Knowledge Pyramid Perspective of the Political Data Ecosystem: A Case

Study of Bhutan

Phub Namgay

1, 2 a

and Pema Wangdi

2 b

Department of Informatics and Media, Uppsala University, Sweden

School of Data Science and Data Analytics, Sherubtse College, Royal University of Bhutan, Bhutan

Keywords:

Political Data Ecosystem, Knowledge Management, Knowledge Pyramid, Democratic Electoral Process,

Bhutan.

Abstract:

This study examines the dynamics of data management and knowledge ﬂow in the political data ecosystem

through the lens of the Knowledge Pyramid. We used open-government electoral documents and polling data

for granular insights into how data, information, knowledge, and wisdom (DIKW) are managed in Bhutan’s

political data ecosystem. Bhutan’s electoral stakeholders and political parties manage and use DIKW of vary-

ing types, sizes, and complexities. In particular, political parties use information systems, websites, and social

media to manage data and construct and use knowledge for political activities. Democracy is still young and

gaining a foothold in Bhutan. The political parties do not employ complex data technologies and rich human

resources to manage DIKW emanating from the political data ecosystem. Thus, scope exists for electoral

stakeholders and political actors to explore and adopt effective and efﬁcient knowledge management infras-

tructures to deal with DIKW elements in the political arena, namely the complex dynamics of turning raw data

into higher elements of the Knowledge Pyramid. In addition to contributing to the knowledge management

literature through an in-depth account of the DIKW aspects in the political space, this paper demonstrated the

analytical and explanatory power of the Knowledge Pyramid for discourse on the political data ecosystem.

1 INTRODUCTION

The conventional political environment is gradually

transitioning to a data-driven political system. Polit-

ical actors use data technologies to harness insights

and knowledge from electorate data to boost cam-

paign activities and electoral success, such as using

big data in the US election (Ruppert et al., 2017).

Cambridge Analytica (Dommett et al., 2023; Micheli

et al., 2020; Ruppert et al., 2017) had a signiﬁcant

impact on the political stakeholders on exploiting the

value of data (Miller and Mork, 2013; Lee, 2017) gen-

erated in the political space. The political actors also

now recognise data as an intangible asset and use the

so-called political technologies (Ruppert et al., 2017)

to tap its value to drive campaign activities and en-

rich decision-making. For instance, during the na-

tional campaigns and electoral processes of countries

such as the United States (US), data management and

analytics were undertaken by a chief analytics ofﬁ-

https://orcid.org/0000-0001-6034-7274

https://orcid.org/0009-0000-8857-7354

cer and his hub of statisticians, engineers, and data

scientists of political parties. Likewise, much of the

data analysis tasks of the electoral activities are in-

creasingly delegated to and driven by complex algo-

rithmic processes (Mittelstadt et al., 2016). In the

present work, the term political was preﬁxed to the

deﬁnition of data ecosystem suggested by Oliveira

and L

oscio (2018) to deﬁne a working deﬁnition of

political data ecosystem—Organisational entities, in-

frastructural technologies, electoral activities, digital

data, and contextual norms in the democratic political

space, where stakeholders concerned, political actors,

general citizens, and democratic values are in an in-

terplay to create and capture value of sociopolitical

data.

The literature on the use of data in democratic pol-

itics, such as data-driven campaign activities and tar-

geting potential voters (Dommett et al., 2023; Ben-

nett and Lyon, 2019), is growing, given the signiﬁ-

cant effect of data analytics, knowledge management,

and social media on the electoral process. Dommett

et al. (2023) highlight that the prior studies on data

use in democratic politics focused on the US and es-

Namgay, P. and Wangdi, P.

Knowledge Pyramid Perspective of the Political Data Ecosystem: A Case Study of Bhutan.

DOI: 10.5220/0013014100003838

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 16th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2024) - Volume 3: KMIS, pages 85-96

ISBN: 978-989-758-716-0; ISSN: 2184-3228

tablished democracies. In a sense, such US-centric in-

formation affects our understanding of how electoral

stakeholders and political parties in different regions

of the world, namely emerging democracies, use data

and related technologies to construct, manage, and

use knowledge in their political space. The use of

data and analytics by political actors of countries

relatively new to democratic values and ethos—that

is, where democracy is recently established, to drive

electoral process and political campaigns has not re-

ceived much scholarly attention and is thus an under-

studied area. For instance, in Bhutan, a small country

in South Asia (Bates, 2009; Metz, 2014), democracy

was established only in 2008, and, on a political time-

line, democracy is only 16 years old. Consequently,

discourses on the use of data and analytics in such a

political space are limited.

It is timely for literature to be augmented with in-

sights into countries other than established democra-

cies (Dommett et al., 2023) on how they use data and

exploit knowledge to facilitate democratic electoral

processes. In the present study, we examine Bhutan’s

political data ecosystem from the perspective of the

Knowledge Pyramid (Jennex, 2017; Frick

e, 2009;

Tuomi, 1999) to determine how electoral stakeholders

and political parties deal with aspects of data, infor-

mation, knowledge, and wisdom (DIKW) for inform-

ing electoral processes and political campaigns. Data

increasingly impact the democratic election, hence

data-driven election (Bennett and Lyon, 2019) and

campaigning (Baldwin-Philippi, 2017). The ability to

turn data into knowledge by contextualising, structur-

ing, and giving meaning to data empowers political

actors to use the value of data for various electoral

or political activities. Thus, the Knowledge Pyramid

(Jennex, 2017; Frick

e, 2009) is an apt framework to

account for data and analytics practices in the politi-

cal data ecosystem of Bhutan. We address the follow-

ing research question: How do electoral stakeholders

and political parties use data, analytics, and knowl-

edge to inform decisions and actions in the political

data ecosystem?

The paper is structured as follows: Section

2 discusses the prior literature on data, politics,

and Knowledge Pyramid. Section 3 overviews the

methodology adopted in the study, and Section 4

presents the study ﬁndings. In Section 5, we discuss

the interpretation of the ﬁndings and connect it with

the extant literature, along with contributions to re-

search and practice, limitations of the study, and fu-

ture work. Finally, Section 6 concludes the paper.

2 RELATED LITERATURE

2.1 Data and Politics

The resources and capabilities for harnessing the po-

litical value of data are fundamental in political en-

deavours. Likewise, intelligent use and related ex-

pertise to analyse and turn data emanating from the

political space into meaningful information and valu-

able knowledge is critical to the success of campaign

activities, such as Barack Obama’s data-driven cam-

paign work in the 2008 and 2012 elections (Baldwin-

Philippi, 2017; Bennett and Lyon, 2019; Jin et al.,

2015). The phenomenon of data-driven politics is in

the mainstream of democratic elections, such as the

hiring of Cambridge Analytica by Donald Trump’s

team during the 2016 US election (Schippers, 2020;

Ruppert et al., 2017; Baldwin-Philippi, 2017). It

is a real-world case of transforming data into value

or extracting value from data (Micheli et al., 2020)

and subsequent practical use of the value in the po-

litical arena. Similarly, instances of using big data

during the US elections (Schippers, 2020; Ruppert

et al., 2017; Dommett et al., 2023) and Brexit refer-

endum (Ruppert et al., 2017) are examples of putting

data value to use. Constructs such as data-driven

campaigning, micro-targeting, voter proﬁling, email

analytics, and data politics (Dommett et al., 2023;

Baldwin-Philippi, 2017; Papakyriakopoulos et al.,

2018; Ruppert et al., 2017) are also added to the lit-

erature. The extant literature discusses how political

actors access and analyse data for insights into cam-

paigns and streamline political activities (Dommett

et al., 2023; Ruppert et al., 2017). Thus, in-depth

knowledge of the processes requires an understanding

of the complex nature of the political data ecosystem.

Furthermore, knowledge of the complexities of

the political data ecosystems and the use of political

technologies (Ruppert et al., 2017) to manage it is a

fundamental intellectual capital (Quintas et al., 1997).

The ubiquity of digitalisation and the intelligent use

of social media (Baldwin-Philippi, 2017), namely an-

alytics of social data (Olteanu et al., 2019), assist

political actors in competing with each other during

democratic elections by producing, analysing, and us-

ing DIKW within context and time frame. For exam-

ple, Baldwin-Philippi (2017) highlights the novel use

of data and analytics for political campaigns and how

corresponding analytical insights inform content pro-

duction and facilitate political advertising (Schippers,

2020; Ruppert et al., 2017; Bennett and Lyon, 2019).

In Bhutan’s context, the social media regulation of the

Election Commission of Bhutan (ECB) also under-

lines social media as a channel to disseminate infor-

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems

mation, communicate content (Election Commission

of Bhutan, 2018), and manage electoral compassing

activities (Dommett et al., 2023). Data in the political

space was also used to develop prediction models for

voters’ personality traits and the likelihood of turn-

ing out to vote (Bennett and Lyon, 2019). Dommett

et al. (2023) suggest using data to formulate cam-

paign strategies and evaluate the effectiveness of a

campaign retrospectively.

However, despite growing political discourses on

data-driven democratic election (Bennett and Lyon,

2019), searches on citation databases, such as Sco-

pus, IEEE Xplore, and ACM Digital Library, with

the search query — ((“Knowledge Pyramid” OR

“Knowledge hierarchy” OR “Information Pyramid”

OR “DIKW hierarchy” OR “Information hierarchy”)

AND (“Political data” OR “politics data” OR “Polit-

ical space” OR “Political data ecosystem*”))— did

not yield useful results

. We inferred it as a gap in

knowledge on using the perspective of the Knowledge

Pyramid (Jennex, 2017) to examine the management

and use of DIKW aspects among electoral stakehold-

ers and political parties for informing administrative

decision-making and shaping campaign activities.

2.2 Knowledge Pyramid

Knowledge Pyramid is a widely used model to illus-

trate the logical relationship between DIKW based on

meaning, context, and value (Tuomi, 1999; Frick

2009; Jennex, 2017). The processes to transform data

into higher elements of the Knowledge Pyramid re-

ﬂect creating and building value, thus data value chain

(Miller and Mork, 2013). Each step higher in the

Knowledge Pyramid answers more questions about

a phenomenon captured in the data. According to

the Knowledge Pyramid, data is a raw or unorgan-

ised discrete collection of facts (Zins, 2007; Frick

2009), such as polling day data of democratic elec-

tion. The next level is information, which is con-

textual, cleansed, processed, and analysis-ready data

(Bellinger et al., 2004; Frick

e, 2009). In the prior

literature (Khatri and Brown, 2010), data and infor-

mation are often used synonymously and do not dif-

ferentiate between the two, and the same applies to

information and knowledge (Wang and Noe, 2010).

Zins (2007) argues that such perception is problem-

atic for articulating data, information, and knowledge.

Similarly, scholars also question whether information

is data or knowledge (Quintas et al., 1997). Knowl-

edge is the consolidation and application of disparate

pieces of information, with underlying meanings and

We adapted the search query across the citation

databases, but the search keyword remained the same.

schemas, to perform tasks and achieve goals. When

one uses data-driven information and knowledge to

solve problems, it is an instance of the possession and

use of wisdom (Frick

e, 2009), which is the apex of the

Knowledge Pyramid (Rowley, 2007). In the knowl-

edge hierarchy discourse, wisdom is succinctly de-

ﬁned as knowledge applied in action, with the caveat

that one appreciates the fallible nature of knowledge

(Frick

e, 2009).

The underlying epistemologies and philosophies

of the elements of the Knowledge Pyramid have been

criticised and critiqued by scholars (Frick

e, 2009;

Tuomi, 1999; Jennex, 2017). However, Jennex (2017)

suggests embracing the debate but does not encour-

age it. Scholars critique the inherent weakness of the

traditional Knowledge Pyramid and propose a revised

version of it (Jennex, 2017). The revised Knowledge

Pyramid (Jennex, 2017) incorporates aspects of the

Internet of Things, data analytics, and big data (Jin

et al., 2015). Some critics of the Knowledge Pyra-

mid argue that data is not the building block of the

higher elements of the Knowledge Pyramid (Tuomi,

1999; Jennex, 2017). They assert that an individ-

ual’s prior wisdom and knowledge play an essential

role in understanding the world, which drives an in-

dividual to gather information to collect data about

a phenomenon of interest. Their perspective implies

that elements of the Knowledge Pyramid are intri-

cately linked, and one could visualise hierarchy ﬂow-

ing downward rather than upward (inverted) (Tuomi,

1999) or in both directions within the context of the

natural or real-world (Jennex, 2017). Some argu-

ments for needing a revised Knowledge Pyramid are

supported by mathematical facts and social reality

(Tuomi, 1999; Frick

e, 2009). Given the dataﬁed re-

ality we are experiencing, individuals, organisations,

and societies continually gather, process, and analyse

DIKW by leveraging their insights and sense-making

capabilities, further complemented by infrastructural

technologies. It is a cyclical process in that one’s prior

wisdom and knowledge shape gathering information

on what aspects of data to collect to accomplish a task.

Although not expressly mentioned, we consider the

revised Knowledge Pyramid in the present work.

3 METHODOLOGY

3.1 Study Setting

Bhutan, a country known for the concept of Gross Na-

tional Happiness (Bates, 2009; Metz, 2014), is rela-

tively new to democracy. The matrix of democratic

values and ethos in Bhutan started only in 2008 (Elec-

Knowledge Pyramid Perspective of the Political Data Ecosystem: A Case Study of Bhutan

tion Commission of Bhutan, 2008). Bhutan is a small

country (38, 394 (14, 824)km

(mi

)) located in South

Asia with roughly about 750,000 population. Re-

garding governance and politics, Bhutan has a demo-

cratic, constitutional monarchy system with a bicam-

eral parliament. The king (Druk Gyalpo) is the head

of the state, and the Prime Minister is the head of the

government. In the bicameral parliament, the upper

house, the National Council, consists of 25 elected

members (20 non-partisan elected plus ﬁve members

appointed by the Druk Gyalpo). The lower house,

the National Assembly, consists of 47 elected mem-

bers from the ruling and opposition parties. The

use of democratic principles to elect candidates and

form a government is a recent phenomenon among

Bhutanese citizens. Democracy empowers Bhutanese

society to elect 47 candidates representing their con-

stituency at the National Assembly. It also allows the

voices and aspirations of the citizens to be heard by

the highest legislative and executive bodies. For in-

stance, the fourth democratic election concluded in

January 2024, and Bhutan has a new democratically

elected government.

The electoral stakeholders such as the ECB (Elec-

tion Commission of Bhutan, 2024) have increasingly

used technologies to streamline democratic electoral

processes. In doing so, it facilitates the management

of coherent and reliable electoral data for free demo-

cratic elections. Data from numerous sources, such

as documents of political parties, demographic data

of candidates, primary-round polling data (vote for

political parties of one’s choice), and general elec-

tion polling data (vote for the candidates ﬁelded by

the two political parties that have made it through

the primary round) is generated at a relatively sig-

niﬁcant scale. Meanwhile, Bhutanese political par-

ties also use data to inform campaign activities and

enrich decision-making, which resonates with Ben-

nett and Lyon (2019) that modern democratic cam-

paigns use data. To illustrate, the political activi-

ties, manifesto documents and party websites sug-

gest that they perform document analysis and de-

scriptive data analytics, with complex analytics del-

egated to external tools such as Facebook ad analyt-

ics (Baldwin-Philippi, 2017) during the parliamentary

election, which is discussed further in the later section

of this paper. It is worth exploring how Bhutan’s elec-

toral stakeholders and political parties manage and

use DIKW aspects through the lens of the Knowledge

Pyramid (Jennex, 2017; Frick

e, 2009; Tuomi, 1999).

3.2 Method

We used qualitative research design to analyse

archival documents and empirical electoral data, al-

beit secondary data, to provide an account of data,

analytics, and knowledge practices in Bhutan’s po-

litical data ecosystem through the lens of Knowl-

edge Pyramid (Jennex, 2017). It is worth mention-

ing that we did not conduct interviews or surveys in

the present work. The snowballing technique gath-

ered open-government electoral documents and sup-

port materials from the ECB (Election Commission

of Bhutan, 2024) and related governmental agencies.

These documents were made available on their web-

site. The Independent Veriﬁcation Committee (Inde-

pendent Evaluation Committee (IEC), as per section

5.4 of Rules of Election Conduct, 2022) veriﬁes these

documents, looking at factual accounts of political

ideologies and ﬁltering out unfeasible promises. We

also consulted the political parties’ 4th Parliamentary

National Assembly election manifestos in Table 2 for

insights into the use of DIKW and related technolo-

gies and their implication on political activities. Like-

wise, to complement the inductive qualitative inquiry

(Thomas, 2006) with quantitative insights and under-

standings, we manually scrapped the aggregated elec-

tion result data from the ECB website

2 3

. Indeed, dif-

ferent data sources were central to triangulating data

and enriching the description of the democratic elec-

toral process in Bhutan with statistical facts and ﬁg-

ures. We also gathered public social media data for

information on the proﬁle of Bhutanese political par-

ties across various social media platforms, such as the

count of followers and type of posts during the up-

coming democratic election.

3.3 Data Analysis

We used a general inductive approach (Thomas,

2006) to analyse the open-access textual documents

and numeric polling data to provide an account of as-

pects of DIKW in the political data ecosystem and

to reﬂect on the data-driven democratic electoral pro-

cess of electing the 4th National Assembly member

in Bhutan. The unit of analysis (Yin, 2014) is the po-

litical data ecosystem of Bhutan. Similarly, the unit

Results of the 4th National Assembly Elections, 2023-

2024 (primary round)— https://www.ecb.bt/declaration-o

f-results-of-the-4th-national-assembly-elections-2023-2

024/

Results of the 4th National Assembly Elections, 2023-

2024 (general election)— https://www.ecb.bt/declaration-o

f-results-of-the-4th-national-assembly-elections-2023-2

024-general-election/

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems

Table 1: Documents consulted for insights into DIKW aspects in the political data ecosystem of Bhutan.

Documents Source Description

Election Act of the Kingdom

of Bhutan, 2008

ECB

Serves as the overall guiding principle for all

election-related activities, including constituency

information, nomination processes and

disqualiﬁcations, memberships and campaigning,

advertising rules, and dispute settlements

Information, Communications

and Media Act of Bhutan 2018

BICMA

Mandates the Bhutan Information Communication and

Media Authority (BICMA) to maintain the required

standards and prudent use of ICT and media facilities

and mechanisms for data sharing and protection

ECB Social Media Rules and

Regulations of the Kingdom

of Bhutan, 2018

ECB

Documents rules and regulations for political advertising,

the use of social media by the ECB and Election Ofﬁcials,

mentions the period of no campaign (48 hours), and

provisions of reporting violations

Election Advertising Regulations

of the Kingdom of Bhutan, 2018

ECB

Lists out election advertising particulars, online election

advertising, the use of posters and banners, and highlights

the use of the Internet or social media only for the election

advertising

Electronic Voting Machine (EVM)

Rules and Regulations of the

Kingdom of Bhutan, 2018

ECB

Serves as a manual for the proper use of electronic voting

machines to train polling personnel and voter education,

and also outlines the procedures for safeguarding

machines and handing-taking over election materials

Strategy for the Implementation of

the Provisions Related to Election

Advertising, 2021

ECB

Provides more clarity on the implementation of the election

advertising provisions in the spirit of fairness and equality

and also speciﬁes the extent of time and ceiling amount

allocated to the political parties for elections

Media Coverage of Elections Rules

and Regulations of the

Kingdom of Bhutan, 2021

ECB

Illustrates provisions for fair and equal access to paid

election advertising and equal allocation of broadcasting

time and space by both ECB and a political party,

including restrictions imposed on each

Rules and Regulations on Content BICMA

Outlines the content speciﬁcs to empower content

providers and ensure accountability and encourage

creativity and innovation

Rules and Regulations on ICT

Facilities and Services in Bhutan

BICMA

Informs the general public of ICT facilities and services

available and protocols/eligibility to obtain licenses for

providing such facilities

e-Governance Policy for the

Royal Government of Bhutan

GovTech

Agency

Aims to leverage existing and emerging information

technology for increasing competitiveness, enhancing

productivity, and improving service delivery through

online services and sustainable governance

of observation is the different information technolo-

gies and related techniques used by various electoral

stakeholders and political actors to manage and use

DIKW for the democratic electoral process. Since a

democratic election is a high-stakes task, aspects of

the electoral processes and associated data in Bhutan

are well documented in the election-related acts, poli-

cies, and guidelines (Table 1). Thus, data sources for

this study are analysis-ready for in-depth insights into

the intricate dynamics of humans, technologies, and

DIKW in the political data ecosystem. All the docu-

ments were closely read and analysed (Thomas, 2006)

to explicate the phenomenon (Yin, 2014)—that is, the

use of DIKW aspects among electoral stakeholders

and political actors to facilitate the democratic elec-

toral system in Bhutan. In doing so, the current work

upholds the rigour and trustworthiness of a qualita-

tive study (Yin, 2014; Nowell et al., 2017). Through

the perspective of the Knowledge Pyramid (Jennex,

2017; Tuomi, 1999), we zoom in on how data ema-

nating from the political space is managed, analysed,

and turned into information, knowledge, and wisdom

Knowledge Pyramid Perspective of the Political Data Ecosystem: A Case Study of Bhutan

among electoral stakeholders and political actors. We

performed qualitative coding (only open code) of the

data set for information on action patterns and DIKW

ﬂows in Bhutan’s political data ecosystem. In order

to consolidate our analytic output, a ﬂow chart was

used to model the complex DIKW-driven interplay of

human, technological, and organisational actors in the

democratic electoral process in Bhutan.

4 FINDINGS

4.1 Political Data Ecosystem of Bhutan

The open-access documents and aggregated polling

data (primary and general elections) from the ECB

are infused with valuable information to examine the

data-driven democratic electoral process in Bhutan.

In fact, the activities of electoral stakeholders and po-

litical parties are well documented by capturing the

subtleties of the electoral process, from the announce-

ment of the upcoming National Assembly election

to declaring primary and general election results and

conﬁrming the ruling and opposition parties. This

end-to-end democratic electoral process is informed

by various data such as party information, candidate

proﬁles, polling data, and election results. Table 2

provides an overview of the registered political par-

ties, the 2023-2024 National Assembly election re-

sults, and the use of various information technolo-

gies in the political data ecosystem. The collation,

analysis, and interpretation of electoral data in the

third and fourth columns and last row of the table

are performed by stakeholders at ECB. The gathering

of the polling data and corresponding rigorous ana-

lytics to transform data into higher elements of the

Knowledge Pyramid, namely information (such as as-

pects of the respective constituency) and knowledge

(such as electoral trends and patterns), of the 4th Na-

tional Assembly general election can be viewed else-

where (https://www.ecb.bt/geresults2024/). Re-

garding the wisdom aspects of the Knowledge Pyra-

mid, upon rigorous analysis of electoral results and

methodical assessment of the situation, the electoral

stakeholders in Bhutan work on tasks of ensuring that

the Bhutanese society has a new government and an

opposition party, as illustrated in Figure 1.

Social media platforms have a signiﬁcant impact

on the political data ecosystem (Bennett and Lyon,

2019; Baldwin-Philippi, 2017). The political par-

ties in Bhutan recognise the opportunities afforded

by social media platforms, albeit wary of the down-

side of such tools on democracy. Bhutanese political

parties use social media, such as Facebook, What-

sApp, and Messenger, extensively for crowdsourc-

ing information on political activities, collecting con-

structive feedback on pledges, and gathering knowl-

edge of community needs. For instance, during the

4th National Assembly election, the BTP conducted

election-related ad campaigns on Facebook (Baldwin-

Philippi, 2017). Meta Ad Library automatically cal-

culates values for metrics such as platform (Facebook,

Instagram, and Audience Network), categories (social

issues, elections or politics), ad audience size, amount

spent, impressions, gender, and location. The merit of

such tools is that ad delivery information is presented

well in a digestible format (in summaries, charts, and

graphs), such as the ‘Impressions’ of the ofﬁcial BTP

Facebook page was 700K–800K. The post mainly ap-

peared in social media posts and feeds of people in

Thimphu dzongkhag

(37%) and mostly covered peo-

ple aged 18-44

—that is, the digital ads of BTP was

displayed on Facebook users in the aforementioned

region and demography (see microtargeted Facebook

ads in Baldwin-Philippi, 2017, p. 629). Similarly, the

political parties also use Kuensel’s

Facebook page

for digital ads during the democratic election.

Additionally, from our lived experiences of demo-

cratic election reality in Bhutan, political parties

use insights assembled from social media for micro-

targeting (Bennett and Lyon, 2019), albeit rudimen-

tarily, through personalised text messages for poten-

tial electorates, audio chats in social media groups,

and emails to potential individual electors (Papakyri-

akopoulos et al., 2018). All these activities to lever-

age social media for political endeavours are guided

by the ECB’s in-house social media policy (Election

Commission of Bhutan, 2018). The policy also stip-

ulates that if the security level of social media plat-

forms and access privilege of the desired information

is uncertain, persons with a stake in the electoral pro-

cess should not share conﬁdential information on such

tools. Considering literacy aspects of potential elec-

tors and other factors such as geography, Bhutanese

political parties generally prefer Facebook posts on

political activities, expressly or implicitly gathering

information and knowledge for political ends. They

or their party workers are also active across social

media groups for insights into and synthesis of the

political debates and discussions among group mem-

bers on needs and demands in the community. In a

Dzongkhag means a district in Dzongkha language

BTP conducted an ad campaign on Facebook during

the 4th National Assembly Election in Bhutan—https://ww

w.facebook.com/ads/library/?active status=all&ad type=al

l&country=ALL&view all page id=101216219504499&

search type=page&media type=all

Kuensel is the national newspaper of Bhutan

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems

Table 2: Overview of the political parties that contested in the 4th National Assembly election in Bhutan.

Political Party Political Slogan

Primary

Election*

General

Election

Remark

(Use of

data analytics,

social platforms,

and other

technologies)

Bhutan Tendrel

Party (BTP)

(Founded:

November 2022)

Your Voice, Your Hope

61,331

(19.58%)

Votes—

147,123 (45.02%)

17 (Opposition)

Conducted ad campaigns

related to social issues,

elections, or politics on

Facebook; 31K followers

on Facebook; 18

followers on X (formally

Twitter); and 132

followers on Instagram

Druk Nyamrup

Tshogpa (DNT)

(Founded:

20 January 2013)

Putting Nation First

41,106

(13.12%)

41K followers on

Facebook, 5.8K+

followers on X, and 52

followers on Instagram

Druk Phuensum

Tshogpa (DPT)

(Founded:

25 July 2007)

Economic Prosperity

and Social Well-being;

Development with

Equity and Justice

46,694

(14.91%)

8.9K followers on

Facebook and 1.6K+

followers on X

Druk Thuendrel

Tshogpa (DTT)

(Founded:

2 May 2022)

Sunomics: Buddhist

Capitalism with the

spirit of GNH

30,814

(9.83%)

34K followers on

Facebook, no X account,

and 459 followers on

Instagram

People’s Democratic

Party (PDP)

(Founded:

24 March 2007)

For a Better Drukyul.

The Promise we will

Deliver.

133,217

(42.53%)

Votes—179,652

(54.98%)

30 (Ruling party)

Maintains an active

political party website

—https://pdp.bt,

61K followers on

Facebook, 7.3K+

followers on X, and

117 followers on

Instagram

*: Eligible Voters—497,058; Total Votes—313,162; Voter Turnout—63% (EVM: 195,719; PB: 117,443)

: Eligible Voters— 498,135; Total Votes— 326,775; Voter Turnout— 65.6% (EVM—218,273; PB—

108,502)

sense, Bhutanese political parties use social media as

an information-eliciting and knowledge-building plat-

form.

4.2 Data Practices Among Political

Stakeholders

Bhutan’s electoral stakeholders and political parties

manage data and associated knowledge for various

political activities. Figure 1 illustrates a ﬂow chart

of the recurrent action patterns in the electoral pro-

cess of political parties ﬁling to contest in the up-

coming democratic election at the ECB through the

primary and general rounds of casting votes to the

declaration of results to the formation of ruling and

opposition parties. The data, that is, votes from

EVM and postal ballots, is consolidated by follow-

ing schema and parameters developed by ECB (see

Fig 5.1 in Election Commission of Bhutan, 2013, p.

16). ECB gathers massive amounts of data during

the election period, such as party registration, elec-

torate registration, and in-country or overseas postal

ballot voters. They also maintain an information sys-

tem (https://dramig.ecb.bt/membership/check) for

registering political party members. The party mem-

bers can use the system to check their membership

status. If it is valid, the system displays information

Knowledge Pyramid Perspective of the Political Data Ecosystem: A Case Study of Bhutan

about their political afﬁliation. ECB also has a web-

based system (https://berms.ecb.bt/enrollment/check)

whereby electorates can check their enrolment sta-

tus to vote in the upcoming democratic election. In

Bhutan, voting is not compulsory, and the electoral

system does not permit proxy voting. If an electorate

wants to cast a vote via postal ballot (in-country or

overseas), the system records the details of elector in-

formation and allows changing postal addresses based

on the voters’ choice.

The political actors of Bhutan now recognise the

importance of tapping the value of data generated in

the political data ecosystem. It is evident from their

campaign activities and manifesto document that they

perform analytics of data on socioeconomic develop-

ments in Bhutan, such as descriptive statistics and ﬁg-

ures, to inform campaign activities, drive decision-

making, and develop party pledges. Such activities

are consistent with Bennett and Lyon (2019) that all

modern democratic election campaigns use data. The

website of some political parties has a feature that al-

lows potential electors to join as party members. For

example, the PDP’s website has a ‘Join Us’ feature.

It asks for CID, name, contact no, gewog (county),

and dzongkhag. After entering this information, the

system automatically fetches the name of the party

candidate in the constituency of an elector. In ef-

fect, such activities are instances of using information

technology, such as voter relationship management

system (Bennett and Lyon, 2019), and DIKW aspects

for informational canvassing and steering political

campaigns through targeting and testing (Baldwin-

Philippi, 2017). Similarly, the DTT’s website uses

cookies to track web trafﬁc and click streams, which

could be used to analyse the digital footprint of po-

tential electors. The actions and dealings with data

suggest that the political parties expressly or implic-

itly elevate data gathered from democratic politics to

the higher elements of the Knowledge Pyramid (Jen-

nex, 2017). However, since democracy in Bhutan is

relatively embryonic, the analytical maturity of polit-

ical parties is rather low in that no political parties use

complex data analytics technologies to analyse polit-

ical data for building predictive models or scoring to

target voters during the democratic election (Bennett

and Lyon, 2019).

4.3 DIKW in the Political Data

Ecosystem

The data-driven democratic electoral activities of the

political actors indicate how they deal with the DIKW

elements of the Knowledge Pyramid (Jennex, 2017).

The strategy to manage knowledge in the data ana-

lytics processes is a critical success factor (Jennex,

2017) as it has implications on exploiting DIKW for

campaign activities and decision-making, such as in-

sights into electorate voting (EVM and postal ballot)

pattern and election trend analysis. Fig. 1 illustrates

the interplay and dynamics of various sociotechni-

cal actors during the democratic electoral process in

Bhutan. It also embodies a complex exchange of

DIKW among numerous stakeholders, such as the

ECB, political parties, and the Ofﬁce of the Returning

Ofﬁcer. Data such as manifesto documents for reg-

istration to contest in the election, primary and gen-

eral round polling day data, and comprehensive data

of the electorate are generated, managed, communi-

cated, and used during the democratic election period.

The political parties also bootstrap available resources

and technologies to gather, manage, and turn data into

higher elements of the Knowledge Pyramid (see Ta-

ble 2). For instance, they use websites, social me-

dia, and other information systems to manage and op-

erationalise DIKW in Bhutan’s political data ecosys-

tem. Their social media posts of the political activities

during the upcoming election, such as photos from

community zomdu (translates to inclusive decision-

making) and updates on personal meetings with party

workers in the 47 constituencies, also illustrate that

the predominant method that political parties use for

gathering aspects of DIKW is in-person campaign-

ing or door-to-door canvassing (Dommett et al., 2023;

Baldwin-Philippi, 2017; Bennett and Lyon, 2019).

Furthermore, polling day is one instance where

data must be as accurate as possible; otherwise, it

would endanger the democratic electoral process. The

demographic information of electorates is carefully

vetted through the voter identity card at the 809

polling stations. ECB follows ﬁrst-past-the-post vot-

ing, where the voter votes for a single candidate, and

the candidate with the maximum votes wins the elec-

tion. On the polling day of the primary and general

elections (Fig. 1), data from 47 constituencies and

809 polling stations across the country are transferred

to the Department of Election of ECB for tallying

the votes. The stakeholders in the department collate,

analyse, and add meaning and context to the polling

day data, which is an instance of dealing with ele-

ments of the Knowledge Pyramid (Jennex, 2017). For

example, analysis of the polling data from the primary

round is used to declare the winning and runners-up

parties where they are eligible to participate in the

general election to form the government, as illustrated

in Fig. 1. Overall, the ﬂowchart captures the logic

of how democratic election decisions are made in the

primary and general rounds to form the ruling and op-

position parties. These recurrent action patterns in the

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems

Figure 1: Dynamics and ﬂow of DIKW in Bhutan’s democratic electoral process.

ﬂow chart every ﬁve years require electoral stakehold-

ers and political parties to wrestle with DIKW tire-

lessly in the political data ecosystem of Bhutan.

5 DISCUSSION

5.1 DIKW-Driven Political Data

Ecosystem

Considering the considerable inﬂuence of and dis-

courses on data and analytics during the democratic

election (Papakyriakopoulos et al., 2018; Baldwin-

Philippi, 2017; Bennett and Lyon, 2019), electoral

stakeholders and political actors must determine the

actual and potential value of DIKW and its implica-

tions on political activities. Moreover, understanding

DIKW (Jennex, 2017; Tuomi, 1999) would avoid du-

plicative efforts and ease the extraction of the value of

data (Micheli et al., 2020) emanating from the politi-

cal data ecosystem. It is also timely for the nascent

political data ecosystem in emerging democracies,

such as Bhutan, to determine barriers and opportu-

nities to share and use political data for facilitating

democratic election and avoid issues of information

asymmetry (Mittelstadt et al., 2016)—that is, unequal

distribution of or access to information. Indeed, doing

so would enrich the open data attitude and culture in

the political space, especially exploring possibilities

for DIKW sharing and reuse among political actors.

Furthermore, individuals entrusted with data analytics

among political parties should conceive data-driven

politics-related activities in terms of DIKW. Political

parties with resource constraints need more technical

know-how to perform complex data analytics and use

analytical results to manage campaign activities and

target electors during democratic election campaigns

(Bennett and Lyon, 2019; Baldwin-Philippi, 2017).

In Bhutan, given the smallness of the constituencies,

geographic-based electoral targeting is a predominant

method among political parties for voting intelligence

of potential electors. They also gather DIKW of po-

tential wins for their candidates in certain strongholds

Knowledge Pyramid Perspective of the Political Data Ecosystem: A Case Study of Bhutan

through their party workers as information conduits,

albeit surprises are not uncommon during elections.

The political data ecosystem needs a systematic

approach to managing the DIKW elements of the

Knowledge Pyramid (Jennex, 2017; Frick

e, 2009).

For instance, a contextual framework to manage

DIKW (Jennex, 2017) is essential to facilitate the

frictionless ﬂow of data within political parties for

decision-making and campaign activities (see Nissen,

2002 on the relational link between data, information,

and knowledge). Likewise, electoral stakeholders and

political parties must know what DIKW is needed and

where it is available. Hence, providing support sys-

tems, such as developing capabilities for acquiring

data from different sources and subsequent transfor-

mation into the higher levels of the Knowledge Pyra-

mid (Jennex, 2017), is also crucial to tap the value of

data emanating from various sources. Moreover, a co-

herent and reliable DIKW will positively affect elec-

toral activities, such as DIKW-informed, analytics-

driven, and cost-effective campaign activities (see Ta-

ble 6 in Dommett et al., 2023). Additionally, mecha-

nisms should be in place to protect DIKW from imi-

tation by others (Gelhaar and Otto, 2020), which calls

for proper inventorying of DIKW in the political en-

vironment (see Miller and Mork, 2013 on data in-

ventory) to avoid the risk of accumulating ineffective

DIKW and mistargeting electors. Regarding voter

analytics and political micro-targeting (Bennett and

Lyon, 2019), clear legislation is fundamental about

the ethics and legitimacy of using electorate data and

related databases for data analytics to shape politi-

cal campaigns and inform decision-making (Dommett

et al., 2023; Baldwin-Philippi, 2017). In Bhutan’s

context, some safeguards, such as acts, policies, and

guidelines, as summarised in Table 1, are in place to

deal with the misuse of so-called political technolo-

gies (Ruppert et al., 2017) during the democratic elec-

tion.

5.2 Knowledge Management in the

Political Data Ecosystem

The complex interplay of various individuals, organ-

isations, and institutions is fundamental for foster-

ing the growth of an ecosystem (Gelhaar and Otto,

2020) and creating and sharing knowledge. Thus, it is

crucial to ensure that tacit (experiences and insights)

and explicit knowledge (knowledge formalised and

codiﬁed in electronic systems) (Choo, 1996; Non-

aka, 2007; Nissen, 2002) is used effectively and ef-

ﬁciently in the political data ecosystem. For example,

in Bhutan, the ECB gathers political documents (such

as manifestos, charters, and party membership) and

polling data and accordingly shares them on its web-

site. Political parties could explore opportunities to

encourage knowledge sharing and use through collab-

oration with other parties. It would provide an avenue

to exchange services whereby political actors share

and exchange skills, capabilities, and knowledge that

beneﬁt each other. Some political data have a short

shelf life in that data value decays over time (Nam-

gay et al., 2023; Lee, 2017). Hence, political actors

must also deal with boundary paradox (Quintas et al.,

1997), where boundaries should be open for DIKW

ﬂow among the relevant stakeholders, but at the same

time, political parties also have to protect and nurture

their DIKW and intellectual capital. Quintas et al.

(1997) argue,‘It is upon the dynamic preservation of

the latter [protect and nurture knowledge base and in-

tellectual capital] that survival depends.’

The issue in the political data ecosystem is how

to manage knowledge using effective and efﬁcient

technologies, which tools and techniques to use for

managing DIKW, and what DIKW related to politics

is needed for decision-making and campaign activi-

ties. Moreover, knowledge generation and applica-

tion is a cyclical process, which alludes to the critique

of knowledge hierarchy (Frick

e, 2009; Tuomi, 1999;

Jennex, 2017). It costs resources, effort, and time to

analyse data and turn it into knowledge. Therefore,

it is imperative to have a practical plan or strategy

to deal with and work on DIKW Jennex (2017), es-

pecially knowledge in the political data ecosystem,

such as identifying metrics for managing and gaug-

ing DIKW. If there is shareable data by de-identifying

certain attributes, such as personally identiﬁable in-

formation, stakeholders such as ECB and other gov-

ernmental organisations should share it with political

parties. Doing so prevents the political parties from

performing needless tasks such as setting up infras-

tructures to collect and analyse data for insights and

knowledge. It is also imperative to incentivise po-

litical parties to use data analytics and foster a cul-

ture of sharing and using knowledge (Wang and Noe,

2010), albeit for sociopolitical good. Otherwise, in

emerging democracies with limited resources, mobil-

ising data and analytics at scale for political ends is

resource intensive, and inaccessibility to rich DIKW

risks data-driven democratic politics favouring estab-

lished political parties (Bennett and Lyon, 2019) with

means to use complex political technologies and em-

ploy human resources to manage aspects of DIKW in

the political data ecosystem.

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems

5.3 Contributions, Limitations, and

Future Work

This study contributes to the literature on knowledge

management via insights into DIKW aspects (Jennex,

2017; Tuomi, 1999) in the political data ecosystem of

emerging democracies that use data and analytics for

political activities. We answer the call by Dommett

et al. (2023) for exposition on data practices in demo-

cratic elections in different countries and contexts, es-

pecially ‘less US-centric studies’. The present work

also contributes to the discourse on the evolution of

the nascent body of knowledge on the political data

ecosystem. It is also timely for researchers inter-

ested in data-driven politics (Bennett and Lyon, 2019;

Baldwin-Philippi, 2017) to rethink how the Knowl-

edge Pyramid elements (Jennex, 2017) are managed

and used in the political space. The study likewise

augments the theory and praxis of the Knowledge

Pyramid in research. Regarding implications for data

stakeholders as well as policymakers, this study pro-

vides a platform to reﬂect on their current data prac-

tices from the frame of the Knowledge Pyramid (Jen-

nex, 2017; Tuomi, 1999; Frick

e, 2009) to unlock the

actual and potential value of data emanating from

the political data ecosystem. The in-depth descrip-

tion of facets of and relational links between DIKW

also sheds light on how political actors can administer

data-driven electoral processes and manage knowl-

edge for decision-making and campaign activities.

We acknowledge that this study has some limi-

tations. The current research is based on a country

where democracy is not even two decades old. Like-

wise, the present work only used open-access gov-

ernmental documents and aggregated polling data to

provide an account of the political data ecosystem

from the perspective of the Knowledge Pyramid (Jen-

nex, 2017). The subjectivity of the DIKW elements

of the Knowledge Pyramid (Frick

e, 2009) could also

affect the interpretations of the ﬁndings and discus-

sions in the present work. An area worth exploring is

using empirical data such as interviews and surveys

to examine data-driven democratic political activities

via the lens of the Knowledge Pyramid. Another re-

search direction is algorithmic processes that under-

lie the data analytics systems in political technolo-

gies, namely the handling of sensitive electorate data.

Moreover, insights into the suggested study will illus-

trate how individuals, political actors, and institutions

in other countries manage and use different facets of

data for electoral and political activities without com-

promising privacy and security.

6 CONCLUSION

In this paper, we analysed Bhutan’s political data

ecosystem through the lens of the Knowledge Pyra-

mid to examine the dynamics of DIKW generated

in the political data ecosystem to facilitate demo-

cratic electoral activities and political campaigns. The

democratic electoral process of the 4th National As-

sembly Election in Bhutan was unpacked for insights

into the practices of transforming data into higher

levels of the Knowledge Pyramid. In Bhutan, the

electoral stakeholders and political parties deal with

DIKW on a relatively small scale and with varying

infrastructural complexities. The political parties use

websites, information systems, and social media to

manage aspects of DIKW, facilitate decision-making,

and administer campaign activities. However, con-

sidering the nascency of democracy in Bhutan, no

extensive data technologies or human resources are

employed to harness the actual and potential value

of data generated in the political space. In this re-

gard, electoral stakeholders and political actors could

explore opportunities to adopt robust data analytics

technologies and knowledge management infrastruc-

ture to manage DIKW emanating from the political

data ecosystem. This study advances the literature

on knowledge management through a ﬁne-grained ac-

count of the democratic election in a country expe-

riencing the signiﬁcant effect of political technolo-

gies to turn political data into higher elements of the

Knowledge Pyramid for various ends in the political

world.

REFERENCES

Baldwin-Philippi, J. (2017). The myths of data-driven cam-

paigning. Political Communication, 34(4):627–633.

Bates, W. (2009). Gross national happiness. Asian-Paciﬁc

Economic Literature, 23(2):1–16.

Bellinger, G., Castro, D., and Mills, A. (2004). Data, infor-

mation, knowledge, and wisdom.

Bennett, C. J. and Lyon, D. (2019). Data-driven elections:

Implications and challenges for democratic societies.

Internet Policy Review, 8(4).

Choo, C. W. (1996). The knowing organization: How orga-

nizations use information to construct meaning, create

knowledge and make decisions. International Journal

of Information Management, 16(5):329–340.

Dommett, K., Barclay, A., and Gibson, R. (2023). Just what

is data-driven campaigning? a systematic review. In-

formation, Communication & Society, 27(1):1–22.

Election Commission of Bhutan (2008). Election Act of the

Kingdom of Bhutan, 2008. Election Commission of

Bhutan, Thimphu.

Knowledge Pyramid Perspective of the Political Data Ecosystem: A Case Study of Bhutan

Election Commission of Bhutan (2013). Handbook

for Counting Supervisor. Election Commission of

Bhutan, Thimphu.

Election Commission of Bhutan (2018). ECB Social Me-

dia Rules and Regulations of the Kingdom of Bhutan,

2018. Election Commission of Bhutan, Thimphu.

Election Commission of Bhutan (2024). Election Com-

mission of Bhutan. Election Commission of Bhutan,

Thimphu.

Frick

e, M. (2009). The knowledge pyramid: A critique of

the DIKW hierarchy. Journal of Information Science,

35(2):131–142.

Gelhaar, J. and Otto, B. (2020). Challenges in the emer-

gence of data ecosystems. In 23rd Paciﬁc Asia Con-

ference on Information Systems, 1–14.

Jennex, M. E. (2017). Big data, the internet of things,

and the revised knowledge pyramid. ACM SIGMIS

Database: The DATABASE for Advances in Informa-

tion Systems, 48(4):69–79.

Jin, X., Wah, B. W., Cheng, X., and Wang, Y. (2015). Sig-

niﬁcance and challenges of big data research. Big

Data Research, 2(2):59–64.

Khatri, V. and Brown, C. V. (2010). Designing data gover-

nance. Communications of the ACM, 53(1):148–152.

Lee, I. (2017). Big data: Dimensions, evolution, impacts,

and challenges. Business Horizons, 60(3):293–303.

Metz, T. (2014). Gross national happiness: A philosophical

appraisal. Ethics and Social Welfare, 8(3):218–232.

Micheli, M., Ponti, M., Craglia, M., and Suman, A. B.

(2020). Emerging models of data governance in the

age of dataﬁcation. Big Data & Society, 7(2).

Miller, H. G. and Mork, P. (2013). From data to decisions:

A value chain for big data. IT Professional, 15(1):57–

59.

Mittelstadt, B. D., Allo, P., Taddeo, M., Wachter, S., and

Floridi, L. (2016). The ethics of algorithms: Mapping

the debate. Big Data & Society, 3(2):1–21.

Namgay, P., Wangdi, P., Thinley, S., and Namgyel, T.

(2023). Fairifying STEM data ecosystem to enhance

data reuse. In 2023 IEEE Frontiers in Education Con-

ference, 1–9.

Nissen, M. E. (2002). An extended model of knowledge-

ﬂow dynamics. Communications of the Association

for Information Systems, 8:251–266.

Nonaka, I. (2007). The knowledge-creating company. Har-

vard Business Review, 85(7/8):162–171.

Nowell, L. S., Norris, J. M., White, D. E., and Moules, N. J.

(2017). Thematic analysis: Striving to meet the trust-

worthiness criteria. International Journal of Qualita-

tive Methods, 16(1).

Oliveira, M. I. S. and L

oscio, B. F. (2018). What is a data

ecosystem. In 19th Annual International Conference

on Digital Government Research: Governance in the

Data Age, 1–10.

Olteanu, A., Castillo, C., Diaz, F., and Kıcıman, E. (2019).

Social data: Biases, methodological pitfalls, and ethi-

cal boundaries. Frontiers in Big Data, 2:1–33.

Papakyriakopoulos, O., Hegelich, S., Shahrezaye, M., and

Serrano, J. C. M. (2018). Social media and microtar-

geting: Political data processing and the consequences

for germany. Big Data & Society, 5(2):1–15.

Quintas, P., Lefrere, P., and Jones, G. (1997). Knowledge

management: A strategic agenda. Long Range Plan-

ning, 30(3):385–391.

Rowley, J. (2007). The wisdom hierarchy: Representations

of the DIKW hierarchy. Journal of Information Sci-

ence, 33(2):163–180.

Ruppert, E., Isin, E., and Bigo, D. (2017). Data politics. Big

Data & Society, 4(2):2053951717717749.

Schippers, B. (2020). Artiﬁcial intelligence and democratic

politics. Political Insight, 11(1):32–35.

Thomas, D. R. (2006). A general inductive approach for an-

alyzing qualitative evaluation data. American Journal

of Evaluation, 27(2):237–246.

Tuomi, I. (1999). Data is more than knowledge: Implica-

tions of the reversed knowledge hierarchy for knowl-

edge management and organizational memory. In

Proceedings of the 32nd Annual Hawaii International

Conference on Systems Sciences, 1–12.

Wang, S. and Noe, R. A. (2010). Knowledge sharing: A

review and directions for future research. Human Re-

source Management Review, 20(2):115–131.

Yin, R. K. (2014). Case Study Research: Design and Meth-

ods. SAGE Publications, Inc., 5 edition.

Zins, C. (2007). Conceptual approaches for deﬁning data,

information, and knowledge. Journal of the Ameri-

can Society for Information Science and Technology,

58(4):479–493.

KMIS 2024 - 16th International Conference on Knowledge Management and Information Systems