GLOBE Review Report

Survey Measures of
Sexual Orientation and
Gender Identity
Results from GLOBE Expert Review

April 6, 2017

Executive Summary
Below is a summary of the expert feedback on sexual orientation and gender identity survey measures.
Feedback came from 31 experts over about one month in early 2017. A table is included for each of the
8 questions for which feedback was obtained, noting the question and response option wording, a oneword summary of the question clarity and response option feedback, and the percentage of
respondents who said that most people would feel “very comfortable” answering this question on a
CATI survey. Also included in the table is the percentage of respondents who named this question as
their “most preferred” way to ask about either sexual orientation or gender identity, as appropriate.
Sexual orientation
Question 1:
Do you consider yourself to be: (select
one)
1 = Straight
2 = Lesbian or gay
3 = Bisexual
4 = Other

Clarity:

High

Categories:

Inadequate

Percentage “very comfortable”:

70%

Percentage most preferred:

68%

Respondents found the question to be clear, though some noted that more categories were needed.
Others mentioned disliking the category “other.” Fully 7 in 10 said that they thought people would be
“very comfortable” answering this question on a CATI survey. Overall, this question was the most
favored of all presented, with 19 out of 25 respondents (68%) saying it is the preferred way of asking
about sexual orientation.
Question 2:
Which of the following best represents how you
think of yourself? (Select one)
1 = Gay
2 = Straight, that is, not gay
3 = Bisexual
4 = Something else

Clarity:

Moderate

Categories:

Inadequate

Percentage “very comfortable”: 48%
Percentage most preferred:

7%

Some respondents mentioned that this question is worded awkwardly, but overall, most found this
question to be clear. Almost half noted that “lesbian” was not included as a response option, indicating
that this would be inadequate for some respondents. Several respondents noted that the category
“something else” is inappropriately worded and does not collect qualitative data on the self-identity.
About half of respondent (48%) said that most respondents would feel “very comfortable” answering
this question on a CATI survey. Overall, this question was the least favored of all presented, with only 1
out of 25 respondents (4%) identifying it as the preferred way of asking about sexual orientation.

1

Question 3:
Which of the following do you consider yourself to
be? (Select one)
1 = Heterosexual
2 = Lesbian or Gay
3 = Bisexual

Clarity:

High

Categories:

Inadequate

Percentage “very comfortable”: 63%
Percentage most preferred:

11%

Most respondents found this question to be clear. However, many mentioned that response options
were not sufficiently exhaustive, including that there was no “other” category for those who do not feel
represented by the three given categories. A few also noted that while “heterosexual” is used,
“homosexual” is not, suggesting that there is an imbalance in response options. That said, 63% of
respondents said that many people would feel “very comfortable” answering the question on a CATI
survey.
Question 4:
Which of the following best represents how you
think of yourself? (Select one)
1 = Gay (for males) / Lesbian or gay (for females)
2 = Straight, that is, not gay
3 = Bisexual
4 = Something else

Clarity:

High

Categories:

Complicated

Percentage “very comfortable”: 42%
Percentage most preferred:

14%

Most respondents said this question was clear. In terms of response options, a few found them to be
overly complicated, while some commented on the use of the phrase “not gay” as being unnecessary to
the question. Still others disliked the “something else” category. Fully 2 in 5 respondents (42%) said
most respondents would feel “very comfortable” answering this question on a CATI survey. Asked about
the skip pattern, the majority of respondents (73%) said that the response options should not be based
on previous gender identity questions, but that all respondents should get all response options to the
question, arguing that by retaining all options for all respondents, the respondent is freer to identify as
s/he sees appropriate, that it is more inclusive to keep all options for all respondents, and that not
depending on a previous response will prevent data errors based on missing or inappropriate responses
to previous questions.
Gender Identity:
Question 5:
Are you…? (select
one)
1 = Male
2 = Female

Clarity:
Categories:
Percentage “very
comfortable”:
Percentage most preferred:

High
Inadequate
64%
4%

Most respondents found the question to be clear, though some said it was overly simplistic and
conflates biological and socially constructed gender identities. Almost all respondents who provided
feedback on this question found the response categories to be too limited, suggesting additional
categories and the inclusion of an “other” category. Still, 64% said that they thought many respondents
would feel comfortable answering this question on a CATI survey, noting that while the majority of
2

people would not find this question troubling, non-cisgender respondents could struggle with a
response. Overall, this question was the least favored of all presented, with only 1 out of 26 respondents
(4%) identifying it as the preferred way of asking about gender identity.
Question 6:
What is your current gender identity? (select all that apply)
1 = Male
2 = Female
3 = Trans male/transman
4 = Trans female/transwoman
5 = Genderqueer/gender non-conforming
6 = Different identity (please state): ___________

Clarity:

High

Categories:

Exhaustive

Percentage “very
comfortable”:
Percentage most
preferred:

71%
27%

Almost all respondents found this question to be clear. Many indicated that the response options were
exhaustive, though some noted they were not necessarily mutually exclusive. Some warned against the
use of the word “queer,” while others said that the trans male and trans female options might be
confusing. Fully 71 percent said that most respondents would be comfortable answering this question
on a CATI survey, however a few noted that cisgender respondents might be overwhelmed by the
number of non-binary categories.
Question 7:
Some people describe themselves as transgender when they
experience a different gender identity from their sex at
birth. For example, a person born into a male body, but who
feels female or lives as a woman. Do you consider yourself
to be transgender? (Select one)
1 = Yes, transgender, male to female
2 = Yes, transgender, female to male
3 = Yes, transgender, gender non-conforming
4 = No

Clarity:

High

Categories:

Clear

Percentage “very comfortable”: 60%
Percentage most preferred:

19%

Many respondents found this question to be clear, though a few said that cisgender respondents might
be overwhelmed by the length of the question stem. About half of respondents said that the response
options were clear, while others said that option 3 seemed unreliable, that there was a need for
additional categories, and that some who have transitioned genders might not identify as transgender in
this question. Three in five (60 percent) said that most respondents would feel “very comfortable”
answering this question on a CATI survey, though some mentioned that genderqueer respondents might
struggle with a response. Note that while some respondents said they thought most people would find
this question uncomfortable to answer, they added the caveat that it was a good and important
question to ask.

3

Question 8:
What sex were you assigned at birth, on your original
birth certificate? (select one)
1 = Male
2 = Female
How do you describe yourself? (select one)
1 = Male
2 = Female
3 = Transgender
4 = Do not identify as female, male, or transgender

Clarity:

Clear

Categories:

Inadequate

Percentage “very
comfortable”:

62%

Percentage most preferred:

50%

All respondents who provided feedback on this question found it to be clear and did not find any of the
words unfamiliar, ambiguous, or otherwise unclear. Some noted that the response options could be
expanded, while others said that the response options were not mutually exclusive. Just over six in ten
respondents (62%) said that most people would feel “very comfortable” answering this question on a
CATI survey. Overall, this is the most favored way of asking about gender identity, with 13 out of 26
respondents (50%) saying this is their preferred question for gender identity.
Preferred Question Stem:
Most respondents (11 out of 25, or 44%) preferred the question stem: “…you consider yourself to be…”
when asking about sexual orientation and gender identity, arguing that it emphasizes individual selfreport and did not imply doubt, noting that it does not use the terms “think” and “describe.” Some (6
out of 25, or 24%) preferred “…best represents,” noting that this question stem is empowering for
respondents, that it acknowledges that the given options may not completely capture their sexual
orientation or gender identity, and that it does not doubt the respondents answer.

Introduction
In consultation with the Centers for Disease Control and Prevention (CDC), ICF methodologists
developed an instrument designed to solicit feedback from experts in the GLOBE community at CDC.
This instrument was not intended to be a survey; rather, it was designed to collect qualitative reflections
from people with expertise in LGBT health issues and/or survey methodology. This goal is reflected in
the opening instructions to respondents, wherein they are asked for their “assessment of the questions
listed below,” and urged not to answer the questions themselves, but rather to “provide feedback on
their shortcomings and how they might be improved.”
Once the instrument was finalized, ICF programmed it into an online platform. This platform generated
a unique URL for respondents to access the instrument. CDC sent out the invitations to the GLOBE
community, encouraging response and including the unique URL, on February 26, 1017. There was one
follow-up email for non-respondents on March 13, 2017, and by March 20, 2017, 31 responses had been
collected and data collection was ended.
In total, respondents were asked to provide feedback on 8 questions related to measuring sexual
orientation or gender identity (SOGI).
4

Table 1. Questions for Review and the Corresponding Construct It Measures
Question Question Text
Number
1
Do you consider yourself to be:
1 = Straight
2 = Lesbian or gay
3 = Bisexual
4 = Other
2
Which of the following best represents how you think
of yourself?
1 = Gay
2 = Straight, that is, not gay
3 = Bisexual
4 = Something else
3
Which of the following do you consider yourself to
be?
1 = Heterosexual
2 = Lesbian or gay
3 = Bisexual
4
Which of the following best represents how you think
of yourself?
1 = Gay (for males)/Lesbian or gay (for females)
2 = Straight, that is, not gay
3 = Bisexual
4 = Something else
5
Are you…?
1 = Male
2 = Female
6
What is your current gender identity?
1 = Male
2 = Female
3 = Trans male/transman
4 = Trans female/transwoman
5 = Genderqueer/gender non-conforming
6 = Different identity (please describe)
7
Some people describe themselves as transgender
when they experience a different gender identity from
their sex at birth. For example, a person born into a
male body, but who feels female or lives as a woman.
Do you consider yourself to be transgender?
1 = yes, transgender, male to female
2 = yes, transgender, female to male
3 = yes, transgender, gender non-conforming
4 = No
8 (series) What sex were you assigned at birth, on your original
birth certificate?
1 = Male

Measuring

Source

Sexual
BRFSS
orientation

Sexual
NHIS
orientation

Sexual
NSDUH
orientation

Sexual
NCSV
orientation

Gender
identity
Gender
identity

GeniUSS

Gender
identity

GeniUSS

Gender
identity

GeniUSS

5

Table 1. Questions for Review and the Corresponding Construct It Measures
Question Question Text
Number
2 = Female

Measuring

Source

How do you describe yourself?
1 = Male
2 = Female
3 = Transgender
4 = Do not identify as female, male, or transgender

Respondents were presented the question text, and then asked a series of reflective questions about
the text, including:





Question clarity: Did this question make sense to you? Were any of the words unfamiliar,
ambiguous, or otherwise unclear?
Response option clarity and adequacy: Are response options mutually exclusive and
exhaustive? Are any of them unclear or using words with which you are unfamiliar?
How comfortable would you feel answering this question during a CATI (telephone) interview?
Additional comments: How comfortable do you think most people would feel answering this
question during a CATI (telephone) interview?

Note that because question 4 depends on a skip pattern to fill the first response option, respondents
were asked an additional question:
For this question, the first response option is determined by how respondents previously
reported their sex. Those who identify as female are given the option "lesbian or gay." Those
that identify as male are given the option "gay."
In your experience, should the response options be based on the previous response to sex,
should respondents get all response options regardless of their self-identified sex, or something
else?
 Based on a previous response to sex
 All respondents get all options
 Something else
Why did you select this response - either based on self-identified sex, all respondents get all
options, or something else?

Respondent Age
A total of 31 people responded to this request for feedback, though not all 31 provided responses for all
questions. Of the 26 who responded to the age question, most were between the ages of 41 and 50
years old, with some being slightly older (51 to 60 years old) and slightly younger (31 to 40 years old).

6

Just 4 respondents reported being ages 18 to 30 years old. No respondents were younger than 18 or
older than 61 years old.
Figure 1. Respondents’ Age
Respondents' Age
Percentage Responding

51 to 60 years old
23%

41 to 50 years old
42%
18 to 30 years old
16%

31 to 40 years old
19%

Sexual Orientation Questions
Question 1:
Do you consider yourself to be: (select one)
1 = Straight
2 = Lesbian or gay
3 = Bisexual
4 = Other
Question Clarity:
In general, respondents found this question to be clear. One commented that the “question makes
sense,” while another said that “the words are familiar and clear.”
Response Option Clarity:
Many respondents (10 out of 29 providing comments on this question) found the response options to be
clear and adequate as they are written. Five respondents, however, mentioned that the response
options should include the words “heterosexual” and “homosexual” along with straight and gay
(respectively). Likewise, several respondents mentioned that additional response options should be
7

added to the question, including pansexual, asexual, queer, polyamorous, questioning, and unknown.
Two respondents suggested including a “decline to answer” category.
Notably, nine respondents commented on the “other” category in this question. Several requested a
text box to capture specifications of ‘other.’ A few recommended the inclusion of examples to prompt
those responding ‘other’ to write in a descriptor, including pansexual, asexual, and other sexual
orientations. Two respondents found the word ‘other’ to be an inappropriate option, stating that the
word “other” might be insensitive to members of the LGBT community.
Response Comfort:
Table 2. Response Comfort for Question 1
Number and Percentage responding
Number
Percentage
Very uncomfortable
3
10%
Somewhat uncomfortable
2
7%
Mostly comfortable
4
13%
Very comfortable
21
70%

When asked about general level of comfort responding to this question, most respondents indicated
that most people would feel comfortable answering this question on a CATI survey. Some mentioned
that particular groups of respondents might be more comfortable than others answering this question,
including those under the age of 50. Three respondents noted that those who are “newly out as
gay/lesbian or bisexual” might not be comfortable answering this question, and that their answer and
comfort level might change depending on where the respondent is during an interview and who may be
hearing their responses. Interestingly, one participant stated that he noticed the numbers at the start of
each response option. This respondent stated that reading the response options with the numbers
would be a way to make participants feel more comfortable, as they would be able to answer with the
number, and not the category label, if not in a completely private space. A few mentioned that while
most people would feel comfortable answering this question, “there are still individuals that will not
answer truthfully about sexual orientation, but really [there is] nothing you can do about this.”

Question 2:
Which of the following best represents how you think of
yourself? (Select one)
1 = Gay
2 = Straight, that is, not gay
3 = Bisexual
4 = Something else

8

Question Clarity:
Overall, most respondents found the question clear. However, a number of respondents commented
that while the question might be clear, it is not well worded: one mentioned that “’how you think of
yourself’ is weird and could be taken the wrong way.” Another commented that the question is “fine,
just didn’t like the wording,” and several others commented that they preferred the previous question’s
wording. Note that one respondent voiced concern that the question stem does not clearly indicate
that the question is about sexual orientation, and that without this indication, respondents might
answer about gender identity in the “something else” field.
Response Option Clarity:
Almost half of respondents who provided comments on this question objected that “lesbian” was not
included as a response option. These respondents indicated that it was not adequate for gay to
represent both lesbians and gay men. One said, “some lesbians might choose ‘something else’ because
they think of ‘gay’ as being a term for men.”
Twelve respondents (of the 23 who commented on this question) commented on the response option
“straight, that is, not gay,” saying that the definition was unnecessary, with one asking “why is there a
clarifying statement for ‘straight’?” Another mentioned that “straight is not the inverse of gay. If you
are bisexual, you are not gay, but you are also not straight.”
The category ‘something else’ also elicited responses. While one respondent appreciated the category
as labeled, several others objected to the use of “something else,“ indicating that it implies that
participants are “things.” In terms of including additional categories, one respondent mentioned
changing ‘something else’ to “some other identity,” while others mentioned adding the terms
homosexual, heterosexual, asexual, and don’t know, or including a text box to capture other responses.
Response Comfort:
Table 3. Response Comfort for Question 3
Number and Percentage responding
Number
Percentage
Very uncomfortable
2
7%
Somewhat uncomfortable
5
19%
Mostly comfortable
7
26%
Very comfortable
13
48%

The majority of respondents mentioned that they would feel comfortable answering the question, and
that they thought that others would also feel comfortable answering this question on a phone survey.
Three respondents mentioned that lesbians might not be comfortable responding to the question, as
the term lesbian is not included in the response options. Likewise, one respondent suggested that
survey respondents might not be comfortable with the question because it “sounds judgmental and

9

collapses important identity distinctions,” while another respondent offered that this item seemed
“more sensitive than the first question.”

Question 3:
Which of the following do you consider yourself to be? (Select one)
1 = Heterosexual
2 = Lesbian or Gay
3 = Bisexual
Question Clarity:
Most respondents stated that the question “makes sense,” though two offered minor changes to the
wording of the question. One suggested removing the words “to be” from the end of the question,
while the other suggested changing the end of the question to read “…to be now” or “…to be currently.”
Response Option Clarity:
Most respondents who gave feedback on this question expressed concern that the response options
were not sufficiently exhaustive. About half suggested additional response options, including pansexual,
asexual, or queer. Likewise, about as many mentioned that the item needed an “other” category in
order to, as one respondent said, “leave room for the respondent to identify their sexual orientation as
anything that is non-binary.” A few respondents noted an imbalance in the response options, saying
that the use of the term heterosexual was “more scientific,” but the use of the terms lesbian or gay are
“less scientific.” Respondents suggested including both response labels for each: heterosexual or
straight and lesbian or gay or homosexual.
Response Comfort:
Table 4. Response Comfort for Question 3
Number and Percentage responding
Number
Percentage
Very uncomfortable
2
7%
Somewhat uncomfortable
3
11%
Mostly comfortable
5
19%
Very comfortable
17
63%
The majority of respondents who commented on the comfort of answering this question during a CATI
survey indicated that most people would feel comfortable. Three respondents mentioned that those
who may have responded with “other” or “something else” to previous questions would not be
comfortable answering this question, as they would not know how to provide a response.

10

Question 4:
Which of the following best represents how you think of yourself? (Select one)
1 = Gay (for males) / Lesbian or gay (for females)
2 = Straight, that is, not gay
3 = Bisexual
4 = Something else
Question Clarity:
Most respondents mentioned that this question was clear, and one called it “excellent.” However, a few
respondents offered that the word “think” in the question stem might not be appropriate. Said one,
“why not just say ‘represents your sexual orientation?’ It’s not how I THINK of myself, it’s my
orientation.” Another suggested that the word “represents” in the question stem was not accurate,
while still another mentioned the need to end the question with “now” or “currently.”
Response Option Clarity:
While a few respondents found the response options to be clear, exclusive, and exhaustive, many more
voiced a number of concerns with the response options for this question. Some found the first response
option to be “overly wordy and complicated,” and that there was no need to explain which terms were
for males and which for females. This may be due to a misunderstanding that this response option is
filled based on previous identification of being male or female, and the parenthetical words are not
displayed to survey respondents.
Others had trouble with the wording of response options for this question. A few find it unnecessary to
explain that straight is ‘not gay,’ while others again mentioned that ‘something else’ is not an ideal
category to capture identities outside of the given categories. A couple of respondents mentioned the
need to expand the categories to include don’t know, or other categories of sexual orientation (i.e.,
asexual).
Response Comfort:
Table 5. Response Comfort for Question 4
Number and Percentage responding
Number
Percentage
Very uncomfortable
3
12%
Somewhat uncomfortable
6
25%
Mostly comfortable
5
21%
Very comfortable
10
42%

Most respondents said that other people would feel comfortable answering this question on a CATI
survey. A few respondents said they think that some would be uncomfortable providing a response to

11

this question, including that a “something else” option might offend some participants. One respondent
suggested reworking the question as a yes/no select all that apply question.
Skip Pattern:
Unlike previous questions, the response categories for question 4 are dependent on respondents’
previously reported sex. GLOBE experts were asked their opinions on this category filling:
For this question, the first response option is determined by how respondents previously
reported their sex. Those who identify as female are given the option “lesbian or gay.” Those
that identify as male are given the option “gay.”
In your experience, should the response options be based on the previous response to sex,
should all respondents get all response options regardless of their self-identified sex, or
something else?
Asked as a close-ended question, almost three in four GLOBE respondents (73 percent) said that “all
respondents should get all options” to the question. Another almost one in five (19 percent) said that
response options should be “based on a previous response to sex.” Just 8 percent said “something else”
should be done with the response categories to this question.
Figure 2. Handling response options to question 4

In your experience, should the response options be based on the previous
response to sex, should all respondents get all response options regardless
of their self-identified sex, or something else?
Percentage responding

Something else
8%

Based on a previous
response to sex
19%

All respondents get
all options
73%

12

As a follow up, respondents were asked “why did you select this response – either based on selfidentified sex, all respondents get all options, or something else?” For those that selected “based on a
previous response to sex,” respondents were motivated mostly by the burden of the response options.
They commented that by using previous data to fill the response options, the question would have
lower burden as a result of eliminating response options that would not be selected anyway and making
the response options to later questions less confusing or more consistent.
For those respondents who selected that “all respondents get all options,” there were three main
reasons for not wanting to use the pre-fill. The first is that it is the simplest for both the respondents
and the interviewer, with one saying “leave it more open, because you don’t want to tell people of
either sex how they should label their sexual orientation.” Another reason some respondents did not
prefer the pre-fill is because it was most inclusive of all different kinds of identities that may not be
captured if a pre-fill is used. For example, one respondent mentioned that “I have trans friends who
consider themselves gay and some consider themselves straight. They all should have the opportunity
to answer this question.” Finally, a few respondents mentioned concerns about getting the cleanest
data by having the question independent of previously collected data, with one saying that this would
result in data that are “clearer and less chance for an error.”

Gender Identity Questions
Question 5:
Are you…? (select one)
1 = Male
2 = Female
Question Clarity:
While most respondents found the question to be clear, two did describe it as “over simplistic,” and
another offered a rewrite of the stem, suggesting: “What do you consider your gender?” Several
respondents mentioned that as written, the question is unclear as to whether the response should be
biological sex at birth or current gender identity, with one saying “the question is unclear. Is it referring
to how I was assigned at birth? Is it referring to how I self-identify?”
Response Option Clarity:
Almost all respondents who provided feedback on the categories for this question found them to be too
limited. Respondents noted that additional options may be useful, suggesting transgender, agender,
nonbinary, or transwoman and transman, while still others mentioned that the question needed an
“other” category with a qualitative capture for respondents to describe their gender identity.
Additionally, one respondent mentioned that asking for one exclusive gender identity is outdated. This
need for expanded response options was summarized by one respondent, saying “there are now over 30
distinct gender identities. We no longer live in a binary world.”

13

Response Comfort:
Table 6. Response Comfort for Question 5
Number and Percentage responding
Number
Percentage
Very uncomfortable
4
16%
Somewhat uncomfortable
2
8%
Mostly comfortable
3
12%
Very comfortable
16
64%

Several respondents said that “most” or the “vast majority” of people would feel comfortable answering
this question. However, many respondents suggested that non-cisgender individuals would not be
comfortable answering this question, with one respondent remarking “transgendered folks would be
very uncomfortable” answering this question.

Question 6:
What is your current gender identity? (select all that apply)
1 = Male
2 = Female
3 = Trans male/transman
4 = Trans female/transwoman
5 = Genderqueer/gender non-conforming
6 = Different identity (please state): ___________

Question Clarity:
Asking about the question stem, only one respondent mentioned that “gender identity” might not be
the best terminology for this question, commenting that “the term ‘gender identity’ can be confusing…I
would just ask ‘are you’ and give them the above options.” All others who offered feedback for this
question found the question to be clear, though some offered alternative wording, including “do you
consider yourself as…,” “which of the following do you best identify as….,” and providing separate
questions for sex at birth and current gender identity.
Response Option Clarity:
Most respondents indicated that the response options were clear. A few described the response
options as “exhaustive,” and some noted that the response options were not mutually exclusive, since
respondents could select more than one category. However, several respondents voiced some concern
about the response options as written. Some of these respondents expressed that “trans
male/transman” and “trans female/transwoman” were not clear, while another warned against using
the term “queer.” One mentioned that a general population survey may not need so many response
categories. Interestingly, whereas in earlier questions, respondents rallied for a qualitative capture for
14

‘other’ categories, in this case, one respondent recommended against providing a space to fill in a
“different identity,” saying that “these [responses] are difficult to analyze because they can’t be
collapsed into an existing category. You could probably just leave it at ‘different identity’ without
allowing them to write in their own reply to avoid the headache” of handling these data.
Response Comfort:
Table 7. Response Comfort for Question 6
Number and Percentage responding
Number
Percentage
Very uncomfortable
3
12%
Somewhat uncomfortable
1
5%
Mostly comfortable
3
12%
Very comfortable
17
71%

Most respondents thought that most people would feel comfortable answering this question in the
context of a CATI interview. However, five respondents offered that cisgender individuals might not be
comfortable answering this question because of the number and text of response categories. Said one,
“cisgender people might be put off by all of the options and hearing the term ‘genderqueer.’”
Additionally, one respondent suggested adding a norming statement similar to question 7 (“Some
people describe themselves as transgender when they experience a different gender identity from their
sex at birth”) to help respondents feel more comfortable with the options to this question.

Question 7:
Some people describe themselves as transgender when they
experience a different gender identity from their sex at birth.
For example, a person born into a male body, but who feels
female or lives as a woman. Do you consider yourself to be
transgender? (Select one)
1 = Yes, transgender, male to female
2 = Yes, transgender, female to male
3 = Yes, transgender, gender non-conforming
4 = No
Question Clarity:
Most respondents found this question to be clear, and one went so far as to say that it has “excellent
wording and clarity.” A few respondents noted cisgender respondents might feel confused by this
question and transgender respondents might find the question to be overly simplistic, with one saying
that the “majority of cisgender people will be confused by this long description, and you run the risk of
angering transgender people who feel like this description over-simplifies their life experience.”

15

Response Option Clarity:
About half of those who provided a response on response option clarity found the response options to
be clear and adequate. However, several noted some confusion about the third response option, “yes,
transgender, gender non-conforming,” noting that they were unfamiliar with the terminology and were
not clear on how it would be applied. This sentiment was expressed by one respondent saying,
“response choice number three seems confusing…you’re transgender but also gender non-conforming?”
Others noted that this question had no other/specify option, and still others suggested additional
response categories, like intersex. Of note, a couple of respondents mentioned that because this is a
‘one step’ gender identity question, some transgender respondents may not be captured by the
response options provided, saying, “many people who have transitioned do NOT identify as transgender,
even if it is described as it is in the question stem, and therefore, this question would not capture these
individuals who have cross-sex experiences.”
Response Comfort:
Table 8. Response Comfort for Question 7
Number and Percentage responding
Number
Percentage
Very uncomfortable
4
16%
Somewhat uncomfortable
2
8%
Mostly comfortable
4
16%
Very comfortable
15
60%

About half of those who provided a comment for this question said that most people would feel
comfortable answering this question. On the other hand, some mentioned that there are instances
where people may not feel comfortable answering this question, including those who are intersex,
heterosexual and cisgender respondents, and, as one respondent commented, “genderqueer folks
would take issue with this question.” Note, though, that some respondents who said this question
might be uncomfortable to answer also noted that it was a good and important question to ask
Question Clarity:
All respondents who provided feedback found this question series to be clear, and did not find any of
the words unfamiliar, ambiguous, or otherwise unclear. Note that two respondents suggested that
asking about an assigned sex at birth might be unfamiliar or upsetting to some, prompting one
respondent to suggest changing the question stem to “what sex was listed on your birth certificate?” A
few respondents mentioned changing the word “describe” to “identify” in the second question, while
still others suggested adding the word “now” to the end of the question.

16

Question 8:
What sex were you assigned at birth, on your original birth
certificate? (select one)
1 = Male
2 = Female
How do you describe yourself? (select one)
1 = Male
2 = Female
3 = Transgender
4 = Do not identify as female, male, or transgender
Response Option Clarity:
While most respondents thought that the response options were clear, many noted that they were not
adequate. Some suggested that more response options were needed, including trans male/transman,
trans female/transwoman, genderqueer/gender non-conforming, intersex, different identity, and not
applicable. Additionally, some respondents suggested that the existing response options were not
mutually exclusive, and that for this question, respondents should be given the option to ‘mark all that
apply.’
Response Comfort:
Table 9. Response Comfort for Question 8
Number and Percentage responding
Number
Percentage
Very uncomfortable
3
13%
Somewhat uncomfortable
2
8%
Mostly comfortable
4
17%
Very comfortable
15
62%
Most respondents noted that they would feel comfortable answering this question series. Two thought
that transgender respondents might feel uncomfortable providing a response, and another two
suggested that both transgender and cisgender respondents might feel uncomfortable answering this
question series.

Preferred Question Stem
Respondents were then asked about their preferred question stem for questions of sexual orientation
and gender identity. Throughout the questions listed on this form, a few different phrases were used as
question stems. Which of the following is your preferred question stem?
1 "...you consider yourself to be..."
2 "...best represents..."
3 "How you think of yourself..."
4 "How you describe yourself..."
17

Most respondents (11 out of 25) preferred the first option, “…you consider yourself to be… as the
question stem for these questions. Some (6 out of 25) preferred“…best represents” as the question
stem. Even fewer preferred “How do you describe yourself,” or “How do you think of yourself” (4 out of
25 respondents each) as the most appropriate questions tem.
Asked why they prefer the stem that they chose, those that chose option one pointed to the emphasis it
places on the individual and self-report. One mentioned that the use of the phrase “consider yourself
seem[ed] the best term that is somewhere between gender identity (how you think of yourself) and
gender expression (how you describe yourself).” Another appreciated that this question stem did not
seem to imply doubt, noting that “describe” and “think” might sound like the responses were untrue.
Figure 3: Preferred Question Stem

Preferred Question Stem
Percentage responding
How do you describe
yourself
16%

How do you think of
yourself
16%

You consider
yourself to be
44%

…best represents
24%

Those that chose ‘best represents’ as their preferred question stem mostly pointed to the phrase as
being empowering. One respondent mentioned that it is their preferred stem because it is not
“doubting the validity of [respondents’] identity or accusing, as if that gender/sexual identity doesn’t
really exist but you ‘think’ you have that identity.” Still another liked “best represents” because it is the
only stem that “recognizes that the correct answer might not be present.”
For those that preferred “how do you describe yourself,” the emphasis is on the word “describe,” with
respondents saying it felt “concrete” and “personal.” Describe, for one respondent, includes “both how
I internally identify and also how I would externally identify myself to others.” Describe, for some, is
more meaningful than ‘consider,’ because ‘consider’ might be “interpreted to mean that the person isn’t
18

really whatever they identify and that they just ‘consider’ themselves as that.” Only one respondent
provided feedback on a preference of “how do you think of yourself,” stating that this one is preferred
because “in my heart I am one way and outwardly I am different. This question stem helps make it
easier to express this.”

Preferred Sexual Orientation Question
Table 10. Preferred Sexual Orientation Question
Number and Percentage responding
Question
Number Question Text
1
Do you consider yourself to be:

Number
Favoring
19

2

1

3

4

Response Options
1 = Straight
2 = Lesbian or gay
3 = Bisexual
4 = Other
Which of the following best represents 1 = Gay
how you think of yourself?
2 = Straight, that is, not gay
3 = Bisexual
4 = Something else
Which one of the following do you
1 = Heterosexual
consider yourself to be?
2 = Lesbian or Gay
3 = Bisexual
Which of the following best represents 1 = Gay (for males) / Lesbian or gay
how you think of yourself?
(for females)
2 = Straight, that is, not gay
3 = Bisexual
4 = Something else

2

3

At the conclusion of the form, respondents were asked to select their preferred sexual orientation
question from the four presented for feedback. Most respondents preferred the first question, “Do you
consider yourself to be: straight, lesbian or gay, bisexual, other.” When asked why, the majority of
respondents mentioned that this question was “straightforward,” “simple and clear,” “to the point,” and
used “plain language.” Respondents said that the response options were inclusive, exhaustive, and
clear. While overall respondents preferred this question, a few did mention that the term ‘other’ could
be replaced with ‘something else’ to soften the language.
The one in three respondents who did not choose the first question as their preferred way of asking
about sexual orientation were also asked a follow-up question on why they selected the question that
they did. Those that chose question two, “Which of the following best represents how you think of
yourself? Gay, straight, that is not gay, bisexual, something else,” mentioned that the way the question
is phrased is what made it the preference, with one saying, “I prefer ‘best represents.’ I would add ‘gay
or lesbian’ as an option, though, instead of just ‘gay’” as a category. Question three – “Which one of the
following do you consider yourself to be? Heterosexual, lesbian or gay, bisexual” – was described as
“good” by one respondent and as straightforward by another. And, for question four – “Which of the
19

following best represents how you think of yourself? Gay/lesbian, straight that is not gay, bisexual,
something else” – was described as being the most inclusive of the options, with one respondent
mentioning the use of “something else” in place of an “other” category was particularly inclusive.
It should be noted that one respondent disliked all of the options presented on this form. In this case, it
is the wording of the question stem that is at odds. This respondent stated: “I would prefer ‘What is
your…?’ To me, [the words] ‘think’ or ‘consider’ implies it is up for debate.”
Figure 4: Preferred Sexual Orientation Question

Preferred Sexual Orientation Question
Percentage responding

Which of the following best
represents how you think of
yourself (with skip fills)
14%
Do you consider
yourself to be
68%

Which one of the following do
you consider yourself to be
11%

Which of the following best
represents how you think of yourself
7%

Preferred Gender Identity Question
Most respondents preferred question series 8, the two-step reporting of gender identity: “What sex
were you assigned at birth, on your original birth certificate? Male, female; How do you describe
yourself? Male, female, transgender, do not identify as female, male, or transgender.” Asked why this is
the preferred way of asking about gender identity, respondents noted that it provided more data than
other questions. Said one, “the two-step approach [to asking gender identity] seems best for
distinguishing biological sex at birth and current gender identity.” Another echoed this sentiment,
saying that this question is “both technically accurate and would completely capture any gender or
sexual identity. The phrasing and response categories are appropriate and sensitive” to all needs.

20

Table 11. Preferred Gender Identity Question
Number and Percentage responding
Question
Number Question Text
Are you…?
5
6

What is your current gender identity?
(select all that apply)

7

Some people describe themselves as
transgender when they experience a
different gender identity from their sex at
birth. For example, a person born into a
male body, but who feels female or lives
as a woman. Do you consider yourself to
be transgender?
Question 8
What sex were you assigned at birth, on
your original birth certificate? (select
one)

8 (series)

How do you describe yourself? (select
one)

Response Options
1 = Male
2 = Female
1 = Male
2 = Female
3 = Trans male/Transman
4 = Trans female/Transwoman
5 = Genderqueer/Gender non-conforming
6 = Different identity (please state):
________
1 = Yes, transgender, male to female
2 = Yes, transgender, female to male
3 = Yes, transgender, gender nonconforming
4 = No

1 = Male
2 = Female

Number
Favoring
1
7

5

13

1 = Male
2 = Female
3 = Transgender
4 = Do not identify as female, male, or
transgender

For other versions of the gender identity question, very few respondents provided any context to why
they chose the question they chose. Asked about question 5, one respondent simply stated: “I am a gay
white male,” indicating that he would have no trouble answering the question. Respondents who
preferred question 6 did so because it is the most inclusive, least confusing question of the set. It should
be noted, though, that one respondent suggested taking “the response options from question six and
mak[ing] them the options for the second question in question 8, and [that would be] the best way to
collect these data.” The few that chose question 7 as their preferred way of asking about gender
identity did so because they thought that the question is clear, with one respondent saying that it
includes “its own internal clarification for respondents who do not know the meaning of transgender.”
Another mentioned that question 7 was the “most personal” way of asking about gender identity.

21

Preferred Gender Identity Question
Percentage responding
Are you…
4%

What sex were you
assigned at birth?
How do you describe
yourself?
50%

What is your current
gender identity
27%

Do you consider
yourself to be
transgender
19%

22