Methodological and Ethical Considerations in LGBTQ-Focused Online Ethnographic Research: A Systematic Review and a Human-Centered Framework

Abstract

Online ethnography is a qualitative research approach particularly effective for studying marginalized or vulnerable communities that are often difficult for researchers to access. This approach is especially relevant for LGBTQ communities, where members often form connections and build communities online. To advance current understanding of online ethnography for LGBTQ populations, this systematic review collected and analyzed 62 Web of Science-indexed LGBTQ-focused online ethnographic studies that directly involve LGBTQ individuals as participants or explicitly examine LGBTQ-related issues. Specifically, this systematic review examines and discusses seven aspects of methodological practices and four ethical considerations. The seven methodological aspects cover populations, online sites, sampling, data collection, data analysis, researcher positionality and reflexivity, and methods for ensuring research rigor. The four ethical considerations include informed consent, respect and trust, privacy and confidentiality, and data security and reuse. Based on the review findings, the authors propose a human-centered framework for conducting LGBTQ-focused online ethnographic research. The authors also identify gaps in current LGBTQ-focused online ethnographic research and propose future directions to enhance methodological rigor, ethical compliance, and the beneficial impact of online ethnography for LGBTQ populations.

Introduction

Ethnography is a qualitative research approach focused on understanding people and their cultures within specific spatial and temporal contexts (Hammersley & Atkinson, 2007; Van Maanen, 2006). It involves prolonged engagement with communities through participant observation and interviews to explore shared behaviors, language, and interactions in natural settings (Creswell & Creswell, 2018; Creswell & Poth, 2017). Through ethnography, researchers can provide rich and contextual insights into social practices and experiences (Hammersley & Atkinson, 2007; Khoo et al., 2012).

The advancement of information and communication technologies, especially the widespread use of social media, has transformed ethnographic research by expanding its scope from physical to digital spaces (Pink et al., 2015; L. Robinson & Schulz, 2009; Wittel, 2000). Researchers have named the ethnography in digital spaces online ethnography (this review uses online ethnography to refer to this method), digital ethnography, virtual ethnography, netnography, cyber ethnography, and Internet ethnography (Kozinets, 2020; Tunçalp & L. Lê, 2014; Udenze, 2019). Online ethnography has gained popularity across diverse fields, such as cultural studies, nursing, education, and tourism (Jensen et al., 2022; Nascimento et al., 2022; Salzmann-Erikson & Eriksson, 2023; Tavakoli & Wijesinghe, 2019; Whalen, 2018; Yu & Xu, 2024; Yu & Zhao, 2021).

Online ethnography is particularly effective for exploring online communities of marginalized or vulnerable populations, where data are naturally generated and are often abundant and accessible (Prior & Miller, 2012; L. Robinson & Schulz, 2009). The Internet has enabled online spaces that empower individuals from marginalized and underrepresented groups, including LGBTQ communities (Gruzd & Haythornthwaite, 2013; Hawkins & Watson, 2017; Mehra et al., 2004; Waterton, 2010). Compared to their heterosexual counterparts, LGBTQ individuals are more likely to seek information, establish social connections, and build communities in online spaces (Brandt & Carmichael, 2020; McInroy et al., 2019). This makes online ethnography an effective means to study LGBTQ topics by capturing the lived experiences, challenges, and resilience of LGBTQ individuals in online contexts.

Although previous literature reviews have addressed LGBTQ-related topics, such as cyberbullying, social media use, and health and well-being (Abreu & Kenny, 2018; Berger et al., 2022; Gahagan & Colpitts, 2017) or the use of online ethnography across disciplines (Jensen et al., 2022; Nascimento et al., 2022; Salzmann-Erikson & Eriksson, 2023; Tavakoli & Wijesinghe, 2019; Whalen, 2018), there is a lack of reviews on the intersection of these two areas. This study aims to fill this gap by conducting a systematic review on LGBTQ-focused online ethnographic research, which refers to empirical online ethnographic studies that directly involve LGBTQ individuals as participants or explicitly examine LGBTQ-related issues. An overarching question is proposed to guide this review: How have researchers approached methodological and ethical considerations in LGBTQ-focused online ethnography? Specifically, this study addresses two research questions (RQs):

• RQ1: How have researchers applied research methods (e.g., participant sampling and recruitment, data collection, and data analysis) in LGBTQ-focused online ethnography?

• RQ2: What ethical concerns have been identified and discussed in LGBTQ-focused online ethnography?

Methods

Systematic reviews are well-suited for synthesizing knowledge within a field, answering questions that require the examination of multiple studies, and identifying future research priorities (Page et al., 2021). A systematic literature review was conducted to understand the methodological approaches and ethical concerns in online ethnographic studies concerning LGBTQ populations.

Search Strategy

The articles reviewed are indexed in the Web of Science Core Collection, which covers multidisciplinary and international peer-reviewed journals, books, and conference proceedings published since 1985 (Clarivate, n.d.). The search query included two core concepts: LGBTQ populations and online ethnography. Recognizing the variant expressions for these two concepts, the research team consulted literature to identify relevant search terms and crafted the query.

Descriptors for LGBTQ populations were based on terminology from the U.S. Centers for Disease Control and Prevention (CDC, 2024), the Homosaurus international LGBTQ+ vocabulary (Homosaurus, n.d.), and relevant scholarly literature (Abreu et al., 2022; Abreu & Kenny, 2018; Berger et al., 2022; Gahagan & Colpitts, 2017; Walch et al., 2020). Variant expressions of online ethnography were identified in relevant literature (Jensen et al., 2022; Kozinets, 2020; Tavakoli & Wijesinghe, 2019; Tunçalp & L. Lê, 2014). The search was conducted on March 8, 2024, using the query in Table 1. Three inclusion criteria were applied: (1) full text available, (2) published in English, and (3) peer-reviewed journal articles including early publications that were accepted, and published online. The search retrieved 85 results.

Table 1.

Search Query

Searched the Topic (TS), which covers title, abstract, keyword plus, and author keywords

TS = ((“lesbian*” OR “gay*” OR “men who have sex with men” OR “MSM” OR “bisexual*” OR “transgender” OR “transsexual*”OR “homosexual*” OR “intersex*” OR “asexual*” OR “agender” OR “two-spirit” OR “polysexual*” OR “pansexual*” OR “sapphic” OR “LGB*” OR “GLB*” OR “queer” OR “sexual minorit*” OR “gender minorit*” OR “non-conforming” OR “nonconforming” OR “non-binary” OR “nonbinary”) AND (“online ethnograph*” OR “digital ethnograph*” OR “virtual ethnograph*” OR “internet ethnograph*” OR “cyber ethnograph*” OR “netnograph*” OR “web-based ethnograph*” OR “computer-mediated ethnograph*” OR “digital anthropology” OR “techno-anthropology” OR “cyber-anthropology” OR “virtual anthropology”))

Screening

Following the search, three phases of screening were conducted. The first phase involved abstract-level screening. After reviewing 85 abstracts, 21 records were excluded due to duplication (1), document type (7), and LGBTQ communities not being a significant focus of the study (13), resulting in 64 eligible articles. Being LGBTQ-focused was the fourth inclusion criterion, adding to the three above-mentioned criteria. LGBTQ-focused studies refer to research that involves LGBTQ individuals as participants or stakeholders (e.g., leaders and volunteers for LGBTQ resource services) or studies that analyze policies, content, or issues that are pertinent to LGBTQ communities.

The second screening phase consisted of full-text pilot coding. During this phase, all research team members reviewed the full texts of a random sample of 11 articles. The fifth inclusion criterion—empirical research employing online ethnography as one of the research methods—was added. As a result, one article was excluded.

The third phase was full-text coding. The first and second authors examined the full texts of the remaining 53 articles. One article published in Spanish with an English abstract was identified and excluded, yielding 62 articles included in the content analysis. Figure 1 presents the PRISMA flowchart of the review process.

Figure 1.

Review process presented in the PRISMA 2020 flowchart

Data Analysis

There were three stages of data analysis. In the first stage, the research team created a literature-informed codebook that covered primary methodological and ethical aspects in qualitative research (Hair et al., 2023; Kozinets, 2020; Liamputtong, 2006). At this stage, all team members independently applied the initial codebook to analyze 11 randomly selected articles for pilot coding. During the pilot coding, the following data were extracted: (1) study information (publication year, title), (2) methodological considerations and relevant literature excerpts, and (3) ethical considerations and relevant literature excerpts. Based on the pilot coding results, the team assessed the applicability and clarity of the codes and revised the codebook iteratively. Codes that were less frequently represented in the literature were removed (e.g., parental consent), and emergent codes that capture recurring themes (e.g., LGBTQ people in general or participants with diverse sexual and gender identities) were added. Code definitions were clarified. Ultimately, the codebook (Table 2) consists of two groups of codes mapping to RQ1 and RQ2: methodological considerations and ethical considerations.

Table 2.

Codebook for Methodological and Ethical Considerations

Methodological considerations
Code category	Definition and examples
Populations	Involvement of human participants and the specific populations studied. Examples include lesbian, gay, bisexual, transgender, or LGBTQ people or people with diverse sexual or gender identities in general
Online sites	The online platform(s) examined in a study and their access level. Examples include facebook, X (formerly twitter), reddit, or discord, categorized as public, private, or mixed
Sampling and recruitment	The strategies used to identify and engage participants in a study. Examples include convenience sampling, purposive sampling, and snowball sampling
Researcher positionality and reflexivity	Researchers’ roles and relationships to the participant community, including site access and researcher identity. Examples include insider (community member), outsider (non-member), or hybrid positions. Reflexivity indicates researchers’ reflection on how their identities, assumptions, and biases influence the research
Data collection	The methods used to gather data in a study. Examples include interviews, focus groups, observations, and collecting existing online data
Data analysis	The methods used to analyze data in a study. Examples include thematic analysis, content analysis, discourse analysis, network analysis, and visual analysis
Research rigor	Measures taken to ensure the validity and reliability of a study. Examples include member checks, debriefing, triangulation, and inter-coder reliability

Ethical considerations
Code category	Definition and examples
Informed consent	Researchers’ efforts regarding communicating study details to participants, including: (1) disclosure of research purpose, procedures, risks, and benefits; (2) statements affirming voluntary participation and the right to withdraw; (3) obtaining informed consent; and (4) waivers of informed consent, if applicable
Respect and trust	Researchers’ efforts to respectfully represent participants and their communities in research, and to foster trust and rapport throughout the research process
Privacy and confidentiality	Measures to protect participants’ identities and personal information, such as anonymity and de-identification
Data security and reuse	Measures to safeguard research data and manage its usage, such as secure data storage and transmission, and policies or limitations on data access, sharing, and reuse

The second stage was the complete full-text coding stage in which the first and second authors each coded half of the remaining 53 articles. Several rounds of iterative coding were conducted, during which the two authors reviewed each other’s results and resolved disagreements through discussions. Additionally, the authors collected representative examples from the analyzed literature for the methodological and ethical considerations identified in the codebook.

The third stage was the synthesizing stage. After compiling the coding results, the first and second authors analyzed every methodological and ethical consideration and summarized relevant examples identified in the reviewed literature. The research team then discussed and proposed a human-centered framework for conducting LGBTQ-focused online ethnographic research. Based on the syntheses and the proposed framework, research gaps in the current literature were identified, and future research directions were discussed.

Findings

Characteristics of Included Studies

The bibliometric characteristics of the 62 studies included in this review highlighted key publication patterns and authorship trends. These LGBTQ-focused online ethnographic studies were published between 2009 and 2024, appearing in 42 journals spanning media studies, sexuality studies, communication, and information science, such as Sexualities, Journal of Homosexuality, New Media & Society, and Media, Culture & Society. Regarding authorship, both single-author and collaborative works were represented, with 31 studies being single-authored and 31 being co-authored by two or more scholars. This trend suggests a balance between single-authored ethnographic research and collaborative, often cross-disciplinary, efforts to explore LGBTQ experiences in digital contexts. Additionally, institutional affiliations spanned across North America, Europe, and Asia, indicating a geographically diverse scholarly engagement with LGBTQ digital cultures.

Methodological Considerations

Populations

The populations in the reviewed articles encompass people with different identities within and outside the LGBTQ spectrum. More than half (33) of the studies included participants who self-identify with LGBTQ or diverse sexual or gender identities, and investigated topics (e.g., information seeking behavior) related to LGBTQ communities in general (e.g., Yue & Lim, 2022). For research focused on specific identities, 16 articles studied the transgender populations, examining topics such as how transgender people navigated their gender identities between online and offline settings (Marciano, 2014) and how social media platforms like Tumblr supported the trans communities (Haimson et al., 2019). These studies recognized the unique challenges and experiences faced by transgender individuals, particularly in digital environments where identity expression and community support were critical. There were 10 studies centered on gay men, and some focused on non-white gay communities, such as Chinese gay influencers on Twitter (S. Wang & Ding, 2022) and Mongolian gay men using Facebook (Dovchin, 2019). The topics investigated include gay men’s usage of dating or hook-up apps or platforms (Ahlm, 2016; MacKee, 2016) and how their sexual behaviors (e.g., barebacking) were discussed in online forums (Garcia‐Iglesias & Ledin, 2021). Notably, only 2 studies (Foeken & Roberts, 2018; Yao et al., 2023) analyzed issues related to lesbian populations, and none of the included studies solely addressed topics concerning bisexual populations. Additionally, some articles covered participants with a wide spectrum of identities, such as cisgender and non-binary (McCormack & Wignall, 2021; McCracken & Brooks–Gordon, 2021).

Besides studies involving people with LGBTQ identities, eight articles included people who did not identify as LGBTQ but were closely related to LGBTQ people. For example, focusing on cyber-HIV/AIDS interventions for men who have sex with men (MSM), Yeo (2009) interviewed resource services leaders, volunteers for these services, and website administrators. Y. Zhou et al. (2021) examined tongqi (i.e., heterosexual women who are unwittingly married to non-straight men) in China, where same-sex marriage is not legally recognized.

Online Sites

The reviewed articles explored diverse online sites. The most popular sites were social media platforms, such as Facebook, X (Twitter), YouTube, and Instagram. Other sites included online forums (e.g., Qruiser forum, Sammyboyforum, Barebackers’ Forum) (Chib et al., 2021; Garcia‐Iglesias & Ledin, 2021; Svensson, 2015), adult-oriented live streaming platforms (e.g., Chaturbate) (Theunissen & Favero, 2021), adult content websites (e.g., Findomme website, Fatclub.com, and Vicesisters.com) (Adams & Berry, 2013; Chib et al., 2021; McCracken & Brooks–Gordon, 2021), online video games (e.g., World of Warcraft) (McKenna & Chughtai, 2020), and dating apps (e.g., Hornet, Butterfly, Jack’d, Grindr, Tinder, and Blued) (Ahlm, 2016; Atuk, 2022; Choy, 2017; McGuire, 2016; S. Wang, 2019). Besides these mainstream platforms, some researchers selected specific sites to capture LGBTQ experiences in various cultural contexts. For instance, Atuk (2022) compared two dating apps, Hornet and Grindr, with an emphasis on the cultural, social, and geographical contexts in Istanbul. Similarly, Yao et al. (2023) selected Zhihu, a Chinese social question-and-answer (Q&A) platform where registered users obtain knowledge from experts, as the site to explore lesbians’ perspectives on family formation with assisted reproductive technology.

Notably, many researchers conducted multi-sited online ethnographic research to understand the topic of interest through variant data sources. For example, Morrison (2022) selected Facebook, X (Twitter), Instagram, and YouTube as the sites to explore queer visibility. Similarly, Taha-Thomure et al. (2022) examined the same sites to investigate gender-based violence and discrimination against transgender individuals.

Besides the cultural contexts and variations of sites, site selection is also impacted by researchers’ ability to access the site(s). Regarding site access, most reviewed studies analyzed publicly accessible data on social media, such as Dovchin (2019) and Quick (2021). A few scholars had access to private sites like closed or private online groups and password-protected forums. For example, Y. Zhou et al. (2021) selected a private online group where members (heterosexual women who unknowingly married non-straight men) shared experiences, sought support, and discussed sensitive issues anonymously. Researchers analyzed data from both public and private sites, such as a combination of public comments on one’s profile picture and private messages (e.g., Adams & Berry, 2013). Some studies examined both online and offline sites, complementing online ethnography with other methods like in-person interviews (Przybylski, 2020), allowing researchers to depict a more holistic picture of the participants and their experiences (Frank, 2020; García-Santesmases Fernández et al., 2017).

Sampling and Recruitment

The literature reviewed applied different approaches for participant sampling and recruitment (Table 3). The approaches include purposive sampling (34), convenience sampling (20), and snowball sampling (13). Each approach has its strengths and limitations in the context of recruiting participants in LGBTQ communities. Purposive sampling allows researchers to intentionally include participants with different perspectives and experiences. Through this approach, mainstream opinions and voices from minority groups could all be included. Considering the variety of LGBTQ communities, purposive sampling is a well-suited choice that ensures the inclusion of diverse voices and balanced representations in research (Hollingshead et al., 2020). To implement purposive sampling, researchers often adopt a criterion sampling strategy identifying specific criteria tailored to research questions and recruit participants who meet those criteria (e.g., age, identity, and experience) (Newton & Southerton, 2021; Yue & Lim, 2022). As researchers have some flexibility to craft the recruitment criteria, purposive sampling could introduce researchers’ biases and result in potential exclusion of certain groups, subjective sampling, and over-representation of certain groups. Purposive sampling is hence suited for studies aiming to emphasize different perspectives, like many LGBTQ online ethnography studies, and less ideal for research pursuing results with high generalizability and replicability (Bryman, 2015; Etikan et al., 2016; Palinkas et al., 2015).

Table 3.

Major Methods for Sampling, Data Collection, and Data Analysis in the Reviewed Studies

Methodological element	Research method	Number of studies	Relevant studies
Sampling methods	Purposive sampling	34	Adams and Berry (2013); Ahlm (2016); Atuk (2022); Avdeeff (2021); Avishai (2020); Cao (2021); Darwin (2017); Dovchin (2019); Foeken and Roberts (2018); Garcia‐Iglesias and Ledin (2021); García-Santesmases Fernández et al. (2017); Haimson et al. (2019); Hollingshead et al. (2020); Y. Huang et al. (2024); Kilic (2024); Kosher et al. (2023); MacKee (2016); McCormack and Wignall (2021); McGuire (2016); McKenna (2019); Newton and Southerton (2021); Numer et al. (2021); Perleberg et al. (2023); Quick (2021); Rodriguez (2022); Schier and Linsenmeyer (2019); Schwartz (2020); Shrodes (2022); Sutherland (2021); Svensson (2015); Tang (2023); Waggoner (2018); Yao et al. (2023); Yue and Lim (2022)
	Convenience sampling	20	Ahlm (2016); Bonocore Morais et al. (2022); Cao (2021); Channon and Matthews (2015); Frank (2020); Hermes and Kopitz (2021); B. Huang et al. (2023); Y. Huang et al. (2024); Kilic (2021, 2024); McCormack and Wignall (2021); McCracken and Brooks–Gordon (2021); McGuire (2016); McKenna and Chughtai (2020); Soini and Eräranta (2022); Taha-Thomure et al. (2022); Theunissen and Favero (2021); Tompkins (2014); Vorobjovas-Pinta and Hardy (2021); Z. Zhou (2024)
	Snowball sampling	13	Ahlm (2016); Avishai (2020); Chib et al. (2021); Choy (2017); Frank (2020); Hollingshead et al. (2020); Kilic (2024); Rand (2022); Ruvio and Belk (2018); Schwartz (2020); Y. Wang and Tan (2023); Y. Zhou et al. (2021); Z. Zhou (2024)
Data collection methods	Searching, selecting, and saving existing online data	42	Adams and Berry (2013); Ahlm (2016); Atuk (2022); Avdeeff (2021); Avishai (2020); Bonocore Morais et al. (2022); Cao (2021); Channon and Matthews (2015); Darwin (2017); Foeken and Roberts (2018); Frank (2020); Garcia‐Iglesias and Ledin (2021); García-Santesmases Fernández et al. (2017); Haimson et al. (2019); Hermes and Kopitz (2021); B. Huang et al. (2023); Y. Huang et al. (2024); Kilic (2024); Kosher et al. (2023); MacKee (2016); Marciano (2014); McCormack and Wignall (2021); McKenna (2019); McKenna and Chughtai (2020); Morrison (2022); Newton and Southerton (2021); Quick (2021); Rodriguez (2022); Schier and Linsenmeyer (2019); Schwartz (2020); Shrodes (2022); Soini and Eräranta (2022); Sutherland (2021); Svensson (2015); Taha-Thomure et al. (2022); Tang (2023); Tompkins (2014); Vorobjovas-Pinta and Hardy (2021); Waggoner (2018); Yao et al. (2023); Yeo (2009); Z. Zhou (2024)
	Interviews	35	Ahlm (2016); Atuk (2022); Avishai (2020); Cao (2021); Chib et al. (2021); Choy (2017); da Silva and Iriart (2012); Dovchin (2019); Frank (2020); García-Santesmases Fernández et al. (2017); Haimson et al. (2019); Hightower (2015); Hollingshead et al. (2020); B. Huang et al. (2023); Y. Huang et al. (2024); Kilic (2024); MacKee (2016); McCormack and Wignall (2021); McGuire (2016); Numer et al. (2021); Perleberg et al. (2023); Rand (2022); Raun (2014); Ruvio and Belk (2018); Schwartz (2020); Svensson (2015); Tang (2023); Theunissen and Favero (2021); S. Wang (2019); S. Wang and Ding (2022); Y. Wang and Tan (2023); Yeo (2009); Yue and Lim (2022); Y. Zhou et al. (2021); Z. Zhou (2024)
	Observation	34	Adams and Berry (2013); Ahlm (2016); Cao (2021); Chib et al. (2021); Choy (2017); da Silva and Iriart (2012); Dovchin (2019); Foeken and Roberts (2018); García-Santesmases Fernández et al. (2017); Hollingshead et al. (2020); B. Huang et al. (2023); Kilic (2024); MacKee (2016)); McCormack and Wignall (2021); McCracken and Brooks–Gordon (2021); McKenna (2019); McKenna and Chughtai (2020); Newton and Southerton (2021); Quick (2021); Raun (2014); Rodriguez (2022); Ruvio and Belk (2018); Schwartz (2020); Shrodes (2022); Soini and Eräranta (2022); Svensson (2015); Theunissen and Favero (2021); Tompkins (2014); Vorobjovas-Pinta and Hardy (2021); Waggoner (2018); S. Wang and Ding (2022); Y. Wang and Tan (2023); Y. Zhou et al. (2021); Z. Zhou (2024)
Data analysis methods	Discourse analysis	15	Adams and Berry (2013); Atuk (2022); Avdeeff (2021); Channon and Matthews (2015); Darwin (2017); Garcia‐Iglesias and Ledin (2021); Kilic (2021); Marciano (2014); Numer et al. (2021); Schwartz (2020); Shrodes (2022); Soini and Eräranta (2022); Tang (2023); Y. Wang and Tan (2023); Yue and Lim (2022)
	Content analysis	14	Atuk (2022); Cao (2021); Frank (2020); Garcia‐Iglesias and Ledin (2021); Hermes and Kopitz (2021); McCracken and Brooks–Gordon (2021); McGuire (2016); McKenna (2019); Morrison (2022); Raun (2014); Rodriguez (2022); Yao et al. (2023); Yeo (2009); Z. Zhou (2024)
	Thematic analysis	12	Ahlm (2016); da Silva and Iriart (2012); Foeken and Roberts (2018); Hollingshead et al. (2020); Y. Huang et al. (2024); Rand (2022); Ruvio and Belk (2018); Svensson (2015); Taha-Thomure et al. (2022); Vorobjovas-Pinta and Hardy (2021); Y. Zhou et al. (2021); Z. Zhou (2024)

Convenience sampling was also a popular method, used in 20 studies. However, recruiting LGBTQ participants via convenience sampling can still be challenging. For example, Theunissen and Favero (2021) were only able to include two male trans* (transgender) cammers on Chaturbate (an adult live-streaming website containing sexually explicit materials) due to the limited number of eligible users they encountered on the platform. This highlights a key limitation of convenience sampling in LGBTQ contexts, that even seemingly “convenient” populations may remain hard to reach due to platform norms, visibility issues, or stigma. In addition, convenience sampling of LGBTQ participants involves more than simply recruiting those who are easily accessible; it also requires alignment with the norms and culture of the community. For example, Ahlm (2016) adhered to the social norms of Grindr by using an in-app recruitment strategy that relied on informal conversations to identify potential participants.

Moreover, snowball sampling was also a valuable method to reach hard-to-access populations within the LGBTQ communities. This method relies on participants’ social networks and interpersonal referrals, enabling researchers to penetrate communities where traditional recruitment methods may fail. For example, Ruvio and Belk (2018) were introduced to transgender members by community leaders and activists, and recruited additional participants by sharing research information through venues such as transgender-related conferences and social haunts.

Data Collection

The most popular data collection methods in the reviewed articles were identified (Table 3). Forty-two studies analyzed existing online data, such as texts and images on social media (Tang, 2023) and online forums (Foeken & Roberts, 2018). Researchers of 35 studies conducted virtual or in-person interviews to collect participants’ experiences, information behaviors, and perspectives. Interview topics include participants’ experiences with a dating app (Ahlm, 2016), transgender sex workers’ online gender performances and communication practices (Chib et al., 2021), and transgender men’s self-representation and behaviors on social media (Tang, 2023). Observation was applied in 34 studies. Through both participatory observation (e.g., Quick, 2021) and non-participatory observation (e.g., Y. Zhou et al., 2021), researchers captured LGBTQ individuals’ behaviors in online spaces, as well as the dynamics of those spaces, such as the cultural norms in online communities. During the observations, researchers took fieldnotes to document contextual details and their interpretations, reflections, and analytical thoughts. For example, S. Wang (2019) reported how his online ethnography study of a gay dating platform impacted him sexually and emotionally. These fieldnotes then provide the foundation for an in-depth analysis of researchers’ engagement with the online environment and the community studied. Besides the above-mentioned approaches, some studies collected data through surveys (B. Huang et al., 2023; McCracken & Brooks–Gordon, 2021; Numer et al., 2021), document analysis (Haimson et al., 2019), and diaries (McCormack & Wignall, 2021).

Data Analysis

Various data analysis methods were employed, reflecting the diversity of the data and varying research objectives (Table 3). The predominant data analysis approaches were discourse analysis (15), content analysis (14), and thematic analysis (12). Discourse analysis allows researchers to critically examine the language, communication patterns, and narratives within LGBTQ communities online. By analyzing multimodal discourses (e.g., textual posts, images, and multimedia content), researchers can uncover how identities, power dynamics, and social norms are constructed and negotiated in digital spaces (Avdeeff, 2021; Tang, 2023).

Content analysis focuses on systematically categorizing and quantifying elements of the data. This method is particularly effective in handling large datasets, such as social media posts or forum discussion posts, enabling researchers to identify interaction patterns, frequencies, and trends within LGBTQ online communities. For example, Rodriguez (2022) used content analysis to examine how YouTube publicly incorporates LGBTQ creators on the platform while privately discriminating against LGBTQ-related users and content.

Thematic analysis emphasizes the identification and interpretation of recurring themes in the data. One example is Hollingshead et al.’s (2020) three-stage study investigating the increase of men who have sex with men (MSM) seeking partners via social networking apps in the Philippines. They identified themes from each stage using inductive thematic analysis, presenting a more comprehensive view of the risks and opportunities of the apps examined.

Besides the above-mentioned data analysis approaches, researchers also used visual analysis (Raun, 2014; Schwartz, 2020; Theunissen & Favero, 2021), grounded theory (McCormack & Wignall, 2021), and specific coding techniques (e.g., axial coding) (McCracken & Brooks–Gordon, 2021). Visual analysis is particularly relevant to studies that examine platforms where visual content, such as images, videos, or memes, plays a central role. Visual analysis enables researchers to examine how visual elements contribute to identity formation, community building, and the expression of shared values within LGBTQ communities. Among the 62 studies reviewed, 12 did not specify the analytical methods used. The lack of explanation regarding data analysis risks undermining research rigor.

Researcher Positionality and Reflexivity

Researcher positionality—whether one engages with a community as an insider or outsider—shapes multiple aspects of the research process, including site access, rapport building, and data interpretation. Among the reviewed studies, several researchers approached LGBTQ communities as outsiders (Darwin, 2017; Dovchin, 2019; Rand, 2022; Sutherland, 2021; Tang, 2023). Their outsider positionality contributed to challenges such as the need for additional effort to understand community dynamics and reduced perceived credibility among participants. For instance, Darwin (2017) recognized his limits in interpreting the data as an outsider to the non-binary community on Reddit.

Two types of hybrid positionality were observed. The first type occurs when individual researchers are simultaneously members of the communities they study, drawing on their direct participation and lived experiences within those spaces. For instance, several studies have researchers who entered the field as both investigators and active community members (Atuk, 2022; Choy, 2017; Newton & Southerton, 2021; Svensson, 2015; Z. Zhou, 2024). Svensson (2015), for example, noted that as a gay man who grew up using Qruiser and has been a member since his twenties, he occupied a dual researcher-member role. This role afforded deeper insights into the forum and its community. The second type of hybrid positionality arises when research teams collectively occupy an in-between position by involving both insiders and outsiders—for example, including LGBTQ-identified researchers alongside non-LGBTQ researchers (e.g., Adams & Berry, 2013; Kosher et al., 2023). This combination integrates emic perspectives from insider researchers with etic perspectives from outsider researchers, producing a balanced understanding of the community under study. For instance, Kosher et al. (2023) emphasized how having both LGBTQ and non-LGBTQ researchers on their team enriched the analysis by combining insiders’ knowledge with outsiders’ critical distance.

Researchers of several reviewed studies acknowledged the impact of their positionality. These scholars included reflexivity in their articles, critically examining how their experiences, biases, and perspectives impact the studies. Some researchers documented and reported their emotions, experiences, and biases (Kilic, 2024; Svensson, 2015; Y. Zhou et al., 2021), while others disclosed themselves as members of the LGBTQ communities (Chib et al., 2021; McCormack & Wignall, 2021; McCracken & Brooks–Gordon, 2021) and reflected on how their positionalities shaped their research (Newton & Southerton, 2021).

Research Rigor

The reviewed articles employed a variety of strategies to uphold research rigor. The predominant strategy is triangulation (46), examining multiple sites and employing various data collection and analysis methods (e.g., Foeken & Roberts, 2018; Morrison, 2022; Quick, 2021; S. Wang & Ding, 2022). For example, McCracken and Brooks–Gordon (2021) collected data from both public online platforms (e.g., social media, websites, blogs) and private conversations and interviews to understand findommes’ work, motivations, and boundary negotiations. Rodriguez (2022) analyzed data collected through observations, fieldnotes, and public websites to study LGBTQ individuals’ perspectives and experiences on YouTube. Atuk (2022) applied both discourse analysis and visual content analysis to explore Hornet users’ identity presentation and social interactions within the virtual cruising environment. Besides triangulation, scholars ensured research rigor through achieving satisfactory inter-coder reliability (Garcia‐Iglesias & Ledin, 2021; Yao et al., 2023), debriefing (Soini & Eräranta, 2022; Vorobjovas-Pinta & Hardy, 2021), and member checking (Choy, 2017; Yao et al., 2023).

Ethical Considerations

Informed Consent

Obtaining informed consent is an ethical practice that protects participants. Eleven reviewed studies mentioned receiving participants’ informed consent or permission through written or verbal acknowledgment of their rights and understanding of the research process (e.g., Chib et al., 2021; McCormack & Wignall, 2021; Ruvio & Belk, 2018; Yao et al., 2023; S. Wang, 2019). Seven studies mentioned that obtaining consent was impractical or the research posed minimal risk, and therefore informed consent waivers were approved by institutional review boards (IRBs) (e.g., Rodriguez, 2022; Schier & Linsenmeyer, 2019; Taha-Thomure et al., 2022). For example, B. Huang et al. (2023) described how obtaining individual informed consent from participants in large, semi-private online communities was logistically impractical. They instead relied on collective announcements with opt-out options to address ethical requirements. Notably, most researchers of the included articles rarely reflected on this ethical tension, suggesting a lack of transparency in how consent-related dilemmas are addressed in practice. Another component of informed consent is research disclosure. Authors of eight reviewed studies explicitly explained to participants the objectives, methods, potential risks, and data usage of their studies (e.g., Ahlm, 2016; Frank, 2020; McCracken & Brooks–Gordon, 2021; S. Wang, 2019). Through research disclosure, researchers invite participants to make informed decisions about their participation. One other component of informed consent is voluntariness. As emphasized in four of the reviewed studies, researchers should ensure that participants’ involvement in the study is free of coercion, and participants can withdraw at any stage of the research (da Silva & Iriart, 2012; McCracken & Brooks–Gordon, 2021; Ruvio & Belk, 2018; Y. Zhou et al., 2021).

Respect and Trust

Respecting participants, protecting their privacy and autonomy, and establishing trust and rapport form the foundation of successful ethnographic research. These aspects are particularly important in research regarding minority groups, including LGBTQ communities. This literature review identifies the approaches researchers applied to account for respect and trust. The approaches include obtaining participants’ informed consent (B. Huang et al., 2023), conforming to the social norms of the participants and their communities (Chib et al., 2021), acknowledging and addressing any language or cultural barriers that may impact trust-building (S. Wang, 2019), and representing participants respectfully, such as using participants’ preferred pronouns (Frank, 2020; Perleberg et al., 2023). Regarding participant representation, while anonymization is a common practice to protect participants’ privacy, in the context of LGBTQ-focused online ethnographic research, there were cases where participants preferred not to be anonymized. Instead, participants preferred to be represented by their legal names or activist nicknames, to gain visibility and amplify their voices (García-Santesmases Fernández et al., 2017).

Privacy and Confidentiality

Risks and harms should be considered in research that involves marginalized populations. Previous studies have noted hostility and discrimination toward both vulnerable participants (e.g., LGBTQ participants) and vulnerable researchers (e.g., researchers who self-disclosed as insiders of the community studied) in online spaces (B. Huang et al., 2023; Newton & Southerton, 2021). Rigorous measures, such as pseudonymization and data anonymization, are required to protect privacy, ensure confidentiality, and mitigate potential risks and harms. Among the reviewed studies, the predominant approach was using pseudonyms to represent participants (S. Wang & Ding, 2022). Some studies achieved anonymity by not collecting identifying information in the first place (Shrodes, 2022), while some removed identifying information after data collection to de-identify participants (Y. Zhou et al., 2021). Researchers may also choose to either avoid citing public posts or omit usernames in order to protect participants’ privacy (Morrison, 2022; Newton & Southerton, 2021).

Data Security and Reuse

Data security and reuse involve the practices and ethical considerations of protecting sensitive research data (e.g., participants’ personal information, interview recordings). Researchers discussed their measures to store data securely, such as storing data in encrypted laptops that only authorized personnel can access (Y. Zhou et al., 2021) and deleting interview recordings after transcripts were generated (Z. Zhou, 2024).

Discussion

A Human-Centered Framework for LGBTQ-Focused Online Ethnographic Research

Ethnography or online ethnography is fundamentally concerned with understanding people and their lived experiences in in-person or online contexts (Hammersley & Atkinson, 2007; Pink et al., 2015; Robinson & Schulz, 2009; Van Maanen, 2006). Notably, our findings repeatedly highlighted how people—both researchers and participants—interacted with other aspects of online ethnographic research (i.e., methods, ethics, and fields), pointing to the need for a human-centered framework. We propose an overarching human-centered framework for conducting and understanding LGBTQ-focused online ethnographic research (Figure 2) by contextualizing the findings of this review with prior discussion of methods and ethics in qualitative and online ethnographic research (Hair et al., 2023; Kozinets, 2020; Liamputtong, 2006). The framework consists of four interrelated components: people (i.e., LGBTQ populations, stakeholders, and researchers), methods (i.e., sampling, data collection, data analysis, and research rigor), ethics, and fields. It highlights how people influence and interact with other aspects of research and identifies key topics within each component that researchers should consider in relation to the communities they study. While grounded in LGBTQ-focused online ethnography, the human-centered emphasis of this framework may also guide qualitative research with other vulnerable or stigmatized populations (e.g., children, people with disabilities, and immigrants).

Figure 2.

A human-centered framework for LGBTQ-focused online ethnographic research

The People component is the center of this framework, encompassing both researchers and participants. Insider researchers are at the intersection of the two groups, having a dual role as both members of the LGBTQ community and researchers who conduct LGBTQ-focused ethnography. The researched group includes not only LGBTQ individuals and their communities but also relevant stakeholders, such as LGBTQ allies. The People component emphasizes the human actors or agents in LGBTQ-focused ethnographic research while underscoring the goal of researching for the good of the people involved.

In LGBTQ-focused online ethnographic research, Methods, Ethics, and Fields are interconnected aspects that shape the research and share an emphasis on People. Methods influence how data are collected, analyzed, and interpreted, shaping the depth and authenticity of engagement with LGBTQ communities. Researchers must ensure that their methodological choices align with ethical principles by obtaining informed consent, protecting participants’ privacy, and managing power dynamics, particularly when working with marginalized groups (Prime, 2024). The Ethics component encompasses the Methods as ethical considerations are intricately linked to methodological decisions throughout the research process. To ensure ethical integrity throughout the research process, researchers must engage in ongoing reflection and remain flexible and responsive in their decision-making. This includes regularly reassessing participants’ consent and well-being, as well as promptly identifying and addressing ethical concerns to uphold participant autonomy and minimize potential harm (Giordano et al., 2007; Ramcharan & Cutcliffe, 2001; Sperling, 2022).

The Fields component in this human-centered framework refers to the wide range of fields—online, offline, or hybrid—where LGBTQ-focused ethnographic research is conducted. These fields not only impact participant engagement and communication (Cannon et al., 2017; Han et al., 2019) but also influence which platforms or spaces researchers can ethically and practically access, what types of data can be collected, and how researchers may navigate visibility, consent, and boundaries. Specifically, Fields are deeply intertwined with Methods, as they influence the selection of sampling strategies and shape data collection methods. Equally important is the reciprocal relationship between Fields and Ethics: ethical considerations not only determine what can be done within a given field but are also shaped by the characteristics of the field itself, which influence whether access or entry is ethically possible in the first place. Thus, Fields are not peripheral but integral, shaping and shaped by ethical and methodological choices.

To sum up, the intertwined relationships among the four components illustrate why the framework adopts a layered structure, with People at the center and each surrounding component—Methods, Ethics, and Fields—interacting not only with the core but also with one another. Researchers must prioritize the welfare and lived experiences of LGBTQ individuals, their communities, stakeholders, and themselves. This focus should guide decisions regarding field sites, ethics, and methods. In doing so, researchers can create a more inclusive and reflexive research process.

Methodological and Ethical Implications

This section outlines key methodological and ethical implications drawn from the reviewed LGBTQ-focused online ethnographic studies. It begins with issues of representation, noting uneven attention to specific LGBTQ identities and the need for more inclusive, intersectional approaches. It then examines how researcher positionality and reflexivity influence access, rapport, and data interpretation. The discussion moves on to research settings, emphasizing hybrid and multi-sited approaches, followed by an overview of commonly used data collection and analysis methods, and the role of triangulation in enhancing rigor. Finally, it addresses core ethical concerns, including informed consent, privacy, rapport building, and respectful representation of LGBTQ participants in digital contexts.

The reviewed articles studied diverse LGBTQ populations, encompassing identities across and beyond the LGBTQ spectrum. A significant body of the analyzed literature involves LGBTQ or queer populations in general, echoing the similar comprehensive population focus of LGBTQ research in other fields, such as human-computer interaction (HCI) (Taylor et al., 2024). While studying the collective LGBTQ population promotes an overarching understanding of their shared experiences shaped by heteronormativity (Robinson, 2016), it is critical to recognize differences within and across queer identities. Our findings indicate uneven representation of LGBTQ identities in the analyzed research, with disproportionate focus on transgender and gay populations and comparatively limited attention given to lesbian and bisexual people. One potential explanation of this phenomenon is the pivotal roles and higher visibility of gay communities and transgender people in LGBTQ rights movements and activism history (Aultman, 2021; Duberman, 2019; Faderman, 2015). Another possible explanation is the greater number of digital sites specifically or primarily designed for gay and transgender people (e.g., Blued, Grindr, Taimi, and Trans) compared to those for other LGBTQ identities (He, 2021; Miao & Chan, 2020; Thompson, 2024; Zheng & Lin, 2024). In response, we advocate not for diminishing the focus on gay and transgender individuals, but for broadening the scope of online ethnographic research to better represent understudied LGBTQ populations, such as lesbian and bisexual individuals. Through expanding the focus to encompass a wider range of experiences, scholars can foster a more balanced and comprehensive understanding of LGBTQ identities in future research.

Intersectionality can enrich queer studies by emphasizing the complexity of how multiple identities (e.g., age, ability, race, class, and national origin) of an individual shape their experiences (Battle & Ashley, 2008; Vaccaro & Koob, 2019). Specifically, we have witnessed a limited number of non-White LGBTQ studies that provide valuable insights into LGBTQ people’s online experiences in non-Western digital contexts, avoiding Western-centrism and contributing to global LGBTQ perspectives in online ethnography. Also related to the complexity of queer studies, LGBTQ-focused online ethnographic research should go beyond LGBTQ participants and include non-LGBTQ participants who are closely related to LGBTQ populations, such as LGBTQ allies, stakeholders in HIV/AIDS interventions, and tongqi. Involving non-LGBTQ participants allows for a broader perspective for understanding societal and relational factors affecting LGBTQ communities.

Researcher positionality significantly influences online LGBTQ ethnography. Insider researchers who identify as LGBTQ typically have easier access to private LGBTQ online communities due to established relationships and familiarity with group culture (Bonner & Tolhurst, 2002; Grove, 2017; Hayfield & Huxley, 2015). However, openly stating an insider positionality can expose researchers to professional and personal risks, including harassment, bullying, physical and mental health impacts, difficulty developing collegial relationships, and lower pay (Veldhuis, 2022). Insider researchers’ dual roles may also complicate objectivity and confidentiality (Saidin & Yaacob, 2016). In contrast, outsider researchers often face significant barriers accessing private LGBTQ spaces, where richer and more sensitive data typically reside (Ellard-Gray et al., 2015; Gatlin & Johnson, 2017). Therefore, collaborative partnerships can be beneficial, where insider researchers contribute an empathetic lens grounded in their lived LGBTQ experiences, while outsider researchers have complementary objectivity and critical distance, together producing culturally sensitive and methodologically sound research (Court & Abbas, 2022). Especially in community-based research, most participants perceived the diversity of the research teams as a strength (Kerstetter, 2012).

Reflexivity is crucial for navigating these complexities introduced by researcher positionality in studies involving vulnerable populations such as LGBTQ individuals (Roffee & Waling, 2017). Reflexivity involves critically examining researchers’ assumptions and biases, as well as their potential impact on the research process. Researcher-participant relationships formed during the study can significantly shape data collection and interpretation (Meneau, 2024). By engaging in reflexivity, researchers can respect LGBTQ voices, minimize potential risks and harm, and enhance the depth and rigor of their studies (Iacono et al., 2021; Pillow, 2003). Deeper reflexivity can be facilitated by collaborative research partnerships involving both insider and outsider researchers, reducing biases and enhancing methodological rigor.

In terms of fields, online spaces offer relatively safe and inclusive platforms for LGBTQ identity expression and community building (Brandt & Carmichael, 2020; McInroy et al., 2019). Our analysis highlights the use of diverse online platforms to capture LGBTQ experiences across contexts. Frequently studied platforms—such as Facebook, X (Twitter), YouTube, and Instagram—align with trends reported in prior online ethnography reviews (Heinonen & Medberg, 2018; Kaur et al., 2021; Tavakoli & Wijesinghe, 2019). In addition to mainstream platforms, it is worth noting that adult-oriented sites and dating apps were widely included in the analyzed studies, shedding light on how LGBTQ people express their identities and behaviors in niche, often more private, digital spaces (Adams & Berry, 2013; Chib et al., 2021; McCracken & Brooks–Gordon, 2021; Theunissen & Favero, 2021). That is, mainstream and niche online sites together underscore the multifaceted nature of online queer experiences. The diverse sites provide a holistic view of the unique ways these digital spaces support LGBTQ identity expression. Moreover, a multi-sited ethnographic approach (Falzon, 2016; Marcus, 2012) enables a more comprehensive understanding of LGBTQ interactions across online and offline spaces, as well as public and private groups. Integrating offline and online data offers a holistic view of LGBTQ lives, echoing Flick’s (2019) call to connect online communication with “real-world activities” (p. 346). Specifically, hybrid methods, such as combining online observations with in-person interviews, allow researchers to bridge the digital and physical spheres (Frank, 2020; Przybylski, 2020). This approach is particularly effective in exploring complex topics like body politics, identity formation, and community dynamics. Given the blurring boundaries between people’s online and offline lives (Auriemma et al., 2024), researchers should continue leveraging this approach to account for the interplay between online interactions and offline realities.

Methodologically, the reviewed studies predominantly used online data collection methods, including interviews, collecting existing online data, and observation. To analyze the diverse data collected, researchers applied various data analysis methods. The main approaches include discourse analysis, content analysis, and thematic analysis. This echoes the findings of previous reviews of online ethnographic research (Heinonen & Medberg, 2018; Tavakoli & Wijesinghe, 2019; Whalen, 2018). Notably, some studies didn’t specify their analysis methods, which is an issue that has been raised by Tavakoli and Wijesinghe (2019). It is crucial to explicitly state data analysis methods to improve the transparency of research (Tuval-Mashiach, 2017).

Additionally, methodological rigor can be improved by employing triangulation strategies. For example, data triangulation involves utilizing multiple data sources, such as different online platforms, diverse user demographics, or various time frames, to validate findings (Alessi et al., 2020; Craig et al., 2021; Earnshaw et al., 2020; Shipman, 2022). Methodological triangulation enhances research rigor by combining different data collection or analysis methods, such as qualitative interviews, ethnographic observations, content analysis, and sentiment analysis (Moore et al., 2021; Opall, 2021; Singh & Shelton, 2011). While triangulation was the most frequently used strategy to ensure trustworthiness and credibility across the reviewed studies, additional measures such as inter-coder reliability, peer debriefing, and member checks could further enhance the rigor and credibility of future LGBTQ online ethnographic research (Brown et al., 2023; Craig et al., 2021; DeChants et al., 2022; Quinn et al., 2015; Shefcik & Tsai, 2023; Vashisht et al., 2024).

LGBTQ-focused online ethnography should align with established research ethics guidelines while also acknowledging and addressing unique challenges inherent to studying marginalized and vulnerable populations in digital contexts. The included studies emphasize the importance of informed consent in maintaining ethical research practices and ensuring that participants understand their role in the research process. To respect LGBTQ participants, researchers should clearly communicate the study purpose, outline potential risks in the informed consent form, and obtain consent whenever feasible (Kozinets, 2020). In addition to obtaining informed consent, establishing non-maleficence, respect, and trust is fundamental to ethical online ethnography (Hair et al., 2023), especially in LGBTQ-focused research. A strong commitment to privacy, transparent disclosure of research information, and rapport building empowers participants to make informed decisions about their involvement while enhancing research credibility and ethical integrity (Z. Zhou, 2024). In establishing rapport, researchers strive to build relationships that encourage the sharing of rich, insightful data (Guillemin & Heggen, 2009). It is crucial to communicate participants’ rights and how to exercise them to create a safe place for sharing personal information (Schmid et al., 2024). Moreover, establishing rapport before recruitment can further enhance engagement (Ellard-Gray et al., 2015), highlighting the potential advantages that LGBTQ ethnographers may have as community insiders of communities under study. Meanwhile, ethical representation of LGBTQ participants and their communities remains a primary aspect of fostering ongoing respect and trust throughout the research process (Lucassen et al., 2017). Specifically, using LGBTQ participants’ preferred pronouns is a recommended way to respect their identity and dignity (James et al., 2016; Prayson & Rowe, 2019). Additionally, visibility is important for many LGBTQ participants who seek to assert their identities and be recognized. When participants request it, using their real names can ethically affirm their agency.

At the same time, privacy and confidentiality are among the key considerations in studying marginalized populations, including LGBTQ communities. To protect vulnerable participants, researchers have adopted robust measures such as pseudonymization and data anonymization to reduce privacy risks and ensure confidentiality (Gibney, 2017; Woods & Hardman, 2022). However, pseudonyms might unintentionally reveal participants’ identities, as community members could still identify individuals based on the context or details provided in selected quotations (Damianakis & Woodford, 2012). Therefore, researchers should acknowledge the risks associated with the LGBTQ community throughout the study and seek participants’ preferences before publishing the research (Roberts, 2015).

Gaps in Current Research and Directions for Future Studies

This review identifies several gaps in current LGBTQ-focused online ethnographic research. First, many studies focus predominantly on general LGBTQ populations, transgender people, and gay people, while the experiences of lesbian and bisexual individuals are understudied. Second, there has been limited research studying private online spaces and a mixture of public and private online spaces, which limits our understanding of LGBTQ experiences in private spaces. Third, the wide adoption of convenience sampling, the reliance on pre-existing online data, and the absent or vague description of analytical methods may undermine research rigor and the depth of analysis. Fourth, current LGBTQ-focused online ethnographic research rarely incorporates rigor-enhancing practices other than triangulation, such as inter-coder reliability, debriefing, and member checks. Lastly, ethical challenges arise when researchers engage with online communities without explicit permission. There is a tension between collecting data (e.g., participants’ online behaviors) in a natural environment and protecting and respecting participants’ rights.

Future researchers can address the above issues through several key strategies. First, employing diverse sampling methods can enhance inclusivity and ensure broader representation of gender and sexual identities, contributing to a more comprehensive understanding of LGBTQ experiences. Second, adopting a multi-sited approach—spanning public and private, online and offline spaces—can provide a more holistic view of LGBTQ digital interactions. Third, applying underutilized methods such as visual analysis and critical techno-cultural discourse analysis can better capture the multimodal nature of communication in queer online spaces. Fourth, incorporating rigor-enhancing practices (e.g., inter-coder reliability, peer debriefing, and member checks) can strengthen research validity and ethical standards. Finally, adhering to ethical guidelines, including informed consent and transparent disclosure, is essential to balancing naturalistic data collection with the protection of participants’ rights.

Limitations of the Study

The research team recognized the limitations of this study. First, the literature reviewed was limited to journal articles written in English, which excluded other resource types (e.g., book chapters, conference proceedings, dissertations) published in other languages. There is also the possibility that the search query failed to comprehensively retrieve all relevant articles, despite the attempt to include variant expressions of the two core concepts, LGBTQ communities and online ethnography. Additionally, making empirical research an inclusion criterion excluded theoretical or conceptual studies that may offer critical insights into LGBTQ-focused issues.

Conclusion

This review underscores the transformative potential of online ethnography in understanding LGBTQ experiences. Significant methodological and ethical issues in LGBTQ-focused online ethnographic research were identified. For instance, the findings reveal a predominant focus on certain populations, such as gay and transgender individuals, leaving other identities (e.g., lesbian and bisexual people) relatively underexplored. Some methodological issues include the overreliance on convenience sampling and the missing or vague description of analytical approaches. Highlights of the identified ethical concerns include obtaining informed consent and privacy risks. More rigorous and transparent practices are needed to address these issues. Based on the findings, a human-centered framework for LGBTQ-focused online ethnographic research was proposed. Future research should prioritize the inclusion of diverse sexual and gender identities, the application of diverse methods (e.g., sampling and rigor-enhancing), and ethical engagement with LGBTQ communities in digital spaces and beyond. By addressing gaps in current literature, researchers can conduct more rigorous and ethical studies, foster meaningful advocacy, and enhance support for LGBTQ populations.