See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/362781397

Research YouTube: Methods, Tools, Analytics

Chapter · September 2022

CITATIONS READS

11 8,383

2 authors:

Laeeq Khan Aqdas Malik

Ohio University Sultan Qaboos University
54 PUBLICATIONS 2,268 CITATIONS 59 PUBLICATIONS 2,110 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Laeeq Khan on 19 August 2022.

The user has requested enhancement of the downloaded file.

 44
Researching YouTube: Methods,
Tools, and Analytics
M. Laeeq Khan and Aqdas Malik

INTRODUCTION into understanding social interactions on YouTube

employing social media analytics techniques.
Social media use has become an integral part of We define social media analytics research as the
the lives of billions across the globe. As a modern use of varied tools and techniques to make sense
media powerhouse, YouTube is the world’s most of online activities and conversations by obtaining,
popular video-sharing site (Alexa, 2021). The refining, analyzing, and visualizing social media
largest market for YouTube is the United States data. Such analyses can include both quantitative
(14.8%), followed by India (8.1%). Over 70% of and qualitative research techniques with the aim
the views on YouTube are from mobile devices of understanding online user behavior, needs, per-
(Cooper, 2019). ceptions, and challenges (Frauke, this volume).
YouTube content is diverse, emanating from Through analyzing social media, researchers
a broad range of audiences, in a number of lan- assess which social media channels are being used
guages, and in a variety of genres (Khan, 2017). the most, what is being discussed by people, how
Videos are one of the most engaging media, and users interact with each other, and why the users
such a high level of activity on this social media exchange and consume specific information.
platform necessitates a thorough analysis of the Given the significance of YouTube as one
content posted in the form of videos and com- of the leading players in the social media land-
ments. In comparison to other leading social media scape (Alexa, 2021), organizations, businesses,
(e.g. Facebook, Twitter, and Instagram), consider- governments, and the public alike have a robust
ing its user base, multitude and volume of con- presence on the platform. It is therefore impera-
tent, and influence on popular culture, YouTube tive for researchers to study YouTube and engage
is still an understudied platform (Allgaier, 2020). in research that advances their understanding of
Of the research that has been conducted in the content and the resulting engagement surround-
realm of YouTube, the focus has been on areas ing that content. For example, YouTube research
such as business, education, and medicine. Major can be about understanding participatory behav-
research methods employed in YouTube research iors and knowing what kind of content garners
have been interviews, surveys, and experiments. the greatest user interest. YouTube research can
A growing number of studies have started delving also be about studying the motivations of online

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 651 23/05/22 5:02 PM

 652 The SAGE Handbook of Social Media Research Methods

participation and consumption (Khan, 2017; with television, YouTube contenet viewership has
Szmuda et al., 2020). Additionally, research may shown a considerable rise(Smith, 2020).
employ various analytic techniques to understand Some of the key features of the platform include
user engagement patterns and behaviors around information searching, creating personal chan-
videos through likes, dislikes, comments, and nels, uploading, liking, commenting, sharing, and
shares. following channels and users. In 2015, YouTube
In this chapter, we shed light on the various Music and YouTube Kids were launched (Ben-
tools for YouTube data collection. These include Yair, 2015). YouTube has also become a major
the more recent social media analytics techniques source of children’s content. YouTube Kids is a
that often employ ‘big data’. In addition, we child-friendly version of the platform that filters
discuss seminal research centered on YouTube, content that is safe for minors and offers parental
highlight limitations, and sum up the various prac- controls such as limited screen time and a disa-
tices to effectively conduct YouTube research. bling search functionality.
Furthermore, we offer two case studies to illumi- YouTube Games is an avenue for live-streaming
nate the latest research on the platform and the games, where gamers can stream their playing ses-
methods employed to answer the research ques- sions to a live audience by interacting and chatting
tions. In discussing the significance of big data, with them in real-time. More recently, YouTube
artificial intelligence (AI), and machine learning, has introduced YouTube Live that supports organi-
we conclude by advocating the need for collabora- zations and independent content creators to stream
tion across varied academic disciplines. live content. This feature facilitates the stream-
ing of live news and social events such as gaming
tournaments, concerts, and global events includ-
ing the Hajj, the Olympics, Super Bowl, and the
US presidential debates. Many leading news and
EVOLUTION OF YOUTUBE
entertainment television channels including Sky
News, ABC News, and Al Jazeera stream their
YouTube has been evolving and redefining the broadcast in real-time.
convergent media landscape for more than a
decade. The site was founded in 2005 in a make-
shift garage office in Silicon Valley (Alexa, 2021).
By 2021, the platform was hosting over five bil- YouTube and Media Convergence
lion videos, and 500 hours of video content is
uploaded every minute (Press, 2021). In the begin- Media convergence can be understood as the
ning, YouTube was predominantly a showcase for blending of various mediums – print, television,
amateur and home videos, but after being acquired radio, and the Internet. Advancements in technol-
by Google in October 2006 for 1.65 billion dol- ogy have propelled the change in media land-
lars, it transformed into a platform channeling a scapes and a shakeup of traditional means of
wide variety of content. The platform is often communication. Today, the way individuals access
referred to as the second-largest search engine entertainment and information is vastly different
behind Google due to the sheer number of searches from a couple of decades ago. In particular, the
and content available on the site (Smith, 2020). Internet revolution has enabled the viewing of
In the early days, the motive of sharing content images, videos, and the reading of textual infor-
on the platform was focused on drawing attention mation to happen on one platform.
and appreciation from the audience, however, it is The boundaries between television and online
now geared toward fame and revenue generation. video have blurred. YouTube offers content shar-
Consequently, YouTube has also focused on more ing and viewing of videos to online users across
monetization features apart from video ads and the globe. Furthermore, YouTube users are also
revamped how content creators develop and share able to engage with such content through likes,
media (Coromina et al., 2020). dislikes, comments, and shares. A typical YouTube
Within the last few years, the platform has channel thus offers entertaining video content as
emerged as a global broadcaster and has become well as text in the form of comments. We argue that
an integral part of the mainstream media. More this juxtaposition of text and videos on YouTube
recently, YouTube has also partnered with main- may be referred to as content convergence. While
stream media news, entertainment, and gaming entertainment is the dominant motive for visiting
channels. The introduction of high-definition videos YouTube (Khan, 2017), about 50% of YouTube
on the platform has resulted in content from niche users indicate that the site is important in the con-
content creators, including gamers and vloggers, to text of learning about new things that they have
posters of high-resolution content. In comparison not done before (Smith, 2020).

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 652 23/05/22 5:02 PM

 Researching YouTube: Methods, Tools, and Analytics 653

Content convergence and the shifting of power indices (number of comments, likes, dislikes,
in the hands of average Internet users has led to and subscribers), and textual data (such as user
the blossoming of a variety of original content cre- comments, subtitles/closed captions, and video
ated and shared by individuals. YouTube has been descriptions). Analyzing the content on YouTube
an enabler of a new cultural phenomenon that not only provides a nuanced understanding of the
includes content such as how-to videos, music, traces of social behaviors on the platform but also
vlog phenomena, and much more. Various diverse the user community surrounding that content.
fields such as education, health, business, enter- Given the sheer amount of video content,
tainment, and politics are thriving on YouTube. engagement indices, and textual data, traditional
There has been widespread discovery of global methods of data analysis may prove limiting.
talent and the rise of influencers reflected on To overcome these limitations, contemporary
YouTube. Online streaming services made pos- algorithms and sophisticated statistical analysis
sible through YouTube are now in direct compe- might provide a better assessment and insights
tition and are even major disruptors within the into YouTube’s big data to understand video and
traditional media industries. YouTube Live and audio content, associated text, user behaviors, and
YouTube TV are also unique services that the plat- relevant interactions. There is a strong need to
form offers to users worldwide. develop automated tools that can efficiently col-
lect relevant data as well as the ability to analyze
video and audio content and textual data avail-
able on the platform. In comparison with other
Why YouTube Research Matters forms of research methods such as interviews and
focus groups, analyzing YouTube content per-
The global popularity and robust user engagement mits researchers to observe users unobtrusively.
on YouTube open up new avenues of data collec- Furthermore, it is easier and, in many cases, more
tion and analysis. Coinciding with the high levels appropriate to understand the given phenom-
of content production (both video and text), the ena by observing content and the interactions of
platform offers researchers diverse and dynamic users/communities that would be difficult to reach
datasets. Traditional quantitative and qualitative otherwise.
research methods have their own set of limitations Over the years, YouTube has often been
that make the collection, storage, analysis, and researched to collect multimodal data and content
visualization of large datasets infeasible. This is samples as well as to examine user behaviors, atti-
where YouTube research attracts researchers to go tudes, and preferences (Jiménez and Vozmediano,
beyond traditional methods and employ social 2020; Khan, 2017; Porreca et al., 2020). Like other
media analytics tools and techniques to extract social media, YouTube is often targeted by fake
meaning and engage in effective data storytelling. news and misinformation that leads researchers to
The proliferation of YouTube as the go-to venue question the trustworthiness and authenticity of the
for information seeking and sharing, learning and content on the platform. Similarly, ethical issues,
teaching, social interaction, and entertainment privacy concerns, methodological shortcomings,
necessitates the investigation of its usage, features, and limited availability of data collection and anal-
and effects.With local versions in more than 100 ysis tools have been cited as some of the reasons
countries and content in 80 different languages, for the limited research on YouTube. Despite some
YouTube has a global reach. The site is logged of the challenges and pitfalls of using YouTub as
into by over two billion users every month mak- a source of data, the significance of the platform
ing it a goldmine of data for researchers (Press, in shaping opinions, delivering information, and
2021). As an extremely fertile source of original reaching publics across the globe makes it worthy
content, YouTube serves a creative outlet that of study. YouTube has also empowered individu-
is in direct competition with more established als to generate, share, and curate content through
traditional outlets such as television and film. its participatory culture. Data from YouTube pro-
YouTube, therefore, holds great significance for vides ample opportunities for qualitative, quantita-
marketers, advertisers, policymakers, businesses, tive, mixed-method, and analytics researchers to
and individuals. explore events, phenomena, and users.
In recent years there has been an increase Social media research involves an understand-
in the use of the platform as a source for social ing of the network structure and identification of
and behavioral research, however, further studies the influencers, opinion makers, and prominent
across academic domains and insights into various personalities that control the flow of information
phenomena are called for. The platform facilitates on such networks. In particular, YouTube research
different forms of content that can be gathered is about understanding user interactions, the flow
and analyzed such as videos, audio, engagement of information, and learning about consumer

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 653 23/05/22 5:02 PM

 654 The SAGE Handbook of Social Media Research Methods

sentiments, attitudes, and opinions (Lee and Yoon, cancer, diabetes, and HIV/AIDS is also an active
2020; Porreca et al., 2020). The extraordinary vol- stream of research (Loeb et al., 2019; Ortiz-
ume of content, geographical reach, and the num- Martinez et al., 2017; Pons-Fuster et al., 2020).
ber of active users and their interactions make the Similar to educational content, research within
platform ideal for researching a wide variety of the public health domain raises concerns about
phenomena. misinformation, fake news, and the credibility of
the content (Loeb et al., 2019). From a psycholog-
ical perspective, researchers have also studied how
users (in particular, vulnerable cohorts of society,
such as children) are affected when using the
SEMINAL RESEARCH ON YOUTUBE platform. Themes such as cyberbullying, trolling,
exposure to scatological, sexual, and violent con-
Earlier research about YouTube centered on more tent have been actively studied (Elias and Sulkin,
traditional research methods to understand user 2017; Burcu et al., 2019). Table 44.1 provides an
behavior. For example, a study by Lange (2007) overview of selected seminal works from various
used an interview method to understand young domains together with studied variables, applied
people’s engagement with online video sharing and methods(s), data collection tools, sample sizes,
the formation of social networks (Lange, 2007). and theoretical foundations.
In more recent years, various computational
research methods and analytics techniques have
been employed to understand user behavior and
interactions on YouTube. Qualitative, quantitative,
and mixed-methods research has been conducted METHODS OF YOUTUBE RESEARCH
on various aspects of YouTube that can be classi-
fied under three key dimensions: study disciplines, Researchers from various scientific disciplines
content forms, and users/communities. such as communication, business, sociology, and
With respect to academic disciplines, most of informatics have applied numerous research
the research has been conducted within public methods to examine and gain insights into differ-
health, education, journalism, global events, and ent perspectives surrounding the platform itself
monetization. YouTube is often referred to as one and its users. These include the need to explore
of the key instructional technology channels for users and communities to understand their moti-
learning and delivering education. As a significant vations for posting and watching, and in and pre-
source of education, studies report increased per- dicting video virality. A study by Snelson et al.
sonal development, learning satisfaction, knowl- (2012) identified seven key areas that needed
edge, and self-confidence from using the platform attention in YouTube research: (1) users, groups,
for educational purposes (Curran et al., 2020; and communities; (2) teaching/learning; (3) social
Jones and Cuthrell, 2011). Conversely, studies and political impact; (4) video creation/produc-
also report accuracy, credibility, and misinforma- tion; (5) legal/ethical concerns; (6) media man-
tion concerns (Jaffar, 2012). Interestingly, much of agement; and (7) commercial interests (Snelson
the conducted research within education concen- et al., 2012). We have adapted and refined these
trates on medical education, where factual, reli- categories to offer the following five major areas
able, and expert information is critical. Given the that need to be explored in YouTube research:
recent disruptions from the COVID-19 pandemic
and its impact on learners and educators, there is 1. User motivations for engagement
a strong need to understand how YouTube is being 2. Social, political, educational, and economic
used for education, and what challenges are being impacts of video content
faced by educators and learners.
3. Impacts of user engagement (views, likes, dis-
Public health is another key theme that has
been actively researched. As YouTube is used by likes, comments, and shares) on user perceptions
patients and their caregivers to find and share and sense-making
content about different illnesses and health con- 4. Formation of groups and communities
cerns, several studies have focused on patient 5. Ethical and legal challenges of video sharing.
education on the platform. Due to the sensitivity
of the content and its potential impact on the life There is a range of options available for research-
and wellbeing of the users, many studies have ers in seeking answers to various questions within
assessed the quality of the health content shared the five major research areas outlined above.
on the platform. Understanding the prevalence of Before discussing the various data collection and
videos around various diseases such as depression, analysis techniques in YouTube research, we first

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 654 23/05/22 5:02 PM

 Researching YouTube: Methods, Tools, and Analytics 655

Table 44.1 Overview of selected seminal work on YouTube

Study Domain/Theme Variables studied Method(s)/Data Sample
collection tool

Khan (2017) General/Engagement Information seeking, Giving Online survey N=1143 (Females:
information, Self- 63%, Males: 37%)
status seeking, Social
interaction, Relaxing
entertainment, Visit
Frequency, Experience,
Gender
Jiménez and General / Most Subject matter and Content analysis N=400 videos
Vozmediano popular teenagers- characteristics of videos
(2020) related videos on sex, bullying, drugs,
and pregnancy
Lee and Yoon Branding/ Viral Number of views, Engagement analysis, N=372 videos
(2020) marketing comments, likes, and Social Network
campaign dislikes, network Analysis
structure, network
centrality
Zhang, Baker, Public Health Content elements, message Content analysis N=341 videos
Pember, and appeal, Content poster,
Bissell (2017) engagement
Porreca, A., Public Health Engagement, Betweenness Co-occurrence N=2720 videos
Scozzari, F., centrality, emotions’ networks and
and Di Nicola, polarity, temporal Sentiment analysis/
M. (2020) sentiment Netvizz
Moon and Lee Public Health Assessment of Popularity, Content analysis / N=105 videos
(2020) Categorization of YouTube API
Source, Assessment of
Usefulness, Reliability,
Quality, and Title–
Content Consistency,
Das, S., Dutta, A., Transportation content types, Most Text mining, Sentiment N=107 videos
Lindheimer, Frequent Terms, analysis, and
T., Jalayer, M., Co-occurrence of terms, Polarity mapping/ R
and Elgart, Z. Subjectivity analysis, software package
(2019) Keywords-sentiment (tuber)
association
Chan (2019) Gig economy/ Self- Becoming an Uber driver/ Interviews, Content Semi-structured
presentation blogger, positioning analysis interviews (N=11),
Uber driver/bloggers as Textual channel
experts descriptions
(n = 34), videos
content (n = 14)

want to shed light on some of the key questions Question 2: What kind of comments are YouTube us-
that have been frequently posed by researchers in ers posting underneath videos?
the realm of YouTube research (Das et al., 2019; Question 3: Why are people viewing certain YouTube
Jiménez and Vozmediano, 2020; Lee and Yoon, videos?
2020; Moon and Lee, 2020; Porreca et al., 2020;
Question 4: How do people derive meaning from
Zhang et al., 2017).
reading comments?
Question 1: What type of YouTube videos are users Question 5: Which videos attract the highest views
posting? (passive engagement)?

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 655 23/05/22 5:02 PM

 656 The SAGE Handbook of Social Media Research Methods

Question 6: Which videos attract the highest active established research techniques described above
engagement from users in the form of likes, dis- have increasingly been adapted for Internet stud-
likes, comments, and shares? ies, often blending them with newer social media
Question 7: What are the roles of YouTube algorithms methods.
and the recommender systems in suggesting vid- When social media platform-generated data is
involved, we often refer to such studies as social
eos to users?
media analytics (SMA). Such analytics provide
valuable insights into online user behavior, atti-
In the following section, we will first discuss the
tude, and action. Various existing social science
traditional methods of data collection and analy-
research techniques have been employed to study
sis. We will then discuss the newer big data and
web and social media data.
analytics techniques that can be used to gather,
analyze, and visualize YouTube data.

Beyond Traditional Methods:

Traditional Data Collection Social Media Analytics
and Analysis on YouTube SMA is an evolving field that is defined as ‘the art
and science of extracting valuable insights from
Commonly employed data collection methods on
vast amounts of semi-structured and unstructured
YouTube in which users are involved include
interviews, focus groups, surveys, content analy- social media data to enable informed and insight-
ses, and experiments. These traditional methods ful decision-making’ (Sponder and Khan, 2017,
can mainly be classified as either quantitative or p. 167). SMA may also be viewed as a combina-
qualitative. In some instances, researchers blend tion of techniques to extract meaning, and to
various quantitative and qualitative methods to measure the performance of social actions in data
understand user attitudes. gathered from social media platforms. Researchers
Qualitative research typically involves collect- have the following options to collect YouTube
ing and making sense of non-numerical data to data: via the YouTube Application Programming
understand people’s opinions, attitudes, beliefs, Interface (API), and manual YouTube data collec-
and experiences (Rasmussen-Pennington, this tion from the YouTube interface.
volume). Qualitative research techniques include YouTube APIs. Automated data collection on
ethnographies, grounded theory, interviews, YouTube usually requires some programming
focus groups, and thematic analysis (Chesebro experience, and these programming languages
and Borisoff, 2007). In ethnographic research, could be R or Python. This data is often in large
researchers immerse themselves in the environ- quantities and can be referred to as big data that
ment to understand things. In thematic analysis, further requires storage and cleaning before carry-
the aim is to locate patterns within the data (Malik ing out a detailed analysis. Researchers who have
et al., 2019). Qualitative methods such as inter- little to no programming expertise in tapping into
views and focus groups provide small but often the YouTube APIs can rely on specialized services
rich data (Latzko-Toth et al., this volume). and companies that provide data from social plat-
Quantitative research methods focus on forms such as YouTube.
measurement using statistical, mathematical, or An API is a set of protocols and specifications
numerical analyses. Quantitative research can be that allows applications to interact and communi-
descriptive (summary of variables), correlational cate with each other. To better understand YouTube
(investigating relationships between variables), data collection, it may be argued that there are
and experimental (examining cause and effect). two types of interfaces: the API and the user inter-
In addition, quantitative techniques include sur- face (UI). The UI is concerned with the user side
veys and content analysis (can be quantitative in the form of the webpage that users encounter
when measuring or counting, and qualitative when they go online. Conversely, API, as the name
when interpreting). For example, survey methods implies, is concerned with the programmers who
were employed to study the uses and gratifica- are adept at dealing with coding, and this is often
tions of engagement on YouTube (Khan, 2017); the backend workings of a website. Simply put, an
the experimental method has been employed to API may be visualized as a power cord connecting
study advertising messages on YouTube (Belanche to an electrical outlet so that electricity can transfer
et al., 2020); and content analysis was carried out and power a device. YouTube API allows program-
to understand conversations surrounding Chagas mers to access video-related data such as video sta-
disease on YouTube (Otieno et al., 2020). Various tistics and actual numbers about user interactions.

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 656 23/05/22 5:02 PM

 Researching YouTube: Methods, Tools, and Analytics 657

Screen scraping. Another way to collect to study diverse phenomena such as charitable
YouTube data is via screen scraping, which causes, activism, COVID-19 vaccine passports,
involves copying information that is being dis- and e-cigarettes (Khan et al., 2018; Malik et al.,
played on a screen. Some browser extensions 2019Khan et al., 2022). In the field of journal-
also allow the scraping of YouTube data. In other ism, Groshek and Tandoc (2017) examined mil-
instances, programming language codes in Python lions of tweets to find important actors using SNA
can also facilitate data gathering via screen scrap- techniques. Other researchers have employed
ing. Screen scraping is often cumbersome and semantic and linguistic analyses to unearth the
time-consuming. The size of data from manual meaning within text (Cervi, 2020; Jelodar et al.,
screen scraping is usually smaller than the data 2021). Such analyses can greatly enrich our under-
obtained through APIs. standing of the text presented as comments under
Once YouTube data has been obtained, it needs YouTube videos.
to be analyzed. Over the years, SMA has emerged Predictive and prescriptive analytics. Using
as a distinct field that builds upon a variety of current data, predictive SMA helps to predict
research methods. Since SMA involves the col- events and deals with questions such as, ‘what will
lection, analysis, and visualization of social media happen?’, by employing forecasting and predict-
data (Zeng et al., 2010), we contend that from a ing modeling techniques (Davenport and Harris,
methods perspective, any research where social 2017). Netflix, Amazon, and YouTube recom-
media data is gathered for analysis, it should be mender systems are good examples of predictive
treated as an SMA study. SMA studies have often analytics in practice. YouTube video recommen-
combined various quantitative and qualitative dations play a key role in what people watch on
research techniques, however, the major thrust the site. The site’s algorithms determine the vid-
of such studies has been quantifying information eos that appear in a user’s feed as well as what
based on a collection of data (often big data). In comes next after a video is watched (Smith, 2020).
other words, SMA is about understanding num- According to Hootsuite, 70% of what people
bers and the sense-making of data, often in large watch on YouTube is determined by its algorithm
quantities (big data). (Cooper, 2019).
Social media use is pervasive, and the interdis- In addition to sentiment analysis, engagement
ciplinary nature of the field comprising elements analysis, and SNA, more advanced analytics tech-
of informatics, computer science, communication, niques include AI, and machine learning. Analysis
business, and education (Khan, 2020), demands of YouTube videos can be challenging, especially
looking at SMA from a fresh perspective. The fast because videos are moving images, and analyzing
flowing, non-static, constantly generating, nature them manually, particularly in large quantities,
of social media data complicates the task of ana- will be time-consuming. Therefore, AI and deep
lyzing it using only traditional quantitative and learning techniques that often involve program-
qualitative techniques, thus giving rise to SMA. ming know-how are employed to analyze big data.
SMA can be further broken down into descriptive, Machine learning and AI are also being employed
predictive, and prescriptive analytics. to automate data collection and analysis. Since
Descriptive analytics. Descriptive SMA, about 500 hours of video is uploaded on YouTube
which currently forms the major portion of ana- every minute, YouTube itself has employed AI and
lytics-based studies tackles questions such as, machine learning to remove objectionable con-
‘what happened?’ or ‘what is happening?’ often tent on its platform (Marr, 2019). Moreover, the
describing the phenomena (Davenport and Harris, YouTube recommendation system also builds its
2017). Various commercial SMA software suites capability on deep learning, thus ensuring that the
and listening tools mine social media for publicly recommended YouTube videos are at scale, rel-
available comments to unearth sentiments and evant, credible, and fresh (Covington et al., 2016).
key themes through text analysis. The examina- Deep learning can be viewed as a form of AI
tion of YouTube data also includes analyzing the that mimics the human brain by processing data,
videos by the number of views, thus indicating learning, and making autonomous decisions. Deep
video popularity. Other techniques of analysis learning methods have proven to be promising in
rely on community detection through social net- automating data analysis, especially that involving
work analysis (SNA), which is used to map the image recognition and classification (Ketonen and
network and understand the relationships between Malik, 2020). Put simply, deep learning algorithms
users, organizations, and online entities. Through aid automation of the extraction of representa-
YouTube data, researchers can also analyze video tions by training the machines to act like humans
content using techniques such as content analy- in finding, observing, analyzing, learning, and
sis. For example, thematic- and content analysis relearning patterns to make decisions. For exam-
techniques have been employed on Twitter data ple, in fields such as health informatics, YouTube

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 657 23/05/22 5:02 PM

 658 The SAGE Handbook of Social Media Research Methods

videos can be analyzed from a health literacy and As YouTube is the second-largest search engine
patient education perspective by applying algo- after Google, millions of users employ YouTube
rithms to improve information dissemination (Liu search to look for video content related to their
et al., 2018). In addition, health informatics, big interests. Various factors are involved in enabling
data, and AI are increasingly being employed in this search process: algorithms, keywords and
sophisticated ways in domains such as national search engine optimization, and search queries by
security and surveillance, customer intelligence, users.
fraud detection, and marketing (Najafabadi
et al., 2015).
Prescriptive SMA suggests the best course
of action and includes optimization and simu- YouTube Data Collection and Analysis
lation modeling (Davenport and Harris, 2017). Tools
Prescriptive analytics involves machine learning
to automate processes so that the best course of One of the key reasons behind the lack of YouTube
action can be automated through the analytics research can be attributed to the limited availabil-
process. Media companies can employ prescrip- ity of tools that can be used for collecting and
tive analytics to capitalize on the best engage- analyzing YouTube data. Most of the studies uti-
ment opportunity as and when it arises on a social lizing data from YouTube have gathered it manu-
platform. ally, which ultimately limits the sample size. To
Limitations of SMA. Researchers have also overcome these challenges, researchers may opt
argued that SMA has limitations, mainly centering for tools such as ‘YouTube Data Tools’ and
on issues such as emerging methods in disparate ‘Mozdeh’. These tools enable researchers and
domains,difficult evaluation, complexity involv- practitioners from various domains to gain insights
ing automatic and manual judgments, and the into the users of YouTube, how they use the plat-
size of datasets (Thelwall, 2018b). Despite these forms, and the prevalence of different topics. The
limitations, we believe that SMA helps to obtain following section provides a brief overview of
valuable insights that might otherwise not be pos- four tools that may be used for gathering and ana-
sible. Moreover, the complexity in analyzing large lyzing YouTube data.
datasets using a combination of techniques such as Mozdeh: Mozdeh is a free desktop application
textual and sentiment analyses adds to the depth developed by the Statistical Cybermetrics Research
of analysis. Group at the University of Wolverhampton, UK
(Thelwall, this volume). In addition to extracting
data from Twitter and Reddit, this Windows-based
program facilitates the gathering of users’ com-
YouTube Analytics ments on YouTube videos. Keyword and advanced
search functionality can be used to gather and
YouTube users who upload videos on their filter user comments. Additionally, the software
YouTube channels are often interested in knowing allows detailed analysis of gathered data such as
about who is watching their videos, which videos sentiment analysis, identifying comments based
are performing well, and identifying the location on user gender, temporal analysis, association
of the people watching the videos. YouTube mining, and network analysis. Mozdeh has been
Analytics provides a powerful dashboard that deployed by several scholarly studies to investigate
makes available all such information bundled with the differences between male and female com-
numerous metrics. Like the Google Analytics menters on YouTube videos (Thelwall, 2018a),
framework, YouTube Analytics provides vital to determine the effectiveness of an educational
intelligence about videos on the site. YouTube channel (Terlumun and Appollm, 2018),
YouTube video metrics are important in pro- and to understand the reactions of users through
viding understanding of video performance. For their comments on videos about Syrian refugees
businesses in a highly competitive environment, in Turkey (Ozduzen, 2020).
YouTube Analytics offers the necessary intel- YouTube Data Tools: Another significant data
ligence to better understand their content reach collection approach is to employ YouTube Data
and visibility. Measurement is important and pro- Tools, developed by Digital Methods Initiative,1
gress can be monitored via YouTube Analytics to University of Amsterdam. Through the web-based
track whether specific goals are being met. These suite of tools, researchers can retrieve data about
goals can be in the form of video views, associ- YouTube channels and videos. More specifically,
ated engagement, and so forth. YouTube Analytics information about channels, channel networks,
dashboards provide various forms of insights video networks, and comments, and statistics
including real-time video metrics. about videos can be extracted. YouTube Data

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 658 23/05/22 5:02 PM

 Researching YouTube: Methods, Tools, and Analytics 659

Tools has been used to gain insights about videos recipes, religious services, and homeschooling.4
on chronic diseases (Zheluk and Maddock, 2020), Given the significant rise of content and searches
COVID-19 pandemic news and information related to the global crisis, this research study
(Marchal et al., 2020), and communication about aimed to investigate the type of content being
local and organic foods (Meza and Yamanaka, shared on the platform about COVID-19.
2020). Data collection: As a search term, ‘coronavirus’
Netlytic2: Netlytic is a convenient cloud-based was used to gather data for the current study. After
tool for social media research (see Gruzd et al., sorting the videos by view-count, 100 videos with
this volume). Developed by the Social Media the highest number of views were selected for the
Lab at Ryerson University in Canada, the online analysis.
tool helps gather data from YouTube (in addition Data analysis: A codebook was established
to Twitter, RSS feeds, text files, etc.). Without based on prior studies on videos about the Zika
requiring programming skills, users can conduct virus and Ebola, and the Centers for Disease
an in-depth exploratory text and network analysis. Control and Prevention factsheet on COVID-19.
YouTube comments can be extracted and ana- Content of the COVID-19 videos was classified
lyzed from any YouTube channel. Netlytic has under five categories that included preventive
been used to gather and analyze data for various behaviors, mortality and fear, symptoms, trans-
social media studies. For example, Netlytic has mission and actual history, and other precautions.
been employed to understand YouTube conversa- Additionally, key characteristics of the analytical
tions about Chagas disease (Otieno et al., 2020), sample such as number of views, length, language,
and discourse on YouTube about the Syrian war presentation style, source, and date of upload
(Dekker and Nguyen, 2020). were also coded. To assess intercoder reliability,
Webometric Analyst3: Webometric Analyst one author coded all the videos (N = 100) while
is able to collect data from YouTube via the per- another coded a randomly selected sample of 10
mitted YouTube API. The tool has been created videos.
by the Statistical Cybermetrics Research Group. Findings and implications: The descriptive
Webometric Analyst features enable video search analysis of the dataset indicated that the videos
by keywords, the download of YouTube channel on COVID-19 were generally short (mean length
statistics, comments on videos, and a network dia- = 6.4 minutes) and received enormous viewer-
gram of the comment replies. ship (mean number of views per video ~1.25 mil-
lion). Most of the videos in the sample were in
the English language and hosted on the channels
of various news agencies. From the perspective
of video content, less than a third of the videos
CASE STUDIES ILLUSTRATING YOUTUBE covered COVID-19 preventive measures and
RESEARCH behaviors such as hand hygiene, social distanc-
ing, and face mask usage. The analysis further
In the following sections, we present two case revealed that most of the videos injected fear and
studies (Basch et al., 2020; Rauchfleisch and anxiety among the viewers by portraying mortal-
Kaiser, 2020) that contextualize what has been ity rates and referencing death. These findings
discussed in this chapter. Through the case stud- highlight several implications for researchers and
ies, we exemplify some of the prominent YouTube practitioners within public health and computer-
research methods. The case studies provide mediated communication domains. Although pre-
insights in relation to data collection, analysis, senting information through videos and images
findings, and implications. on social media platforms is considered a highly
valuable and effective means to reach public and
niche communities (e.g. high-risk groups and peo-
ple with low levels of literacy), less than a third
Case Study 1: Content Analysis of of the videos highlighted the behavioral recom-
YouTube Videos About the COVID-19 mendations related to COVID-19. Likewise, the
Pandemic (Basch et al., 2020) extensive portrayal of death, suffering, and fear in
the sample data may result in socially and mor-
Since the recent outbreak of the COVID-19 pan- ally irresponsible behaviors, including hoarding or
demic, online users have been actively accessing black-marketing essential items such as food and
various social media to seek and share informa- medical supplies. Finally, with the rapidly chang-
tion. Likewise, user activity on YouTube in rela- ing nature of the pandemic, the sheer amount of
tion to COVID-19 videos has surged sharply as content, and the surrounding misinformation, it
they have been searching for face masks, food is of utmost importance for YouTube (as well as

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 659 23/05/22 5:02 PM

 660 The SAGE Handbook of Social Media Research Methods

other social media) to demarcate reliable, valid, identifying the network structure and communi-
and up-to-date content. Doing so will ensure that ties, the study identified the topics being discussed
users can make conscious decisions to protect in the comments in each community. Topic mod-
themselves and their loved ones through the right eling analysis revealed the following topics, in
information. order of salience, to be refugees, German politics,
insulting language, Russia and Ukraine, Merkel,
lying press, Euro, vaccination & health, Islam,
conspiracy theories, and anti-feminism. The over-
Case Study 2: Mapping the German all common thread within the comments of the
Far-right Communities on YouTube German far-right on YouTube was the refugee
(Rauchfleisch and Kaiser, 2020) crisis and associated problems. Some channels
also actively promoted anti-Semitism. It was also
The Internet provides a space for dissenting concluded that the German far-right communi-
views. Far-right groups and individuals often ties on YouTube have created a small but vocal
express a xenophobic ideology that is anti-immi- and cohesive alternative media and a ‘networked
grant, anti-refugee, anti-Islam, and anti-feminism, counterpublic’.
openly expressing racial superiority. Particularly
in the EU context, German far-right groups such
as Pegida, Identitarian Movement, and AfD
(Alternative for Germany) have also found a vent
on video-sharing sites like YouTube. In various CONCLUSION
EU countries, such extremist groups are taking a
more central political role, often challenging the This chapter provides a much-needed roadmap for
status quo and entering the political landscape social media researchers in answering important
through the political process. Germany has seen a questions surrounding user activity on YouTube. It
strong and alarming rise in the popularity of such is clear from the methods described above that
groups as they have become vocal on various YouTube research involves a range of tools and
social platforms. This study maps the creation of techniques that have a learning curve. However,
far-right online communities on YouTube. the sheer amount of activity on the most popular
Data collection: Data was downloaded via the video-sharing platform in the world makes its
YouTube API. A total of 116 far-right YouTube study worthwhile. The two case studies have pro-
channels in the German language were selected vided insights into some of the approaches that
using ‘snowball crawling’ of the YouTube recom- can be adapted and further improved on to not
mender system. Only the YouTube channels that only study text but also identify communities and
allowed user comments under the videos were important online actors.
selected for analysis. A total of 2.2 million com- YouTube research is interdisciplinary, and a
ments and 276,142 replies were downloaded for single method may not prove sufficient in answer-
analysis. ing the research questions posed. A more prag-
Data analysis: Users who had commented on matic approach to studying YouTube would be to
multiple channels were identified. For SNA in employ a combination of research methods. In a
Gephi software, a directed network emerged with similar vein, we argue in favor of a cross-discipli-
channels as nodes. The final network consisted of nary approach. It is often noted that informatics
113 nodes and 1,097 edges. and computer science researchers are undertak-
Findings and implications: The study ing important work in their own domains that is
researched the following questions: (1) the kind analytics-oriented, yet it lacks the social science
of comment overlap user communities (repre- perspective and theoretical enrichment. Similarly,
senting the old and the new far-right) that could social science, business, and humanities research-
be identified within the overall German far-right ers employ various research techniques to under-
YouTube sphere, (2) the topics that are discussed stand YouTube interactions, however, they often
in these YouTube far-right communities, and lack the necessary programming skills, especially
(3) their temporal patterns. SNA revealed the in the field of deep learning and AI. This calls for
emergence of seven key communities or clusters. academic and research collaborations across fields
These were also the major far-right YouTube chan- to ensure a more diverse set of research methods,
nels: AfD, new right, alternative media, national- tools, and techniques to understand the increas-
ists and music, far-right conspiracies, old right, ingly complex nature of emerging issues such as
and historical channels. Many of the commenters information virality and misinformation.
had connections with IM. The analysis revealed Future research should also focus on under-
a tight-knit community of commenters. After standing the affordances of mobile devices that

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 660 23/05/22 5:02 PM

 Researching YouTube: Methods, Tools, and Analytics 661

could enable the sharing of video content on the Cooper P (2019) 23 YouTube Statistics That Matter to
go. There is also a need for research on the impacts Marketers in 2022. Available at: https://blog.hoot-
of video sharing in the social and political realm, suite.com/youtube-stats-marketers/.
particularly in understanding the implications for Coromina O, Matamoros-Fernández A and Rieder B
democracy and civic participation. (2020) Follow the money: A large scale investiga-
tion of monetization and optimization on You-
Tube. In: AoIR Selected Papers of Internet Research,
The 21st Annual Conference of the Association of
Internet Researchers Virtual Event, 2020. https://
Notes
doi.org/10.5210/spir.v2020i0.11194
1  https://wiki.digitalmethods.net/Dmi/DmiAbout Covington P, Adams J and Sargin E (2016) Deep
2  www.netlytic.org neural networks for YouTube recommendations.
3  http://lexiurl.wlv.ac.uk/searcher/youtube.html In: Proceedings of the 10th ACM Conference on
4  www.youtube.com/trends/articles/covid-impact/ Recommender Systems, Bilbao, Spain, pp. 191–
198. https://doi.org/10.1145/2959100.2959190
Curran V, Simmons K, Matthews L, et al. (2020) You-
Tube as an Educational resource in medical educa-
tion: A scoping review. Medical Science Educator
REFERENCES 30(4): 1775–1782. https://doi.org/10.1007/s40670-
020-01016-w
Alexa (2021) The Top 500 Sites on the Web. Availa- Das S, Dutta A, Lindheimer T, et al. (2019) YouTube
ble at: www.alexa.com/topsites. as a source of information in understanding
Allgaier J (2020) Science and medicine on YouTube. autonomous vehicle consumers: Natural language
In: Hunsinger J., Allen M., Klastrup L. (eds) Second processing study. Transportation research record
International Handbook of Internet Research, 2673(8): 242–253. http://dx.doi.org/10.1177/
pp. 7–27. Springer, Dordrecht. https://doi.org/ 0361198119842110
10.1007/978-94-024-1555-1_1 Davenport T and Harris J (2017) Competing on Ana-
Basch C H, Hillyer G C, Meleo-Erwin Z C, Jaime, C., lytics: Updated, with a New Introduction: The New
Mohlman, J., & Basch, C. (2020) Preventive behav- Science of Winning. Cambridge, MA: Harvard
iors conveyed on YouTube to mitigate transmission Business Press.
of COVID-19: Cross-sectional study. JMIR Public Dekker I and Nguyen D (2020) Combining qualitative
Health and Surveillance 6(2): e18807. http:// and digital methods for exploratory framing analy-
dx.doi.org/10.2196/18807 ses: The case of alternative video coverage of the
Belanche D, Flavián C and Pérez-Rueda A (2020) Syrian War on YouTube. In: Nguyen D., Dekker I.,
Brand recall of skippable vs non-skippable ads in Nguyen S. (eds) Understanding Media and
YouTube. Online Information Review. http:// Society in the Age of Digitalisation. Cham: Pal-
dx.doi.org/10.1108/OIR-01-2019-0035 grave Macmillan, pp. 103–130. https://doi.org/
Ben-Yair S (2015) Introducing the Newest Member of 10.1007/978-3-030-38577-4_6
Our Family, the YouTube Kids App–Available on Elias N and Sulkin I (2017) YouTube viewers in dia-
Google Play and the App Store. Available at: pers: An exploration of factors associated with
https://blog.youtube/news-and-events/youtube- amount of toddlers’ online viewing. Cyberpsychol-
kids. ogy: Journal of Psychosocial Research on Cyber-
Burcu, I., Jones, I., Özdemir, T. B., Alktebi, L., Bakir, D. space 11(3). https://doi.org/10.5817/CP2017-3-2
(2019). Youtube & young children: research, con- Groshek J and Tandoc E 2017) The affordance effect:
cerns and new directions. In R. Brito & P. Dias Gatekeeping and (non) reciprocal journalism on
(Coords.), Crianças, famílias e tecnologias. Que Twitter. Computers in Human Behavior, 66, 201–
desafios? Que caminhos?. (pp. 81–92). Lisboa: 210. https://doi.org/10.1016/j.chb.2016.09.020
Centro Interdisciplinar de Estudos Educacionais. Jaffar A A (2012) YouTube: An emerging tool in
ISBN 978-989-8912-09-1 DOI: https://doi. anatomy education. Anatomical sciences educa-
org/10.34629/ipl.eselx.cap.livros.017 tion 5(3): 158–164. http://dx.doi.org/10.1002/
Cervi, L. (2020). Exclusionary Populism and Islamo- ase.1268
phobia: A comparative analysis of Italy and Spain. Jiménez A and Vozmediano M (2020) Subject matter
Religions, 11(10), 516. http://dx.doi.org/10.3390/ of videos for teens on YouTube. International Jour-
rel11100516 nal of Adolescence and Youth 25(1): 63–78.
Chesebro J W and Borisoff D J (2007) What makes https://doi.org/10.1080/02673843.2019.1590850
qualitative research qualitative? Qualitative Jelodar, H., Wang, Y., Rabbani, M., Ahmadi, S. B. B.,
research reports in communication 8(1): 3–14. Boukela, L., Zhao, R., & Larik, R. S. A. (2021). A
http://dx.doi.org/10.1080/17459430701617846 NLP framework based on meaningful latent-topic

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 661 23/05/22 5:02 PM

 662 The SAGE Handbook of Social Media Research Methods

detection and sentiment analysis via fuzzy lattice Marchal N, Au H and Howard PN (2020) Coronavirus
reasoning on youtube comments. Multimedia news and information on YouTube. Health 1(1):
Tools and Applications, 80(3), 4155-4181. http:// 0–3.
dx.doi.org/10.1007/s11042-020-09755-z Marr B (2019) The Amazing Ways YouTube Uses
Jones T and Cuthrell K (2011) YouTube: Educational Artificial Intelligence and Machine Learning. Avail-
potentials and pitfalls. Computers in the Schools able at: www.forbes.com/sites/bernardmarr/
28(1): 75–85. http://dx.doi.org/10.1080/0738056 2019/08/23/the-amazing-ways-youtube-uses-
9.2011.553149 artificial-intelligence-and-machine-learning.
Ketonen V and Malik A (2020) Characterizing vaping Meza X V and Yamanaka T (2020) Food communica-
posts on Instagram by using unsupervised machine tion and its related sentiment in local and organic
learning. International Journal of Medical Infor- food videos on YouTube. Journal of Medical Inter-
matics 141: 104223. http://dx.doi.org/10.1016/j. net Research 22(8): e16761. https://doi.org/
ijmedinf.2020.104223 10.2196/16761
Khan M L (2017) Social media engagement: What Moon H and Lee G H (2020) Evaluation of Korean-
motivates user participation and consumption on language COVID-19-related medical information
YouTube? Computers in Human Behavior 66: on YouTube: Cross-sectional infodemiology study.
236–247. https://doi.org/10.1016/j.chb.2016. Journal of medical Internet research 22(8):
09.024 e20775. https://doi.org/10.2196/20775
Khan, M. Laeeq, Malik, A., Ruhi, U., Al-Busaidi, A., Najafabadi M M, Villanustre F and Khoshgoftaar TM
(2022). Conflicting attitudes: Analyzing social (2015) Deep learning applications and challenges
media data to understand early discourse on in big data analytics. Journal of Big Data 2, 1. DOI:
COVID-19 passports, Technology in Society, 68, 10.1186/s40537-014-0007-7.
101830. https://doi.org/10.1016/j.techsoc.2021. Ortiz-Martinez Y, Ali-Salloum W, González-Ferreira F,
101830 et al. (2017) HIV videos on YouTube: Helpful or
Khan M L (2020) Big data and entrepreneurship. In: harmful? Sexually Transmitted Infections 93(7):
Mahoney LM and Tang T (eds) Handbook of Media 481–481. http://dx.doi.org/10.1136/sextrans-2017-
Management and Business. Lanham, MD: 053197
Rowman & Littlefield, pp. 391–406. Otieno A W, Roark J, Khan M L, et al. (2020) The kiss
Khan M L, Zaher Z and Gao B (2018) Communicating of death – Unearthing conversations surrounding
on twitter for charity: Understanding the wall of Chagas disease on YouTube. Cogent Social Sci-
kindness initiative in Afghanistan, Iran, and ences 7(1): 1858561. http://dx.doi.org/10.1080/2
Pakistan. International Journal of Communication 3311886.2020.1858561
12: 25. Ozduzen O (2020) ‘We are not Arabs and Taksim is
Lange P G (2007) Publicly private and privately ours’: YouTubed political agency, place-making
public: Social networking on YouTube. Journal of and social exclusion of Syrians. City 24(5–6),
computer-mediated communication 13(1): 361– 741–758. https://doi.org/10.1080/13604813.202
380. http://dx.doi.org/10.1111/j.1083-6101.2007. 0.1833538
00400.x Pons-Fuster E, Roca J R, Tvarijonaviciute A, et al.
Lee M and Yoon H J (2020) When brand activism (2020) YouTube information about diabetes and
advertising campaign goes viral: An analysis of oral healthcare. Odontology 108(1): 84–90.
Always #LikeAGirl video networks on YouTube. https://doi.org/10.1007/s10266-019-00445-3
International Journal of Advanced Culture Tech- Porreca A, Scozzari F and Di Nicola M (2020) Using
nology 8(2): 146–158. https://doi.org/10.17703/ text mining and sentiment analysis to analyse You-
IJACT.2020.8.2.146 Tube Italian videos concerning vaccination. BMC
Liu X, Zhang B, Susarla A, et al. (2018) YouTube for Public Health 20(1): 1–9. http://dx.doi.org/
patient education: A deep learning approach for 10.1186/s12889-020-8342-4
understanding medical knowledge from user- Press Y (2021) YouTube. Available at: https://blog.
generated videos. https://arxiv.org/abs/1807.03179 youtube/press.
Loeb S, Sengupta S, Butaney M, et al. (2019) Dis- Rauchfleisch A and Kaiser J (2020) The German far-
semination of misinformative and biased informa- right on YouTube: An analysis of user overlap and
tion about prostate cancer on YouTube. European user comments. Journal of Broadcasting & Elec-
Urology 75(4): 564–567. DOI: 10.1016/j.eururo. tronic Media 64(3): 373–396. DOI: 10.1080/
2018.10.056. 08838151.2020.1799690.
Malik A, Li Y, Karbasian H, et al. (2019) Live, love, Smith K (2020) 48 Fascinating and Incredible You-
Juul: User and content analysis of Twitter posts Tube Statistics. Available at: www.brandwatch.
about Juul. American Journal of Health Behavior com/blog/youtube-stats/.
43(2): 326–336. DOI: https://doi.org/10.5993/ Snelson C, Rice K and Wyzard C (2012) Research
AJHB.43.2.9 priorities for YouTube and video-sharing

BK-SAGE-QUAN_HASSE-220012-Chp44.indd 662 23/05/22 5:02 PM

 Researching YouTube: Methods, Tools, and Analytics 663

technologies: A delphi study. British Journal of Thelwall M (2018b) Social media analytics for You-
Educational Technology 43(1): 119–129. http:// Tube comments: Potential and limitations. Interna-
dx.doi.org/10.1111/j.1467-8535.2010.01168.x tional Journal of Social Research Methodology
Sponder M and Khan GF (2017) Digital Analytics for 21(3): 303–316. http://dx.doi.org/10.1080/13645
Marketing. Abingdon, Oxford: Routledge. http:// 579.2017.1381821
dx.doi.org/10.4324/9781315640914 Zeng D, Chen H, Lusch R, et al. (2010) Social media
Szmuda T, Syed M T, Singh A, et al. (2020) YouTube analytics and intelligence. IEEE Intelligent Systems
as a source of patient information for Coronavirus 25(6): 13–16. http://dx.doi.org/10.1109/MIS.
Disease (COVID-19): A content-quality and audi- 2010.151
ence engagement analysis. Reviews in Medical Zhang X, Baker K, Pember S, et al. (2017) Persuading
Virology 30(5): e 2132. http://dx.doi.org/10.1002/ me to eat healthy: A content analysis of YouTube
rmv.2132 public service announcements grounded in the
Terlumun I T and Appollm Y I (2018) Determining the health belief model. Southern Communication
effectiveness of Youtube videos in teaching and Journal 82(1): 38–51. http://dx.doi.org/10.1080/1
learning with Mozdeh algorithm. International Jour- 041794X.2016.1278259
nal of Education and Evaluation 4(4): 100–110. Zheluk A and Maddock J (2020) Plausibility of using
Thelwall M (2018a) Can museums find male or a checklist with YouTube to facilitate the discovery
female audiences online with YouTube? Aslib Jour- of acute low back pain self-management content:
nal of Information Management 70(5): 481–497. Exploratory study. JMIR formative research 4(11):
http://dx.doi.org/10.1108/AJIM-06-2018-0146 23366. http://dx.doi.org/10.2196/23366

BK-SAGE-QUAN_HASSE-220012-Chp44.indd
View publication stats 663 23/05/22 5:02 PM