THE REPORT OF THE

About the National Center for Transgender Equality The National Center for Transgender Equality (NCTE) is the nation’s leading social justice policy advocacy organization devoted to ending discrimination and violence against transgender people. NCTE was founded in 2003 by transgender activists who recognized the urgent need for policy change to advance transgender equality. NCTE now has an extensive record winning life-saving changes for transgender people.

NCTE works by educating the public and by influencing local, state, and federal policymakers to change policies and laws to improve the lives of transgender people. By empowering transgender people and our allies, NCTE creates a strong and clear voice for transgender equality in our nation’s capital and around the country. © 2016 The National Center for Transgender Equality. We encourage and grant permission for the reproduction and distribution of this publication in whole or in part, provided that it is done with attribution to the National Center for Transgender Equality.

Further written permission is not required. RECOMMENDED CITATION James, S. E., Herman, J. L., Rankin, S., Keisling, M., Mottet, L., & Anafi, M. (2016). The Report of the 2015 U.S. Transgender Survey. Washington, DC: National Center for Transgender Equality.

The Report of the 2015 U.S. Transgender Survey by: Sandy E. James Jody L. Herman Susan Rankin Mara Keisling Lisa Mottet Ma’ayan Anafi December 2016

Acknowledgements ............................................................................................................... 1 Executive

Summary ............................................................................................................... 3

Chapter 1: Introduction ....................................................................................................... 18

Chapter 2: Methodology ..................................................................................................... 21

Chapter 3: Guide to Report and Terminology ................................................................... 39

Chapter 4: Portrait of USTS Respondents ......................................................................... 43

Chapter 5: Family Life and Faith Communities ................................................................. 64

Chapter 6: Identity Documents .......................................................................................... 81

Chapter 7: Health ............................................................................................................... 92

Chapter 8: Experiences at School .................................................................................... 130

Chapter 9: Income and Employment Status .................................................................... 139

Chapter 10: Employment and the Workplace .................................................................. 147

Chapter 11: Sex Work and Other Underground Economy Work ..................................... 157

Chapter 12: Military Service ............................................................................................. 166

Chapter 13: Housing, Homelessness, and Shelter Access ............................................. 175

Chapter 14: Police, Prisons, and Immigration Detention ................................................ 184

Chapter 15: Harassment and Violence ........................................................................... 197

Chapter 16: Places of Public Accommodation and Airport Security .............................. 212

Chapter 17: Experiences in Restrooms ............................................................................ 224

Chapter 18: Civic Participation and Policy Priorities ...................................................... 231 About the Authors ............................................................................................................. 241 Appendix A: Characteristics of the Sample ..................................................................... 243 Appendix B: Survey Instrument (Questionnaire) ........................................................... 250 Appendix C: Detailed Methodology ................................................................................ 291 Table of Contents

ACKNOWLEDGEMENTS 1 T he report authors extend our gratitude to all the members of the transgender community who participated in the 2015 U.S. Transgender Survey and the hundreds of individuals and organizations that made this survey report possible. The following individuals in particular are recognized for their contributions during one or more stages of the process, including survey development, implementation, distribution, data preparation and analysis, and reporting: Osman Ahmed M. V.

Lee Badgett Kellan Baker Genny Beemyn Aaron Belkin Walter Bockting Kyler Broadus David Chae Cecilia Chung Loree Cook Daniels Kerith Conron Ruby Corado Andrew Cray Kate D’Adamo Laura Durso Jake Eleazer Gary J.

Gates Jaime Grant Emily Greytak Ann Haas Jack Harrison-Quintana Mark Hatzenbuehler Darby Hickey Lourdes Ashley Hunter Chai Jindasurat JoAnne Keatley Elliot Kennedy Paul Lillig Emilia Lombardi Jenifer McGuire Ilan Meyer Shannon Price Minter Aaron Morris Asaf Orr Dylan Orr Sari Reisner Gabriel Rodenborn Shelby Rowe Brad Sears Jama Shelton Jillian Shipherd Sharon Stapel Michael Steinberger Catalina Velasquez Bianca Wilson The authors extend special gratitude to the following individuals for their contributions to the project: Ignacio Rivera, Survey Outreach Coordinator, for leading the survey outreach efforts and building a network of individuals and organizations that led to the unprecedented levels of participation in this survey.

Andrew Flores for assistance with questionnaire development and the weighting procedures to prepare the data for analysis. Daniel Merson for continued work in cleaning the data set, re-coding the data, and providing population numbers for comparative purposes in the final report. Michael Rauch for programming and implementation of the complex survey.

Appreciation is also extended to NCTE interns, law fellows, staff, and volunteers who assisted with the project, including outreach across the country, development of outreach materials, and translation: Bre Kidman Cyres Gibson Mati Gonzalez Gil Romeo Jackson Kory Masen Imari Moon Cullen O’Keefe Nowmee Shehab Acknowledgements

2015 U.S.

TRANSGENDER SURVEY 2 A special thanks is also extended to USTS Interns, Fellows, and Assistants for their work at various stages of the project: Rodrigo Aguayo-Romero Willem Miller Shabab Mirza Davida Schiffer Jeymee Semiti Danielle Stevens Venus Selenite The authors also thank the USTS Advisory Committee (UAC) members, who devoted their time to the project by giving valuable recommendations around project development and community outreach: Danni Askini Cherno Biko Thomas Coughlin Brooke Cerda Guzman Trudie Jackson Andrea Jenkins Angelica Ross Nowmee Shehab Stephanie Skora Brynn Tannehill Additional acknowledgement goes to the more than 300 transgender, LGBT, and allied organizations that promoted and distributed the survey to its members throughout the country for completion.

The authors also acknowledge current and former NCTE staff for their work on the project, particularly: Theo George and Vincent Villano for their pivotal work in survey project development and distribution through their respective roles in digital media and communication strategy. Arli Christian, Joanna Cifredo, K’ai Smith, and Harper Jean Tobin for their contributions as report co-writers, including lending their subject-matter expertise and analysis to the findings included in this report.

Various individuals and firms assisted in spreading the word about the survey, both prior to and during the data collection phase, and they also assisted with survey translation, design, and reporting. Thanks goes to: Sean Carlson Leah Furumo Molly Haigh Anna Zuccaro Design Action Collective Dewey Square Group ThoughtWorks TransTech Social Enterprises ZERN Additionally, NCTE would like to express special appreciation to an anonymous donor for providing the largest share of the funds needed to conduct and report on the U.S. Transgender Survey.

Other important funders were the Arcus Foundation, the Gill Foundation, the Human Rights Campaign Foundation, and the David Bohnett Foundation. NCTE is also grateful to its other funders, many of which supported this project through general operating support, including the Ford Foundation, the Evelyn & Walter Haas, Jr. Fund, and the Tides Foundation.

Finally, NCTE would like to give special thanks to the National LGBTQ Task Force, for its previous partnership in conducting the National Transgender Discrimination Survey as a joint project from 2008 to 2011, as well as for its support of NCTE’s re-development of it as the U.S. Transgender Survey.

EXECUTIVE

SUMMARY 3

2015 U.S. TRANSGENDER SURVEY 4 USTS Executive

Summary T he 2015 U.S. Transgender Survey (USTS) is the largest survey examining the experiences of transgender people in the United States, with 27,715 respondents from all fifty states, the District of Columbia, American Samoa, Guam, Puerto Rico, and U.S. military bases overseas. Conducted in the summer of 2015 by the National Center for Transgender Equality, the USTS was an anonymous, online survey for transgender adults (18 and older) in the United States, available in English and Spanish.

The USTS serves as a follow-up to the groundbreaking 2008–09 National Transgender Discrimination Survey (NTDS), which helped to shift how the public and policymakers view the lives of transgender people and the challenges they face. The report of the 2015 USTS provides a detailed look at the experiences of transgender people across a wide range of categories, such as education, employment, family life, health, housing, and interactions with the criminal justice system.

The findings reveal disturbing patterns of mistreatment and discrimination and startling disparities between transgender people in the survey and the U.S. population when it comes to the most basic elements of life, such as finding a job, having a place to live, accessing medical care, and enjoying the support of family and community. Survey respondents also experienced harassment and violence at alarmingly high rates. Several themes emerge from the thousands of data points presented in the full survey report.

Pervasive Mistreatment and Violence Respondents reported high levels of mistreatment, harassment, and violence in every aspect of life. One in ten (10%) of those who were out to their immediate family reported that a family member was violent towards them because they were transgender, and 8% were kicked out of the house because they were transgender. The majority of respondents who were out or perceived as transgender while in school (K–12) experienced some form of mistreatment, including being verbally harassed (54%), physically attacked (24%), and sexually assaulted (13%) because they were transgender.

Further, 17% experienced such severe mistreatment that they left a school as a result. In the year prior to completing the survey, 30% of respondents who had a job reported being fired, denied a promotion, or experiencing some other form of mistreatment in the workplace due to their gender identity or expression, such as being verbally harassed or physically or sexually assaulted at work.

EXECUTIVE

SUMMARY 5 In the year prior to completing the survey, 46% of respondents were verbally harassed and 9% were physically attacked because of being transgender. During that same time period, 10% of respondents were sexually assaulted, and nearly half (47%) were sexually assaulted at some point in their lifetime. Severe Economic Hardship and Instability The findings show large economic disparities between transgender people in the survey and the U.S. population. Nearly one-third (29%) of respondents were living in poverty, compared to 12% in the U.S. population.

A major contributor to the high rate of poverty is likely respondents’ 15% unemployment rate—three times higher than the unemployment rate in the U.S. population at the time of the survey (5%). Respondents were also far less likely to own a home, with only 16% of respondents reporting homeownership, compared to 63% of the U.S. population. Even more concerning, nearly one-third (30%) of respondents have experienced homelessness at some point in their lifetime, and 12% reported experiencing homelessness in the year prior to completing the survey because they were transgender.

Harmful Effects on Physical and Mental Health The findings paint a troubling picture of the impact of stigma and discrimination on the health of many transgender people. A staggering 39% of respondents experienced serious psychological distress in the month prior to completing the survey, compared with only 5% of the U.S. population. Among the starkest findings is that 40% of respondents have attempted suicide in their lifetime—nearly nine times the attempted suicide rate in the U.S. population (4.6%). Respondents also encountered high levels of mistreatment when seeking health care.

In the year prior to completing the survey, one-third (33%) of those who saw a health care provider had at least one negative experience related to being transgender, such as being verbally harassed or refused treatment because of their gender identity. Additionally, nearly one-quarter (23%) of respondents reported that they did not seek the health care they needed in the year prior to completing the survey due to fear of being mistreated as a transgender person, and 33% did not go to a health care provider when needed because they could not afford it.

2015 U.S. TRANSGENDER SURVEY 6 The Compounding Impact of Other Forms of Discrimination When respondents’ experiences are examined by race and ethnicity, a clear and disturbing pattern is revealed: transgender people of color experience deeper and broader patterns of discrimination than white respondents and the U.S. population.

While respondents in the USTS sample overall were more than twice as likely as the U.S. population to be living in poverty, people of color, including Latino/a (43%), American Indian (41%), multiracial (40%), and Black (38%) respondents, were more than three times as likely as the U.S. population (12%) to be living in poverty. The unemployment rate among transgender people of color (20%) was four times higher than the U.S. unemployment rate (5%). People of color also experienced greater health disparities.

While 1.4% of all respondents were living with HIV— nearly five times the rate in the U.S. population (0.3%)—the rate among Black respondents (6.7%) was substantially higher, and the rate for Black transgender women was a staggering 19%. Undocumented respondents were also more likely to face severe economic hardship and violence than other respondents. In the year prior to completing the survey, nearly one - quarter (24%) of undocumented respondents were physically attacked.

Additionally, one - half (50%) of undocumented respondents have experienced homelessness in their lifetime, and 68% have faced intimate partner violence. Respondents with disabilities also faced higher rates of economic instability and mistreatment. Nearly one-quarter (24%) were unemployed, and 45% were living in poverty. Transgender people with disabilities were more likely to be currently experiencing serious psychological distress (59%) and more likely to have attempted suicide in their lifetime (54%). They also reported higher rates of mistreatment by health care providers (42%).

Increased Visibility and Growing Acceptance Despite the undeniable hardships faced by transgender people, respondents’ experiences also show some of the positive impacts of growing visibility and acceptance of transgender people in the United States. One such indication is that an unprecedented number of transgender people—nearly 28,000—completed the survey, more than four times the number of respondents in the 2008–09 NTDS. This number of transgender people who elevated their voices reflects the historic growth in visibility that the transgender community has seen in recent years.

Additionally, this growing visibility has lifted up not only the voices of transgender men and women, but also people who are non-binary, which is a term that is often used to describe

EXECUTIVE

SUMMARY 7 people whose gender identity is not exclusively male or female, including those who identify as having no gender, a gender other than male or female, or more than one gender. With non-binary people making up over one-third of the sample, the need for advocacy that is inclusive of all identities in the transgender community is clearer than ever. Respondents’ experiences also suggest growing acceptance by family members, colleagues, classmates, and other people in their lives.

More than half (60%) of respondents who were out to their immediate family reported that their family was supportive of them as a transgender person. More than two-thirds (68%) of those who were out to their coworkers reported that their coworkers were supportive. Of students who were out to their classmates, more than half (56%) reported that their classmates supported them as a transgender person. O verall, the report provides evidence of hardships and barriers faced by transgender people on a day-to-day basis.

It portrays the challenges that transgender people must overcome and the complex systems that they are often forced to navigate in multiple areas of their lives in order to survive and thrive. Given this evidence, governmental and private institutions throughout the United States should address these disparities and ensure that transgender people are able to live fulfilling lives in an inclusive society.

This includes eliminating barriers to quality, affordable health care, putting an end to discrimination in schools, the workplace, and other areas of public life, and creating systems of support at the municipal, state, and federal levels that meet the needs of transgender people and reduce the hardships they face. As the national conversation about transgender people continues to evolve, public education efforts to improve understanding and acceptance of transgender people are crucial.

The rates of suicide attempts, poverty, unemployment, and violence must serve as an immediate call to action, and their reduction must be a priority. Despite policy improvements over the last several years, it is clear that there is still much work ahead to ensure that transgender people can live without fear of discrimination and violence.

2015 U.S.

TRANSGENDER SURVEY 8 Overview of Key Findings Family Life and Faith Communities • A majority of respondents (60%) who were out to the immediate family they grew up with said that their family was generally supportive of their transgender identity, while 18% said that their family was unsupportive, and 22% said that their family was neither supportive nor unsupportive. • Those who said that their immediate families were supportive were less likely to report a variety of negative experiences related to economic stability and health, such as experiencing homelessness, attempting suicide, or experiencing serious psychological distress.

Experienced homelessness Attempted suicide Currently experiencing serious psychological distress 0% 10% 20% 30% 40% 50% 60% 27% 45% 37% 54% 31% 50% Negative experiences among those with supportive and unsupportive families % of respondents whose families were supportive % of respondents whose families were unsupportive • One in ten (10%) respondents who were out to their immediate family reported that a family member was violent towards them because they were transgender. • One in twelve (8%) respondents who were out to their immediate family were kicked out of the house, and one in ten (10%) ran away from home. • Nineteen percent (19%) of respondents who had ever been part of a spiritual or religious community left due to rejection.

Forty-two percent (42%) of those who left later found a welcoming spiritual or religious community.

EXECUTIVE

SUMMARY 9 Identity Documents • Only 11% of respondents reported that all of their IDs had the name and gender they preferred, while more than two-thirds (68%) reported that none of their IDs had the name and gender they preferred.

Driver’s license/ state-issued ID Social Security records Student records (current or last school attended) Passport Birth certificate Updated name or gender on ID OUT OF THOSE WHO HAD ID AND WANTED TO UPDATE IT (%) Updated name Updated gender 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 44% 29% 43% 23% 31% 18% 28% 18% 18% 9% • The cost of changing ID documents was one of the main barriers respondents faced, with 35% of those who have not changed their legal name and 32% of those who have not updated the gender on their IDs reporting that it was because they could not afford it. • Nearly one-third (32%) of respondents who have shown an ID with a name or gender that did not match their gender presentation were verbally harassed, denied benefits or service, asked to leave, or assaulted.

10 Health Insurance and Health Care • One in four (25%) respondents experienced a problem in the past year with their insurance related to being transgender, such as being denied coverage for care related to gender transition or being denied coverage for routine care because they were transgender. • More than half (55%) of those who sought coverage for transition-related surgery in the past year were denied, and 25% of those who sought coverage for hormones in the past year were denied. • One-third (33%) of those who saw a health care provider in the past year reported having at least one negative experience related to being transgender, with higher rates for people of color and people with disabilities.

This included being refused treatment, verbally harassed, or physically or sexually assaulted, or having to teach the provider about transgender people in order to get appropriate care. • In the past year, 23% of respondents did not see a doctor when they needed to because of fear of being mistreated as a transgender person, and 33% did not see a doctor when needed because they could not afford it.

Psychological Distress and Attempted Suicide • Thirty-nine percent (39%) of respondents experienced serious psychological distress in the month before completing the survey (based on the Kessler 6 Psychological Distress Scale), compared with only 5% of the U.S. population. • Forty percent (40%) have attempted suicide in their lifetime, nearly nine times the rate in the U.S. population (4.6%). • Seven percent (7%) attempted suicide in the past year—nearly twelve times the rate in the U.S. population (0.6%).

HIV • Respondents were living with HIV (1.4%) at nearly five times the rate in the U.S. population (0.3%). • HIV rates were higher among transgender women (3.4%), especially transgender women of color. Nearly one in five (19%) Black transgender women were living with HIV, and American Indian (4.6%) and Latina (4.4%) women also reported higher rates.

EXECUTIVE

SUMMARY 11 Experiences in Schools • More than three-quarters (77%) of those who were out or perceived as transgender at some point between Kindergarten and Grade 12 (K–12) experienced some form of mistreatment, such as being verbally harassed, prohibited from dressing according to their gender identity, disciplined more harshly, or physically or sexually assaulted because people thought they were transgender. • Fifty-four percent (54%) of those who were out or perceived as transgender in K–12 were verbally harassed, nearly one-quarter (24%) were physically attacked, and 13% were sexually assaulted in K–12 because of being transgender. • Seventeen percent (17%) faced such severe mistreatment as a transgender person that they left a K–12 school. • Nearly one-quarter (24%) of people who were out or perceived as transgender in college or vocational school were verbally, physically, or sexually harassed.

Experiences of people who were out as transgender in K–12 or believed classmates, teachers, or school staff thought they were transgender EXPERIENCES % OF THOSE WHO WERE OUT OR PERCEIVED AS TRANSGENDER Verbally harassed because people thought they were transgender 54% Not allowed to dress in a way that fit their gender identity or expression 52% Disciplined for fighting back against bullies 36% Physically attacked because people thought they were transgender 24% Believe they were disciplined more harshly because teachers or staff thought they were transgender 20% Left a school because the mistreatment was so bad 17% Sexually assaulted because people thought they were transgender 13% Expelled from school 6% One or more experiences listed 77%

2015 U.S. TRANSGENDER SURVEY 12 Unemployment rate RACE/ETHNICITY (%) Income and Employment Status • The unemployment rate among respondents (15%) was three times higher than the unemployment rate in the U.S. population (5%), with Middle Eastern, American Indian, multiracial, Latino/a, and Black respondents experiencing higher rates of unemployment.

Overall American Indian Asian Black Latino/a Middle Eastern* Multiracial White 5% 15% 12% 23% 4% 10% 10% 20% 7% 21% 35% 9% 22% 4% 12% 0% 5% 10% 15% 20% 25% 30% 35% % in USTS (supplemental survey weight applied) % in U.S. population (CPS) * U.S. population data for Middle Eastern people alone is unavailable in the CPS. • Nearly one-third (29%) were living in poverty, more than twice the rate in the U.S. population (12%).

Employment and the Workplace • One in six (16%) respondents who have ever been employed—or 13% of all respondents in the sample— reported losing a job because of their gender identity or expression in their lifetime. • In the past year, 27% of those who held or applied for a job during that year—19% of all respondents— reported being fired, denied a promotion, or not being hired for a job they applied for because of their gender identity or expression. • Fifteen percent (15%) of respondents who had a job in the past year were verbally harassed, physically attacked, and/or sexually assaulted at work because of their gender identity or expression. • Nearly one-quarter (23%) of those who had a job in the past year reported other forms of mistreatment based on their gender identity or expression during that year,

EXECUTIVE

SUMMARY 13 such as being forced to use a restroom that did not match their gender identity, being told to present in the wrong gender in order to keep their job, or having a boss or coworker share private information about their transgender status without their permission. • Overall, 30% of respondents who had a job in the past year reported being fired, denied a promotion, or experiencing some other form of mistreatment related to their gender identity or expression. • More than three-quarters (77%) of respondents who had a job in the past year took steps to avoid mistreatment in the workplace, such as hiding or delaying their gender transition or quitting their job. • Nearly one-quarter (23%) of respondents experienced some form of housing discrimination in the past year, such as being evicted from their home or denied a home or apartment because of being transgender. • Nearly one-third (30%) of respondents have experienced homelessness at some point in their lives. • In the past year, one in eight (12%) respondents experienced homelessness because of being transgender. • More than one-quarter (26%) of those who experienced homelessness in the past year avoided staying in a shelter because they feared being mistreated as a transgender person.

Those who did stay in a shelter reported high levels of mistreatment: seven out of ten (70%) respondents who stayed in a shelter in the past year reported some form of mistreatment, including being harassed, sexually or physically assaulted, or kicked out because of being transgender. Seven out of ten respondents who stayed in a shelter in the past year reported being mistreated because of being transgender. • Respondents were nearly four times less likely to own a home (16%) compared to the U.S. population (63%). Housing, Homelessness, and Shelter Access

2015 U.S. TRANSGENDER SURVEY 14 Sex Work and Other Underground Economy Work • Respondents reported high rates of experience in the underground economy, including sex work, drug sales, and other work that is currently criminalized.

One in five (20%) have participated in the underground economy for income at some point in their lives— including 12% who have done sex work in exchange for income—and 9% did so in the past year, with higher rates among women of color. • Respondents who interacted with the police either while doing sex work or while the police mistakenly thought they were doing sex work reported high rates of police harassment, abuse, or mistreatment, with nearly nine out of ten (86%) reporting being harassed, attacked, sexually assaulted, or mistreated in some other way by police. • Those who have done income-based sex work were also more likely to have experienced violence.

More than three-quarters (77%) have experienced intimate partner violence and 72% have been sexually assaulted, a substantially higher rate than the overall sample. Out of those who were working in the underground economy at the time they took the survey, nearly half (41%) were physically attacked in the past year and over one-third (36%) were sexually assaulted during that year. Police Interactions and Prisons • Respondents experienced high levels of mistreatment and harassment by police.

In the past year, of respondents who interacted with police or law enforcement officers who thought or knew they were transgender, more than half (58%) experienced some form of mistreatment. This included being verbally harassed, repeatedly referred to as the wrong gender, physically assaulted, or sexually assaulted, including being forced by officers to engage in sexual activity to avoid arrest. • Police frequently assumed that respondents—particularly transgender women of color— were sex workers.

In the past year, of those who interacted with law enforcement officers who thought or knew they were transgender, one-third (33%) of Black transgender women and 30% of multiracial women said that an officer assumed they were sex workers. • More than half (57%) of respondents said they would feel uncomfortable asking the police for help if they needed it. • Of those who were arrested in the past year (2%), nearly one-quarter (22%) believed they were arrested because they were transgender.

EXECUTIVE

SUMMARY 15 • Respondents who were held in jail, prison, or juvenile detention in the past year faced high rates of physical and sexual assault by facility staff and other inmates. In the past year, nearly one-quarter (23%) were physically assaulted by staff or other inmates, and one in five (20%) were sexually assaulted. Respondents were over five times more likely to be sexually assaulted by facility staff than the U.S. population in jails and prisons, and over nine times more likely to be sexually assaulted by other inmates.

Harassment and Violence • Nearly half (46%) of respondents were verbally harassed in the past year because of being transgender. • Nearly one in ten (9%) respondents were physically attacked in the past year because of being transgender. • Nearly half (47%) of respondents were sexually assaulted at some point in their lifetime and one in ten (10%) were sexually assaulted in the past year.

Respondents who have done sex work (72%), those who have experienced homelessness (65%), and people with disabilities (61%) were more likely to have been sexually assaulted in their lifetime. • More than half (54%) experienced some form of intimate partner violence, including acts involving coercive control and physical harm. • Nearly one-quarter (24%) have experienced severe physical violence by an intimate partner, compared to 18% in the U.S. population.

Transgender women reporting that police assumed they were sex workers in the past year (out of those who interacted with officers who thought they were transgender) RACE/ETHNICITY (%) Overall* American Indian women Asian women Middle Eastern women** Multiracial women Black women Latina women White women *Represents respondents of all genders who interacted with officers who thought they were transgender **Sample size too low to report 0% 5% 10% 15% 20% 25% 30% 35% 11% 23% 20% 33% 25% 30% 11%

2015 U.S.

TRANSGENDER SURVEY 16 LOCATION VISITED % OF THOSE WHO SAID STAFF KNEW OR THOUGHT THEY WERE TRANSGENDER Public transportation 34% Retail store, restaurant, hotel, or theater 31% Drug or alcohol treatment program 22% Domestic violence shelter or program or rape crisis center 22% Gym or health club 18% Public assistance or government benefit office 17% Department of Motor Vehicles (DMV) 14% Nursing home or extended care facility 14% Court or courthouse 13% Social Security office 11% Legal services from an attorney, clinic, or legal professional 6% Denied equal treatment or service, verbally harassed, or physically attacked in public accommodations in the past year because of being transgender Places of Public Accommodation • Respondents reported being denied equal treatment or service, verbally harassed, or physically attacked at many places of public accommodation—places that provide services to the public, like retail stores, hotels, and government offices.

Out of respondents who visited a place of public accommodation where staff or employees thought or knew they were transgender, nearly one-third (31%) experienced at least one type of mistreatment in the past year in a place of public accommodation. This included 14% who were denied equal treatment or service, 24% who were verbally harassed, and 2% who were physically attacked because of being transgender. • One in five (20%) respondents did not use at least one type of public accommodation in the past year because they feared they would be mistreated as a transgender person.

Experiences in Restrooms The survey data was collected before transgender people’s restroom use became the subject of increasingly intense and often harmful public scrutiny in the national media and legislatures around the country in 2016.

Y et respondents reported facing frequent harassment and barriers when using restrooms at school, work, or in public places. • Nearly one in ten (9%) respondents reported that someone denied them access to a restroom in the past year. • In the past year, respondents reported being verbally harassed (12%), physically attacked (1%), or sexually assaulted (1%) when accessing a restroom.

EXECUTIVE

SUMMARY 17 • More than half (59%) of respondents avoided using a public restroom in the past year because they were afraid of confrontations or other problems they might experience. • Nearly one-third (32%) of respondents limited the amount that they ate and drank to avoid using the restroom in the past year. • Eight percent (8%) reported having a urinary tract infection, kidney infection, or another kidney-related problem in the past year as a result of avoiding restrooms.

Civic Participation and Party Affiliation • More than three-quarters (76%) of U.S. citizens of voting age in the sample reported that they were registered to vote in the November 2014 midterm election, compared to 65% in the U.S. population. • More than half (54%) of U.S. citizens of voting age reported that they had voted in the midterm election, compared to 42% in the U.S. population. • Half (50%) of respondents identified as Democrats, 48% identified as Independents, and 2% identified as Republicans, compared to 27%, 43%, and 27% in the U.S. population, respectively.

Political party affiliation POLITICAL PARTY % IN USTS % IN U.S. POPULATION (GALLUP) Democrat 50% 27% Independent 48% 43% Republican 2% 27% More than half (59%) of respondents avoided using a public restroom in the past year because they were afraid of confrontations or other problems they might experience.

2015 U.S. TRANSGENDER SURVEY 18 T his report presents the findings of the 2015 U.S. Transgender Survey (USTS), a study conducted by the National Center for Transgender Equality (NCTE). With 27,715 respondents, it is the largest - ever survey examining the lives of transgender people in the United States. The USTS provides a detailed portrait of the experiences of transgender people across many areas, including health, family life, employment, and interactions with the criminal justice system.

The USTS serves as a follow-up to the National Transgender Discrimination Survey (NTDS), which was developed by NCTE and the National LGBTQ Task Force and conducted in 2008–09. The NTDS was the first comprehensive survey examining the lives and experiences of transgender and gender nonconforming people in the United States. With 6,456 respondents reporting on a range of experiences throughout their lives, the NTDS was a groundbreaking study.

The results were published in the 2011 report, Injustice at Every Turn, and showed that discrimination against transgender people was pervasive in many areas of life, including education, employment, health care, and housing. The report also highlighted the resilience of transgender people in the face of such discrimination and found that family and peer support could have a substantially positive impact on a transgender person’s quality of life.

The report quickly became a vital source of information about transgender people and continues to serve as an important resource for advocates, policymakers, educators, service providers, media, and the general public.

CHAPTER 1 Introduction

INTRODUCTION 19 This report demonstrates that transgender people continue to face discrimination in numerous areas that significantly impact quality of life, financial stability, and emotional wellbeing, including employment, education, housing, and health care. Furthermore, many respondents experienced discrimination in multiple areas of their lives, the cumulative effect of which leads to severe economic and emotional hardship and can in turn have devastating effects on other outcome areas, such as health and safety.

Although issues impacting transgender people have become more visible in the years since the NTDS was published, the data overwhelmingly demonstrates that there is still a long way to go towards eliminating harmful discrimination and providing sustainable systems of support for transgender people throughout their lives. These findings are presented with the recognition that advocates, researchers, and transgender communities will greatly benefit from additional research conducted using this extensive data source.

The authors encourage subsequent analyses to delve into areas of the data that this report is unable to address, and as before, will strive to make the dataset available for such analyses. Much has changed since the NTDS was conducted in 2008–09 and results were published in 2011, including increased visibility of transgender people in the media and in society in general. Despite making significant strides in the five years since the report was published, there is still a substantial amount of work to be done to address critical needs in transgender communities throughout the United States.

Transgender people continue to experience discrimination and anti-transgender bias in virtually all areas of life. The 2015 U.S. Transgender Survey was developed by the National Center for Transgender Equality to provide updated and more detailed data to inform a wide range of audiences about the experiences of transgender people, how things are changing, and what can be done to improve the lives of transgender individuals in the United States. It is the largest survey of transgender people conducted to date, far surpassing the previous survey, with 27,715 respondents.

This study explores a wider range of topics than the previous survey and more deeply examines specific issue areas where transgender people are disparately impacted, such as health care, HIV/AIDS, housing, workplace discrimination, immigration, sex work, and police interactions. Additionally, by closely mirroring questions from federal and other existing surveys, this study seeks to fill in the gaps left by the lack of data collected about transgender people in national surveys.

Since federal survey data is often used by government agencies to make key determinations about policies and programs that affect individuals in many areas of life, such as employment and health, it is important to provide specific data on the potential impact of such policies on transgender people. This report on the U.S. Transgender Survey data draws comparisons between transgender people and the U.S. population and examines disparities across multiple issue areas.

2015 U.S. TRANSGENDER SURVEY 20 Report Roadmap The next

chapter of the report will give an overview of the study’s methodology, which will be followed by a guide to this report, including information about terminology used throughout.

These will be followed by chapters discussing respondents’ experiences across a range of areas that impact transgender people’s lives: • Portrait of USTS Respondents • Family Life and Faith Communities • Identity Documents • Health • Experiences at School • Income and Employment Status • Employment and the Workplace • Sex Work and Other Underground Economy Work • Military Service • Housing, Homelessness, and Shelter Access • Police, Prisons, and Immigration Detention • Harassment and Violence • Places of Public Accommodation and Airport Security • Experiences in Restrooms • Civic Participation and Policy Priorities The report also contains three appendices, which offer more detailed information related to the study: Appendix A: Characteristics of the Sample Appendix B: Survey Instrument (Questionnaire) Appendix C: Detailed Methodology

METHODOLOGY 21

CHAPTER 2 Methodology T he U.S. Transgender Survey is the largest survey ever conducted to examine the experiences of transgender people in the United States.

The survey instrument was comprised of thirty-two sections reflecting 1,140 distinct variables that covered a broad array of topics, such as health and health care access, and experiences around employment, education, housing, law enforcement, and public accommodation. 1 The survey was developed by a team of researchers and advocates and administered online to transgender adults residing in the United States. 2 The survey was accessible via any web-enabled device (e.g., computer, tablet, netbook, smart phone), accessible for respondents with disabilities (e.g., through screen readers), and made available in English and Spanish.

Rankin & Associates Consulting hosted the survey on several secure servers. The survey was accessed exclusively through a website created specifically for the promotion and distribution of the survey. 3 Data was collected over a 34-day period in the summer of 2015, 4 and the final sample included 27,715 respondents from all fifty states, the District of Columbia, American Samoa, Guam, Puerto Rico, and U.S. military bases overseas. The survey contained mainly closed-ended questions, but respondents were also offered the opportunity to provide write-in responses in fifty-three of the survey questions.

Over 80,000 write-in responses were provided by respondents.

2015 U.S. TRANSGENDER SURVEY 22 I. About the U.S. Transgender Survey The U.S. Transgender Survey (USTS) was developed as the follow-up to the groundbreaking National Transgender Discrimination Survey (NTDS), which was the first study to comprehensively measure experiences and life outcomes of transgender people in the United States. Fielded in late 2008 to early 2009 by the National Center for Transgender Equality (NCTE) and the National LGBTQ Task Force (“the Task Force”), the NTDS provided data that has informed policymakers, advocates, and educators since its publication in 2011.

However, the NTDS report acknowledged that the study had “just scratched the surface of this extensive data source” and encouraged advocates and researchers to conduct additional research to continue collecting data aimed at identifying and addressing the needs of transgender people. 5 The NTDS authors also examined the survey instrument and concluded that there were “imperfections” in the manner in which several questions had been posed. 6 The authors addressed areas for potential improvement with respect to both survey question design and substantive content in an “issues and analysis”

section of the report. 7 These recommendations were considered in the development of the U.S. Transgender Survey. In subsequent years, researchers have performed additional analyses using the NTDS public use dataset provided by NCTE and the Task Force. These analyses provided further insight into the experiences of transgender people, but also increased awareness of the questions that remained unanswered after the NTDS report was published.

In some instances, there was insufficient information to draw nuanced comparisons between life outcomes of transgender people collected in the NTDS and the U.S. general population. In other cases, the ability to form additional conclusions was limited due to a lack of follow-up questions.

For example, the NTDS asked a single question about suicide attempts, which did not allow for a clear examination of suicidal thoughts and behaviors. 8 Additionally, given the deficiency of longitudinal data on outcomes specific to transgender people, there remained a need to collect data that could speak to the experiences of transgender people over time and how outcomes may have changed in the years since the NTDS was published.

In these respects, the NTDS provided an important platform upon which to build the USTS to address identified areas for improvement and collect data that would enable new insights to be drawn about transgender people in the United States. The study was renamed the U.S. Transgender Survey for several reasons. One was to clarify the geographical location of the intended study sample both during the data collection period and following report publication.

The use of “U.S.” signaled that this study was developed with the unique needs of transgender people in the United States and U.S. territories in mind, considering relevant policies, procedures, and practices applicable to residents of the United States at the time of the study in areas such as health care and insurance, income, employment, housing, and education. Recognizing the contextual differences between the experiences of transgender people in the U.S. and in other parts of the world, the research team sought to dispel any confusion arising from the use of “national” in the title.

The new name was also intended to reflect the depth and breadth of the experiences of transgender people in the U.S. and elevate a variety of narratives beyond discrimination, including the resilience and resourcefulness of the transgender community in the face of hardship, as well as experiences of acceptance and affirmation. “Discrimination” was removed from the title to clarify that the survey was designed to capture all such experiences. Additionally, removing the word

METHODOLOGY 23 reduced potential bias in respondents’ answers or resulting from primarily attracting respondents who felt they had experienced discrimination. II. USTS Respondents The study population included individuals who identified as transgender, trans, genderqueer, non-binary, and other identities on the transgender identity spectrum, in order to encompass a wide range of transgender identities, regardless of terminology used by the respondent.

Although “transgender” was defined broadly for the purposes of this study as being inclusive of a wide range of identities—such as genderqueer, non-binary, and crossdresser—the research team recognized that many individuals for whom the study was intended may have used different terminology or

definitions and might have assumed that the term “transgender” did not include them. To address this, promotional materials affirmed that the survey was inclusive of all transgender, trans, genderqueer, and non - binary people. Additionally, materials specified that the survey was for adults at any stage of their lives, journey, or transition to encourage participation among individuals with diverse experiences regarding their transgender identity. An in-depth description of survey respondents is available in the Portrait of USTS Respondents chapter.

The study included individuals aged 18 and older at the time of survey completion, as did the NTDS. The study was not offered to individuals under the age of 18 due to limitations created by specific risk factors and recommendations associated with research involving minors.

These considerations, including requirements for parental/guardian consent, would have impacted the survey’s scope and content and also reduced the literacy level at which the survey could be offered. 9 Furthermore, the current experiences and needs of transgender youth often differ from those of adults in a number of key areas, including experiences related to education, employment, accessing health care, and updating identity documents, and many of these experiences or needs could not be adequately captured in a survey that was not specifically tailored to transgender people under the age of 18.

The sample was limited to individuals currently residing in a U.S. state or territory, or on a U.S. military base overseas, since the study focused on the experiences of people who were subject to U.S. laws and policies at the time they completed the survey. Individuals residing outside of the U.S. may have vastly different experiences across a number of outcome measures based on each respective country’s laws, policies, and culture, particularly in the areas of education, employment, housing, and health care.

Additionally, many survey questions were taken from U.S. federal government surveys that also limit their sample population to individuals in the U.S., and the research team sought to examine a similar population with regard to geographical location to allow for comparisons to the U.S. general population. III. Developing the Survey Instrument The USTS survey instrument was developed over the course of a year by a core team of researchers and advocates in collaboration with dozens of individuals with lived experience, advocacy and research experience, and subject-matter expertise.

When developing the survey instrument, the research team focused on creating a questionnaire that could provide data to address both current and emerging needs of transgender people while gathering information about disparities

2015 U.S. TRANSGENDER SURVEY 24 that often exist between transgender people and non-transgender people throughout the U.S. To achieve this, questions were included that would allow comparisons between the USTS sample and known benchmarks for the U.S. population as a whole or populations within the U.S.

Consequently, questions were selected to best match those previously asked in federal government or other national surveys on a number of measures, such as measures related to income and health. 10 Changes were made to the language of comparable questions whenever it was required to more appropriately reflect issues pertaining to transgender people and language in common use in the transgender community while maintaining comparability to the best extent possible.

However, in many cases, language was preserved to ensure that responses to a USTS question would maintain maximum comparability with surveys such as the U.S. Census Bureau’s American Community Survey and Current Population Survey. Several questions were also included in an attempt to provide comparability between the NTDS, where possible, to determine how certain outcomes may have changed since the NTDS data was collected in 2008–09.

While the USTS provides crucial updated data, it is important to note that many of the questions asked in the NTDS were either not included in the USTS, or they were asked in a manner that reduced comparability with the NTDS. For example, many USTS questions asked about whether certain experiences occurred within the past year instead of asking whether those experiences occurred at any point during an individual’s lifetime.

These questions were included for both comparability with federal government or other national surveys and also to yield improved data regarding changing experiences in future iterations of the USTS. In such instances, the NTDS continues to provide the best available data regarding experiences that occurred over respondents’ lifetime. The authors suggest referring to both the USTS and the NTDS to gain a full picture of issues impacting transgender people.

The survey instrument was reviewed by researchers, members of the transgender community, and transgender advocates at multiple intervals throughout the development process. This included thorough reviews of sections that addressed specific subject matter and the entire questionnaire. The questionnaire was revised based on feedback from dozens of reviewers. a. Pilot Study Prior to finalizing the survey instrument and launching the survey in field, a pilot study was conducted to evaluate the questionnaire.

The pilot study was conducted among a small group of individuals with characteristics that were representative of the sample the study was intended to survey. The pilot study was administered through an online test site using the same platform and format in which the final survey later appeared. The purpose of the pilot study was to provide both a substantive and technical evaluation of the survey. Approximately 100 individuals were invited to complete and evaluate the survey online during a specified period of time.

In order to receive access to the pilot study test site, invitees were required to confirm their participation by indicating that they met the following pilot study criteria: they were (1) 18 years or older, (2) transgender, (3) willing to provide feedback that would be used to make improvements to the survey, (4) available to take the survey online during specified dates, and (5) agreeing to not share the questions in the pilot study with anyone so as to not compromise the study. Forty (40) individuals confirmed their participation and received access to the pilot study test site.

Thirty-two (32) people completed the study and submitted feedback on the questionnaire, including participants in fifteen states ranging in age from 19 to 78. Participants

METHODOLOGY 25 reported identifying with a range of gender identities 11 and racial and ethnic identities, including 34% who identified as people of color. 12 In addition to providing general feedback on individual questions and the entire questionnaire, pilot study participants were asked to address specific questions as part of their evaluation, including: (1) how long it took to complete the survey, (2) what they thought about the length of the survey, (3) whether any existing questions were confusing or difficult to answer, (4) whether they found any questions offensive or thought they should be removed or fixed, (5) whether they experienced technical or computer issues while taking the survey, and (6) what they thought about the statement explaining why the term “trans” was used throughout the survey. 13 All participant feedback was compiled, discussed, and used to further develop the questionnaire, such as through the revision of language and the addition of questions to more thoroughly examine an issue. b.

Length The final survey questionnaire contained a total of 324 possible questions in thirty-two discrete sections addressing a variety of subjects, such as experiences related to health and health care access, employment, education, housing, interactions with law enforcement, and places of public accommodation. The online survey platform allowed respondents to move seamlessly through the questionnaire and ensured they only received questions that were appropriate based on previous answers.

This was accomplished using skip logic, which created unique pathways through the questionnaire, with each next step in a pathway being dependent on an individual respondent’s answer choices. For example, respondents who reported that they had served in the U.S. Armed Forces, Reserves, or National Guard received a series of questions about their military service, but those who had not served did not receive those questions. Due to the customized nature of the survey, the length varied greatly between respondents, and no respondent received all possible questions.

Prior to the pilot study, estimates indicated a survey-completion time of 30–45 minutes. The completion-time estimate was extended to 60 minutes based on feedback from pilot study participants, and it was consistent with many reports during the fielding period. 14 Despite observations about survey length discussed in the NTDS, 15 evolving data needs relating to issues affecting transgender people required an in-depth treatment of multiple issue areas. This often required multiple questions to thoroughly assess an issue—including in areas where the NTDS asked only one question—and resulted in a lengthier survey.

Survey instrument length was assessed throughout its development to ensure it would be manageable for as many participants as possible. Furthermore, through multiple reviews and evaluations of the survey instrument—including the pilot study—survey takers reported that the length was appropriate for a survey addressing such a wide range of issues and the need for data outweighed concerns about the overall length of the survey. IV.

Survey Distribution and Sample Limitations The survey was produced and distributed in an online-only format after a determination that it would not be feasible to offer it in paper format due to the length and the complexity of the skip logic required to move through the questionnaire. With so many unique possibilities for a customized survey experience for each respondent, the intricate level of navigation through the survey

2015 U.S. TRANSGENDER SURVEY 26 would have created an undue burden and confusion for many respondents. This could have led to questions being answered unnecessarily or being skipped completely, which could have increased the potential for missing data in the final dataset. 16 This made online programming the best option for ensuring that respondents received all of the questions that were appropriate based on their prior answers, decreasing the probability of missing data.

However, the potential impact of internet survey bias on obtaining a diverse sample has been well documented in survey research, 17 with findings that online and paper surveys may reach transgender respondents with “vastly different health and life experiences.” 18 With those considerations in mind, outreach efforts were focused on addressing potential demographic disparities in our final sample that could result from online bias and other issues relating to limited access.

Although the intention was to recruit a sample that was as representative as possible of transgender people in the U.S., it is important to note that respondents in this study were not randomly sampled and the actual population characteristics of transgender people in the U.S. are not known. Therefore, it is not appropriate to generalize the findings in this study to all transgender people. V. Outreach The main outreach objective was to provide opportunities to access the survey for as many transgender individuals as possible in different communities across the U.S. and its territories.

Additionally, outreach efforts focused on reaching people who may have had limited access to the online platform and who were at increased risk of being underrepresented in such survey research. This included, but was not limited to, people of color, seniors, people residing in rural areas, and low-income individuals. The outreach strategy was a multi-pronged approach to reach transgender people through various connections and points-of - access, including transgender - or LGBTQ-specific organizations, support groups, health centers, and online communities.

Outreach efforts began approximately six months prior to the launch of the data-collection period with a variety of tactics designed to raise awareness of the survey, inform people when it would be available, and generate opportunities for community engagement, participation, and support.

A full-time Outreach Coordinator worked for a period of six months to develop and implement the outreach strategy along with a team of paid and volunteer interns and fellows. 19 An initial phase of outreach involved developing lists of active transgender, LGBTQ, and allied organizations who served transgender people and would eventually support the survey by spreading the word through multiple communication platforms and in some cases providing direct access to the survey at their offices or facilities.

Establishing this network of “supporting organizations” was an essential component of reaching a wide, diverse sample of transgender people. Over 800 organizations were contacted by email, phone, and social media, and they were asked if they would support the survey by sharing information about it with their members and contacts. Specifically, supporting organizations were asked to share information through email blasts and social media channels, and the research team provided language and graphics for organizations to use in an effort to recruit appropriate respondents into the study.

Of the organizations contacted, approximately half responded to requests for support, resulting in direct recruitment correspondence with nearly 400 organizations at regular intervals during the pre-data-collection period and while the survey was in the field. 20,21 These organizations

METHODOLOGY 27 performed outreach that contributed to the far reach of the survey and unprecedented number of respondents. 22 The organizations were also featured on the survey website so potential respondents could determine whether organizations they knew and trusted had pledged support for the survey. Nearly 400 organizations responded to outreach and confirmed their support for the survey. The remaining organizations did not respond directly to invitations to learn more about the survey and become supporters.

Consequently, these organizations did not receive correspondence aimed at directly recruiting respondents prior to the survey launch or during the data-collection period. It is possible, however, that survey respondents were still made aware of the survey through those organizations. Since there is no information regarding whether these organizations shared information about the survey through their channels, it is difficult to assess the full scope of the outreach efforts. a. Advisory Committee A significant element of outreach involved convening a USTS Advisory Committee (UAC).

The UAC was created to increase community engagement in the survey project and raise awareness by connecting with transgender people in communities across the country through a variety of networks. The UAC was comprised of eleven individuals with advocacy, research, and lived experience from a wide range of geographical locations. 23 Members were invited to join the committee as advisors on survey outreach to facilitate the collection of survey data that would best reflect the range of narratives and experiences of transgender people in the U.S.

Each member brought unique skills and expertise to contribute to the committee’s objectives. UAC members participated in five monthly calls with members of the USTS outreach team from May to September 2015. UAC monthly calls focused on providing project updates and identifying pathways by which outreach could be conducted to increase the survey’s reach and promote participation from a diverse sample.

Members suggested organizations, individuals, and other avenues through which to conduct outreach, shared ideas and strategies for improving outreach to specific populations of transgender people, and spread the word about the survey through their professional and personal networks. b. Survey-Taking Events In an effort to increase accessibility of the survey, the outreach team worked with organizations across the country to organize events or venues where people could complete the survey.

Survey - Taking Events, 24 or “survey events,” were spaces in which organizations offered resources to provide access to the survey, such as computers or other web-enabled devices. These organizations provided a location in which to take the survey at one particular time or over an extended period of time, such as during specified hours over the course of several days. 25 The events were created with the intention of providing access to individuals with limited or no computer or internet access, those who may have needed assistance when completing the survey, or those who needed a safe place to take the survey.

Additionally, the population that had previously been identified as being more likely to take a paper survey than an online survey were considered, 26 and the events were developed to target those individuals. Given the potential variety of these survey events—including the types of available resources and times at which they were conducted— guidelines were needed to maintain consistency across the events and preserve the integrity of the data-collection process. A protocol was developed outlining the rules for hosting a survey event to advise hosts on best practices for ensuring

2015 U.S. TRANSGENDER SURVEY 28 a successful data-collection process, including guidelines to prevent the introduction of bias into survey responses. The protocols described the steps for becoming a survey-event host and tips for how to conduct outreach about the event. The protocol also specified that hosts should inform NCTE of their event prior to hosting and report on how many people attended the event and how many people completed and submitted the survey. This was helpful information for evaluating the relative success and benefits of these events.

All confirmed supporting organizations were invited to become survey event hosts, and those who accepted the invitation were sent the protocol. Seventy-one (71) organizations accepted the invitation and confirmed the date(

s) and time(

s) of their events. 27 Survey events were promoted on the survey website and given a specific designation on the supporting organization map (described further in the “Survey Website” section), including information about where and when people could attend. Hosts were encouraged to promote their event through multiple channels and consider outreach methods beyond online avenues, such as direct mail or flyers, to better reach transgender people with limited or no internet access.

Additionally, hosts were provided with flyer templates so they could promote the events in their facilities or through communications with their members or constituents. Of the organizations who confirmed their survey events, 46 reported information about attendance at the event. The hosts reported that 341 people attended their events, including transgender and non-transgender friends, family, and volunteers.

Approximately 199 respondents completed the survey at these events. 28 However, survey responses indicate that additional unreported survey events or similar gatherings may have been held where participants had an opportunity to complete the survey. 29 Event-related information submitted by organizations following the fielding period was not comprehensive enough to make a thorough determination as to whether the events had achieved their previously stated objectives. 30 c. Incentives As an incentive for completing the survey, participants were offered a cash-prize drawing.

Incentives, such as cash prizes are widely accepted as a means by which to encourage and increase participation in survey research. 31 Studies have shown that such incentives may have a positive effect on survey response rate, which is the proportion of individuals in the population of interest that participates in the survey. 32 Research has also found that lottery-style cash drawings may be beneficial in online surveys, 33 since they offer a practical method for providing incentives in surveys with a large number of respondents by eliminating the potential high cost of both the cash incentive and prize distribution. 34 USTS respondents were offered the opportunity to enter into a drawing for one of three cash prizes upon completion of the survey, including one $500 cash prize and two $250 cash prizes. 35 After completing and submitting their anonymous survey responses, USTS respondents were re - directed away from the survey hosting site 36 to a web page on the NCTE-hosted USTS website.

In addition to being thanked for their participation on this page, respondents received a message confirming that their survey had been submitted and any further information they gave would not be connected to their survey responses. Only individuals who completed and submitted the survey were eligible for one of the cash prizes. To enter into the prize drawing, respondents were required to check a box giving their consent to be entered. 37 Respondents were also asked to provide their contact information in order to be notified if selected in the drawing. The final drawing contained 17,683 entrants.

Each entrant was assigned a number, and six numbers were randomly chosen by a non-NCTE party: three

METHODOLOGY 29 numbers for the prize winners and three for alternates if necessary. The three prize winners were contacted and awarded their prizes upon acceptance. VI. Communications Communications for the survey required a multifaceted approach and a coordinated effort with the outreach strategy to most effectively reach a wide range of transgender people and ensure a robust sample size.

The goals of survey communications were to: (1) inform people that NCTE would be conducting a survey to further the understanding of the experiences of transgender people in the U.S initially gleaned through the NTDS, (2) communicate when the survey would be available to complete and how it could be accessed, and (3) find creative ways of reaching diverse populations of potential respondents. This involved raising awareness of the survey through several communication methods, including email, social media, and print media, as well as through additional unique campaigns.

Many survey promotional materials were produced in English and Spanish to increase the accessibility of the survey. 38 a.

Survey Website A website was created and designed specifically for the promotion and distribution of the survey. 39 This website served as a platform for providing information about the survey starting several months prior to its release in the field, such as a description of the survey, information about the team working on the survey, frequently asked questions, and sample language and graphics for individuals and organizations to use for email and social media communications, including sample Facebook and Twitter postings.

The website also featured an interactive map, which included information about organizations that had pledged to support the survey. Additionally, the map distinctly indicated information about organizations that were hosting survey-taking events, including the date, time, and location of such events. The website later served as the only platform through which the survey could be accessed and provided English and Spanish links to enter the survey, since there was no direct link available to the off-site hosting platform. b.

Survey Pledge The survey pledge campaign was developed to raise awareness about the survey and generate investment in the project. The campaign engaged potential participants and allies by inviting them to pledge to take the survey and/or spread the word about the survey. The survey pledge was a critical method of both informing people that the survey would be launching and sustaining engagement with potential respondents in the months leading up to the fielding period. Pledges received reminders about the survey launch date and availability through email communications.

Beginning in January 2015, pledge palm cards were distributed at a variety of events across the country, including conferences and speaking engagements. The cards contained information about the upcoming survey and asked people to sign up to help by committing to: (1) spread the word about the survey; and/or (2) take the survey. Transgender and non-transgender individuals were asked to complete the pledge information, either through a palm card or directly online through the survey website. Individuals who completed pledge information received email communications throughout the pre-data-collection phase.

Pledge information was collected continuously for several months, and by the time of the survey launch, over 14,000 people had pledged to take the survey. Additionally, more than 500 people pledged to promote the survey among their transgender friends and family. 40 The pledge proved to be

2015 U.S. TRANSGENDER SURVEY 30 an effective method of assessing how many people had learned about the survey and were interested in completing it, where potential survey respondents were distributed geographically, and how more potential respondents could be effectively engaged. c. Photo Booth Campaign In January 2015, a photo booth campaign was launched as another method for engaging people and raising awareness about the survey.

Individuals and groups were asked to take photos holding one of two signs with messages expressing support for the survey. 41 USTS photo booths were conducted at several conferences and events across the country. More than 300 photos were collected and shared directly through NCTE’s Facebook page. Photos were also sent to most participants so they could conduct their own promotion using their photos. d. Social Media With the increased use of social media in the years since the previous survey (the NTDS), it was important to engage via these outlets to further the reach of the survey.

Facebook and Twitter 42 became the primary social media outlets used throughout the survey project, and their use significantly amplified awareness, increasing the number of people who were exposed to the survey. A series of postings provided the ability to rapidly and succinctly communicate with individuals and groups who had an interest in contributing to the survey’s success by completing the survey and spreading the word about it.

Although social media reach fluctuated during the months leading up to the survey launch, over 96,000 Facebook users were estimated to have received NCTE’s post announcing that the survey was live and available for completion on August 19, 2015. e.

USTS Awareness Week Prior to launching the survey in the field, communication was maintained with thousands of individuals and organizations who fell into three categories: (1) people who had signed up to take or spread the word about the survey (“pledge list”), (2) organizations that had committed to support the survey through outreach efforts (“supporting organization list”), and (3) people who had signed up to be in communication with NCTE about the organization’s work and projects (“NCTE list”).

Communication with the individuals and groups on these lists through targeted messages occurred at various intervals; however, one of the most important methods for promoting the survey was through USTS Awareness Week. This campaign was designed to share a significant amount of information about the survey over a concentrated period of time in close proximity to the launch of the survey. Awareness Week occurred during the week of July 27, 2015 and highlighted different aspects of the survey focusing on a different medium each day, including social media, email, and blogs.

Awareness Week was introduced to the communication lists on July 15, and recipients were invited to access and download a planning kit for the campaign, which was available on the survey website. The planning kit included language and graphics for email and social media communications. Communications were sent on each of the days devoted to social media, 43 email, 44 and blogs 45 with appeals for organizations to share the information with their membership and individuals to share the information through their personal networks.

Awareness Week proved to be one of the most effective methods for increasing the number of individuals who pledged to take the survey and likely increased the number of eventual respondents. 46

METHODOLOGY 31 f. Additional Communications Methods The overall approach to survey communications was diverse and captured many media forms. In addition to the previously stated campaigns and projects, communications involved working with a variety of individuals such as bloggers, artists, advocates, and others to create print blogs and videos promoting the survey. Op-eds were another medium that contributed to survey promotion, and media consultants and traditional media sources aided in expanding the survey’s reach even further.

Approximately 50 articles, blogs, and op-eds focused on the survey were produced and distributed by organizations, including NCTE, and individuals prior to the launch of the survey and during the data-collection period. The wide variety of approaches contributed to the number of individuals who were reached through all communications and likely impacted the final number of respondents in the sample. VII. Language and Translation Throughout the survey questionnaire, the use of accessible language was balanced with preserving the meaning of each question to the greatest extent possible.

This was of particular importance in maintaining comparability with questions from existing surveys that allowed conclusions to be drawn about how the experiences of the USTS sample compares to the U.S. population. In order to make assessments about USTS survey respondents in relation to the U.S. population, it was important that USTS respondents had similar

interpretations of questions taken from other surveys as non-transgender survey takers had to those questions in federal surveys. In many places, language was revised to use terminology that would most appropriately speak to individuals in the many communities for which the survey was intended. However, several areas required difficult choices about keeping language that may have caused discomfort for some respondents.

Throughout the questionnaire, language was avoided that could be interpreted as stigmatizing or characterized as a value judgment wherever possible while maintaining objectivity in crafting sound research questions. For example, at times survey questions referred to work or activities that were “currently considered illegal.” Such deliberate language was used in an attempt to separate the issue of criminalization from the activity in question while maintaining comparability with other surveys. This was a difficult balance to achieve throughout the survey.

Eliminating technical language was also necessary, unless it was widely used and accepted in transgender communities, such as some medical terminology. Short descriptions or parenthetical explanations were provided whenever technical language was required for those who may not have been familiar with the language. Additionally, hyperlinked explanations of specific terms were included when those terms could be interpreted in several ways or if similar explanations were provided in the federal surveys from which the questions were taken.

For example, explanations were provided for the terms “active duty” when asking about military service and “household” when asking about income. The research team remained conscious of individual and collective identities throughout the survey instrument drafting process, and attempted to use language that acknowledged the breadth and significance of individual identities while also making the questions accessible to the widest range of transgender people possible across the U.S. and in the territories.

The questionnaire was reviewed and revised for consistent readability at an eighth-grade literacy level where possible, 47 although several

2015 U.S. TRANSGENDER SURVEY 32 terms used in the survey were at a considerably higher literacy level. This included places where language was preserved for comparability with other surveys and when language describing transgender-specific experiences or procedures was used. Additionally, community members and researchers reviewed the survey and suggested revised language throughout the development process. This collaborative process was beneficial in providing collective insight on the best language to use in each particular instance based on lived experience and research expertise.

The research team acknowledges, however, a continuing need to work towards identifying suitably inclusive terminology within an evolving language and community for future iterations of the survey. The questionnaire was translated into Spanish by a translation service, and several native-Spanish - speaking community members and NCTE staff and interns reviewed and revised the language to use terminology that was most prevalent in Spanish - speaking transgender communities in the U.S.

In many instances, it was difficult to find language that accurately captured the meaning of a question or specific terms, but in each case language was selected to convey

interpretations as close to the English-language question

interpretations as possible. VIII. Institutional Review The study was vetted through an Institutional Review Board (IRB) process, which is meant to ensure confidentiality and protect the rights and welfare of individuals participating in a research study. The USTS underwent a full board review by the University of California Los Angeles (UCLA) IRB.

As a requirement of approval, the questionnaire began with a study information sheet describing aspects of the study and rights of individuals as participants in the study. 48 To be included in the study, participants were required to indicate their consent at the end of the information sheet. This process established that participants were fully informed about the risks and benefits of participating in the study and that their participation was voluntary.

IRB review also required the submission of all recruitment materials leading up to the launch of the survey and throughout the time the survey was in the field. 49 This required the production of a large volume of messaging for the many different types of media through which people were invited to participate in the survey in both English and Spanish. It also required anticipating how messaging might need to change while the survey was in the field and submitting this language for pre-approval for later use as needed. IX.

Survey Hosting The survey was hosted online by Rankin & Associates Consulting, under the supervision of USTS research team member, Dr. Susan Rankin. Access to the survey was provided exclusively through the USTS website. All programming of the questionnaire and online administration of the survey was handled through Rankin & Associates Consulting, which managed the process of collecting the survey data throughout the 34-day fielding period.

The survey was anonymous, and maintaining privacy and confidentiality in the collection and maintenance of survey data was an important component of preserving participants’ anonymity. Furthermore, as a condition of IRB approval, the research team was required to ensure that confidentiality protections were in place for the study and demonstrate sufficiency of data security protocols. Accordingly, data from online

METHODOLOGY 33 participants was submitted through seven secure firewalled servers with forced 256-bit SSL (Secure Sockets Layer) security and Security-Enhanced Linux (SELinux) security extensions to encrypt and protect the survey data. Given the volume of traffic on the seven servers during the initial launch of the survey, an eighth server was added. The survey was stored in an SQL database that could only be accessed locally. The servers themselves were only accessible using encrypted SSH (Secure Shell) connections originating from the local network.

The servers were also in RAID (Redundant Array of Inexpensive Disks), which is a data storage virtualization technology that combines multiple physical disk drive components into a single logical unit for the purposes of data redundancy, performance improvement, or both, to reduce the chance of any data loss due to hardware failure. The servers performed nightly security audits from data acquired via the system logs and notified the system administrators.

Despite a successful data-collection period evidenced by the large final sample size, it is important to note issues that occurred in the initial days of the survey data-collection period, given the potential impact on the data collection and the final sample. Prior to the survey launch, the online platform had been assessed and capacity was predicted for the seven dedicated servers based on reasonable estimated response rates.

However, in the first days of the data-collection period, exceptionally high levels of traffic to the survey far exceeded the predicted response rates and overwhelmed the capacity of the servers, causing significant delays in accessing and completing the survey.

The resulting server delays occurred within hours of the survey launch on August 19, 2015, producing unusually long page - loading times and may have served as a barrier to completing the survey. 50 The survey team notified potential respondents of the delays through email and social media communication and updated the first page of the online survey questionnaire with a note about the issues and information about the continued availability of the survey. 51 The hosting team added a server to process the high level of traffic and returned the survey to normal loading speeds within a couple days of the initial reports.

Although high numbers of survey submissions were received throughout these days, it is likely that the server delays affected the completion and submission of some surveys or may have discouraged individuals from attempting to take the survey. X. Cleaning the Data The dataset was cleaned following collection to remove survey responses that did not belong in the final sample. 52 Data cleaning is the process of detecting and removing some survey responses (e.g., duplicate responses, incomplete responses, illogical responses) in order to improve the quality of the sample.

This dataset was “cleaned” using commonly accepted procedures. 53 The first step was to remove survey responses from individuals who did not consent to take the survey and those who did not meet the eligibility criteria, such as not being at least 18 years of age and not residing in the U.S. These survey respondents had been automatically sent to a disqualification page, 54 but their responses were included in the initial dataset. Incomplete responses were then removed from the sample based on a requirement that respondents minimally complete specific questions in

Section 2 of the questionnaire to be included in the final dataset. 55 Duplicate survey responses were removed next, as were those with illogical responses, such as those with contradictory responses to related questions. Missing-data analyses were then conducted to determine the percentage of missing data. 56 The next step of the process was recoding data, including re-categorization of answer choices

2015 U.S. TRANSGENDER SURVEY 34 in several questions for improved analysis or to match existing categories for comparison to other surveys. Answers were evaluated for those questions that allowed a write-in response when the selected option was “not listed above.” In some cases, these answers were recoded into existing answer choices where appropriate, and in other cases, new answer categories were created for write-in responses that were frequently repeated. The recoding process included two coding teams.

The first coding team conducted initial data recoding, and the second team reviewed the recoding and flagged areas of disagreement.

A simple percent agreement score was calculated to determine inter-rater reliability. 57 Several survey weights were developed for presentation of results in the report. 58 A race and ethnicity weight was developed based on the Census Bureau’s 2014 American Community Survey (ACS). 59 Additionally, given the disproportionally large number of respondents who reported an age of 18 years old, a weight was created to balance the representation in the sample of those respondents in relation to the rest of the sample. 60 The race and ethnicity weight and the 18-year-old weight were both included in a “standard weight” applied to the dataset.

All results presented in this report are weighted based on the standard weight unless otherwise noted. Additional survey weights were created for the purposes of comparability with federal government and national data sources, including weights for age and educational attainment. 61 These weights were applied in addition to the standard weight when comparing the USTS sample to the U.S. population for items that are sensitive to age and educational attainment, such as individual and household income, and are noted accordingly as the “supplemental weight.” XI.

Data Analysis and Presentation of Findings The data was first analyzed to tabulate individual responses to each of the questions in the survey. The respondents included in each tabulation differed throughout the survey due to certain questions only being asked of a particular set of respondents and/or due to some respondents choosing not to answer a question.

Analyses were performed to explore how survey responses differed based on demographic characteristics— such as race, gender, and income—and non - demographic factors—such as experience with sex work, HIV status, and experiences of family support or rejection. All findings in the report are presented as weighted percentages of the entire sample or of the subgroups being examined. For example, educational attainment is presented as a percentage of the whole sample, while much of the data related to HIV care represent percentages of those respondents who are living with HIV.

In limited instances, unweighted frequencies are included where the additional information could be informative and to provide context for the weighted percentages reported. Percentages are rounded to whole numbers, except in cases where a more exact comparison to national data sources was desired or where more precision was needed due to the reported percentages being small.

When rounding to whole numbers, the following convention was generally followed: findings containing decimals of 0.50 and above were rounded up, and findings with 0.49 and below were rounded down (e.g., 1.50% was rounded to 2% and 1.49% was rounded to 1%). Additionally, a finding of 0.49% and below was generally labeled “less than 1%” or “<1%.” Throughout the report, results are presented in

METHODOLOGY 35 various figures and tables. The percentages in these figures and tables do not always add up to 100% due to respondents being able to select more than one answer to a question (“mark all that apply”) or due to rounding. Throughout the report, U.S. population findings are provided for comparison to USTS findings or to provide context for USTS findings, where available and/or applicable.

Where USTS data is compared to data from existing research, the 1 The survey included questions related to the following topics (in alphabetical order): accessing restrooms; airport security; civic participation; counseling; education; employment; faith; family and peer support; health and health insurance; HIV; housing and homelessness; identity documents; immigration; income; intimate partner violence; military service; police and incarceration; policy priorities; public accommodations; sex work; sexual assault; substance use; suicidal thoughts and behaviors; unequal treatment, harassment, and physical attack; and voting. 2 Detailed information about survey methodology is available in Appendix C (Detailed Methodology). 3 www.USTransSurvey.org 4 The survey was in the field between August 19 and September 21, 2015. 5 Grant, J.

M., Mottet, L. A., Tanis, J., Harrison, J., Herman, J. L., & Keisling, M. (2011). Injustice at Every Turn: A Report of the National Transgender Discrimination Survey. (p. 11). DC: National Center for Transgender Equality and National Gay and Lesbian Task Force. 6 Grant et al., p. 182. 7 Grant et al. 8 See Haas, A. P., Rodgers, P. L., & Herman, J. L. (2014). Suicide Attempts Among Transgender and Gender Non - Conforming Adults. New Y ork, NY & Los Angeles, CA: American Foundation for Suicide Prevention & Williams Institute. data source is specified.

When providing U.S. population comparisons, the research team made efforts to limit the comparisons to adults (18 years and older) to most appropriately match the USTS sample. Whenever that was not possible, notes as to age ranges or other limitations are provided. Additionally, calculations made by the research team when necessary to present U.S. population findings are noted. Data in this report is generally presented without information regarding statistical testing. 62 9 See e.g., The GenIUSS Group. (2014). In J. L.

Herman (Ed.), Best Practices for Asking Questions to Identify Transgender and Other Gender Minority Respondents on Population-Based Surveys (p. vii). Los Angeles, CA: Williams Institute. (“Adolescents may have particular difficulties with complex vocabulary and sentences. Therefore, questions designed for adolescents should take extra care to use plain language and simple sentences.

Terms used in measures of sex and gender should be defined since adolescents, and cisgender (non-transgender) adolescents in particular, conflate the terms sex and gender, and have varying understanding of the term transgender, masculine, and feminine.”).

Given the need to collect data about the unique experiences of transgender youth, it is important to design and conduct future studies focusing on the issue areas and needs most applicable to transgender youth. 10 Information about the source of survey questions used for comparison to the U.S. population can be found in Appendix C (Detailed Methodology). 11 Forty-four (44%) of pilot participants identified as a woman or trans woman (MTF), 41% as a man or trans man (FTM), and 16% as non-binary or genderqueer. 12 These pilot participants identified as American Indian, Asian, multiracial, Black, Latino/a, and a racial/ethnic identity not listed above, in addition to 66% who identified as white. 13 The following statement was provided to explain why the word “trans” was used throughout the survey: We know that not everyone is comfortable with the word “transgender,” but for this survey, we must use one word to refer to all trans and non-binary identities.

Because of this we will use the word “trans” in this survey to refer to all trans and non-binary identities.” ENDNOTES |

CHAPTER 2: METHODOLOGY

2015 U.S. TRANSGENDER SURVEY 36 14 A notable exception to the 30–60 minute estimate for completing the survey occurred during the first days of the survey’s availability, when a high volume of survey takers overwhelmed multiple servers, causing lengthy delays when completing the survey. This is discussed further in the “Survey Hosting” section. 15 Grant et al., p. 13. 16 Post-NTDS analysis of respondents who had completed that survey online or in paper format found that surveys completed online were less likely to have missing data, providing further support for the decision to only offer the survey online.

See Reisner, et al. (2014). Comparing in - person and online survey respondents in the U.S. National Transgender Discrimination Survey: Implications for transgender health research. LGBT Health, 1(2), 98–106. 17 See Dillman, D. A., Smyth, J. D., & Christian, L. M. (2014). Internet, Phone, Mail, and Mixed-Mode Surveys: The Tailored Design Method (4th ed.). Hoboken, NJ: John Wiley & Sons. 18 Reisner et al., p. 98. See note 16.

This analysis also found that “[a] higher proportion of in-person respondents were young, male-to-female, people of color, publicly insured, with lower incomes and lower educational attainment than online respondents (all p<0.05).

In-person respondents also were more likely than online respondents to be current daily smokers, to endorse substance use to cope with mistreatment, and to self-report as HIV-positive (all p<0.05).” 19 Although outreach efforts were instrumental in obtaining the largest sample of transgender respondents ever collected, a longer outreach period may have resulted in reaching more individuals in communities that are often underrepresented in online surveys. 20 A total of 827 organizations received at least one outreach email, and organizations received additional outreach emails and/or phone calls if no response was received.

Out of those organizations, 392 confirmed their support, and 435 did not respond to any communications. 21 Correspondence included almost one dozen emails with asks to spread the word about the survey and with various information about the availability of the survey. 22 The research team attempted to ascertain the level of outreach engagement of supporting organizations; however, the limited amount of information received about the outreach did not allow a calculation of a response rate.

Of the 392 organizations that pledged their support, 58 (15%) reported information on their outreach activities and estimated reaching over 20,000 transgender people through their channels. In the future, researchers are encouraged to collect consistent outreach activity data from supporting organizations that will help to better assess the effectiveness of outreach and response rate estimates. 23 Information about UAC members can be found in the Acknowledgements

section of the report. 24 These events were promoted as “Survey-Taking Events” on recruitment materials and described accordingly (see note 25). However, it is possible that the name did not appropriately capture the nature of these vastly differing events.

A lack of clarity may have decreased the number of people who attempted to access the survey through organizations who offered space or computers to complete the survey online. 25 Survey-Taking Events were described as “a function in which an organization or group opens its doors and provides access to its facilities (such as community centers and office buildings) to allow trans survey participants use of its resources (including computers, tablets, and internet access) to complete the USTS. This will occur during specified periods of time or throughout the time the survey is available on a drop-in basis.

For example, a community center might participate by setting aside one Saturday from 9am–6pm where some or all of its computers are available for survey takers to use, or it might host people on Monday–Friday from 5pm–9pm each evening for a week, or longer.” 26 A total of 435 NTDS respondents completed the survey in paper format (7% of the sample) and were found to differ from online survey takers in sociodemographic characteristics, health outcomes, and life experiences. Reisner et al., p. 98, 103.

See note 16. 27 Although only 71 organizations confirmed their events, based on information reported at various intervals throughout the data-collection period, it appeared that more organizations hosted survey events or similar gatherings to complete the survey without reporting them to the survey outreach team. Additionally, it is also possible that individuals and organizations held informal parties where groups of friends could gather to complete the survey at the same time.

Data regarding this sort of activity was not collected or received. 28 This completion rate is a conservative estimate based on reports that some individuals started the survey at the event and then left to complete it on their own at a later time. 29 Four hundred and seventeen (417) respondents answered “yes” in response to the following survey question: “Are you taking this survey at a survey event or meeting, such as one hosted by an LGBTQ or Trans organization or meeting?” 30 In future iterations of the USTS and other research studies, the research team suggests a more robust approach towards organizing, conducting, and monitoring survey events to increase the reach and availability of such events in providing access to the survey.

Researchers are also encouraged to conduct follow-up analyses to

METHODOLOGY 37 determine the demographic characteristics of individuals who completed the survey at events and whether these events were successful in capturing a similar demographic to those who had completed paper surveys in the previous survey. See Reisner, et al. (discussing the demographics of online and paper respondents in the NTDS). 31 See e.g., Göritz, A. S. (2006). Incentives in web studies: Methodological issues and a review.

International Journal of Internet Science, 1(1), 58–70. (finding that “material incentives increase the odds of a person responding by 19% over the odds without incentives”). 32 Pedersen, M. J. & Nielsen, C. V. (2016). Improving survey response rates in online panels: Effects of low-cost incentives and cost-free text appeal interventions. Social Science Computer Review, 34(2), 229–243. 33 Pedersen et al., pp. 237–238. 34 Singer, E. & Y e, C. (2013). The use and effects of incentives in surveys.

The ANNALS of the American Academy of Political and Social Science, 645(1), 123–124. 35 Participants were informed of the cash prize incentives in several ways. The study information sheet placed at the beginning of the survey prior to obtaining each respondent’s consent to enter the survey contained the following information in response to the question of whether respondents would be paid for their participation: “Y ou will receive no payment for your participation.

Y ou will have the option to voluntarily enter a drawing to win one of three cash prizes: one prize of $500 and two prizes of $250.” The frequently asked questions

section of the survey website also offered the following statement: “When you complete the survey, you will have the option to enter a drawing to win one of three cash prizes: one prize of $500 and two prizes of $250. Because thousands of trans people across the country will complete the survey, we cannot offer payment to each participant.” Additionally, some recruitment materials mentioned the cash-prize drawing, including email blasts. 36 The survey was hosted by Rankin & Associates Consulting.

Further details are described in the “Survey Hosting” section. 37 The check box stated: “Enter me in the drawing for one of three cash prizes: one prize of $500 and two prizes of $250!” 38 Due to limited funding, it was not possible to translate all survey materials, such as email communications.

Translation of all promotional materials may positively impact the response rate amongst respondents with limited English proficiency in future iterations of the study. 39 www.USTransSurvey.org 40 Final pledge numbers were 14,005 and 561 for survey takers and promoters, respectively. 41 Photo booth participants could choose from one of two signs indicating that the survey was coming in the summer of 2015 and stating the following: (1) “My Voice Counts: I’m Taking the #USTransSurvey” or (2) “Every Voice Counts: Spread the Word About the #USTransSurvey.” 42 The Twitter hashtag used to promote the survey was #USTransSurvey. 43 For social media day, recipients received one of the following requests, based on whether they were organizations or individuals: (1) “Use the hashtag #USTransSurvey on social media asking your social networks to join us” or (2) “Please join Social Media day.

We have sample copy and a variety of photos and graphics.” 44 For email day, recipients received one of the following requests, based on whether they were organizations or individuals: (1) “Email a friend explaining why this is so important to you” or (2) “Download the sample email and send it to your membership list today.” 45 For blog day, recipients were invited to share a blog written by Outreach Coordinator, Ignacio Rivera, cross post the blog on an organization’s blog site, or draft a blog about the importance of the survey. 46 The number of individuals who pledged to take the survey on the pledge list increased from approximately 7,700 when the initial Awareness Week email was sent on July 15 to over 14,000 at the time the survey launched in the field.

The 82% increase in the numbers of survey pledges is likely due to the increased exposure generated by Awareness Week communications. 47 The initial literacy level review and revision was conducted by a certified copy editor proficient in reading levels, and the questionnaire was determined to be at an eighth grade reading level. 48 Due to IRB requirements, the language in the study information sheet was generally at a higher literacy level than the rest of the questionnaire. 49 This included all materials aimed at “recruiting” or getting people to participate in a research study, such as website pages, flyers, emails, and social media messages. 50 The research team received reports that it took some individuals up to several hours to complete the survey on the first day, and others reported that they were not able to complete or submit their survey at all due to the technical issues.

2015 U.S. TRANSGENDER SURVEY 38 51 The following note was added to the first page of the survey (in English and Spanish) to notify respondents of the delay: “Our servers have been overwhelmed by the number of enthusiastic participants and some are experiencing unusual delays. We apologize for the inconvenience as we work to address this issue. Y ou can complete the survey now but may experience delays. However, the survey will be available to complete through at least September 21 st . If you experience delays, we encourage you to return to this site in the coming days.

If the survey is slow to respond, you can leave the page open and return later. If the survey times out, you can hit the ‘back’ button. However, if you close your browser, you may have to restart the survey.” 52 A detailed description of the cleaning process is included in Appendix C (Detailed Methodology). 53 Rahm, E. & Do, H. H. (2000). Data cleaning: Problems and current approaches.

IEEE Data Engineering Bulletin, 23(4), 3–13. 54 Ineligible respondents were sent to one of two disqualification pages notifying them of their ineligibility and providing either an opportunity to visit the survey website for more information or giving information about their gender identity or expression and experiences related to gender identity or expression. 55 See Appendix C (Detailed Methodology) for more information on the

Section 2 questions that were required to remain in the sample. 56 Missing-data analyses determined that there was less than 5% missing data on all but two questions. Therefore, the research team did not impute the missing data. See Appendix C (Detailed Methodology) for more information. 57 A modified version of an inter-rater reliability metric was used by the two teams that conducted the review. Each team included a principal researcher and an outside researcher. One researcher on each team conducted the initial coding and the other researcher reviewed the coding for approval or revisions.

See Appendix C (Detailed Methodology) for more information. 58 “Weighting” is a common statistical technique used to adjust data with disproportionate sample sizes to be more representative of the population from which the sample was drawn. For example, the proportion of respondents aged 18–24 and 25–44 in a survey sample taken in the U.S. may differ from the proportion of those age groups in the total U.S. population. Therefore, weights are applied to survey data in order to make comparisons between the collected survey data and the total population.

See Appendix C (Detailed Methodology) for more detailed information about weights applied to the survey data. 59 Studies using representative samples of transgender adults have found that transgender adults differ from the general population in regard to race and ethnicity, with transgender people more likely to be people of color. See e.g., Flores, A. R., Brown, T. N. T., & Herman, J. L. (2016). Race and Ethnicity of Adults who Identify as Transgender in the United States. Los Angeles, CA: Williams Institute; Conron, K. J., Scott, G., Stowell, G. S., & Landers, S. J. (2012).

Transgender health in Massachusetts: Results from a household probability sample of adults. American Journal of Public Health, 102(1), 118–122. However, the USTS sample has a higher percentage of white respondents than the U.S. general population. To help correct for this sampling bias, the research team applied U.S. population weights for race and ethnicity.

While this may still over-represent white respondents, this weighting procedure brings the sample closer to what is estimated to be the true population distribution for race and ethnicity for transgender people. 60 The weight for 18-year-old respondents was created with propensity scores developed using a regression discontinuity model.

For more information on this process and other weighting procedures, see Appendix C (Detailed Methodology). 61 The age, race, and educational attainment weights were created based on the Census Bureau’s 2014 American Community Survey (ACS). 62 Due to the large sample size, bivariate statistical tests largely result in statistically significant differences among the groups being compared. Small group differences often will be found to be statistically significant, even when the differences are small and, therefore, not particularly meaningful.

In writing the findings to this report, the research team considered other measures when pointing out meaningful differences among groups, such as a particular cell’s contribution to an overall chi-square test statistic and effect sizes. These tests are on file with the research team. Future researchers are encouraged to use additional bivariate and multivariate modeling to provide more nuanced understanding of group differences.

GUIDE TO REPORT AND TERMINOLOGY 39 T hroughout the report, the authors use a variety of terminology to refer to respondents in the sample or experiences that respondents reported. The authors also applied several conventions in the reporting of results. While explanations are often included in chapters to provide context and clarity, several terms and conventions that are used widely throughout the report are outlined in this

chapter to make the report more accessible to a broad range of audiences.

CHAPTER 3 Guide to Report and Terminology

2015 U.S. TRANSGENDER SURVEY 40 term transgender is used for the purposes of this report to represent the diverse identities of the individuals who made their voices heard by completing the survey. II. Other Transgender - Specific Terminology Non-binary: This term is used by some to describe people whose gender is not exclusively male or female, including those who identify as having no gender, a gender other than male or female, or more than one gender. In this report, “non-binary respondents” refers to respondents who said that the term “non-binary/genderqueer” best describes their current gender identity in response to Q. 2.3. Crossdresser: While

definitions of “crossdresser” vary, many use this term to describe a person who dresses in a way that is typically associated with a gender different from the one they were thought to be at birth, but who may not identify with that gender or intend to live full time as that gender. In this report, the term “crossdressers” refers to respondents who said that the term “cross-dresser” best described their current gender identity in response to Q. 2.3. Gender transition: This is a process in which a person begins to live according to their gender identity, rather than the gender they were thought to be at birth.

Not all transgender people have transitioned or intend to do so, but many do. Gender transition looks different for every person. Possible steps in a gender transition may or may not include changing one’s clothing, appearance, name I. Use of the Term “Transgender” in this Report The term “transgender” is often used to describe people whose gender identity or expression differs from what is associated with the gender they were thought to be at birth.

Although this term has often been described as an “umbrella term” that encompasses the spectrum of identities and captures the diversity of transgender people, the authors recognize that one term cannot reflect each individual’s unique identity and some people prefer to use other terms to describe their gender identity. However, in order to make the report’s findings clear and accessible, it was important to select a single term for consistent use throughout this report that could best represent the range of identities expressed in the USTS survey sample.

In promotional materials, the survey was described as being inclusive of all “transgender, trans, genderqueer, and non-binary” people, so that those who might have assumed that “transgender” did not include them would know their voice was welcomed. The survey also acknowledged the limitation of current language and used “trans”—a shorthand term that is widely accepted amongst transgender people—consistently throughout the questions.

While respondents in this study identified with a wide range of terms— including more than 500 unique terms that were reported in response to survey questions—88% of respondents thought of themselves as transgender, and 86% expressed that they were “very comfortable,” “somewhat comfortable,” or “neutral” when asked how comfortable they were with the word “transgender” being used to describe them. This included 82% percent of non-binary respondents. This provides evidence of the term’s continued broad usage and general acceptance. Based on this information, the

GUIDE TO REPORT AND TERMINOLOGY 41 and identity documents (for example, a driver’s license), or undergoing medical procedures such as hormone therapy to change one’s physical characteristics. This report refers to gender transition in several places when discussing steps that may be included in one’s gender transition, such as updating the name and gender on identity documents.

Additionally, the report includes a variety of terms to refer to therapy/counseling, hormone therapy, surgical treatments, and other health services transgender people may undergo as part of their transition, including “health care related to gender transition” or “transition-related care.” In this report, the term “respondents who have transitioned” refers to respondents who reported that they are living full time in response to Q. 1.12 (see below).

Living full time: Respondents in the sample who were described in the report as “living full time” are those who reported that they lived full time in a gender different than the gender they were thought to be at birth in response to Q. 1.12. For many people, living full time may include changing one’s name, clothing, and/or appearance, or taking other actions related to their gender transition.

Gender identity or expression: Several questions throughout the report asked whether respondents thought that an experience had occurred due to their “transgender status/ gender identity” and/or “gender expression/ appearance.” Both answer choices were included so that respondents could select what they felt best represented their experience.

Since there was a substantial overlap of respondents who selected both reasons, and because these terms are commonly used interchangeably or with very similar meanings, responses of those who selected one or both of these reasons were collapsed for reporting in one “gender identity/ expression” category. Additionally, several phrases are used interchangeably to describe experiences that respondents had as a result of biases due to being known or perceived to be transgender.

These include, for example: “because they were transgender,” “because of their transgender status,” or “because of their gender identity or expression.” III. Additional Terms and Conventions Used in the Report Sexual assault: In this report, the term “sexual assault” refers to a variety of experiences of unwanted sexual contact. These may include, but are not limited to, oral, genital, or anal contact or penetration, forced fondling, and rape. Respondents were asked about their experiences with unwanted sexual contact or sexual assault in a number of different contexts.

Definitions of these terms varied in some questions based on the context or, in some cases, on the national survey from which a question was adapted. Where applicable, the definition provided for “sexual assault” or “unwanted sexual contact” in each question is included in the report. Underground economy: This terminology refers to fields of work that, in general, are currently criminalized in the United States. In this report, this term includes income - based sex work (including forms of work in the sex trade that are not criminalized, such as pornography), drug sales, and other income-based work that is currently criminalized. See Sex Work and Other Underground Economy Income chapter.

2015 U.S. TRANSGENDER SURVEY 42 U.S. population comparisons and other resources: References to experiences of the U.S. population are included in the report for comparison and to provide context for findings where feasible. References to other research are also provided as resources in several places throughout the report. However, the list of references is not exhaustive, and should be not be treated as a comprehensive list of sources on any particular subject presented in this report.

Stories included in the report: Throughout the report, excerpts of stories are included in sections titled “In Our Own Voices.” These stories, which were submitted by respondents after they completed the survey, are provided to support the findings of the report and offer important anecdotal evidence and context for respondents’ reported experiences. These stories have been edited for length and clarity.

Time period of reported experiences: In the survey, respondents answered questions about experiences that occurred within a period of time prior to having taken the survey, such as in the past year or the past 30 days. The report refers to the time when these experiences occurred in comparison to the time when the respondent completed the survey.

For example, respondents who had certain experiences within the 12 months prior to completing the survey were reported as having those experiences “in the past 12 months” or “in the past year.” If a respondent had an experience that occurred within the 30 days prior to completing the survey, the experience was referred to as occurring “in the past mont