Proceedings of the 55th Hawaii International Conference on System Sciences | 2022

It’s the Methodology For Me: A Systematic Review of Early Approaches to

Studying TikTok

Shaheen Kanthawala Kelley Cotter Kali Foyle Julia R. DeCook

University of Alabama The Pennsylvania State Loyola University Loyola University
skanthawala@ua.edu University Chicago Chicago
kcotter@psu.edu kfoyle@luc.edu jdecook@luc.edu

Abstract WeChat [39]. With the app’s continuing growth and

Research on TikTok has grown along with the app’s unique forms of user content, it has attracted attention
rapidly rising global popularity. In this systematic from policymakers, academia, and other industry and
review, we investigate 58 articles examining TikTok, its government sectors. Despite its share of controversy,
users, and its content. Focusing on articles published in including Former U.S. President Donald Trump’s threat
journals and proceedings across the domains of human- to ban the app and other countries’ ban of the app in
computer interaction, communication, and other related 2020 [15], the app continues to attract users and grow in
disciplines, we analyze the methods being used to study its influence.
TikTok, as well as ethical considerations. Based on our Research on TikTok is in its nascent stages, and the
analysis, we found that research on TikTok tends to use app has begun to attract more scholarly attention due to
content analysis as their primary method and mainly its rapidly felt cultural impact, which has resulted in
focus on user behavior and culture, effects of use, the other social media giants cloning the app’s functionality
platform’s policies and governance, and very few (e.g., Instagram’s Reels [31]). In this study, we conduct
articles discuss the ethical implications of collecting a systematic review to understand how scholars in the
and analyzing such data. Additionally, most studies fields of human-computer interaction (HCI),
employ traditional forms of data collection when the communication, and other related disciplines are
affordances of TikTok tend to differ from other social approaching the app as a venue for academic inquiry.
media platforms. We conclude with a discussion about While TikTok shares many similarities with other social
possible future directions and contribute to ongoing media platforms, it is more heavily dependent on
conversations about ethics and social media data. algorithmic curation than most other platforms. This
difference raises challenges that researchers must
1. Introduction respond to by determining the best way to study the
platform and its users. Examining the formative
research conducted within this space allows us to
With videos that contain everything from
observe gaps in methodological approaches for this
choreographed dances, lip syncs, sea shanties, makeup
novel platform, opportunities, challenges, and
tutorials, cute animals, political commentary,
recommend future directions for researchers.
conspiracy theories, and other memetic content,
Furthermore, the value of ethical considerations
TikTok’s popularity has exploded in recent years and
when studying social media data cannot be overlooked
become a strong competitor to other social media giants
and is one that must be grappled with [1]. This research
like Facebook, Twitter. Originating in China as the apps
allows us insights into whether/how scholarship tangles
musical.ly in 2014 and Douyin in 2016, the Chinese tech
with the ethics of user data on a platform predominantly
giant ByteDance bought Musical.Ly and folded the
used by younger users (including many minors). Our
app’s affordances into the existing Douyin app, creating
systematic review focuses on the variety of research
TikTok and taking the app global in 2018. In just a few
undertaken to study the app including methodology, use
short years, TikTok has amassed nearly 2 billion
of human subjects, topical focus, and discussions of the
downloads, became the 7th largest social network in the
ethical or legal issues of studying TikTok.
world, and boasted 689 million active users in January
2021 [29].
Although not available in China, when combined
with its Chinese counterpart, Douyin, TikTok becomes
the 5th largest social network in the world, ahead of
Instagram and the Chinese social media platform

URI: https://hdl.handle.net/10125/79716
978-0-9981331-5-7 Page 3105
(CC BY-NC-ND 4.0)
2. Background they represent a particular moment in the history of the
platforms. In other words, studies can quickly become
2.1 Social Media Research outdated as platforms alter features, functionality, and
governance structures. The surge in popularity of
Research on TikTok follows a now well-established TikTok as a medium utilized by consumers has led to its
line of social media inquiry examining both “questions rapid growth as a research focus. While TikTok might
related to social media itself” and “questions that inform exhibit transience by evolving, the current interest in the
our understanding of social phenomena” (McCay-Peet platform, by users and researchers alike, necessitates an
& Quan-Haase, 2018, p. 19-20). Social media research examination of how it is being explored.
requires certain methodological considerations Social media research also carries novel ethical
particular to the domain. These mainly include concerns related to the privacy, anonymity, and consent
questions of representativeness, access, transience, and of human subjects, with differences across researchers
ethics. In terms of representativeness, past work has in how to approach these [37]. While the public nature
demonstrated that social media user populations are of social media content has led many researchers to treat
skewed in different ways. In general, platforms tend to it as “fair game” for data collection and analysis, many
oversample those of higher socioeconomic status [13]. scholars have noted that the distinction between public
Different social media also demonstrate differences in vs. private does not neatly apply to online environments
gender and age composition--for example, women are [10]. Instead, social media spaces are characterized by
more likely to use Facebook, and men are more likely to “unsteady forms of publicness” [2]. Private groups and
use Reddit [13]. Further, a small minority of users tend restricted forums are not public, but represent spaces
to produce the majority of content circulating on social where people might “reasonably expect to be observed
media [14,27], which creates challenges for sampling. by strangers” [35:8], suggesting a more complex sense
Beyond this, social media trace data are generated by of privacy. Even when social media data is publicly
platforms with certain organizational goals in mind, available, some users expect or prefer that researchers
namely enhancing the performance and profitability. ask for permission before using their content [9].
These “measurement conditions” shape the insights Finally, many social media users do not have a full
such data may give rise to as a result of the feedback and complete understanding of the extent to which their
loop between users and platform governance and design online activity is tracked, which complicates privacy
[40]. concerns [10,41]. Even with such understanding, people
Related to representativeness, social media may not anticipate that what they post online for
platforms do not typically grant unfettered access to data particular audiences may be extracted from its original
generated by their users. Platforms regulate which data context to be placed under the proverbial microscope
may be accessed and by whom through their APIs and [42]. Certainly, gaining consent from social media users
other tools (e.g., Facebook’s CrowdTangle). These APIs for the use of their data presents many practical
and tools are subject to constant change, often with little problems [10]. Although big data collected from social
forewarning. Partnerships between platforms and media may be anonymized, without careful oversight,
academics have offered one avenue through which individuals can be easily re-identified if a dataset is
researchers can pursue various lines of inquiry [25]. Yet, made public and/or aggregated with other datasets
across APIs, tools, and partnerships, some have [10,41]. This issue is particularly salient with TikTok
expressed concerns about the ways that platforms subtly where a large volume of users are under the age of 18.
influence research and scholarship produced from social
media data, particularly as platforms supply funding for 2.2 POV: You’re a Researcher Interested in
such work [1,5]. Where, to our knowledge, TikTok has Studying TikTok
not developed active associations with researchers so
far, the challenges arising from the affordances of the TikTok shares various features with other popular
platforms also raise issues of representativeness. social media platforms--for example, the presentation of
Not only can it be difficult to access social media content in feeds, the ability to “like” and comment on
data, but social media platforms evolve iteratively, content, and the aggregation of content via hashtags.
which makes them unstable targets for research. Social What sets TikTok apart from most other social media, is
media platforms should be considered “evolving how the content on the app is delivered to users and its
organizations” [11], which exhibit transience, or emphasis on challenges, repeating sounds and visual
“change, often dramatically and in short periods of time, motifs, and other memetic activity. Whereas most social
in their policies, procedures, and affordances” [4:2]. media platforms heavily depend on users to specify
Such transience often means that even when studies of which accounts they wish to see, often centering
specific social media platforms offer valuable insights, existing relationships with friends and family members,

Page 3106
TikTok’s main feed (the “For You Page” [FYP]) mainly Communication and Mass Media Complete (CMMC),
consists of algorithmically recommended content. This Communication Abstracts, and Web of Science in
may potentially give the platform more of an authentic August 2021. These searches, which were narrowed to
or organic feel, as it is likely easier to go viral on TikTok scholarly journal articles only, yielded 274 articles in
than other platforms where number of followers matters total. This initial search was supplemented with an
more [22]. Features like the ability to reuse other users’ additional search of the same terms in the top 50
audio and “stitch'' a reply to another user’s video have communication journals and top 50 HCI journals
rapidly fostered a variety of unique norms and practices identified by SCImago Journal & Country Rank
on the platform [26] – for example, the oft-cited trend of (https://www.scimagojr.com), which returned another
participating in the latest dance challenge. 164 studies.
These unique affordances make TikTok unlike
other social media platforms, and potentially require Selection of Studies
distinctive methodological approaches of examination. From the 438 articles found in the aforementioned
By conducting this systematic review, our goal is to databases and top journals, 380 were eliminated based
provide an easy to navigate roadmap and reference point on the following criteria:
for researchers interested in studying TikTok and the ● The research did not study TikTok, users of TikTok,
emerging cultural phenomena that come from the app, or did not use any TikTok data
while also noting gaps in scholarship. We additionally ● The study only mentioned TikTok as an example
aim to contribute to larger conversations in the fields of (e.g. “Social media platforms like Instagram and
HCI and communication pertaining to the ethical TikTok”)
complexities of studying social media platforms. We are ● The study was not in English
guided by the following research questions: ● Duplicates studies
RQ1: How are scholars exploring TikTok and its users? After eliminating based on these criteria and
RQ2: What research methods are scholars using to study general data cleaning, we were left with a total of 58
TikTok? studies (n = 58) for coding. A full flowchart of the study
RQ3: What challenges and opportunities are there in selection process can be found in Figure 1.
studying TikTok and its platform culture?
RQ4: What are the ethical challenges in researching Assessment of Methodological Quality/Analysis
TikTok? The final sample was coded based on a codebook
developed by the authors which focused on
3. Methods methodological approach and the focus of the study.
Surveys, interviews, participatory, observational,
To begin our systematic review, we first determined ethnographic, computational, digital, experimental or
the parameters of this work and identified certain quasi-experimental, content analysis and critical or
inclusion and exclusion criteria outlined below along cultural analysis were among the methods by which
with our search strategies, coding, and analysis. studies were categorized. For computational
approaches, we considered studies that mostly relied on
Inclusion Criteria methods such as large-scale analysis (i.e. text and image
We narrowed the scope of our study to work in the analysis) whereas digital methods may involve scraping
domains of HCI and communication, targeting relevant data but use a different analytical framework (e.g.
sources accordingly (as elaborated below). Beyond this, content analysis of scraped data) or techniques like the
our primary inclusion criteria was scholarship that walkthrough method [20]. We separated participant
substantively studied TikTok or Douyin. In particular, observation and ethnographic methods from
studies that either focused on the platform, such as those observation because they typically involved interacting
examining platform characteristics or those that looked with participants, whereas observation does not
at users, were included. In order for the focus on TikTok typically involve direct interaction [3]. These categories
to be considered substantive, studies needed to include were also separated mainly based on the amount of time
discussion specific to TikTok (and not just “social spent using these methods. Although participant
media” generally) or needed to use TikTok data in their observation is a tool that can be used in ethnography, not
analysis (e.g. videos and/or other TikTok content). all ethnographies use participant observation, and
participant observation and general observation can
Search Methods and Exclusion Criteria
We conducted a search for the terms “TikTok,”
“Tik Tok,” and “Douyin” in relevant databases, namely
the Association for Computing Machinery (ACM),

Page 3107
 the characteristics of both platforms. Finally, we coded
whether studies discussed ethical considerations related
to data collection and/or analysis. A search for
keywords, including “ethics” and “Institutional Review
Board” was conducted within each manuscript to
determine if and how researchers approached ethical
considerations. Upon manual review of search results,
articles that included any discussion of research ethics
were coded accordingly.

4. Findings
The first observation within our data was the rapid
increase in research focusing on TikTok since 2019,
particularly since mid-2020. Figure 2 depicts this
interest, indicating the importance of examining the
Figure 1. A Flowchart depicting the selection platform and the work being done on it. Based on our
process analysis, we found that emerging studies on TikTok
occur in more short-term studies whereas ethnography tend to use content analysis as their primary method
tends to be more long term [4]. and have been mainly interested in the app’s user
Aside from methodology, studies were coded based behavior and culture, effects of use, and the platform’s
on the following variables: primary or secondary policies and governance. All the studies in our sample
research, study focus, age of participants, whether there conducted primary research, with no secondary data
were human subjects, a focus on TikTok or Douyin; and analyses. Out of the 58 studies, only 18 studies
discussion of ethics. Excepting categories for human mentioned either obtaining Institutional Review Board
subjects and IRB/Ethics discussion, categories were not (IRB) or Ethics Board approvals for their studies.
mutually exclusive, i.e., some studies were coded for Current TikTok research included in this review
multiple dimensions of a single variable (e.g., an article reflects mainstream concerns about the cultural impact
might be coded as focusing on both TikTok and of the app, and conversations about content moderation
Douyin). We also determined whether a study used and the way the platform governs itself are in line with
primary or secondary research. If data was collected by the growing trend of research on platform governance
the authors of a study, it was considered primary; if data writ large. However, the number of studies that did not
was collected for use in a previous study and repurposed address ethics of data collection is concerning, given
in a new one, it was coded as secondary research. Study the app’s popularity among minors. A breakdown of
focus was another area of consideration when analyzing the results by category can be found in the table in the
the dataset. The focus of the studies could include the supplemental files alongside a numerical list of all the
following categories: user behavior; user culture, studies used [S1-S58].
community and movements; algorithms and software;
design and interface; critical analysis or cultural
commentary; and media effects/media psychology.
Criterion for the presence of human subjects was
also included. Other studies, such as those that
performed digital methods research on the platform, but
did not include human subjects, were coded
accordingly. Particularly for those studies that focused
on TikTok users, we noted the age of populations under
study based on three groups: youth (12 and under), teens
(13-17) and adults (18 and above). Studies that either
did not provide distinction of the age of users or did not
examine data where age was known or irrelevant, (such
as with studying the platform interface itself rather than
users), were coded as “not available”. Figure 2. A graph depicting the uptick in
We also distinguished whether the studies focused TikTok research
on TikTok, the global version of the app, or Douyin, the
Chinese version of the app, due to slight variations in Exploring TikTok and its users

Page 3108
 Our first research question asked how scholars were Zeng [S22] examined the parallel platformization of
exploring TikTok and its users. Based on our findings, Douyin and TikTok.
we observed about one-third (18/58) of the TikTok Around one-fifth of the studies (11/58) focused on
studies in our dataset focus on users and their behaviors. the effects of TikTok on users individually and as a
This refers specifically to how individual users use the collective. For instance, Ge, Sui, Zhao and Li [S12]
platform and behaviors associated with this use. For studied the effects of short video ads (on sales.
instance, Chen, Min, Zhang, Ma, and Evans [S7] Additionally, in a letter to the editor Ostrovsky and
examined citizen engagement with government TikTok Chen [S35] explained their analysis of COVID-19
accounts during the COVID-19 pandemic, and Bucknell information on the platform and how the demand for
Bossen and Kottasz [S6] explored the Uses and health content was being surpassed by the supply.
Gratifications of using TikTok among young children. Lastly, a little over one-eighth of the studies (8/58)
In contrast, close to half of the studies (26/58) focus focused on TikTok’s platform or policy governance. For
on user culture. Here, user culture refers to the example, Jia and Ruan [S20] compared Chineses apps’
production of cultural norms and practices on the data and privacy governance and Weimann and Masri
platform that are not focused on individual users, but [S50] examined the spread of hate content on TikTok
instead look at users’ collective actions generative of while critiquing the platform for the lack of application
content genres or patterns of behavior on the platform. of their own Terms of Service to prevent spread of such
For instance, Vázquez-Herrero and colleagues [S47] hate. Out of the 58 studies in our data set, only 27
examine how news organizations are adapting to specified the age of their human subject participants - of
TikTok by tailoring their content to fit TikTok’s content these, 17 had adult participants, 7 focused on teens, and
style. Additionally, Zulli and Zulli [S58] extended the only 3 had participants below the age of 12.
literature of meme-related research and argue that
TikTok encourages imitation and replication (memesis), Methods used to explore TikTok
thus making it a basis of sociality on the app. A few Our second research question asked about what
studies (5/58) focused specifically on organizational methods were being employed to study TikTok. As
TikTok accounts, indicating an interest in how TikTok mentioned above, the most commonly employed
is being harnessed for professional and promotional use. methodology was content analysis (23/58). These
Pan and Chi [S36] focused on TikTok accounts of studies largely examined the content of posts on TikTok,
Chinese airlines. Basch, Fera, Pierce and Basch [S2] most commonly by searching for hashtags or keywords
analyzed videos with the hashtag #WearAMask [e.g., S3, S15, S48, S53] and examining top posts
alongside videos put forth by the World Health returned [e.g., S48] or by or focusing on content from
Organization (WHO) in order to explore TikTok’s role specific kinds of TikTok accounts. These studies
in community mitigation of the coronavirus. examined the post content in detail and contemplated
Less than one-tenth of the studies (4/58) focused possible impacts.
on algorithmic and software related research. For Digital methods (12/58), critical analyses (9/58),
instance, both Khoa, Duy, Hoang, Hien and Pham [S18] surveys (11/58), and observational studies (5/58) were
and Domingues, Nogueira, Francisco and Frade [S9] next in frequency. Examples of digital methods utilized
[7]conducted a forensic analysis of the artifacts left included an app walkthrough [e.g., S20, S22, S55] and
behind by TikTok on Android devices. Meanwhile, the use of a web crawler tool [e.g., S7, S15]. Critical
Simpson and Seeman [S41] explored how TikTok’s analyses covered a range of subjects, for example
FYP algorithm constructs contradictory identity spaces possibilities for “playful political participation” on
that support and violate the identities of LGBTQ+ users. TikTok [S48] and how the platform’s affordances give
Ten studies focused on design and interface, which rise to affective publics [S15]. All studies that employed
tended to highlight the affordances provided by the surveys focused on users, particularly in consumption
platform as well as how features were used. Habibi and rather than production roles. All but one of the
Salim [S14] examined TikTok (in addition to Instagram) observational studies used TikTok as a site for exploring
to see how different types of educational science content phenomena, analyzing content depicting relevant
led to more user engagement. Meanwhile, Vijay and material. The final observation study consisted of a
Gekkar [S48] examined how TikTok’s platform design clinical case report exploring an anorexia patient’s use
shapes the way politics is performed on it. Seven studies of TikTok. Lastly, there were no more than a couple
also offered critical analyses or cultural commentaries. interviews (3/58), focus groups (1/58), ethnographies
For instance, Gray [S13] critically analysed the (3/58), and experimental studies (2/58).
geopolitics of TikTok through a content analysis of Due to the novel way content on TikTok gets
government and company sources and Kaye, Chen and pushed to users and the dissimilar affordances of the
platform with regard to potential data collection, we also

Page 3109
examined if the studies utilized the pre-existing While TikTok’s algorithm makes the platform
affordances of the platform to assist with data collection. uniquely appealing to audiences [33], the algorithm is,
Overall, researchers are not harnessing the pre-existing in essence, a black box. We note through our findings
affordances of the platform to collect data. We do, that only a small number of studies (4/58) target
however, note a few exceptions. For instance, Habibi studying TikTok’s algorithm and related software.
and Salim [S14] explored static vs. dynamic message TikTok has endeavored to be transparent about how its
delivery for science communication by creating and algorithm works [21]. In some ways, the platform has
disseminating posts on TikTok (and Instagram). They been more transparent than other platforms, for example
measured the “reach” of individual posts, i.e., its views by promising to allow experts to examine its algorithm’s
and other engagement by posting the videos to the source code [23]. Yet, these transparency efforts cannot
platform and allowing them to reach their audience fully overcome the challenges of studying highly
through the platform’s affordances. This is a novel complex algorithms such as TikTok’s. The relatively
approach that folds in the algorithmic curation and small amount of work exploring TikTok’s algorithm
affordances of the platform, rather than attempting to could indicate that researchers have not yet formulated
reach “all users” or a specified sample even, since this effective strategies for such inquiries. Future work in
is a rough approximation of how content would reach this area might look to past work recommending
audiences on the platform authentically. Additionally, methodological approaches to studying algorithms “in
Schellewald [S38] describes the different forms of the wild” [30,31].
communication on TikTok by using the platforms for Further, while a large volume of our sample were
30-60 minutes each day for 6 months. Individual content analyses, wherein researchers collected content
affordances of the platform were explored, such as from hashtag or search pages, many of these studies did
popular sounds and trends. This approach was employed not substantively discuss how algorithmic curation
keeping in mind the ephemeral nature of the content on impacts the representativeness of their samples.
this platform. TikTok’s algorithm is heavily dependent on a user’s
Aside from these, most studies (including those that profile (e.g., sign-up information, mobile device
examined the platform’s design or affordances as the information, IP address) and user interactions (e.g.,
central focus of the research) utilized traditional liking, commenting on, or sharing videos) (TikTok,
methods such as surveys, interviews, or even content 2019). In past HCI and communication scholarship,
analyses, but there largely was no justification for the researchers have attempted to approximate a ground
methodological decisions or explication for why certain truth experience by creating fresh user accounts
methods were appropriate for specific studies. specifically for the purposes of their studies, sometimes
We also examined the difference in approaches using brand new devices [8]. [S32] implemented this
between HCI and communication research. Based on strategy in exploring how “Booktok” can be integrated
publication outlets, author affiliation, and into librarians’ reader advisory work “so that the
methodological approach, we determined if each study exposure to videos would not be shaped by the
employed primarily an HCI approach or a researcher’s own personal preferences” (p. 3). However,
communication studies approach. Eleven out of 58 this strategy still creates issues for studies seeking
studies were published in predominantly HCI avenues, generalizability. For example, as [S39] noted in their
whereas the remaining fell under the communication study of eating disorder recovery content on TikTok, the
studies umbrella using scholarship from algorithm poses challenges for demonstrating the
communication, journalism, advertising, internet timeliness and relative visibility of certain content on the
studies, and even health behavior studies. The HCI platform. Relatedly, content analyses like this cannot
research (much like communication studies) employed speak to the relative prevalence of different kinds of
a combination of methods and foci. Three out of the four content. For example, while [S50] were able to
studies focusing on algorithms came from the HCI field. demonstrate the presence of far-right extremist content
Overall, the most notable aberrations from past research on TikTok, the authors were careful to avoid
on major social media platforms [16,19,30,36,38] (as quantifications of the magnitude of this presence. As a
represented in our sample) are relatively more reliance platform so heavily dependent on algorithmic curation,
on content analysis and relatively less reliance on it is impossible to create a “default” user profile meant
surveys and computational approaches. to capture an “average” stream of content. This problem
is not wholly unique to TikTok. Other social media such
Challenges and Opportunities as Instagram with its “Top Posts” on its Discover page
Based on our findings we identified a number of share this challenge. However, unlike TikTok,
challenges with studying TikTok and its users as well as Instagram users primarily encounter content via the
a number of opportunities for future work. platform’s main feed, which is generated from accounts

Page 3110
a user follows. Thus, it is easier to estimate exposure to, might prove to be a valuable resource for academic
and perhaps pervasiveness of, content. Researchers research within this arena.
interested in studying questions related to TikTok Additionally, TikTok has a very unique culture, or,
culture, exposure to content, or effects of exposure must more accurately, a unique federation of subcultures. The
take the highly personalized nature of the platform as an platform encourages development of norms and
inherent limitation. Alternatively, the differences in the memetic replication of content through affordances like
platform’s affordances can be harnessed for data the ability to re-use others’ audio and “stitch” replies to
collection in the future, instead of relying on keyword, others’ videos, as [S38] demonstrated. This leads to
hashtag, or top video searches. organic community building across not just a user-
To this end however, while some studies in our generated network, but all users of the platform. This
sample offered useful examples of strategies for has not only been embraced by users, but actively
identifying specific kinds of users and content on encouraged by the platform [34]. TikTok’s multiplicity
TikTok [e.g., S17, S30, S50], these strategies have and rich well of community connections deserves more
limitations related to platform affordances. For attention from both the perspective of content creators
example, many studies that targeted specific hashtags as well as consumers.
for data collection, raising questions about the extent to
which those hashtags can stand in for all content The Ethics of it All
addressing a topic or theme. For example, [S30] Our last research question was about the ethical
explained that TikTok limits the number of search challenges about studying TikTok and its users. The
results for hashtag queries, but they were unsure of the most significant challenge faced by researchers of the
exact limit, as their multiple searches returned different platform is the age of content creators. A very high
numbers of results. Moreover, the researchers went on volume of Tiktok’s users are minors [32]. As noted in
to note that though they attempted to expand their our results, only 23 studies specified the age of
dataset by including “duets”1 with videos returned by participants, with 17 focusing on adults, 7 on teens and
their hashtag queries, TikTok does not offer this search only 3 on people below 12 (a handful of studies had
functionality. Thus, the researchers instead chose to more than one age group e.g., [S6, S42]). This creates a
search by sound,2 which TikTok does afford. At a more discrepancy in the population that most uses the
basic level, researchers’ decisions about which platform and subset of users being studied. Studying
keywords, hashtags, or sounds to search for will shape younger populations is challenging, but if we want to
the data they collect in ways that also limit understand TikTok, its impact, and its user culture, more
generalizability. research should focus on the most prevalent users of the
Finally, there are also geographical challenges. platform. Related to this, 36 articles in our sample did
TikTok is not accessible in all parts of the world (for not use human subjects data, instead relying on data
instance, the platform is banned in India and used as from the platform. Some research did discuss the ethical
Douyin in China). Additionally, not all the affordances challenges that go along with this. For example, a few
and filters of the apps are available everywhere. studies noted that they only used publicly available user
Moreover, as [S17, S53] noted, studies examining data, blurred faces in figures, and only collected raw
English-language hashtags or posts are not data, such as likes and shares, instead of user profile
representative of TikTok’s global user base. Thus, there pictures and handles [S14, S15, S30]. However, this
is no universal TikTok experience, and researchers must gives rise to two ethical challenges - first, a not
take care to acknowledge the variability of use across insignificant amount of data being collected presumably
countries and contextualize findings within the comes from minors. Although publicly available,
geographical settings of their studies. special care and consideration should be given to the
Based on our work we can also note a number of collection, handling, and publication of such data.
opportunities for TikTok research in the future. TikTok Recognizing the majority of their subjects were
is a newer platform as compared to most of the existing minors, one study in our sample [S30] combated this
social media platforms. Our findings indicate an issue by deleting user data after analysis. This is a good
inherited interest in TikTok policy and governance, measure for ensuring the anonymity of minors, but
extending challenges gleaned from research on other leaves other challenges, namely that of consent. Fiesler
platforms. As the platform has endeavored to be more and Proferes [9] showed that few users are
open about their policies and technologies, TikTok knowledgeable about how their content (in this case,
tweets) could be utilized for research, and most felt that

1 Duets are video responses to other videos on the platform, wherein 2 TikTok allows users to make the audio of their videos reusable by
the “responding” video appears side-by-side with the original video. other users

Page 3111
this should not be allowed without consent. The authors to policy and governance, TikTok integrates live
acknowledged that their findings were dependent on content, which is notoriously difficult to moderate and
context among other factors, such as who is conducting control [12]. In our dataset, surprisingly few articles
and disseminating the research and why. In the context addressed questions about users’ rights or content
of TikTok, one study we reviewed [S17] noted that the moderation. Still, this could be more indicative of the
terms of service accepted by users when they join the pace of academic publishing and scope of our dataset
app can be considered consent for collecting user data than a shortcoming of extant TikTok research.
for research. However, users’ awareness and In addition, we acknowledge certain limitations of
understanding of this point requires additional empirical our research. We primarily targeted HCI and
exploration. communication databases and journals. Certainly, other
related fields will have relevant research that we did not
5. Discussion & Future Directions assess. TikTok is still a newer platform and research
within this space is still on-going. An update to this kind
As an early analysis of the emerging research on of systematic review would be useful in the future and
TikTok, our systematic review was limited by a encouraged to gain a better understanding of how the
relatively small sample, but we believe this systematic research has evolved. Another limitation is that we
review is a good baseline for tracking scholarly interest included only studies that were in English, which is only
in the platform over time. In our analysis, we generally a small portion of the worldwide academic research
observed various common threads from past research on available. Other valuable research on TikTok has been
other platforms. For example, like past research on conducted in other languages, and these works should
Facebook, Instagram, and Twitter [6,19,36], our sample be included in future research on the topic.
emphasized general inquiries about how and why people
use TikTok. We also noticed some oft-repeated trends 6. Conclusion
in social media research that present unique challenges
and opportunities for the exploration of research In this study, we conducted a systematic analysis of
questions related to TikTok and the evolving social research on TikTok in the HCI and communication
media landscape itself. In recent years, restrictions on disciplines to understand how scholarly research is
social media platforms’ APIs, including TikTok’s, make approaching the study of this increasingly popular social
it difficult to collect data. In our analysis, the few studies media platform. Our results indicated that most research
that mentioned scraping data from TikTok did not fully on the app focuses on the emergent culture on the app
articulate the mechanisms by which they accomplished (i.e. the organic norms and trends by users).
this. Currently, TikTok's API offers highly limited Furthermore, researchers also demonstrated some
access to platform data, which has meant a greater interest in the policy and platform governance of
emphasis on third-party tools (e.g., [S53] used a web TikTok, an important area with TikTok’s promises for
crawler to access publicly available metadata for future transparency and the high volumes of complex
TikTok videos). Most, however, used more traditional forms of content created (and viewed) by minors.
methods such as keyword searches, hashtags, and data Methodologically, most of the research employed
collection from the For You page. This points to a lack content analysis, indicating an interest in studying
of a standardized way to collect data from TikTok and creative output and expression on the platform.
study its content. Future research should explore ways However, because of the unique centrality of TikTok’s
to systematically approach this issue, since this problem algorithmically curated FYP, the methods used raise
is not unique to the TikTok platform (e.g. Instagram’s questions about how systematic such data collection is
“top results'' vary from user to user). and what could be done to standardize such exploration
TikTok has become a fierce competitor in the social in the future. Finally, from an ethical standpoint,
media world, more research on its policy and studying TikTok raises multiple challenges, the biggest
governance is critically important. For example, as we being its popularity among minors. Because of this,
were preparing this manuscript, news broke that TikTok much of the content on the platform is created and
would start collecting biometric data from its users in viewed by minors. This presents challenges for
the form of faceprints and voiceprints [24], raising researchers who must either collect primary data from
further questions about data collection and privacy on minors or attempt to exclude this important subset of
the app. Interestingly, while past social media research TikTok users. Further, in studies using publicly
has emphasized concerns related to information available TikTok data, the ethical concerns of whether
disclosure, privacy, and security [6,39], in our dataset, the content creators are aware of their data being used
only two studies [S21, S20] focused on these topics. As for research arises. Future work should particularly
another example of the need for more research related focus on the ethical challenges faced while studying

Page 3112
TikTok as a platform, and not conflate it with other of platform governance. Big Data & Society 7, 1
social media platforms due to its unique affordances. (January 2020), 2053951719897945.
[13] Eszter Hargittai. 2020. Potential Biases in Big Data:
Omitted Voices on Social Media. Social Science
7. References Computer Review 38, 1 (February 2020), 10–24.
[14] Eszter Hargittai and Gina Walejko. 2008. THE
[1] Mohamed Abdalla and Moustafa Abdalla. 2021. The PARTICIPATION DIVIDE: Content creation and
Grey Hoodie Project: Big Tobacco, Big Tech, and the sharing in the digital age. Information,
threat on academic integrity. arXiv:2009.13676 [cs] Communication & Society 11, 2 (March 2008), 239–
(April 2021). 256.
[2] Giovanni Boccia Artieri. 2017. Social Media and the DOI:https://doi.org/10.1080/13691180801946150
Challenge of Big Data/Deep Data Approach. In Data [15] Jennifer Hassan and Ruby Mellen. 2020. It’s not just
Science and Social Research. Springer, 57–65. the United States: These governments also see TikTok
[3] Paul Atkinson and Martyn Hammersley. 1994. as a problem. Washington Post. Retrieved September
Ethnography and participant observation. In 4, 2021 from
Handbook of Qualitative Research. Sage Publications, https://www.washingtonpost.com/world/2020/08/03/it
Inc, Thousand Oaks, CA, US, 248–261. s-not-just-united-states-these-governments-see-tiktok-
[4] Bridget Barrett and Daniel Kreiss. 2019. Platform growing-problem/
transience: changes in Facebook’s policies, [16] Kawaljeet Kaur Kapoor, Kuttimani Tamilmani,
procedures, and affordances in global electoral Nripendra P. Rana, Pushp Patil, Yogesh K. Dwivedi,
politics. Internet Policy Review 8, 4 (December 2019). and Sridhar Nerur. 2018. Advances in Social Media
Retrieved June 14, 2021 from Research: Past, Present and Future. Inf Syst Front 20,
https://policyreview.info/articles/analysis/platform- 3 (June 2018), 531–558.
transience-changes-facebooks-policies-procedures- [17] Rob Kitchin. 2017. Thinking critically about and
and-affordances-global researching algorithms. Information, Communication
[5] danah boyd and Kate Crawford. 2012. Critical & Society 20, 1 (January 2017), 14–29.
Questions for Big Data. Information, Communication [18] Joshua A. Kroll. 2018. The fallacy of inscrutability.
& Society 15, 5 (June 2012), 662–679. Philosophical Transactions of the Royal Society A:
DOI:https://doi.org/10.1080/1369118X.2012.678878 Mathematical, Physical and Engineering Sciences
[6] Ralf Caers, Tim De Feyter, Marijke De Couck, Talia 376, 2133 (November 2018).
Stough, Claudia Vigna, and Cind Du Bois. 2013. [19] Linnea Laestadius. 2016. Instagram. In The SAGE
Facebook: A literature review. New Media & Society Handbook of Social Media Research Methods, Luke
15, 6 (September 2013), 982–1002. Sloan and Anabel Quan-Haase (eds.). SAGE, London,
[7] Patricio Domingues, Ruben Nogueira, José Carlos UK, 573–592.
Francisco, and Miguel Frade. 2020. Post-mortem [20] Ben Light, Jean Burgess, and Stefanie Duguay. 2018.
digital forensic artifacts of TikTok Android App. In The walkthrough method: An approach to the study of
Proceedings of the 15th International Conference on apps. New Media & Society 20, 3 (March 2018), 881–
Availability, Reliability and Security (ARES ’20), 900.
Association for Computing Machinery, New York, [21] Louise Matsakis. 2020. How TikTok’s “For You”
NY, USA, 1–8. Algorithm Actually Works. Wired. Retrieved June 13,
[8] William H. Dutton, Bianca Reisdorf, Elizabeth 2021 from https://www.wired.com/story/tiktok-
Dubois, and Grant Blank. 2017. Search and Politics: finally-explains-for-you-algorithm-works
The Uses and Impacts of Search in Britain, France, [22] McGlew. 2020. This is How the TikTok Algorithm
Germany, Italy, Poland, Spain, and the United States. Works. Later Blog. Retrieved June 15, 2021 from
Social Science Research Network, Rochester, NY. https://later.com/blog/tiktok-algorithm/
DOI:https://doi.org/10.2139/ssrn.2960697 [23] Pappas. 2019. TikTok to launch Transparency Center
[9] Casey Fiesler and Nicholas Proferes. 2018. for moderation and data practices. TikTok Newsroom.
“Participant” Perceptions of Twitter Research Ethics. Retrieved June 14, 2021 from
Social Media + Society 4, 1 (January 2018). https://newsroom.tiktok.com/en-us/tiktok-to-launch-
[10] aline shakti franzke, Anja Bechmann, Michael transparency-center-for-moderation-and-data-
Zimmer, Charles Ess, and Association of Internet practices
Researchers. 2020. Internet Research: Ethical [24] Sarah Perez. 2021. TikTok just gave itself permission
Guidelines 3.0. Association of Internet Researchers. to collect biometric data on US users, including
Retrieved from https://aoir.org/reports/ethics3.pdf ‘faceprints and voiceprints.’ TechCrunch. Retrieved
[11] Annabelle Gawer. 2014. Bridging differing June 14, 2021 from
perspectives on technological platforms: Toward an https://social.techcrunch.com/2021/06/03/tiktok-just-
integrative framework. Research Policy 43, 7 gave-itself-permission-to-collect-biometric-data-on-u-
(September 2014), 1239–1249. s-users-including-faceprints-and-voiceprints/
[12] Robert Gorwa, Reuben Binns, and Christian [25] Cornelius Puschmann. 2019. An end to the wild west
Katzenbach. 2020. Algorithmic content moderation: of social media research: a response to Axel Bruns.
Technical and political challenges in the automation Information, Communication & Society 22, 11

Page 3113
 (September 2019), 1582–1589. Using behavioral trace data against platform episteme.
[26] A Schellewald. 2021. Communicative Forms on New Media & Society (July 2020),
TikTok: Perspectives From Digital Ethnography. [41] Michael Zimmer. 2010. “But the data is already
International Journal of Communication 15, (2021), public”: on the ethics of research in Facebook. Ethics
1437–1457. Inf Technol 12, 4 (December 2010), 313–325.
[27] Jen Schradie. 2011. The digital production gap: The [42] Michael Zimmer. 2018. Addressing Conceptual Gaps
digital divide and Web 2.0 collide. Poetics 39, 2 in Big Data Research Ethics: An Application of
(April 2011), 145–168. Contextual Integrity. Social Media + Society 4, 2
[28] Nick Seaver. 2017. Algorithms as culture: Some (April 2018).
tactics for the ethnography of algorithmic systems.
Big Data & Society 4, 2 (December 2017).
[29] Katie Sehl. 2021. 23 Important TikTok Stats
Marketers Need to Know in 2021. Hootsuite Blog.
Retrieved June 14, 2021 from
https://blog.hootsuite.com/tiktok-stats/
[30] Chareen L. Snelson. 2016. Qualitative and Mixed
Methods Social Media Research: A Review of the
Literature. International Journal of Qualitative
Methods 15, 1 (December 2016).
[31] Victoria Song. 2021. Instagram, Honey, Please Stop
Copying TikTok. Gizmodo. Retrieved June 14, 2021
from https://gizmodo.com/instagram-honey-please-
stop-copying-tiktok-1846190644
[32] Statista. 2021. U.S. TikTok users by age 2021.
Statista. Retrieved June 14, 2021 from
https://www.statista.com/statistics/1095186/tiktok-us-
users-age/
[33] Joe Tidy and Sophia Smith Galer. 2020. TikTok: The
story of a social media giant. BBC News. Retrieved
June 14, 2021 from
https://www.bbc.com/news/technology-53640724
[34] TikTok. 2021. Subcultures are the new demographics.
TikTok for Business. Retrieved June 14, 2021 from
https://www.tiktok.com/business/en-
US/blog/subcultures-are-the-new-demographics
[35] Leanne Townsend and Claire Wallace. 2016. Social
Media Research: A Guide to Ethics. The University of
Aberdeen.
[36] Jessica Vitak. 2016. Facebook as a Research Tool in
the Social and Computer Sciences. In The SAGE
Handbook of Social Media Research Methods, Luke
Sloan and Anabel Quan-Haase (eds.). SAGE, London,
UK, 627–642.
[37] Jessica Vitak, Katie Shilton, and Zahra Ashktorab.
2016. Beyond the Belmont Principles: Ethical
Challenges, Practices, and Beliefs in the Online Data
Research Community. Proceedings of the 19th ACM
Conference on Computer Supported Cooperative
Work and Social Computing (February 2016).
DOI:https://doi.org/10.1145/2818048.2820078
[38] Shirley A. Williams, Melissa M. Terras, and Claire
Warwick. 2013. What do people study when they
study Twitter? Classifying Twitter related academic
papers. Journal of Documentation 69, 3 (May 2013),
384–410.
[39] Robert E. Wilson, Samuel D. Gosling, and Lindsay T.
Graham. 2012. A Review of Facebook Research in the
Social Sciences. Perspect Psychol Sci 7, 3 (May
2012), 203–220.
[40] Angela Xiao Wu and Harsh Taneja. 2020. Platform
enclosure of human behavior and its measurement:

Page 3114
 Appendix A: List of Studies of Medical Internet Research 23(2), 2021, pp.
1-13.
LIST OF ARTICLES
[S8] Chen, Z., Q. He, Z. Mao, H.-M. Chung,
[S1] Avdeeff, M.K., “TikTok, Twitter, and and S. Maharjan, “A Study on The
Platform-Specific Technocultural Discourse Characteristics of Douyin Short Videos And
in Response to Taylor Swift’s LGBTQ+ Implications For Edge Caching””,
Allyship in ‘You Need to Calm Down”, Proceedings of the ACM Turing Celebration
Contemporary Music Review, 2021, pp. 1–21. Conference - China, ACM (2019), pp. 1–6.

[S2] Basch, C.H., J. Fera, I. Pierce, and C.E. [S9] Domingues, P., R. Nogueira, J.C.
Basch, “Promoting Mask Use on TikTok: Francisco, and M. Frade, “Post-mortem
Descriptive, Cross-sectional Study”, JMIR Digital Forensic Artifacts of TikTok Android
Public Health and Surveillance 7(2), 2021, App”, Proceedings of the 15th International
pp. 1-7. Conference on Availability, Reliability and
Security, ACM (2020), pp. 1–8.
[S3] Basch, C.H., J. Fera, and N. Quinones,
“A Content Analysis of Direct-To-Consumer [S10] Du, X., T. Liechty, C.A. Santos, and J.
DNA Testing on TikTok”, Journal of Park, “I Want to Record and Share My
Community Genetics 12(3), 2021, pp. 489– Wonderful Journey’: Chinese Millennials’
492. Production and Sharing of Short-Form Travel
Videos on TikTok or Douyin”, Current Issues
[S4] Basch, C.H., Z. Meleo-Erwin, J. Fera, C. in Tourism, 2020, pp. 1–13.
Jaime, and C.E. Basch, “A Global Pandemic
in The Time of Viral Memes: COVID-19 [S11] Francisco, M.E.Z., and S. Ruhela,
Vaccine Misinformation and Disinformation “Investigating TikTok as an AI User
on TikTok”, Human Vaccines & Platform”, 2021 2nd International
Immunotherapeutics, 2021, pp. 1–5. Conference on Computation, Automation and
Knowledge Management (ICCAKM), IEEE
[S5] Boatman, D.D., S. Eason, M.E. Conn, (2021), pp. 293–298.
and S.K. Kennedy-Rea, “Human
Papillomavirus Vaccine Messaging on [S12] Ge, J., Y. Sui, X. Zhou, and G. Li,
TikTok: Social Media Content Analysis”, “Effect of Short Video Ads on Sales Through
Health Promotion Practice, 2021, pp. 1-6. Social Media: The Role of Advertisement
Content Generators”, International Journal of
[S6] Bucknell Bossen, C., and R. Kottasz, Advertising 40(6), 2021, pp. 870–896.
“Uses and Gratifications Sought by Pre-
Adolescent and Adolescent Tiktok [S13] Gray, J.E., “The Geopolitics Of
Consumers”, Young Consumers 21(4), 2020, ‘Platforms’: The TikTok Challenge”, Internet
pp. 463–478. Policy Review 10(2), 2021, pp. 1-18.

[S7] Chen, Q., C. Min, W. Zhang, X. Ma, and [S14] Habibi, S.A., and L. Salim, “Static vs.
R. Evans, “Factors Driving Citizen Dynamic Methods of Delivery for Science
Engagement with Government TikTok Communication: A Critical Analysis of User
Accounts During the COVID-19 Pandemic: Engagement with Science on Social Media”,
Model Development and Analysis”, Journal PLOS ONE 16(3), 2021, pp. 1-15.

Page 3115
[S15] Hautea, S., P. Parks, B. Takahashi, and Mediation”, Internet Research ahead-of-
J. Zeng, “Showing They Care (Or Don’t): print(ahead-of-print), 2021, pp. 1-23.
Affective Publics and Ambivalent Climate
Activism on TikTok”, Social Media + Society [S22] Kaye, D.B.V., X. Chen, and J. Zeng,
7(2), 2021, pp. 1-14. “The Co-Evolution of Two Chinese Mobile
Short Video Apps: Parallel Platformization of
[S16] Hayes, C., K. Stott, K.J. Lamb, and Douyin and TikTok”, Mobile Media &
G.A. Hurst, “‘Making Every Second Count’: Communication 9(2), 2020, pp. 229–253.
Utilizing TikTok and Systems Thinking to
Facilitate Scientific Public Engagement and [S23] Klug, D., Y. Qin, M. Evans, and G.
Contextualization of Chemistry at Home”, Kaufman, “Trick and Please. A Mixed-
Journal of Chemical Education 97(10), 2020, Method Study on User Assumptions about the
pp. 3858–3866. TikTok Algorithm”, 13th ACM Web Science
Conference 2021, ACM (2021), pp. 84–92.
[S17] Herrick, S.S.C., L. Hallward, and L.R.
Duncan, “‘This is Just How I Cope’: An [S24] Li, Y., X. Xu, B. Song, and H. He,
Inductive Thematic Analysis Of Eating “Impact of Short Food Videos on the Tourist
Disorder Recovery Content Created And Destination Image—Take Chengdu as an
Shared on TikTok using # Example”, Sustainability 12(17), 2020, pp. 1-
EDRECOVERY””, International Journal of 13.
Eating Disorders 54(4), 2021, pp. 516–526.
[S25] Literat, I., “On Research and Hope, in
[S18] Hoang Khoa, N., P. The Duy, H. Do an America Aflame: Sketching Youth Civic
Hoang, D. Thi Thu Hien, and V.-H. Pham, Futures as a Mother and a Researcher”,
“Forensic Analysis of TikTok Application to Journal of Children and Media 15(1), 2021,
Seek Digital Artifacts on Android pp. 109–111.
Smartphone”, 2020 RIVF International
Conference on Computing and [S26] Litzinger, R., and Y. Ni, “Inside the
Communication Technologies (RIVF), IEEE Wuhan Cabin Hospital: Contending
(2020), 1–5. Narratives During the COVID-19 Pandemic”,
China Information, 2021, pp. 1-20.
[S19] Iodice, R., and C. Papapicco, “To be a
TikToker in COVID-19 Era: An Experience [S27] Logrieco, G., M.R. Marchili, M.
of Social Influence”, Online Journal of Roversi, and A. Villani, “The Paradox of Tik
Communication and Media Technologies Tok Anti-Pro-Anorexia Videos: How Social
11(1), 2021, pp. 1-12. Media Can Promote Non-Suicidal Self-Injury
and Anorexia”, International Journal of
[S20] Jia, L., and L. Ruan, “Going global: Environmental Research and Public Health
Comparing Chinese Mobile Applications’ 18(3), 2021, pp. 1-4.
Data and User Privacy Governance at Home
and Abroad”, Internet Policy Review, 2020, [S28] Ma, L., J. Feng, Z. Feng, and L. Wang,
pp. 1-22. “Research on User Loyalty of Short Video
App Based on Perceived Value—Take Tik
[S21] Kang, H., W. Shin, and J. Huang, Tok as an Example”, 2019 16th International
“Teens’ Privacy Management on Video- Conference on Service Systems and Service
Sharing Social Media: The Roles of Management (ICSSSM), IEEE (2019), 1–6.
Perceived Privacy Risk and Parental

Page 3116
[S29] Masciantonio, A., D. Bourguignon, P. of Chinese Airline”, 2020 IEEE 2nd
Bouchat, M. Balty, and B. Rimé, “Don’t Put International Conference on Civil Aviation
All Social Network Sites in One Basket: Safety and Information Technology
Facebook, Instagram, Twitter, TikTok, and (ICCASIT, IEEE (2020), pp. 936–939.
Their Relations with Well-Being During the
COVID-19 Pandemic”, PLOS ONE 16(3), [S37] Rogers, R., “Marginalizing the
2021, pp. 1-14. Mainstream: How Social Media Privilege
Political Information”, Frontiers in big Data
[S30] Medina Serrano, J.C., O. 4, 2021, pp. 1-10.
Papakyriakopoulos, and S. Hegelich,
“Dancing to the Partisan Beat: A First [S38] Schellewald, A., “Communicative
Analysis of Political Communication on Forms on TikTok: Perspectives from Digital
TikTok”, 12th ACM Conference on Web Ethnography”, International Journal of
Science, ACM (2020), pp. 257–266. Communication; Vol 15 (2021), 202, pp.
1437-1457.
[S31] Meng, K.S., and L. Leung, “Factors
Influencing TikTok Engagement Behaviors in [S39] Scherr, S., and K. Wang, “Explaining
China: An Examination of Gratifications the Success of Social Media with
Sought, Narcissism, and the Big Five Gratification Niches: Motivations Behind
Personality Traits”, Telecommunications Daytime, Nighttime, and Active Use of
Policy 45(7), 2021, pp. 1-15. TikTok in China”, Computers in Human
Behavior 124, 2021, pp. 1-9.
[S32] Merga, M.K., “How Can BookTok on
TikTok Inform Readers’ Advisory Services [S40] Shuhan Gu, J. Ping, M. Xu, and Y.
for Young People?”, Library & Information Zhou, “TikTok browsing for anxiety relief in
Science Research 43(2), 2021, pp. 1-10. the preoperative period: A randomized
clinical trial”, Complementary Therapies in
[S33] Ni, Y., M. Fabbri, C. Zhang, and K.A. Medicine 60, 2021, pp. 1-5.
Stewart, “Reciprocity in Quarantine:
Observations from Wuhan’s COVID-19 [S41] Simpson, E., and B. Semaan, “For You,
Digital Landscapes”, Asian Bioethics Review or For 'You’?: Everyday LGBTQ+
12(4), 2020, pp. 435–457. Encounters with TikTok”, Proceedings of the
ACM on Human-Computer Interaction
[S34] Nouwen, M., and M. Hermine Christine 4(CSCW3), 2021, pp. 1–34.
Marie Ghislaine D, “TikTok as a Data
Gathering Space: The Case of Grandchildren [S42] Song, S., Y.C. Zhao, X. Yao, Z. Ba,
and Grandparents During The COVID-19 and Q. Zhu, “Short video apps as a Health
pandemic”, Interaction Design and Children, Information Source: An Investigation of
ACM (2021), pp. 498–502. Affordances, User Experience and Users’
Intention to Continue the Use of TikTok”,
[S35] Ostrovsky, A.M., and J.R. Chen, Internet Research ahead-of-print(ahead-of-
“TikTok and Its Role in COVID-19 print), 2021.
Information Propagation”, Journal of
Adolescent Health 67(5), 2020, pp. 730. [S43] Southwick, L., S.C. Guntuku, E.V.
Klinger, E. Seltzer, H.J. McCalpin, and R.M.
[S36] Pan, C., and R. Chi, “Analysis and Merchant, “Characterizing COVID-19
Research on Operation of Tik Tok Accounts Content Posted to TikTok: Public Sentiment

Page 3117
and Response During the First Phase of the [S51] Yang, M., S. Hu, B.E. Kpandika, and L.
COVID-19 Pandemic”, Journal of Adolescent Liu, “Effects of Social Attachment on Social
Health 69(2), 2021, pp. 234–241. Media Continuous Usage Intention: The
Mediating Role of Affective Commitment”,
[S44] Su, Y., B.J. Baker, J.P. Doyle, and M. Human Systems Management(Preprint), , pp.
Yan, “Fan Engagement in 15 seconds: 1–13.
Athletes’ Relationship Marketing During a
Pandemic via TikTok”, International Journal [S52] Yu, Z., “An Empirical Study of
of Sport Communication 13(3), 2020, pp. Consumer Video Activism in China:
436–446. Protesting Against Businesses with Short
Videos”, Chinese Journal of Communication
[S45] Sun, L., H. Zhang, S. Zhang, and J. 14(3), 2021, pp. 297–312.
Luo, “Content-based Analysis of the Cultural
Differences between TikTok and Douyin”, [S53] Zeng, J., and C. Abidin, “‘#OkBoomer,
2020 IEEE International Conference on Big Time To Meet The Zoomers’: Studying The
Data (Big Data), IEEE (2020), pp. 4779– Memefication of Intergenerational Politics on
4786. TikTok”, Information, Communication &
Society, 2021, pp. 1–23.
[S46] Tang, X., W. Zou, Z. Hu, and L. Tang,
“The Recreation of Gender Stereotypes in [S54] Zhang, Z., “Infrastructuralization of Tik
Male Cross-Dressing Performances on Tok: Transformation, Power Relationships,
Douyin”, Journal of Broadcasting & And Platformization of Video Entertainment
Electronic Media, 2021, pp. 1–19. in China”, Media, Culture & Society 43(2),
2020, pp. 219–236.
[S47] Vázquez-Herrero, J., M.-C. Negreira-
Rey, and X. López-García, “Let’s Dance the [S55] Zhen Ye, “Streamer-Interface-Viewer
News! How the News Media are Adapting to Entanglement in Popular Chinese Social
the Logic of TikTok”, Journalism, 2020, pp. Media Apps: An Analysis of the Discursive
1-19. and Affective Live-Streaming Chatroom
Interfaces”, MedieKultur: Journal of Media
[S48] Vijay, D., and A. Gekker, “Playing & Communication Research 37(70), 2021,
Politics: How Sabarimala Played Out on pp. 131–150.
TikTok”, American Behavioral Scientist
65(5), 2021, pp. 712–734. [S56] Zheng, D.X., A.Y. Ning, M.A.
Levoska, L. Xiang, C. Wong, and J.F. Scott,
[S49] Wang, Y., “Humor and Camera View “Acne and Social Media: A Cross‐Sectional
on Mobile Short-Form Video Apps Influence Study of Content Quality on Tiktok”,
User Experience and Technology-Adoption Pediatric Dermatology 38(1), 2021, pp. 336–
Intent, an Example of TikTok (DouYin)”, 338.
Computers in Human Behavior 110, 2020, pp.
1-9. [S57] Zhu, C., X. Xu, W. Zhang, J. Chen, and
R. Evans, “How Health Communication via
[S50] Weimann, G., and N. Masri, “Research Tik Tok Makes a Difference: A Content
Note: Spreading Hate on TikTok”, Studies in Analysis of Tik Tok Accounts Run by
Conflict & Terrorism, 2020, pp. 1–14. Chinese Provincial Health Committees”,
International Journal of Environmental

Page 3118
Research and Public Health 17(1), 2019, pp. on the TikTok Platform”, New Media &
192. Society, 2020, pp. 1-19.

[S58] Zulli, D., and D.J. Zulli, “Extending the

Internet Meme: Conceptualizing
Technological Mimesis and Imitation Publics

Page 3119
 Appendix B
Results by category
Categories are not mutually exclusive, excepting Human Subjects Research and IRB/Ethics Discussed

Category Code Total

TikTok 47
App Focus Douyin 15
Both apps 4
Primary Research 58
Type of Study
Secondary Data Analysis 0
Survey 11
Interviews 3
Participatory 0
Experimental 2
Focus Groups 1
Methodology Observational 5
Ethnographic 3
Computational 5
Digital Methods 12
Content Analysis 23
Critical/Cultural Analysis 9

Children (12 and under) 3

Age Teens 7
Adults 17
User Behavior 18
User Culture 26
Organizations 5
Study Focus
Algorithm/Software 4
Design and Interface 10
Critical Analysis/Cultural Commentary 9

Page 3120
 Media Effects/Media Psychology 11
Platform/Policy Governance 8
Yes 22
Human Subjects Research
No 36
Yes 18
IRB/Ethics Discussed
No 40

HCI 11
Discipline
Communication 47

Page 3121