE News. Available at: https://
www.rte.ie/news/ireland/2018/0219/941956-dna-data/ [Last
accessed: 24.03.2020].
[32]
IBM. (2017). 10 Key Marketing Trends for 2017 and Ideas for
Exceeding Customer Expectations. IBM Marketing Cloud.
Available at: http://anyflip.com/vdoi/wpza/basic [last ac-
cessed: 24.03.2020].
[33]
United Nations Economic Commission for Europe (2011).
Using Administrative and Secondary Sources for Official
Statistics: A Handbook of Principles and Practices. New
York and Geneva, 2011. Available at: https://www.unece.org/
fileadmin/DAM/stats/publications/Using_Administrative_
Sources_Final_for_web.pdf [last accessed: 29.10.2019].
[34]
MacFeely, S. (2018). Big Data and Official Statistics. In
Kruger, S. and Kruger, M. (Eds.) in Big Data Governance
and Perspectives in Knowledge Management, IGI Global,
Hershey, PA, pp. 25–54.
[35]
Sangeetha, S., Sreeja, A.K. (2015). No science no humans,
no new technologies no changes “big data a great revolution”.
International Journal of Computer Science and Information
Technologies, 6(4), 3269–3274.
[36]
MacFeely, S. (2017). Measuring the sustainable development
goals: What does it mean for ireland? Administration 65(4),
41–71.
[37]
Conan, H., Valot, I. (2019). Official statistics are a public
good for all. Euractiv, 14 June, 2019. Available at: https://
www.euractiv.com/section/politics/opinion/la-statistique-
officielle-est-un-bien-public-qui-appartient-a-tous/1350058/
[last accessed 05.11.2019].
[38]
Jones, K., Millard, F., Twigg, L. (1990). The Right to Know:
Government and Information. In Savage, S.P. and Robins,
L. Public Policy under Thatcher. MacMillan Education. St.

1092 S. MacFeely / In search of the data revolution
Martin’s Press Inc. New York.
[39]
McDermott, P. (2008). Who Needs to Know? – The State of
Public Access to Federal Government Information. Bernan
Press, Maryland, USA.
[40]
Van Belle, J.P. (2018). Africa Data Revolution Report 2018
– Status and Emerging Impact of Open Data in Africa.
Available at: http://webfoundation.org/docs/2018/10/African-
data-revolution-report-brochure-publish.pdf [last accessed:
20.02.2020].
[41]
Wilkinson, M.D., Dumontier, M., Aalbersberg, J., Appleton,
A., Axton, M., Baak, A., Blomberg, N., Boiten, J., Bonino da
Silva Santos, L., Bourne, P.E., Bouwman, J., Brookes, A.J.,
Clark, T., Crosas, M., Dillo, I., Dumon, O., Edmunds, S.,
Evelo, C.T., Finkers, R., Gonzalez-Beltran, A., Gray, A.J.G.,
Groth, P., Goble, C., Grethe, J.S., Heringa, J., ’t Hoen, P.A.C.,
Hooft, R., Kuhn, T., Kok, R., Kok, J., Lusher, S.J., Martone,
M.E., Mons, A., Packer, A.L., Persson, B., Rocca-Serra, P.,
Roos, M., van Schaik, R., Sansone, S., Schultes, E., Sen-
gstag, T., Slater, T., Strawn, G., Swertz, M.A., Thompson,
M., van der Lei, J., van Mulligen, E., Velterop, J., Waag-
meester, A., Wittenburg, P., Wolstencroft, K., Zhao, J., Mons,
B. (2016). The FAIR Guiding Principles for scientific data
management and stewardship. Scientific Data, 3. 15 March,
2016. 160018 (2016) doi: 10.1038/sdata.2016.18. Available
at: https://www.nature.com/articles/sdata201618#citeas [last
accessed: 20.01.2020].
[42]
Berry, D. (2015). The Libre Culture Manifesto. Free Software
Magazine. 4 April, 2005. Available at: http://freesoftwarema
gazine.com/articles/libre_manifesto/ [20.01.2020].
[43]
Sen, A. (2000). Development as Freedom. Anchor Books,
New York.
[44]
United Nations (2014). The Fundamental Principles of Of-
ficial Statistics. Resolution 68/261 adopted by the Sixty-
eighth session of the General Assembly on 29 January 2014.
A/RES/68/261. Available at: https://unstats.un.org/unsd/dnss/
gp/FP-New-E.pdf [last accessed: 17.01.2020].
[45]
Samuelson, P.A. (1954). The pure theory of public expendi-
ture. Review of Economics and Statistics, 36(4), 387–389.
[46]
Stiglitz, J.E. (1999). Knowledge as a public good. In Kaul, I.,
Grunberg, I. and Stern, M. (Eds). Global Public Goods: Inter-
national Cooperation in the 21
st
Century, pp. 308–352. Ox-
ford Scholarship Online. doi: 10.1093/0195130529.001.0001.
[47]
Coyle, D. (2019). Valuing data is tricky but crucial for the
public good. Financial Times. November 20, 2019. Available
at: https://www.ft.com/content/b5bc3762-09f9-11ea-8fb7-
8fcec0c3b0f9 [last accessed 17.01.2020].
[48]
Warsh, D. (2006). Knowledge and the Wealth of Nations: A
story of economic discovery. W.W. Norton & Company, New
York and London.
[49]
Jordan, T. (2015). Information Politics: Liberation and Ex-
ploitation in the Digital Society. Pluto Press, London.
[50]
Rolland, A. (2017). The concept and commodity of official
statistics. Statistical Journal of the IAOS, 33(2), 373–385.
[51]
Froman, M., Shah, R. (2020). A Data Revolution for All.
Project Syndicate. January 20
th
, 2020. Available at: https://
www.project-syndicate.org/commentary/dataorg-relaunch-
boosts-social-entrepreneurship-by-michael-froman-and-
rajiv-shah-1-2020-01 [last accessed 23.01.2020].
[52]
The Economist (2020). Are data more like oils or sunlight?
Special Report, February 20
th
, 2020. Available at: https://
www.economist.com/special-report/2020/02/20/are-data-
more-like-oil-or-sunlight [last accessed: 25.02.2020].
[53]
Chignard, S. (2013). A brief history of Open Data. Paris
Innovation Review. 29 March, 2013. Available at: http://
parisinnovationreview.com/articles-en/a-brief-history-of-
open-data [last accessed 05.11.2019].
[54]
Fretwell, L. (2014). A brief history of open data. FCW, June
09, 2014. Available at: https://fcw.com/articles/2014/06/09/
exec-tech-brief-history-of-open-data.aspx [last accessed 05.
11.2019].
[55]
European Union (2013). Directive 2013/37/EU of the Euro-
pean Parliament and of the Council of 26 June 2013 amend-
ing Directive 2003/98/EC on the re-use of public sector in-
formation June. Available at: https://eur-lex.europa.eu/legal-
content/EN/TXT/PDF/?uri=CELEX:32013L0037&from=FR
[last accessed 13.05.2019].
[56]
Gray, J. (2014). Towards a Genealogy of Open Data (Septem-
ber 3, 2014). The paper was given at the General Conference
of the European Consortium for Political Research in
Glasgow, 3-6th September, 2014. Available at SSRN: https://
papers.ssrn.com/sol3/papers.cfm?abstract_id=2605828
[24.03.2020].
[57]
Morozov, E. (2013). The Meme Hustler: Tim O’Reilly’s
crazy talk. The Baffler, No. 22, April 2013. Available at:
https://thebaffler.com/salvos/the-meme-hustler [last accessed:
20.02.2020].
[58]
HM Government (2012). Open Data White Paper: Unleash-
ing the Potential. Presented to the Parliament by the Minister
of State for the Cabinet Office and Paymaster General by
command of Her Majesty. June 2012. Available at: https://
assets.publishing.service.gov.uk/government/uploads/system
/uploads/attachment_data/file/78946/CM8353_acc.pdf [last
accessed: 20.01.2010].
[59]
Australian Government (2015). Australian Government Pub-
lic Data Policy Statement. 7th December 2015. Available at:
https://www.pmc.gov.au/sites/default/files/publications/aust_
govt_public_data_policy_statement_1.pdf [last accessed: 20.
01.2020].
[60]
Government of Philippines (2014): Joint Memorandum Cir-
cular 2014-01 (Open Data Philippines). 22 January, 2014.
Available at: https://data.gov.ph/about/joint-memorandum-
circular-2014-01-open-data-philippines [last accessed 20.01.
2020].
[61]
OECD (2019). Open Data Government. Available at: https://
www.oecd.org/gov/digital-government/open-government-
data.htm [last accessed: 20.01.2020].
[62]
United Nations Conference on Trade and Development
(2019b). The Many Faces of Inequality. UNCTAD SDG Pulse.
Available at: https://sdgpulse.unctad.org/in-focus-inequality/
[last accessed: 20.01.2020].
[63]
Burri, M. (2017). The governance of data and data flows in
trade agreements: The pitfalls of legal adaptation. University
of California, Davies, Law Review, 51, 65–132.
[64]
Zuboff, S. (2019). The Age of Surveillance Capitalism. Public
Affairs, New York.
[65]
Ibarra, I.A., Goff, L., Jiménez Hernández, D., Lanier, J.,
Weyl, E.G. (2017). Should we treat data as labor? Moving be-
yond ‘Free’. American Economic Association Papers & Pro-
ceedings, 1(1), Forthcoming. December 27, 2017. Available
at SSRN: https://ssrn.com/abstract=3093683 [last accessed
20.01.2020].
[66]
Taylor, L. (2006). The ethics of big data as a public good:
which public? Whose good? The Royal Society. Philo-
sophical transactions. Series A, Mathematical, physical,
and engineering sciences. 2016 Dec 28; No. 374. Avail-
able at: http://doi.org/10.1098/rsta.2016.0126 [last accessed:
09.03.2020].
[67]
Stepanov, I. (2020). Introducing a property right over data in

S. MacFeely / In search of the data revolution 1093
the EU: The data producer’s right – an evaluation. Journal
of International Review of Law, Computers and Technology,
31(Issue 1), 65–86.
[68]
Deichmann, J., Heineke, K., Reinbacher, T., Wee, D. (2016).
Creating a successful Internet of Things data marketplace.
McKinsey Digital. October 2016. Available at: https://www.
mckinsey.com/business-functions/mckinsey-digital/our-
insights/creating-a-successful-internet-of-things-data-
marketplace [last accessed: 14.04.2020].
[69]
Prendergast, T. (2018). The next cold war is here, and it’s all
about data. Wired. 03 September, 2018. Available at: https://
www.wired.com/story/opinion-new-data-cold-war/ [last ac-
cessed: 20.01.2020].
[70]
Davis, W. (2017). How statistics lost their power – and why
we should fear what comes next. The Guardian, Thursday 19.
Available at: https://www.theguardian.com/politics/2017/jan/
19/crisis-of-statistics-big-data-democracy [last accessed: 01.
11.2019].
[71]
Citro, C. (2020). Federal Statistical Agencies Struggle to
Maintain Their Vital Role in the U.S. Data Infrastructure –
Why and Whence? LinkedIn.Com. April 30, 2020. Available
at: https://www.linkedin.com/pulse/federal-statistical-agenc
ies-struggle-maintain-vital-role-citro/?trackingId=5FNA2
FuHQ%2BGX27h1J%2BRj1A%3D%3D [last accessed 12.
06.2020].
[72]
Rugaber, C. (2020). How’s the economy? Fed increasingly
turns to private data. The Seattle Times. February 6, 2020.
Available at: https://www.seattletimes.com/business/hows-
the-economy-fed-increasingly-turns-to-private-data/ [last ac-
cessed: 20.02.2020].
[73]
Hodson, H. (2016). Baidu uses millions of users’ location data
to make predictions. New Scientist. 20 July, 2016. Available
at: https://www.newscientist.com/article/2098206-baidu-
uses-millions-of-users-location-data-to-make-predictions/
[last accessed: 20.04.2020].
[74]
Open Data Watch (2019). Open Data Inventory 2018/2019
Annual Report – A Progress Report on Open Data. Available
at: http://odin.opendatawatch.com/Report/annualReport [last
accessed: 16.12.2019].
[75]
United Nations (2020a). Data Strategy of the Secretary-
General for Action by Everyone, Everywhere: With Insight,
Impact and Integrity 2020-22. Available at: https://www.un.
org/en/content/datastrategy/images/pdf/UN_SG_Data-
Strategy.pdf [last accessed: 22.06.2020].
[76]
United Nations (2020b). System-wide Roadmap for Inno-
vating UN Data and Statistics. United Nations System Chief
Executives Board for Coordination. CEB/2020/1/Add.1. 14
May 2020. Available at: https://unstats.un.org/unsd/unsystem/
documents/CEB-Segment2-Roadmap.pdf [last accessed 12.
06.2020].
[77]
United Nations (2015). ‘Transforming our world: the
2030 Agenda for Sustainable Development’ Resolution
A/RES/70/1 adopted by the General Assembly on 25 Septem-
ber 2015, 21 October 2015. Available at: http://www.un.org/
ga/search/view_doc.asp?symbol=A/RES/70/1&Lang=E [last
accessed: 20.01.2020].
[78]
Boaz, A., Grayson, L., Levitt, R., Solesbury, W. (2018). Does
evidence-based policy work? Learning from the UK experi-
ence. Evidence & Policy A Journal of Research Debate and
Practice, 4(2), 233–253.
[79]
Kotz, S. (2005). Reflections on early history of official statis-
tics and a modest proposal for global coordination. Journal
of Official Statistics, 21(2), 139–144.
[80]
Bernstein, P.L. (1996). Against the Gods: The remarkable
story of risk. John Wiley & Sons, NY.
[81]
Rostow, W.W. (1959). The stages of economic growth. The
economic history review. Second Series, 12(1), 1–16.
[82]
Agar, J. (2001). Turing and the Universal Machine – The
making of the modern computer. Icon Books Ltd. London.
[83]
Keynes, J.M. (1940). How to Pay for the War: A radical plan
for the Chancellor of the Exchequer. Macmillan and Co. Ltd.
London. Available at: https://archive.org/details/in.ernet.dli.
2015.499597/page/n1 [last accessed: 16.01.2020].
[84]
Lepenies, P. (2016). The Power of a Single Number: A polit-
ical History of GDP. Columbia University Press, New York.
[85]
Fogel, R., Fogel. E.M., Guglielmo, M., Grotte, N. (2013). Po-
litical Arithmetic: Simon Kuznets and the Empirical Tradition
in Economics. University of Chicago Press, Chicago.
[86]
Lyons, G.M. (1969). The Uneasy Partnership – Social Science
and the Federal Government in the Twentieth Century. Russel
Sage Foundation. New York, N.Y.
[87]
Karabell, Z. (2014). The Leading Indicators – A short history
of the numbers that rule our world. Simon and Schuster, New
York.
[88]
Masood, E. (2016). The Great Invention: The story of GDP
and the making and unmaking of the modern world. Pegasus
Books Ltd, New York.
[89]
Patinkin, D. (1976). Keynes and econometrics: On the inter-
action between the macroeconomic revolutions of the inter-
war period. Econometrica, 44(6), 1091–1123.
[90] Muller, J.Z. (2018). The Tyranny of Metrics. Princeton Uni-
versity Press, New Jersey.
[91]
Desrosières, A. (2010). Words and Numbers: For a Sociology
of the Statistical Argument. In, Saetnan, A.R., Lomell, H.M.
and Hammer, S. (Eds.) (2010). The Mutual Construction of
Statistics and the Society. London: Taylor & Francis Group.
[92]
Desrosières, A. (2003). Managing the Economy In (Eds.)
(2003). Porter, T.M. and Ross, D. The Cambridge History
of Science – Vol 7: The Modern Social Sciences Cambridge
University Press, Cambridge.
[93]
Soros, G. (2013). Fallibility, reflexivity, and the human uncer-
tainty principle. Journal of Economic Metholodology, 20(4),
309–329. doi: 10.1080/1350178X.2013.859415.
[94]
Tenner, Edward. 2018. The Efficiency Paradox: What Big
Data Can’t Do. Alfred A. Knopf.
[95]
Lewis, M. (2003). Moneyball – the art of winning an unfair
game. W.W. Norton & Company, New York.
[96]
Lohr, S. (2015). Data-ism: Inside the Big Data Revolution,
Oneworld Publications, London.
[97]
Casalini, F., López González, J. (2019). Trade and Cross-
Border Data Flows. OECD Trade Policy Papers, No. 220,
OECD Publishing, Paris. Available at: doi: 10.1787/b2023a
47-en [last accessed: 13.11.2019].
[98]
Rogoff, K. (2019). The High Stakes of the Coming Digital
Currency War. Project Syndicate. Nov 11, 2019. Available
at: https://www.project-syndicate.org/commentary/global-
battle-for-digital-currency-supremacy-by-kenneth-rogoff-
2019-11 [last accessed: 18.11.2019].
[99]
United Nations Conference on Trade and Development
(2019a). Digital Economy Report 2019 – Value Creation and
Capture: Implications for Developing Countries. Available
at: https://unctad.org/en/PublicationsLibrary/der2019_en.pdf
[last accessed: 07.01.2020].
[100]
Postman, N. (1986). Amusing Ourselves to Death: Public
Discourse in the Age of Show Business. Penguin Books Ltd.
London.
[101]
Fukuyama, F. (2017). The Emergence of a Post Fact
World. Project Syndicate. Available at: https://www.project-

1094 S. MacFeely / In search of the data revolution
syndicate.org/onpoint/the-emergence-of-a-post-fact-world-
by-francis-fukuyama-2017-01 [last accessed 24.03.2020].
[102]
Stiglitz, J.E. (2019). The End of Neoliberalism and the Re-
birth of History. Project Syndicate, November 4, 2019. Avail-
able at: https://www.project-syndicate.org/commentary/end
-of-neoliberalism-unfettered-markets-fail-by-joseph-e-stigli
tz-201911?utm_source=Project+Syndicate+Newsletter&utm
_campaign=324044c49a-sunday_newsletter_10_11_2019&
utm_medium=email&utm_term=0_73bad5b7d8-324044c4
9a-105505185&mc_cid=324044c49a&mc_eid=8772e9ab0c
[last accessed: 13.11.2019].
[103]
Davis, E. (2017). Post Truth: Peak Bullshit and what we can
do about it. Abacus, London.
[104]
Le Carre, J. (2020). John le Carre on Brexit: ‘It’s break-
ing my heart’. The Guardian, Sat 1 Feb, 2020. Available
at: https://www.theguardian.com/books/2020/feb/01/john-le-
carre-breaking-heart-brexit [last accessed 03.02.2020].
[105]
O’Neill, C. (2016). Weapons of Math Destruction – How big
data increases inequality and threatens democracy. London:
Allen Lane.
[106]
Berners-Lee, T. (2014). Tim Berners-Lee on the Web at 25:
the past, present and future. Wired, August 23. Available
from: http://www.wired.co.uk/article/tim-berners-lee [last ac-
cessed: 24.03.2020].
[107]
United Nations (2019b). Report of the Friends of the Chair
group on the Fundamental Principles of Official Statistics
– Note by the Secretary-General. 51
st
session of the Sta-
tistical Commission, 3–6 March 2020, New York. Doc.
E/CN.3/2020/21. Available at: https://unstats.un.org/unsd/
statcom/51st-session/documents/2020-21-FPOS-E.pdf [last
accessed 19.03.2020].
[108]
Kirkpatrick, M. (2010). Facebook’s Zuckerberg Says the Age
of Privacy is Over. Available at: https://readwrite.com/2010/
01/09/facebooks_zuckerberg_says_the_age_of_privacy_is_
ov/ [last accessed: 24.03.2020].
[109]
Noyes, K. (2015). Scott McNealy on privacy: You still don’t
have any. PC World. Available at: https://www.pcworld.com/
article/2941052/scott-mcnealy-on-privacy-you-still-dont-
have-any.html [last accessed: 24.03.2020].
[110]
McAfee, J. (2015). Untitled posting on Facebook. Available
at: https://www.facebook.com/officialmcafee/posts/464114
187078100:0 [last accessed: 24.03.2020].
[111]
Pearson, E. (2013). Growing Up Digital. Presentation to the
OSS Statistics System Seminar Big Data and Statistics New
Zealand: A seminar for Statistics NZ staff, Wellington, New
Zealand. Available at: https://www.youtube.com/watch?v=
lRgEMSqcKXA [last accessed: 24.03.2020].
[112]
Payton, T., Claypoole, T. (2015). Privacy in the Age of Big
Data – Recognising the Threats Defending Your Rights and
Protecting Your Family. Lanham, MD: Rowman & Littlefield.
[113]
Taplin, J. (2017). Move Fast and Break things – How Face-
book, Google and Amazon cornered culture and undermined
democracy. New York: Little, Brown and Company.
[114]
Lyon, D. (2001). Surveillance after september 9/11. Socio-
logical Research Online, 6(3), 116–121.
[115]
Cozzens, T. (2020). 19 Countries Track Mobile Location to
Fight Covid-19. GPS World, march 26, 2020. Available at:
https://www.gpsworld.com/19-countries-track-mobile-loca
tions-to-fight-covid-19/ [last accessed 05.04.2020].
[116]
Jones, S. (2020). Liechtenstein rolls out radical Covid-19
bracelet programme. Financial Times, 16 April, 2020. Avail-
able at: https://www.ft.com/content/06b7e6f3-a725-4eda-91
53-e0af48040e30 [last accessed: 17.04.2020].
[117]
Sudhir, K. (2020). Can Big Data Fight a Pandemic? Yale
Insights. April 8, 2020. Available at: https://insights.som.yale.
edu/insights/can-big-data-fight-pandemic [last accessed: 23.
04.2020].
[118] Kelion, L. (2020). Coronavirus: Google reveals travel habits
during the pandemic. BBC News. 3 April, 2020. Available
at: https://www.bbc.com/news/technology-52138076 [last ac-
cessed: 04.04.2020].
[119]
McGrath, P. (2020). Google publishes location data on peo-
ple’s movements during pandemic. RTE News. Friday 3
April, 2020. Available at: https://www.rte.ie/news/coronavir
us/2020/0403/1128229-google-location-data/ [last accessed:
04.04.2020].
[120]
Doffman, Z. (2020). COVID-19 Phone Location Tracking:
Yes, It’s Happening Now – Here’s What You Should Know.
Forbes, March 27, 2020. Available at: https://www.forbes.
com/sites/zakdoffman/2020/03/27/covid-19-phone-location
-tracking-its-moving-fast-this-is-whats-happening-now/#3c
007dc811d3 [last accessed: 05.04.2020].
[121]
United Nations Statistics Division and World Bank (2020).
Monitoring the state of statistical operations under the
COVID-19 Pandemic: Highlights from a global COVID-19
survey of National Statistical Offices. Available at: https://
unstats.un.org/unsd/covid19-response/covid19-nso-survey-
report.pdf [last accessed: 11/06/2020].
[122]
Diaz Muñoz, P. (2020). The Importance of Figures. Société
Luxembourgeoise de Statistique. 21 April, 2020. Available
at: https://luxstat.lu/opinions/ [last accessed: 11/06/2020].
[123]
UNCTAD (2020). COVID-19: Some reflections from a sta-
tistical viewpoint. SDG Pulse – 2020. Available from: https://
sdgpulse.unctad.org/ [last accessed 01.07.2020].
[124]
Cheung, P. (2020). Impact of COVID-19 on Official Statistics
(2) – Is Official Statistics Non-Essential Service? Linkedin.
Available at: https://www.linkedin.com/pulse/impact-covid
-19-official-statistics-2-non-essential-paul-cheung/?tracking
Id=AWBHyvCKQqWOsV4f98T22Q%3D%3D [Last
accessed 19.06.2020].
[125]
Di Gennaro Splendore, L. (2020). COVID-19: Unprecedented
situation, unprecedented official statistics. Journal of Official
Statistics, 36(2), 229–235.
[126]
Radermacher, W. (2020). How Statistics Can Help – Going
Beyond COVID-19. Medium.com – Data & Policy. May 18,
2020. Available at: https://medium.com/data-policy/how-
statistics-can-help-going-beyond-covid-19-22bb2ce92440
[last accessed: 12.06.2020].
[127]
Waters, R. (2020). Microsoft throws weight behind open data
movement. The Financial Times, April 21, 2020. Available
at: https://www.ft.com/content/661b16ff-f86c-4dad-a557-
2e231501bf58 [last accessed: 21.04.2020].
[128]
Larsen, C. (2018). Who needs democracy when you have
data? MIT Technology Review. August 20, 2018. Available
at: https://www.technologyreview.com/2018/08/20/240293/
who-needs-democracy-when-you-have-data/ [last accessed:
15.04.2020].
[129]
MacFeely, S., Dunne, J. (2014). Joining up public service
information: The rationale for a national data infrastructure.
Administration, 61(4), 93–107.
[130]
Runde, D. (2017). The Data Revolution in Developing Coun-
tries has a long way to go. Forbes, February 25, 2017. Avail-
able at: https://www.forbes.com/sites/danielrunde/2017/02/
25/the-data-revolution-in-developing-countries-has-a-long-
way-to-go/#4c4b8a561bfc [last accessed: 01.11.2019].