Abstract

On 20 November 2017, Beijing LGBT Center and the Department of Sociology at Peking University released a nationwide online survey concerning transgender and gender-nonconforming people in mainland China. Its English edition was titled Chinese Transgender Population General Survey Report: Transgender and Gender-Nonconforming People. The report records 5,677 submitted questionnaires by September 2017, of which 2,060 met four validity criteria: every question completed, a unique IP address, more than eight minutes of completion time, and self-reported gender identity or expression within the study scope. Recruitment circulated through Weibo, WeChat, Zhihu, community service centers, educational organizations, media accounts, and other groups serving sexual-orientation and gender-identity or expression minorities across China (Beijing LGBT Center and Department of Sociology, Peking University 2017). The dataset later became a durable base for research. A 2019 suicidality study selected 1,309 transgender men and women from the 2,060 valid questionnaires; a 2020 study of gender-affirming hormone therapy and surgery analyzed 1,304 participants; a 2022 school-discrimination and mental-health study used an analytic sample of 1,106 (Chen et al. 2019; Liu et al. 2020; Spielmann et al. 2022).

This entry interprets the survey as an object in the history of knowledge infrastructure. Its historical significance rests in an evidence-infrastructure chain: community recruitment → validity gate → report tables → public circulation → analytic subset → peer-reviewed reuse → policy and advocacy citation → digital preservation → cross-survey comparison. The same questionnaire collection therefore generated several valid denominators for different questions. The figures 2,060, 1,309, 1,304, and 1,106 correspond to distinct sample states. This entry also proposes a number-provenance ledger that records, for every cited statistic, the source edition, recruitment frame, denominator, identity subset, item or measure, analytic filters, geographic and temporal scope, and inferential scope. A percentage then becomes a traceable data state.

National geographic reach and sample composition belong in the same reading. The report received responses from every mainland provincial-level jurisdiction, while 42.4% of respondents came from Beijing, Guangdong, Shanghai, Shandong, and Zhejiang. Respondents were predominantly young, urban, and highly educated: 87.8% were under 30, 82.7% lived in cities, and 54.3% held a bachelor’s degree or higher (Beijing LGBT Center and Department of Sociology, Peking University 2017). These characteristics make the survey especially informative about a large set of people reached through digital and community networks. Population size, population prevalence, and temporal change each require designs matched to those inferential tasks. The 2021 Chinese Transgender Health Survey and its peer-reviewed outputs formed a later generation of evidence under new recruitment, measurement, and analytic filters, while using 2017 as an important comparison point (Wang et al. 2023; Hou et al. 2026).

Keywords: mainland Chinese transgender history; 2017 Chinese Transgender Population General Survey; community survey; survey methods; data history; evidence infrastructure; analytic denominator; digital archive

1. Why a survey belongs in transgender history

Transgender histories often begin with people, organizations, policies, medical procedures, court cases, or media events. Surveys also deserve historical treatment because they determine which experiences become questions, which identities become categories, and which observations become figures that circulate through journalism, scholarship, policy work, and community memory. The 2017 survey occupies a particularly revealing moment. UNDP’s 2014 China country report had already assembled broad evidence on sexual and gender minority life, and the 2016 Being LGBTI in China survey established a wider baseline about social attitudes and lived conditions. In 2017, transgender issues acquired a dedicated large-scale quantitative project jointly connected to a community organization and a university sociology department (UNDP 2014, 2016; Beijing LGBT Center and Department of Sociology, Peking University 2017).

The survey stated an explicitly public-facing research purpose. Its English report described government departments, international groups, nonprofit organizations, and for-profit organizations as potential users. It sought a current statistical foundation for understanding China’s transgender population, greater visibility for relevant social issues, laws and policies responsive to community needs, protection of rights, and opposition to discrimination (Beijing LGBT Center and Department of Sociology, Peking University 2017). The project therefore connected three kinds of knowledge practice from the outset: systematic collection of community experience, quantitative organization through an academic partnership, and production of figures suited to public and policy circulation.

That connection took a concrete infrastructural form. Participants opened a questionnaire on phones or computers. Recruitment posts moved through Weibo, WeChat, Zhihu, community service networks, educational organizations, and media accounts. Questions about sex assigned at birth, internal gender identity, and lived or expressed gender fed classification rules. Four quality screens then turned 5,677 submissions into 2,060 valid questionnaires. Tables and charts turned individual responses into percentages. Journalists selected figures concerning health care, self-harm, family experience, and school life for wider circulation. Researchers later extracted subsets aligned with specific hypotheses and statistical models (Beijing LGBT Center and Department of Sociology, Peking University 2017; China Development Brief 2017; China Daily 2017a, 2017b).

The survey’s importance also comes from reuse over time. A data collection conducted in 2017 continued to appear in journal articles, policy reviews, legal scholarship, advocacy reports, digital archives, and later survey comparisons through the following decade. A “2017 number” can therefore have several later scholarly lives. Historical reading requires attention to where the number was first generated, which filters shaped a later analytic sample, and which denominator supports a quoted percentage. This is where data history and transgender history directly meet (Liu et al. 2020; Spielmann et al. 2022; Fappani 2024; Hou et al. 2026).

2. From 5,677 to 2,060: recruitment, screening, and classification

The original report provides unusually concrete method documentation. By September 2017 the project had received 5,677 online questionnaires and accepted 2,060, or 36.3%, as valid. Four conditions formed the validity gate: completion of every survey question; use of a unique IP address; completion time exceeding eight minutes; and self-reported gender identity or gender expression within the study scope, defined as transgender man, transgender woman, genderqueer person, or cross-dresser (Beijing LGBT Center and Department of Sociology, Peking University 2017). Those conditions transformed a site visit and submission into an analytic record.

Recruitment shaped the social form of the database. The report lists Weibo, WeChat, Zhihu, community service centers, educational organizations, media accounts, and other organizations serving sexual-orientation, gender-identity, and gender-expression minorities across China. This route is especially effective at reaching people already connected, directly or indirectly, to digital information networks and community organizations, people able to complete a substantial online questionnaire, and people willing to describe their gender through the available survey categories. Community participation also helped center concrete concerns including health care, family, school, work, and public life (Beijing LGBT Center and Department of Sociology, Peking University 2017; Guo 2018; Engebretsen, Schroeder, and Bao 2015).

Identity classification combined several questions. The survey asked sex assigned at birth, internal gender identity, and whether a participant had lived or identified as a gender different from the one assigned at birth. Different response combinations mapped respondents into transgender man, transgender woman, genderqueer, and cross-dresser categories. The historical report also used language common in its period, including FTM, MTF, and “biological sex.” Contemporary historical writing can preserve those terms when discussing questionnaire wording while using current descriptive language such as sex assigned at birth, transgender man, transgender woman, and gender-diverse people in editorial narration (Beijing LGBT Center and Department of Sociology, Peking University 2017; Reed et al. 2016; Kronk et al. 2022).

This classification design directly shaped later reanalysis. Papers focused on transgender men and women selected those respondents from the four-category valid-questionnaire dataset. A paper on school experience also required usable responses to school-related items. One survey thus generated several analytic samples appropriate to distinct research questions. Data here behave as a stateful object: submitted record, valid record, identity subset, complete-case subset, and model sample are different states in the same lineage. Preserving the state in a citation keeps one study’s denominator attached to its own question.

3. National reach and sample structure: two facts in one record

The report states that questionnaires came from every province, autonomous region, and municipality in mainland China. That geographic breadth was rare for transgender research in China at the time. Participation also concentrated in eastern and metropolitan areas: Beijing, Guangdong, Shanghai, Shandong, and Zhejiang together accounted for 42.4% of respondents. The demographic pattern was equally distinctive. Among respondents, 87.8% were under 30, 54.3% had a bachelor’s degree or higher, and 82.7% lived in cities (Beijing LGBT Center and Department of Sociology, Peking University 2017). These figures form the first methodological map for reading every later percentage in the report.

“Nationwide reach” answers where respondents came from. Population representativeness asks how the sample corresponds to the distribution of the target population. The 2017 project achieved an unusually wide geographic network while reflecting the young, urban, educated structure produced by online community recruitment. Contemporary China Daily coverage stated that estimation of the total number of transgender people in China lay outside the report’s scope. The later medical-care analysis similarly described an online questionnaire suited to the participants’ health-care needs and access patterns, while population prevalence and broader epidemiologic estimates require other sampling designs (China Daily 2017a; Liu et al. 2020).

This distinction adds historical information. It shows the communication topology of transgender public knowledge in mid-2010s China. People reached through Weibo, WeChat, Zhihu, and community organizations constituted a geographically dispersed yet strongly digital public. Sha Liu’s work on digital gender practices among Chinese transgender people documents the continuing role of platforms in identity performance, authenticity, and visibility. Shaohua Guo’s history of the Chinese internet similarly demonstrates how platform and institutional architectures organize public visibility (Liu 2023; Guo 2020). The 2017 survey can therefore be read as a quantitative cross-section of a digitally networked transgender public.

Sample composition itself becomes historical evidence. The concentration among younger, urban, educated, and eastern respondents indicates who had easier access to community information networks, stable internet participation, the survey’s conceptual vocabulary, and the time and safety needed to complete a long questionnaire. It also identifies directions later research could expand: age, locality, class, rural access, and recruitment beyond established digital networks. Methodological boundaries thus become information about the period’s social infrastructure.

4. What the survey measured: from embodied experience to institutional interfaces

The report covered a broad range of domains. It asked about puberty and embodied experience, development of gender identity, desire for hormone therapy and surgery, access to medical resources, family relationships, school and workplace experiences, discrimination in public space, self-harm, and suicide-related experience. That architecture placed transgender life inside a connected system of family, education, labor, health care, social services, and public safety (Beijing LGBT Center and Department of Sociology, Peking University 2017).

Health-care figures received especially wide public circulation. The report indicated that 62% of relevant respondents wanted hormone therapy and 51% wanted gender-reassignment-related surgery. Among people seeking hormone therapy, 71% placed access to reliable information and medically supervised treatment in the difficult-to-highest-difficulty range. China Daily covered the health-care gap on both the release day and the following day, quoting Beijing LGBT Center staff on the limited supply of professional services (China Daily 2017a, 2017b). Amnesty International’s 2019 report on barriers to gender-affirming care in China subsequently used the 2017 survey as an important evidentiary baseline (Amnesty International 2019).

Family, school, and mental-health items entered a different evidence route. Figures concerning family acceptance, school violence, self-harm, and suicide-related experience circulated through media and community publications and later became variables in peer-reviewed analysis. Chen and colleagues used the transgender-men and transgender-women subset to model suicidal ideation and suicide attempts alongside family conflict, depression, self-harm, and mental-health service seeking. Spielmann and colleagues used school-related variables to examine discrimination, environmental support, mental health, and self-harm (Chen et al. 2019; Spielmann et al. 2022).

A descriptive report table and a peer-reviewed statistical model therefore perform distinct jobs. The report describes distributions across a broad participant set. A journal article establishes a research question, variable definitions, inclusion rules, and a specific model. Journalism performs another translation, selecting a small number of figures that can enter public discussion. Historical reading benefits from identifying which layer currently carries a number.

5. Release and public circulation: how statistics entered public language

The report was released on 20 November 2017, Transgender Day of Remembrance. China Development Brief described the project as launched by Beijing LGBT Center and the Department of Sociology at Peking University, with support including the Dutch Embassy, and characterized its 2,060 valid questionnaires as the largest quantitative transgender research database in China at the time (China Development Brief 2017). The release date, partnership, and venue connected research production directly to a public advocacy moment.

News coverage rapidly selected high-impact domains such as medical access, self-harm, and family violence. China Daily emphasized gaps between demand and access for hormones and surgery and then extended the discussion to medication safety and professional care. People’s Daily Online later placed the 2017 project alongside a smaller 2018 family-violence survey, creating a public evidence pattern of a broad baseline plus issue-specific follow-up (China Daily 2017a, 2017b; People’s Daily Online 2018).

Public circulation also produced an important methodological phenomenon: different outlets and later studies selected different identity groups and questionnaire items, so related topics could produce different percentages. Suicide-related figures in the original report and the lifetime suicidal-ideation and attempt estimates calculated by Chen and colleagues for transgender men and women use different question and sample states. When figures vary, denominator, item wording, time frame, and identity filter provide the first comparison points (Beijing LGBT Center and Department of Sociology, Peking University 2017; Chen et al. 2019).

By 2022, Common Language’s Chinese introduction to a Williams Institute public-attitudes study explicitly described the 2017 transgender survey as using a community-recruited sample and distinguished it from a separate survey of public attitudes. That distinction illustrates a mature evidence practice: multiple surveys can address one broad topic while retaining different target populations and inferential tasks (Common Language 2022; Williams Institute 2021).

6. How 2,060 became 1,309: the suicidality analytic subset

Chen and colleagues published a 2019 study in the Journal of Affective Disorders using 1,309 transgender men and women from the 2017 database, representing respondents across 32 provincial-level jurisdictions. The paper selected participants aligned with its identity definitions and variable requirements, then used logistic regression to examine suicidal ideation, suicide attempts, and associated factors. Within that analytic sample it reported lifetime suicidal ideation of 56.4% and suicide attempts of 16.1%, alongside associations involving depression, family conflict, self-harm, and mental-health service use (Chen et al. 2019).

The historically important point is that denominator change expresses a research decision. The 2,060 valid-questionnaire database included transgender men, transgender women, genderqueer people, and cross-dressers. Chen and colleagues focused on transgender men and women and then applied the needs of their statistical analysis. The figure 1,309 means “the analytic sample for this paper’s questions and variables”; 2,060 means “the valid-questionnaire database in the original report.” They belong to the same data lineage while retaining separate meanings.

The paper opened another citation route. Later systematic reviews of self-harm and suicidality included the Chinese sample, and clinical and public-health work cited its associated factors. A questionnaire item recruited through a community survey had therefore crossed community organizing, a university report, psychiatric and psychological research, evidence synthesis, and clinical discussion. Survey impact can be understood through the length and diversity of that citation chain.

7. How 2,060 became 1,304 and 1,106: medical and school reanalysis

Liu and colleagues’ 2020 paper in The Journal of Sexual Medicine selected 1,304 transgender men and women from the 2,060 valid questionnaires: 626 transgender men and 678 transgender women. It examined desire for gender-affirming hormone therapy and surgery, actual treatment status, access channels, and medical monitoring. The paper reported that 79.4% of the analytic participants expressed desire for hormone therapy and documented difficulty obtaining physician-supported care, use of informal medication channels, and limited regular monitoring (Liu et al. 2020). The analysis converted the report’s broad health-care chapter into a more tightly specified medical-services study.

Spielmann and colleagues formed a further analytic sample of 1,106. Their work focused on school discrimination and environmental support and examined associations with mental health and self-harm, explicitly identifying the 2017 survey as its data source (Spielmann et al. 2022). The same database thus supported a social-psychological and school-environment research question.

Placed side by side, 1,309, 1,304, and 1,106 reveal a survey functioning as a recomposable evidence base. Each denominator follows from a research question: suicidality, health-care access, and school environment require different identity ranges and valid variables. A secondary citation gains precision when it identifies the paper, denominator, and subset instead of compressing every result into a generic reference to “the 2017 national survey.”

8. From 2017 to 2021: how a survey became a comparison baseline

A new Chinese Transgender Health Survey was conducted in 2021 and generated a larger later wave of data. Wang and colleagues’ Nature Mental Health paper used that survey to study gender-identity conversion practices, mental health, substance use, and suicidality. Hou and colleagues’ JAMA Network Open paper used a transgender-men and transgender-women subset to examine milestones of gender identity development and hormone utilization (Wang et al. 2023; Hou et al. 2026). Each paper again applied an analytic denominator tailored to its own question.

Hou and colleagues make the 2017 survey’s infrastructural legacy especially visible. Their methods describe a two-step measurement of sex assigned at birth and current gender identity and state that the 2021 survey’s target sample size drew on experience from the 2017 survey. Their analysis also compares 2021 hormone desire, use, and prescription access with published 2017 results for transgender men and women (Hou et al. 2026). The 2017 project therefore became both a data baseline and a design reference.

The appropriate historical unit for this comparison is the survey wave. The 2017 and 2021 projects recruited different participants at different times under evolving questionnaires and network conditions. Differences between the two waves can describe contrasts observed across survey samples and generate questions about changing health-care access. Individual developmental trajectories require longitudinal follow-up of the same people. Keeping these two levels separate gives cross-year comparison a clear interpretive scope.

The 2021 work also shows growth in community research infrastructure: larger recruitment networks, more systematic identity measurement, richer health-care and mental-health items, explicit quality controls, and a broader peer-reviewed publication pipeline. In this sense, 2017 is both a substantive baseline and an organizational-methodological prehistory for later transgender survey research in China.

9. An original framework: the evidence-infrastructure chain

To describe the survey’s long life, this entry proposes an evidence-infrastructure chain:

community recruitment → validity gate → report tables → public circulation → analytic subset → peer-reviewed reuse → policy and advocacy citation → digital preservation → cross-survey comparison

The first link is community recruitment. Beijing LGBT Center and partner networks carried the questionnaire into platforms and organizations able to reach participants. The second is the validity gate, which converted 5,677 submissions into 2,060 valid questionnaires. The third is report production, where individual responses became percentages, figures, and tables. The fourth is public circulation, where health care, family, school, and mental-health statistics became part of media and organizational discourse.

The fifth and sixth links are academic reanalysis. Researchers defined question-specific samples for suicidality, medical care, and school experience, producing denominators such as 1,309, 1,304, and 1,106, and moved those analyses through peer review. The seventh link is citation in policy, law, advocacy, and institutional publications. Survey figures acquired relevance for discussions of health-care standards, legal recognition, discrimination, and service design. The eighth is digital preservation. As original websites changed structure and publication links moved, archives such as the Chinese Transgender Digital Archive preserved files, metadata, and original-link information so the report remained discoverable (Chinese Transgender Digital Archive 2025).

The ninth link is cross-survey comparison. The 2021 survey and later scholarship brought 2017 results into a new design context, giving the earlier data continuing comparative value. Each link alters context. Statistical evidence therefore has a version history in much the same way that laws, websites, and archival records have version histories.

This framework also makes community organizations visible as knowledge institutions. Community networks contribute issue identification and participant reach; academic partners contribute survey design and analytic interfaces; journalists, international organizations, and policy researchers extend circulation. Survey knowledge is built through this institutional combination.

10. An original tool: the number-provenance ledger

Readers, journalists, and researchers can keep a number-provenance ledger for every important statistic. At minimum it records eight fields:

FieldQuestion to record
Source editionDoes the number come from the 2017 report, a derived paper, or the 2021 survey?
Recruitment frameWhich platforms, organizations, and networks recruited participants?
DenominatorDoes the percentage use N=2,060, 1,309, 1,304, 1,106, or another count?
Identity subsetWhich identity categories are included?
Item or measureWhat did the questionnaire ask, and how did a paper define the variable?
Analytic filtersDid inclusion require complete variables, a particular age range, or an identity definition?
Time and geographyWhen were data collected, and which places were represented?
Inferential scopeWhich participant set and which research question does the number describe?

For example, “79.4% expressed desire for hormone therapy” comes from Liu and colleagues’ 1,304-person analytic sample of transgender men and women. The original report’s “62% wanted hormone therapy” uses a broader report classification and denominator. The figures describe different participant frames. A provenance ledger makes the distinction visible in the denominator, identity-subset, and measure fields (Beijing LGBT Center and Department of Sociology, Peking University 2017; Liu et al. 2020).

The same logic applies to suicide-related statistics. Chen and colleagues’ 56.4% lifetime suicidal-ideation estimate belongs to their 1,309-person transgender-men and transgender-women analytic sample. The original report and contemporary journalism used different items and participant frames. Labeling the data state moves the reader from asking which percentage is larger to asking who, what, and when each percentage measures.

The ledger also improves 2017–2021 comparisons. When Hou and colleagues align hormone-related results from the two surveys, their analysis performs substantial work to align identity subsets and measures. Secondary writing should still mark the two waves as different participant samples. Cross-year findings can then be described as observed differences between survey waves while methodology and interpretation remain linked (Hou et al. 2026).

11. Five recurrent citation mismatches and stronger reading practices

A first mismatch treats nationwide geographic coverage as a statement about the target population’s demographic distribution. A stronger citation records both the nationwide provincial reach and the sample’s young, urban, highly educated, eastern concentration. Geographic reach describes where recruitment succeeded; sample structure describes who participated (Beijing LGBT Center and Department of Sociology, Peking University 2017).

A second mismatch reads the historical phrase “Population General Survey” as an estimate of the total transgender population. A stronger citation keeps the report’s historical title and simultaneously records its online community and platform recruitment, 2,060 valid questionnaires, and descriptive purpose. Population-size estimation is a separate inferential task (China Daily 2017a; Liu et al. 2020).

A third mismatch merges different analytic denominators. The 1,309, 1,304, and 1,106 samples come from suicidality, health-care, and school studies. A stronger citation places the paper, denominator, and identity subset next to each result (Chen et al. 2019; Liu et al. 2020; Spielmann et al. 2022).

A fourth mismatch treats comparison between the 2017 and 2021 cross-sectional surveys as an individual longitudinal trajectory. A stronger description calls it a comparison between two survey waves and records collection time, recruitment, and identity measurement for each (Wang et al. 2023; Hou et al. 2026).

A fifth mismatch transfers historical questionnaire language directly into present-day editorial narration. The original report’s FTM, MTF, and “biological sex” language is important evidence about 2017 measurement practice. Current narration can preserve those words in titles, quotations, and method discussion while using current terminology elsewhere (Beijing LGBT Center and Department of Sociology, Peking University 2017; Kronk et al. 2022).

These boundaries increase the survey’s long-term utility. Clear scope makes a dataset easier to reuse because later readers can determine which questions it supports, which comparisons are appropriate, and which additional sources should be paired with it.

12. Historical significance: community knowledge entering academic and institutional interfaces

One durable contribution of the 2017 survey was the conversion of dispersed experiences into data that could move across institutional interfaces. Family conflict, school experience, health-care demand, mental health, and public-space encounters became structured questions and comparable records. Community organizations could use the resulting evidence to convert service observations into public knowledge, while academic researchers could build focused analyses on the same base (Beijing LGBT Center and Department of Sociology, Peking University 2017; Chen et al. 2019; Spielmann et al. 2022).

The project also documents a community–university model of knowledge production. Beijing LGBT Center brought sustained community contact, service experience, and issue recognition; the Peking University sociology partnership supplied a research and academic interface. Guo’s study of the Beijing Tongzhi Center and the broader scholarship collected in Queer/Tongzhi China provide organizational context for understanding this form of collaboration (Guo 2018; Engebretsen, Schroeder, and Bao 2015). The survey sits at the intersection of organizing, research, and public communication.

Peer-reviewed reuse subsequently changed the report’s scholarly position. It became a data source for research on mental health, gender-affirming medical care, and school environments. Amnesty International’s health-care report, later legal scholarship on transgender workers, and other policy materials continued to cite it, carrying the survey into legal and institutional knowledge chains (Amnesty International 2019; Fappani 2024).

Digital preservation created another historical layer. After publication pages and download paths changed, archival projects preserved the report file, metadata, and original source URL, leaving the 2017 material directly discoverable in 2025 and 2026 (Chinese Transgender Digital Archive 2025). Survey methods, organizational partnerships, web preservation, and scholarly reanalysis thus form one evidence ecology in the history of transgender knowledge in China.

13. A practical reading and citation guide

General readers can begin with the original report’s methods page, then read the sample-composition page, and then choose health care, family, school, or mental health. The methods page explains the transition from 5,677 submissions to 2,060 valid records. The sample page describes age, education, urban residence, and regional concentration. Together they define the interpretive range for later percentages (Beijing LGBT Center and Department of Sociology, Peking University 2017).

Journalists and public writers can attach “source + denominator + population” to every major number. A sentence such as “A 2020 study using the 2017 survey analyzed 1,304 transgender men and women and found…” preserves much more methodological information than a generic attribution to a national survey (Liu et al. 2020).

Academic researchers can distinguish original descriptive reporting, derived analytic samples, and the later 2021 survey wave. For a specific statistical association, the peer-reviewed paper that performed the analysis is usually the closest source. For the history of the survey, community organization, and public circulation, the original report and release materials serve as primary historical records. A number-provenance ledger can sit alongside research notes to preserve relationships among variables, denominators, and versions.

Archivists can preserve the report file, release page, contemporary media coverage, derived scholarship, and later surveys as a connected set. File preservation protects the text. Preservation of publication date, organizational attribution, original URL, mirror date, and later citation relationships reconstructs the full evidence-infrastructure chain.

Conclusion

The 2017 Chinese Transgender Population General Survey left two connected legacies in the history of transgender knowledge in mainland China. The first is substantive: 2,060 valid questionnaires brought health care, family, school, work, public-space, and mental-health experience into one large nationwide-reach quantitative record. The second is infrastructural: the dataset continued to be reorganized by media, community organizations, international institutions, and academic researchers, producing analytic samples such as 1,309, 1,304, and 1,106 and eventually serving as a comparison baseline for the 2021 survey and scholarship published through 2026.

The evidence-infrastructure chain shows how a questionnaire becomes public knowledge. The number-provenance ledger provides a daily citation practice. Together they emphasize a simple historical principle: statistical figures have provenance histories. Recording where a number came from, which filters it passed through, and which people and question it describes allows transgender history to remain readable, verifiable, and methodologically transparent.

The survey’s historical value continues to develop. It records a set of transgender and gender-diverse participants reached by digital community networks in 2017, and it records how community organizations, universities, journalism, scholarly journals, policy institutions, and digital archives jointly constructed a public evidence base that has lasted for nearly a decade. Writing that construction process into history helps explain how transgender knowledge acquires scale, citation, and preservation.

References

  1. Chen, Runsen, Xuequan Zhu, L. Wright, Jack Drescher, Y. Gao, L. Wu, X. Ying, J. Qi, C. Chen, Y. Xi, L. Ji, H. Zhao, J. Ou, and Matthew R. Broome. 2019. “Suicidal Ideation and Attempted Suicide Amongst Chinese Transgender Persons: National Population Study.” Journal of Affective Disorders 245: 1126–1134. https://doi.org/10.1016/j.jad.2018.12.011
  2. Liu, Ye, Runsen Chen, Jianjun Ou, Xuequan Zhu, Lijuan Wu, Yuanyuan Wang, et al. 2020. “The Desire and Status of Gender-Affirming Hormone Therapy and Surgery in Transgender Men and Women in China: A National Population Study.” The Journal of Sexual Medicine 17(11): 2291–2298. https://doi.org/10.1016/j.jsxm.2020.07.081
  3. Spielmann, Julia, Siqi Feng, Daniel A. Briley, and Chadly Stern. 2022. “Mental Health Contributors Among Transgender People in a Non-WEIRD Society: Evidence From China.” Social Psychological and Personality Science 13(3): 747–757. https://doi.org/10.1177/19485506211039388
  4. Hou, Jiayu, et al. 2026. “Gender Identity Milestones and Hormone Utilization in Transgender Men and Women in China.” JAMA Network Open 9(1): e2552440. https://doi.org/10.1001/jamanetworkopen.2025.52440
  5. Wang, Yuanyuan, Zhishan Hu, Ke Peng, Joanne Rechdan, Yuan Yang, Lijuan Wu, et al. 2023. “A National Transgender Health Survey from China Assessing Gender Identity Conversion Practice, Mental Health, Substance Use and Suicidality.” Nature Mental Health 1: 254–265. https://doi.org/10.1038/s44220-023-00041-z
  6. Chen, Y., S. Chen, S. Arayasirikul, et al. 2020. “A Cross-Sectional Study of Mental Health, Suicidal Ideation and Suicide Attempt among Transgender Women in Jiangsu Province, China.” Journal of Affective Disorders 277: 869–874. https://doi.org/10.1016/j.jad.2020.09.002
  7. Yan, H., et al. 2019. “High HIV Prevalence and Associated Risk Factors among Transgender Women in China: A Cross-Sectional Survey.” Journal of the International AIDS Society 22: e25417. https://doi.org/10.1002/jia2.25417
  8. Yan, Zi-Han, et al. 2019. “Identity, Stigma, and HIV Risk among Transgender Women: A Qualitative Study in Jiangsu Province, China.” Infectious Diseases of Poverty. https://doi.org/10.1186/s40249-019-0606-9
  9. Wang, Liying, et al. 2023. “Health Service Utilization and Its Associations with Depression and Sexual Risk Behaviors Among Transgender Women in Shanghai, China.” Transgender Health 8(6): 516–525. https://doi.org/10.1089/trgh.2021.0009
  10. Yang, Xiaoshi, et al. 2015. “Sex Partnership and Self-Efficacy Influence Depression in Chinese Transgender Women: A Cross-Sectional Study.” PLOS ONE 10(9): e0136975. https://doi.org/10.1371/journal.pone.0136975
  11. Lin, T., Y. Cheng, T. L. Hughes, et al. 2022. “LGBTQ Health and Wellbeing in China: A Trend Analysis of English- and Chinese-Language Research, 2011–2018.” https://www.moderntreatise.com/s/LGBTQ-Health-and-Wellbeing-in-China-A-Trend-Analysis-of-English-and-Chinese-Language-Research-201120.pdf
  12. Chiang, Howard. 2018. After Eunuchs: Science, Medicine, and the Transformation of Sex in Modern China. Columbia University Press. https://cup.columbia.edu/book/after-eunuchs/9780231185790
  13. Chiang, Howard. 2021. Transtopia in the Sinophone Pacific. Columbia University Press. https://howardchiang.faculty.eastasian.ucsb.edu/publications/books/
  14. Liu, Sha. 2023. Performative Authenticity: Chinese Transgender People’s Digital Gender Practices. PhD thesis, London School of Economics. https://etheses.lse.ac.uk/4608/
  15. Engebretsen, Elisabeth L., William F. Schroeder, and Hongwei Bao, eds. 2015. Queer/Tongzhi China: New Perspectives on Research, Activism and Media Cultures. NIAS Press. https://cnlgbtdata.com/files/uploads/2024/06/Queer_Tongzhi_China.pdf
  16. Guo, Lifu. 2018. “Tongzhi Solidarity under Chinese Authoritarian Government: A Case Study of the Beijing Tongzhi Center.” https://repository.dl.itc.u-tokyo.ac.jp/record/49786/
  17. Bao, Hongwei. 2018. “From ‘Celluloid Comrades’ to ‘Digital Video Activism’: Queer Filmmaking in Postsocialist China.” https://eprints.nottingham.ac.uk/49485/
  18. Bai, Tianyi. 2026. “No-One Signed Up: Fatigue, Failure, and Fragile Optimism in Chinese Trans Advocacy.” https://doi.org/10.22459/mic.11.01.2026.13
  19. Meyer, Ilan H. 2003. “Prejudice, Social Stress, and Mental Health in Lesbian, Gay, and Bisexual Populations: Conceptual Issues and Research Evidence.” Psychological Bulletin 129(5): 674–697. https://doi.org/10.1037/0033-2909.129.5.674
  20. Henrich, Joseph, Steven J. Heine, and Ara Norenzayan. 2010. “Beyond WEIRD: Towards a Broad-Based Behavioral Science.” Behavioral and Brain Sciences 33(2–3): 111–135. https://doi.org/10.1017/S0140525X10000725
  21. Guo, Shaohua. 2020. The Evolution of the Chinese Internet: Creative Visibility in the Digital Public. Stanford University Press. https://www.sup.org/books/asian-studies/evolution-chinese-internet
  22. Reed, Geoffrey M., Jack Drescher, Richard B. Krueger, Elham Atalla, Susan D. Cochran, Michael B. First, Peggy T. Cohen-Kettenis, Iván Arango-de Montis, Sharon J. Parish, Sara Cottler, Peer Briken, and Shekhar Saxena. 2016. “Disorders Related to Sexuality and Gender Identity in the ICD-11: Revising the ICD-10 Classification Based on Current Scientific Evidence, Best Clinical Practices, and Human Rights Considerations.” World Psychiatry 15(3): 205–221. https://doi.org/10.1002/wps.20354
  23. Fappani, Valeria. 2024. “The Protection of Transgender Workers in China: Between Law and Society.” Rivista di Diritto dell’Asia Orientale 1: 51–74. https://doi.org/10.30687/RIDAO/4713-223X/2024/01/003
  24. Zhang, Q., et al. 2020. “Epidemiological Considerations in Transgender Health: A Systematic Review with Focus on Higher Quality Data.” International Journal of Transgender Health 21(2): 125–137. https://doi.org/10.1080/26895269.2020.1753136
  25. Kronk, Clair A., et al. 2022. “Transgender Data Collection in the Electronic Health Record: Current Concepts and Issues.” Journal of the American Medical Informatics Association 29(2): 271–284. https://doi.org/10.1093/jamia/ocab136
  26. Beijing LGBT Center and Department of Sociology, Peking University. 2017. Chinese Transgender Population General Survey Report: Transgender and Gender-Nonconforming People. https://digital.transchinese.org/社群及NGO文件/统计报告/Chinese_Transgender_Population_General_Survey_Report.pdf
  27. China Development Brief. 2017. “2017 Chinese Transgender Population General Survey Report.” https://chinadevelopmentbrief.org/publications/2017-chinese-transgender-population-general-survey-report/
  28. China Development Brief. 2017. “中国跨性别群体生存现状调查报告圆满发布.” https://www.chinadevelopmentbrief.org.cn/news/detail/17772.html
  29. China Daily. 2017a. “Medical Needs Unmet for Many in Transgender Community.” https://www.chinadaily.com.cn/china/2017-11/20/content_34778453.htm
  30. China Daily. 2017b. “Medical Services for Transgender People Needed, Survey Finds.” https://www.chinadaily.com.cn/china/2017-11/21/content_34798757.htm
  31. People’s Daily Online. 2018. “90% of Chinese Transgenders Threatened and Insulted, Received ‘Therapy’: Survey.” https://en.people.cn/n3/2018/0329/c90000-9442973.html
  32. Chinese Transgender Digital Archive. 2025. “Chinese_Transgender_Population_General_Survey_Report.” https://digital.transchinese.org/社群及NGO文件/统计报告/Chinese_Transgender_Population_General_Survey_Report_page/
  33. Chinese Transgender Digital Archive. 2025. “中国跨性别群体—生存现状调查报告.” https://digital.transchinese.org/社群及NGO文件/统计报告/中国跨性别群体-_生存现状调查报告_page/
  34. Transgender Resource Center Hong Kong. 2017. “2017中国跨性别群体生存现状调查报告.” https://www.tgr.org.hk/index.php/zh/2017/11/20/2017-chinese-transgender-population-general-survey-report/
  35. Equal Eyes. 2017. “China: 2017 Chinese Transgender Population General Survey Report.” https://equal-eyes.org/database/2017/12/11/china-2017-chinese-transgender-population-general-survey-report
  36. Voice of America Chinese. 2017. “年终报道：2017年中国LGBT人群现况与展望.” https://www.voachinese.com/a/year-ender-china-lgbt-20171229/4184970.html
  37. UNDP. 2014. Being LGBT in Asia: China Country Report. https://www.undp.org/china/publications/being-lgbt-asia-china-country-report
  38. UNDP. 2016. Being LGBTI in China: A National Survey on Social Attitudes towards Sexual Orientation, Gender Identity and Gender Expression. https://www.undp.org/china/publications/being-lgbti-china
  39. UNDP. 2018. Legal Gender Recognition in China: A Legal and Policy Review. https://www.undp.org/china/publications/legal-gender-recognition-china-legal-and-policy-review
  40. Amnesty International. 2019. China: “I Need My Parents’ Consent to Be Myself”: Barriers to Gender-Affirming Treatments for Transgender People in China. https://www.amnesty.org/en/documents/asa17/0269/2019/en/
  41. Williams Institute. 2021. Public Opinion of Transgender Rights in China. https://williamsinstitute.law.ucla.edu/publications/opinion-trans-rights-china/
  42. Common Language. 2022. “Ta山 | 数据报告出炉！中国公众对跨性别者权利怎么看？” https://commonlanguage.github.io/TYarchives2022-2/20220331_1Ta山数据报告出炉！中国公众对跨性别者权利怎么看？.html
  43. Rainbow Lawyers. 2019. 跨性别证件修改手册. https://digital.transchinese.org/社群及NGO文件/手册指南/跨性别证件修改手册_page/
  44. Project Trans. 2026. “中国大陆跨性别相关法律法规变迁.” https://project-trans.org/china-legal/
  45. General Office of the Ministry of Health of the PRC. 2009. “关于印发《变性手术技术管理规范（试行）》的通知（卫办医政发〔2009〕185号）.” https://www.nhc.gov.cn/wjw/gfxwj/200911/6341a0f92cad4f09acfae1e5d5b56477.shtml
  46. General Office of the National Health Commission of the PRC. 2022. “国家限制类技术目录和临床应用管理规范（2022年版）.” https://www.nhc.gov.cn/yzygj/c100068/202204/2655831f6f444b00b3e50604e67531f5.shtml
  47. Ministry of Public Security of the PRC. 2021. “户口居民身份证管理工作规范（试行）（公通字〔2021〕12号）.” https://gaj.cq.gov.cn/zwgk/jczwgk/zcwj_224735/gabzcwj/202111/t20211112_9968734.html
  48. Beijing LGBT Center. 2021. 2021全国跨性别健康调研报告. https://digital.transchinese.org/社群及NGO文件/统计报告/2021全国跨性别健康调研报告_page/
  49. Beijing Impact Law Firm, Guangzhou Gender and Sexuality Education Center, and Guangzhou Trans Center. 2017. 跨性别校园与证件修改调查报告. https://digital.transchinese.org/社群及NGO文件/统计报告/跨性别校园与证件修改调查报告_page/
  50. Chinese Transgender Digital Archive. 2025. “2020年《跨性別就業與職場調查》報告.” https://digital.transchinese.org/社群及NGO文件/统计报告/2020年《跨性別就業與職場調查》報告.docx_page/