Content uploaded by Lori Brotto
Author content
All content in this area was uploaded by Lori Brotto on Feb 10, 2015
Content may be subject to copyright.
Psychological Assessment
A Validated Measure of No Sexual Attraction: The
Asexuality Identification Scale
Morag A. Yule, Lori A. Brotto, and Boris B. Gorzalka
Online First Publication, November 10, 2014. http://dx.doi.org/10.1037/a0038196
CITATION
Yule, M. A., Brotto, L. A., & Gorzalka, B. B. (2014, November 10). A Validated Measure of No
Sexual Attraction: The Asexuality Identification Scale. Psychological Assessment. Advance
online publication. http://dx.doi.org/10.1037/a0038196
A Validated Measure of No Sexual Attraction:
The Asexuality Identification Scale
Morag A. Yule, Lori A. Brotto, and Boris B. Gorzalka
University of British Columbia
Human asexuality has been described as a lack of sexual attraction toward anyone or anything. One
percent of the adult population is thought to be asexual, and research suggests that asexuality is best
conceptualized as a sexual orientation. A serious limitation in past research on asexuality has been the
complete lack of a validated tool to measure asexuality. Due to limitations in recruiting sufficiently
powered local samples, most studies have relied on recruiting via online web-based asexual communities.
This is problematic because it limits the sample to individuals who have been recruited through
established asexuality networks/communities. The present study aimed to develop and validate a
self-report questionnaire to assess asexuality. The questionnaire was intended to provide a valid measure
independent of whether the individual self-identified as asexual and was developed in several stages,
including: development and administration of open-ended questions (209 participants: 139 asexual and
70 sexual); administration and analysis of resulting 111 items (917 participants: 165 asexual and 752
sexual); administration and analysis of 37 retained items (1,242 participants: 316 asexual and 926
sexual); and validity analysis of the final items. The resulting Asexuality Identification Scale (AIS), a
12-item questionnaire, is a brief, valid, and reliable self-report instrument for assessing asexuality. It is
psychometrically sound, easy to administer, and has demonstrated ability to discriminate between sexual
and asexual individuals. It should prove useful to allow researchers to recruit more representative samples
of the asexual population, permitting for an increased understanding of asexuality.
Keywords: asexuality, low sexual desire, sexual orientation, questionnaire development, factor analysis
Sexual attraction has always been thought be to a ubiquitous
aspect of the human experience and necessary for propagation of
species. However, there is mounting evidence that at least 1% of
the adult human population lack any attraction toward others
(Bogaert, 2004; Lucassen et al., 2011; Poston & Baumle, 2010).
Human asexuality has been defined as a lack of sexual attraction to
anyone, or anything, at all (Bogaert, 2004; Brotto, Knudson, In-
skip, Rhodes, & Erskine, 2010; Jay, 2008), and has been concep-
tualized as a sexual orientation (Berkey, Perelman-Hall, & Kurdek,
1990; Brotto & Yule, 2011; Brotto et al., 2010; Storms, 1979).
While this definition is the most widely endorsed among asexual
individuals, asexuality has been described by others in terms of a
lack of sexual behavior (Rothblum & Brehony, 1993), as well as
low sexual excitation (Prause & Graham, 2007), and even as a
sexual dysfunction (Bogaert, 2004, 2006; Prause & Graham, 2007;
Brotto, 2013). More colloquially, there are pervasive stereotypes
that assume all asexual individuals are aromantic (i.e., that they
don’t experience romantic attraction—described as “a feeling that
causes people to desire a romantic relationship with a specific
other person” (Asexuality Visibility and Education Network
[AVEN] Wiki, 2013), female, afraid of sex, highly religious,
disabled, victims of sexual trauma, or as making a conscious
choice to be asexual (e.g., celibacy) (Neth, 2011; NextStepCake,
2011; Sloan, 2006; Walters & Geddie, 2006). There is also strong
evidence of bias and discrimination against asexual individuals
(MacInnis & Hodson, 2012). How asexuality is conceptualized,
therefore, has profound clinical, academic, and sociocultural im-
plications.
Although asexuality has appeared sporadically throughout the
scientific literature since Kinsey, Pomeroy, and Martin’s (1948)
Sexual Behavior in the Human Male, it is only in the past decade
that it began to receive serious academic attention, beginning with
Bogaert’s (2004) finding that approximately 1% of a large British
sample reported a lack of sexual attraction. Researchers then
focused on conceptualizing and defining the construct (Bogaert,
2006; Brotto et al., 2010; Prause & Graham, 2007), and a flurry of
publications revealed new data on the characteristics of asexual
populations. For example, asexual individuals were found to differ
significantly from sexual groups on several characteristics, such
that they experience poorer health (Bogaert, 2004; Poston &
Baumle, 2010), have shorter stature and later age of menarche
(Bogaert, 2004), have increased levels of depression and anxiety as
well as decreased self-esteem (Nurius, 1983; Yule, Brotto &
Gorzalka, 2013), have lower socioeconomic status and higher
religiosity (Bogaert, 2004), and decreased levels of sexual func-
Morag A. Yule, Department of Psychology, University of British Co-
lumbia; Lori A. Brotto, Department of Obstetrics and Gynaecology, Uni-
versity of British Columbia; Boris B. Gorzalka, Department of Psychology,
University of British Columbia.
This research was supported in part by a Frederick Banting and Charles
Best Canada Graduate Scholarship Master’s Award awarded to Morag
Yule by the Canadian Institutes of Health Research.
Correspondence concerning this article should be addressed to Morag
Yule, 2775 Laurel Street, 6th Floor, Department of Gynaecology, Univer-
sity of British Columbia, Vancouver, BC, Canada, V5Z 1M9. E-mail:
moragy@gmail.com
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Psychological Assessment © 2014 American Psychological Association
2014, Vol. 26, No. 4, 000 1040-3590/14/$12.00 http://dx.doi.org/10.1037/a0038196
1
tioning (Brotto et al., 2010; Prause & Graham, 2007). Asexual
subjects have been the focus in sexual psychophysiological re-
search aimed at testing whether a deficiency in the capacity for
physical sexual response may underlie their reported lack of at-
traction (Brotto & Yule, 2011). Finally, there has been recent
interest in biological markers (e.g., digit ratios, handedness, fra-
ternal birth order), which are associated with the development of
homosexuality, in asexual men and women (Yule, Brotto & Gor-
zalka, 2014a).
Previous research has identified asexual individuals using cri-
teria such as self-identification or agreement with the statement “I
have never felt sexually attracted to anyone at all” (Bogaert, 2004;
Brotto et al., 2010) or have recruited asexuals through various
asexuality online organizations. These methods of recruitment are
problematic because those individuals who may lack sexual attrac-
tion but who have not yet identified as asexual may be missed.
Further, those asexuals recruited via an asexuality website may
have different features and experiences than those recruited
through other means. In particular for those asexual subjects not
recruited through asexuality websites, it has been hypothesized
that they may have responded to the question about sexual orien-
tation with either “heterosexual” or “homosexual” as they found
that this best fits their romantic attraction (Hinderliter, 2009) but
doing so might result in their being miscategorized as sexual. The
Asexuality and Visibility Education Network (AVEN; http://www
.asexuality.org) has been described as an important aspect in the
identification process of asexual individuals, as it is a venue for
validating one’s experiences, where asexuals can discuss their lack
of sexual attraction, and where they can find a sense of community
(Brotto & Yule, 2009). Very little is known, on the other hand,
about the experiences of individuals who lack sexual attraction, but
who have not yet “come out” and adopted the label of “asexual.”
In fact, it is unknown what proportion of those individuals might
proceed on to self-identify as asexual. Some have speculated that
those individuals who report a lack of sexual attraction, but have
not yet discovered “asexuality” may be more isolated, distressed,
or confused than those who belong to the asexual community
(Brotto & Yule, 2009). Certainly, those belonging to AVEN may
be a distinct group within the asexual population, as they have
already acknowledged their asexuality as an identity, and they may
have different motivating factors to join such an online community
(e.g., distress) that make them different from those who are not
part of an online community (Brotto et al., 2010). Some AVEN
members also acknowledge that the recognition of an asexual
identity “enables asexuals to form a new self-image” (Radloff,
2008), suggesting that their membership might influence or change
the way that an individual lacking sexual attraction might respond
to research questions.
Methodologically speaking, research on asexuality should in-
clude both individuals from online asexual communities and those
who had not yet come across these online communities. However,
a serious limitation in past research on asexuality has been the
complete lack of a validated tool to identify asexuality. This is
problematic because, due to limitations in recruiting sufficiently
powered local samples, most studies have relied on recruiting via
online web-based asexual communities (e.g., AVEN; Brotto et al.,
2010; Brotto, Yule, & Gorzalka, 2014b; Prause & Graham, 2007),
and results in participants who already identify as being part of a
“community” (Brotto & Yule, 2009; Hinderliter, 2009).
A study by Prause and Graham (2007) provides a good illus-
tration of the problem of asexual identification. In that study,
individuals who selected “asexual” from a multiple choice sexual
orientation prompt did not consistently choose “asexual” in a
free-response question regarding sexual orientation. It has been
suggested that this group were likely not recruited from AVEN
(Hinderliter, 2009). On the other hand, those who did write asexual
to the free-response item likely were recruited through AVEN.
Based on this hypothesis, asexual activist Hinderliter (2009) has
made three suggestions; (a) those individuals who identify as
asexual in a free-response query as to their sexual orientation will
have spent more time thinking about whether or not they are
asexual, (b) asexual individuals recruited from AVEN will more
closely fit the definition of asexuality as defined by AVEN, and (c)
those recruited from online asexual communities will be strongly
influenced by online discourses of asexuality, and the categories
used to describe the experiences of asexual individuals. Because
asexuality research is still in its infancy, it is crucial that recruit-
ment be as broad as possible to ensure representativeness in the
sample and it should not be constrained by whether or not subjects
belong to a well-defined online community. The development of a
validated self-report measure of asexuality would facilitate the
acquisition of representative samples and would overcome previ-
ous shortcomings regarding participants having already self-
identified as asexual and/or belonging to a community.
The present study was aimed at developing and validating a
self-report questionnaire of asexuality, as defined by the research-
ers as an individual who lacks sexual attraction. While we con-
ceptualize (a)sexuality as a continuous construct (as has been
suggested elsewhere [McClave, 2013]), we aimed to create a
cut-off score to indicate whether or not a participant was asexual.
The questionnaire was intended to provide a valid measure inde-
pendent of whether the individual self-identified as asexual. Al-
though the measure may be used in future studies to explore
correlates of asexuality, this was not deemed to be part of the aim
of this study. Instead, our goal was to create a measure that could
reliably differentiate an asexual from a sexual individual.
Method
The questionnaire was developed in three Stages: Stage I—de-
velopment and administration of open-ended questions; Stage II—
administration and analysis of 111 items in order to determine
which items to retain and which to exclude; Stage III—adminis-
tration and analysis of 33 retained items, and validity analysis of
the final items. Participants were not restricted to taking part in
only one of the three stages, nor were participants in early stages
targeted to participate in later stages. Because participants in this
study identified with a wide range of gender identities (including
transgender), we did not analyze the findings according to tradi-
tional sex (e.g., male vs. female) categories given that several
participants did not identify their gender according to male versus
female. Further, due to the large number of different genders that
participants identified with (e.g., androgyne, neutrois, gender-
queer, pangendered, etc.) we did not conduct analyses by gender.
We therefore carried out analyses on the full group of participants
(independent of their reported sex or gender). All analyses were
carried out using IBM’s statistical software Statistical Package for
the Social Sciences (SPSS), version 17.0.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
2YULE, BROTTO, AND GORZALKA
Stage I: Development and Administration of
Open-Ended Items
The initial open-ended item pool was generated through discus-
sions between the first two authors. The intent of this initial stage
was to generate open-ended questions that might best discriminate
asexual from sexual individuals. The resulting eight open-ended
questions focused on definitions of asexuality, sexual attraction,
sexual desire, and romantic attraction, and included solicitations of
asexual individuals’ opinions on factors that initially led them to
identify as asexual, how they distinguish asexuality from low
sexual desire, how they might have described their sexuality before
they came across the term asexual, and what questions they would
use to identify an individual as asexual. There were no space
limitations, and participants were encouraged to answer in as much
or as little detail as they felt necessary. (See Table 1 for initial item
list).
Participants
Two samples were used to identify themes in response to the
eight open-ended questions.
Asexual sample. One hundred thirty-nine individuals who
self-identified as asexual were recruited from the AVEN website
during the months of August and September 2009 and presented
with the eight open-ended questions developed above in an online
survey. Age ranged from 18–56 years. The mean age was 27.3
(SD ⫽2.8) years. Thirty-six (26%) participants were male, 76
(55%) were female, two (1.4%) were transgendered, 14 (10%)
described their gender as being undefined or ‘other’ (e.g., andro-
gyne, gender neutral, genderqueer, etc.), and 11 (8%) gave no
response.
Sexual sample. To identify how responses differ between
sexual and asexual participants, a modified version of these eight
open-ended questions was administered to 70 sexual individuals
(who self-identified as heterosexual, bisexual or homosexual [gay
or lesbian]) via online recruitment through Craigslist and Face-
book. Age range of sexual participants was 18–75 years. The mean
age was 26.0 (SD ⫽1.3) years. Seventeen (24%) sexual partici-
pants were male, 51 (73%) were female, one (1%) was transgen-
dered, and one (1%) did not answer.
Responses to the open-ended questions were examined by us
(M.A.Y. and L.A.B.) to identify prevalent themes. There was
saturation in the themes with the current sample. This information
was used to generate 111 multiple choice items that made up the
Asexuality Identification Scale–111 (AIS-111). The items ad-
dressed 15 concepts, including: 1–Sexual Attraction/Desire (16
items assessing sexual attraction and sexual desire);
2–Masturbation (five items assessing desire to masturbate and
pleasure obtained from masturbation); 3–Sexual Fantasy (seven
items assessing presence and frequency of sexual fantasy);
4–Erotica (three items assessing enjoyment and arousal derived
from viewing erotic films); 5–Sex-Related Distress (15 items
assessing level of distress experienced in relation to the partici-
pant’s sexuality); 6–Sexual Activity (15 items assessing enjoy-
ment of sexual activity); 7–Sexual Identity (nine items assessing
how the sexuality of participants is related to their identity);
8–Sex-Related Disgust (four items assessing repulsion or disgust
of sexual activity); 9–Self-Reported Sexual Arousal (five items
assessing self-reported sexual arousal or sex drive); 10–Inability to
Relate to Others’ Sexuality (seven items); 11–Disinterest in Sex
(10 items); 12–Religion (two items assessing impact of religiosity
on sexuality); 13–Sexual Avoidance (two items); 14–Sex in Re-
lationships (six items assessing beliefs about sexuality within
relationships); and 15–Romantic Attraction and Intimacy (five
items assessing the beliefs about romantic and sexual attraction as
distinct constructs). Each item was scored on a 5-point Likert
scale, with higher scores reflecting experiences more typical of
asexual respondents than sexual ones. Several items dependent on
sexual activity contained an opt-out alternative response (i.e., “I
have never experienced sexual activity”). Concept names and item
numbers were retained throughout this process to ensure continuity
and to simplify discussion.
Stage II: Administration and Analysis of AIS-111
Administration of the AIS-111
Study design. The initial items in the AIS-111 developed
during Stage I were administered to 165 individuals who self-
identified as asexual and 752 sexual individuals (who self-
identified as heterosexual, bisexual, or homosexual [gay or les-
bian]) in an online survey. Participants were recruited through
several separate avenues, including postings on international web-
sites (e.g., Craigslist), on the AVEN online web-community,
through our university’s human subjects pool, and through web-
Table 1
Initial Open-Ended Item Pool
Item number Item
1 How would you define/describe asexuality?
2 How would you define/describe sexual attraction?
3 How would you define/describe sexual desire?
4 How would you define/describe romantic attraction?
5 What are some factors that initially led you to consider yourself as an asexual?/What are some factors that might lead you to
think that someone is asexual?
6 How would you distinguish asexuality from a sexual dysfunction such as sexual desire disorder?
7 How might you describe your sexuality BEFORE you came across the term “asexual”?/How might you describe your sexuality?
8 What questions would you use (without describing or using the term “asexual”) to identify an individual who might be asexual
but has not yet come across the term?
Note. Items modified for sexual participants are included where appropriate.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
3
ASEXUALITY IDENTIFICATION SCALE
sites that host psychological online studies (e.g., Hanover Col-
lege’s “Psychological Research on the Net” website).
Participants.
Asexual. Asexual participants ranged from 18 to 69 years old,
with a mean age of 26.6 (SD ⫽9.5) years. Eighty-two percent
noted that they were not in a relationship, and the majority (87%)
had at least some postsecondary education. Most participants re-
ported being Caucasian (85%). Thirty-one (19%) were male, 106
(64%) were female, 13 (8%) reported being transgender, and 15
(9%) described their gender as being undefined or “other” (e.g.,
androgyne, gender neutral, genderqueer, etc.).
Sexual. Sexual participants ranged from 18 to 80 years old,
with a mean age of 24.5 (SD ⫽8.8) years. Six hundred and 10
self-identified as heterosexual, 98 were bisexual, and 44 were gay
or lesbian. Thirty-nine percent reported being single and the ma-
jority (85%) had at least some postsecondary education. A large
proportion of participants reported being Caucasian (52%) or East
Asian (28%). Two hundred forty-seven (33%) were male, 495
(66%) were female, five (0.7%) reported being transgendered, and
five (0.7%) described their gender as being undefined or “other”
(e.g., androgyne, gender neutral, etc.).
Analysis of the AIS-111
Statistical analysis. Exploratory maximum-likelihood factor
analysis with direct oblimin rotation was conducted to determine
which items should be retained. This method of factor analysis was
selected as it is robust to violations of normality and allows factors
to correlate (Costello & Osborne, 2005) and is appropriate for
analyses that are theoretically grounded (Tabachnick & Fidell,
2001). In maximum-likelihood factor analysis (unlike principal
components analysis), unique variance and error variance are
excluded so that only shared variance is available for analysis, and
the total variance explained is less than in principal components
analysis (Costello & Osborne, 2005; Tabachnick & Fidell, 2001).
The minimum sample required for the factor analysis was 75, a
criterion that was more than satisfied with our sample of 917
participants. This figure is based on the rule-of-thumb that a ratio
of at least five times as many respondents should be used to the
number of items in the questionnaire for factor analytic purposes
(McDowell, 1996), which in this case is the 15 concepts listed
above. A priori criteria for domain and item retention were: (a)
Eigenvalues ⬎1.0, and (b) items with factor loadings at least 0.4.
A greater amount of information can be gained from questions
without an opt-out choice (e.g., those questions that can be an-
swered whether or not the participant has experienced masturba-
tion or other sexual activity). Therefore, most items that necessar-
ily included an opt-out response were removed at this point,
leaving 96 items. However, two items (Items 9 and 10), which
addressed the concepts of Masturbation and Erotica, necessarily
included an opt-out response, and thus were retained as these items
were deemed necessary to obtain an accurate measure of these two
concepts.
Domain structure. Item scores in each concept were aver-
aged, and an exploratory maximum-likelihood factor analysis with
direct oblimin rotation was used to assess the factor structure of the
15 concepts (96 items) for all participants. The technique of
averaging item scores used here is known as parceling, a recog-
nized technique for use with exploratory factor analysis. There are
several proposed advantages to this method (Little, Cunningham,
Shahar & Widaman, 2002). Disadvantages of using item-level data
include lower reliability, lower communality, a smaller ratio of
common-to-unique factor variance, and a greater likelihood of
distributional violations. Further, items have fewer, larger, and less
equal intervals between scale points than do parcels. According to
MacCallum, Widaman, Zhang, and Hong (1999) models based on
parceled data are more parsimonious, have fewer chances for
residuals to be correlated or for dual loadings to emerge, and lead
to reductions in various sources of sampling error. It may be that
subsets of items will share specific sources of variance, and that
these are unlikely to be hypothesized by the researcher a priori,
which can lead to systematic sources of common variance. It is
thought that parceling a large number of items into smaller groups
can reduce or even eliminate the number of unwanted sources of
variance and lead to a better model fit than if all items were used.
Parceling is thought to be particularly effective when the construct
is unidimensional, which we believe asexuality to be.
A three-factor solution was produced using the 15 concepts
outlined above, and labeled: Traits of Asexuality, Sex-Related
Distress, and Masturbation. These three factors together explained
74% of the variance in the model.
Further analysis of the items and their relationship to each other
indicated that a one-factor solution may be more appropriate.
Factor 1 (Traits of Asexuality) had 12 concepts that loaded above
0.40, an eigenvalue of 8.69, and accounted for 58% of the variance
in the model. Factor 2 (Sex-Related Distress) had one concept that
loaded above 0.40, an eigenvalue of 1.33, and accounted for 9% of
the variance. Factor 3 (Masturbation) also had a single loading
concept, an eigenvalue of 1.10, and accounted for 7% of the
variance in the model. In addition to the rule of thumb that a factor
with three or fewer concepts with a loading greater than 0.40 is
generally weak and unstable (Costello & Osborne, 2005), the three
factors did not correlate with one another, further suggesting that
the instrument may represent a single factor. Additionally, the
scree test (Costello & Osborne, 2005) suggested that a single factor
should be retained. Further, mean Sex-Related Distress scores for
sexual and asexual participants were highly similar to one another
(sexual participants: M⫽3.93, SD ⫽0.73; asexual participants:
M⫽4.09, SD ⫽0.78, t(912) ⫽⫺2.53, p⫽.012, Cohen’s d⫽
.21). While this finding was significant, the effect size is small. To
put this in context, it can be compared to the highly significant
difference between mean scores for Concept 1 (Sexual Attraction/
Desire), for which sexual participants scored 1.91 (SD 0.56), and
asexual participants scored 3.88 (SD 0.44), t(911) ⫽⫺42.13, p⬍
.001, Cohen’s d⫽5.51, which can be considered a very large
effect size.
This small difference between mean distress scores may be
explained by the observation that asexual individuals do not gen-
erally experience distress in relation to their asexuality (Brotto et
al., 2010). Lack of distress is fundamental to the conceptualization
of asexuality. Thus, it makes theoretical sense that items in this
concept would be incapable of distinguishing between sexual and
asexual participants. Since the goal of this instrument was to allow
for distinction between asexual and sexual individuals, we elimi-
nated all Sex-Related Distress items. The concept of Masturbation
was retained at this point, as the data on whether masturbation
rates differ between asexual and sexual individuals is equivocal
(Brotto et al., 2010). Thus, we wanted to confirm that this concept
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
4YULE, BROTTO, AND GORZALKA
was not useful in differentiating between asexual and sexual par-
ticipants in subsequent analyses. Further, mean Masturbation
scores for sexual and asexual participants were significantly dif-
ferent (sexual participants: M⫽2.61, SD ⫽1.17; asexual partic-
ipants: M⫽3.55, SD ⫽1.23, t(910) ⫽⫺9.277, p⬍.001, Cohen’s
d⫽.78), supporting our decision to retain the Masturbation
concept at this stage. Concept 12 (Religion) did not load onto any
factor and was thus removed from further analyses.
We next examined the reliability (internal consistency) of each
concept, as well as how well each individual item contributed to
each concept. Reliability of each concept was clearly demon-
strated. Most concepts, with the exception of Self-Reported Sexual
Arousal, had good reliability (␣⬎.50), with eight concepts
demonstrating a high reliability of ␣⬎.80. For most concepts, the
majority of their items contributed strongly to them, with only
Concept 6 (Sexual Activity) having less than half of the items
contribute. Overall reliability of the AIS-111 was very high (␣⬎
.98).
Individual items were selected from each concept based on how
well they contributed to the concept’s reliability. Further, for item
pairs that were very closely worded or very similar in meaning,
one of the items was removed to increase brevity of the question-
naire. The resulting intermediate version of the Asexuality Iden-
tification Scale (AIS-37) retained 13 concepts and contained 37
items: Sexual Attraction/Desire (Items 5, 35, 36, 74, 82, 88, 94),
Masturbation (Items 10 and 40), Fantasy (Items 20 and 103),
Erotica (Items 9 and 104), Sexual Activity (Items 17, 27, 66, 89,
95), Sexual Identity (Items 50 and 87), Sex-Related Disgust (Items
18 and 51), Self-Reported Sexual Arousal (Items 34 and 38),
Inability to Relate to Others’ Sexuality (Items 55, 90, 98), Sexual
Interest (Items 14, 19, 46, 93), Sexual Avoidance (Items 57 and
100), Relationships (Items 28 and 56), and Romantic Attraction
and Intimacy (Items 2 and 110).
Stage III: Administration and Analysis of the AIS-37
Study Design
The 37 intermediate items (AIS-37) developed during Stage II
were administered to 316 individuals who self-identified as asex-
ual, and 926 sexual individuals who self-identified as heterosexual,
bisexual, or gay or lesbian in an online survey. Participants were
recruited through several avenues, including postings on interna-
tional websites (e.g., Craigslist), on the AVEN online web-
community, through the university’s human subject pool, and
through websites that host psychological online studies (e.g., Ha-
nover College’s “Psychological Research on the Net” website).
The current sample was comparable to the sample assessed in
Stage II above with respect to proportion of participants recruited
from each of these sources.
Participants
Asexual. Asexual participants ranged from 18 to 72 years old,
with a mean age of 24.8 (SD ⫽7.6) years. Fourteen percent noted
that they were in a relationship, either committed or noncommit-
ted, and the majority (88%) had at least some postsecondary
education. Most participants reported being Caucasian (88%).
Fifty-four (17%) participants were male, 206 (65%) were female,
10 (3%) were transgendered, and 46 (15%) described their gender
as being undefined or “other” (e.g., androgyne, gender neutral,
genderqueer, etc.).
Sexual. Sexual participants ranged from 18 to 69 years old,
with a mean age of 25.0 (SD ⫽8.3) years. Six hundred and 75
self-identified as heterosexual, 130 were bisexual, and 121 were
gay or lesbian. Fifty-four percent reported being in a relationship
and the majority (84%) had at least some postsecondary education.
A large proportion reported being either Caucasian (54%) or East
Asian (27%). Two hundred forty-one (26%) were male, 672 (73%)
were female, six (0.6%) were transgendered, and seven (0.8%)
described their gender as being undefined or “other” (e.g., andro-
gyne, gender neutral, genderqueer, etc.).
Measures
All questions pertaining to sex, gender, and sexual orientation
were in a free response format given the finding that asexual
subjects respond differently when queried using free response
versus forced choice formats (Brotto et al., 2010; Prause & Gra-
ham, 2007). Those questions pertaining to ethnicity, education,
relationship status, and masturbation frequency, had forced-choice
response options.
Statistical Analysis
Defining domains and item retention summary. As in Stage
II, exploratory maximum-likelihood factor analysis with direct
oblimin rotation was conducted to determine which items from the
AIS-37 should be retained, and which should be discarded. The
minimum sample for the planned factor analysis was 65 asexual
participants based on an ideal 5:1 respondent-to-item ratio (which
was the case for the 13 concepts listed above) (McDowell, 1996).
As in Stage II, a priori criteria for domain and item retention were:
(a) Eigenvalues ⬎1.0, and (b) items with factor loadings at least
0.4.
Domain structure. Exploratory maximum-likelihood factor
analysis with direct oblimin rotation was used to assess the factor
structure of the 13 concepts of the AIS-37 for all participants. A
two-factor solution was produced using the 13 concepts outlined
above. Factor 1 had eight loading concepts, and was labeled
Interpersonal Sexuality. Factor 2 contained four loading concepts
and was labeled Intrapersonal Sexuality. These two factors to-
gether explained 72% of the variance in the model (see Table 2).
Again, concept names and numbers were retained throughout this
process to ensure continuity and to simplify discussion. The Inter-
personal Factor of the AIS-37 consisted of Concept 1 (Sexual
Attraction/Desire), Concept 6 (Sexual Activity), Concept 7 (Sex-
ual Identity), Concept 8 (Sex-Related Disgust), Concept 10 (In-
ability to Relate), Concept 11 (Disinterest in Sex), Concept 13
(Sexual Avoidance), and Concept 14 (Relationships), all of which
pertained to sexuality in relation to another individual. The Intrap-
ersonal Factor, on the other hand, consisted of Concept 2 (Mas-
turbation), Concept 3 (Fantasy), Concept 4 (Erotica), and Concept
9 (Self-Reported Sexual Arousal), all of which, it might be argued,
are not dependent on presence of a partner. Previous research has
found variance in interpersonal behavior (e.g., petting, sexual
intercourse) to be independent of variance in intrapersonal behav-
ior (e.g., fantasy, masturbation) (Meston, Trapnell, & Gorzalka,
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
5
ASEXUALITY IDENTIFICATION SCALE
1996), which supports our findings here. Concept 15 (Romantic
Attraction and Intimacy) did not load onto either factor and was
thus removed from further analyses.
Further analysis of the concepts and their relationship to each
other indicated that, again, a one-factor solution is more appropri-
ate than a two-factor model. Factor 2 (Intrapersonal) had an
eigenvalue very close to 1, and accounted for a relatively small
amount of variance, indicating that this factor should be elimi-
nated. Additionally, the scree test (Costello & Osborne, 2005)
suggested that a single factor should be retained. Due to evidence
that asexual individuals experience genital sexual arousal (Brotto
& Yule, 2011), that they experience sexual fantasies (Yule, Brotto,
& Gorzalka, 2014b; although these are often not focused on other
people), and that they have similar rates of masturbation to their
sexual counterparts (Brotto et al., 2010), we would argue that the
intrapersonal aspects of sexuality, which map onto Factor 2 in this
model, are not unique to asexuality—the latter being conceptual-
ized as a lack of attraction toward others, with perhaps, minimal to
no impact on self-sexual behavior. Based on this evidence, as well
as the results of our factor analysis, we excluded those items that
made up the concepts (2, 3, 4 and 9) that mapped onto Factor 2
from further analyses.
Once it had been determined that the AIS-37 consisted of a
single factor we examined the reliability (internal consistency) of
the remaining concepts (1, 6, 7, 8, 10, 11, 13, and 14), as well as
how well each individual item contributed to each concept. Reli-
ability of each concept was clearly demonstrated (see Table 3).
Most concepts, with the exception of Sex-Related Disgust, Inabil-
ity to Relate, and Sexual Avoidance, had very high reliability (␣⬎
.90). Many concepts had the majority of their items contribute
strongly to them, with only Concept 6 (Activity) having less than
half of the items contribute.
Item retention for the 12-item AIS. For those concepts that
consisted of two items only (and thus had identical correlations
with the total for each item), mean differences were calculated to
determine which item more strongly differentiated asexual from
sexual participants (e.g., for the Sexual Avoidance concept, there
was a larger difference between mean scores for asexual and
sexual participants for item 38 than for item 4, and therefore item
38 was retained and item 4 discarded). Items containing an opt-out
alternative were removed whenever there was a choice between an
opt-out item and a non-opt-out item. (See the Appendix for the
final items included in the AIS-12).
Psychometric analyses then focused on the validity of the
AIS-12 total score. Analyses were based on the AIS-12 items,
which were completed by the participants in Stage II (i.e., those
who completed the AIS-111), as the validity questionnaires were
not included in the final stages of the study.
1
In order to demonstrate that the questionnaire items on the
AIS-12 were measuring the construct of asexuality, a series of
validation techniques were conducted as follows:
1. Known-groups validity. If two groups are known to differ
on a given condition, then one would expect differing
results on a questionnaire that captures the condition. In
this instance, we compared AIS-12 scores between indi-
viduals who do and do not self-identify as asexual, and
performed ttests to determine whether any differences
were statistically significant.
One hundred sixty-four asexual participants and 716 sex-
ual participants completed the AIS-12. Levene’s test re-
vealed that the sample variances were not equal, thus
necessitating the use of an independent samples ttest in
which equal variances was not assumed. The AIS-12
1
We feel that this is a reasonable estimation of the AIS-12’s validity,
and acknowledge that these analyses should be repeated in a new sample
using only the items on the AIS-12.
Table 2
Maximum Likelihood Analysis With Direct Oblimin Rotation of Concepts of the Asexuality
Identification Survey—37: Factor Loadings
ⴱ
Concept Number of
items
Factor 1
(Interpersonal
sexuality)
Factor 2
(Intrapersonal
sexuality)
1. Sexual Attraction/Desire 7 .810 .151
2. Masturbation 2 ⫺.089 .760
3. Sexual Fantasy 2 .164 .733
4. Erotica 2 .371 .477
6. Sexual Activity 5 .968 ⫺.001
7. Sexual Identity 2 .941 ⫺.156
8. Sex-Related Disgust 2 .799 ⫺.005
9. Self-Reported Sexual Arousal 2 .212 .695
10. Inability to Relate to Others’ Sexuality 3 .747 .127
11. Sexual Interest 4 .971 ⫺.029
13. Sexual Avoidance 2 .610 .128
14. Relationships 2 .940 ⫺.029
15. Romantic Attraction and Intimacy 2 .241 .066
Relative explanatory power of individual
factors (Eigenvalue) 8.226 1.124
Percent variance accounted for by
individual factors 63.27% 8.65%
ⴱ
Items with the highest loadings within each factor are in bold type.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
6YULE, BROTTO, AND GORZALKA
Total score differed significantly between sexual and
asexual groups, t(312.8) ⫽49.60, p⬍.001.
2. Convergent validity. To assess whether the questionnaire
correlates a construct thought to overlap with asexuality
(i.e., sexual desire), summed AIS-12 Total scores were
correlated with Dyadic and Solitary scores on the Sexual
Desire Inventory (SDI; Spector, Carey & Steinberg,
1996), a measure that captures the cognitive (rather than
behavioral) aspect of sexual desire. Many of the items on
the SDI refer to desire in response to an “attractive
person,” or otherwise assume the presence of sexual
attraction. Because lack of sexual attraction would very
likely lead to a lack of sexual desire, we would expect a
low score on the SDI (indicating low desire) to corre-
spond with a high score on the AIS-12 for asexual indi-
viduals. Specifically, we predicted a negative association
between the Dyadic, but not the Solitary subscale of the
SDI and the AIS-12 given that items pertaining to intrap-
ersonal sexuality were removed from the AIS. For sexual
individuals, we would also expect negative correlations
between the AIS-12 and the SDI, for the opposite reason;
we would expect sexual individuals to obtain low scores
on the AIS-12, and high scores on the SDI.
Convergent validity was assessed by correlating total
scores of the AIS-12 with scores on the Solitary and
Dyadic subscales of the Sexual Desire Inventory. As
Table 3
AIS-37: Individual Items and Concept Reliability
ⴱ
Concept
Standardized
Cronbach’s
Alpha Item number Correlation
with total
Cronbach’s
Alpha if
item deleted
1. Sexual Attraction/Desire .925 5–I know what it is like to experience sexual attraction .831 .907
35–I experience sexual attraction towards other people .907 .898
36–Compared to most people I know, I would describe
my level of sexual attraction to others as being .767 .915
74–I have lost the feeling of sexual desire towards others .379 .947
82–I find other people to be sexually appealing .808 .909
88–I find myself experiencing sexual attraction
towards another person .867 .905
94–I have never felt drawn toward another person
sexually .837 .906
6. Sexual Activity .941 17–I would prefer to be sexually inactive .837 .928
27–I lack interest in sexual activity .850 .925
66–I find sexual activity pleasant .825 .931
89–I have never wanted to engage in sexual activity with
another person .809 .932
95–I would be relieved if I was told that I never had to
engage in any sort of sexual activity again .890 .918
7. Sexual Identity .703 50–I don’t feel that I fit the conventional categories of
sexual orientations such as heterosexual, homosexual,
or bisexual
.542 n/a
87–The term “nonsexual” would be an accurate
description of my sexuality .542 n/a
8. Sex-Related Disgust .921 18–I am repulsed by sex .854 n/a
51–The thought of sexual activity repulses me .854 n/a
10. Inability to Relate .776 55–I find it hard to understand or relate to terms such as
“hot” or “sexy” .622 .688
90–I am confused by how much interest and time other
people put into sexual relationships .639 .669
98–I feel that other people think about sex much more
than I do .579 .734
11. Disinterest in Sex .919 14–I have always found sex boring .792 .899
19–I find sex interesting .753 .912
46–I would be content if I never had sex again .852 .880
93–Sex has no place in my life .864 .874
13. Sexual Avoidance .788 57–I go to great lengths to avoid situations where sex
might be expected of me .651 n/a
100–I try to avoid situations in which I might be
approached for sex .651 n/a
14. Relationships .926 28–Given the choice between being in a sexual
relationship and not being in a sexual relationship, I
would prefer to NOT be in a sexual relationship
.861 n/a
56–My ideal relationship would not involve sexual
activity .861 n/a
Note. n/a Reliability statistic cannot be calculated as there is only one item for this concept.
ⴱ
Items included in final AIS-12 are bolded.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
7
ASEXUALITY IDENTIFICATION SCALE
expected, for asexual individuals, the SDI Solitary sub-
scale correlated only weakly (␣⫽⫺.19) with total scores
on the AIS-12. The SDI Dyadic subscale did have a
moderate negative correlation with the AIS-12 Total
score (␣⫽⫺.57). Sexual individuals also had weak
negative correlations between the AIS-12 and the SDI
Solitary Subscale (␣⫽⫺.31), and moderate negative
correlations for the SDI Dyadic Subscale (␣⫽⫺.76).
3. Approximated incremental validity. In order to determine
whether the instrument is a more accurate measure of
asexuality than existing measures of sexual orientation,
total scores on the AIS-12 were compared with partici-
pant scores on the Klein Scale (Klein, Sepekoff, & Wolf,
1985), which was adapted to allow it to include asexu-
ality as a sexual orientation. While comparison to an
adapted measure may not be a true estimate of incremen-
tal validity, we wished to demonstrate that the AIS-12
could assess asexuality over and above an easily adapted
existing measure. The original Klein scale requires par-
ticipants to select their “choice partner or associate” (in
the past, in the present and their ideal partner) on a
number of measures, including sexual attraction, sexual
behavior, sexual fantasies, emotional preference and so-
cial preference. Participants are given seven choices,
ranging from “other sex only” to “same sex only.” An
additional category was added to allow the selection of
“neither sex” for these items. A second section of the
Klein Scale asks participants about their “identity or
lifestyle” (again past, present, and ideal), including life-
style preference, social identity and political identity.
Participants are asked to select their choice of seven
items, ranging from “heterosexual only” to “homosexual
only.” In order to make these items accessible to asexual
participants, we added options allowing participants to
select an asexual identity for these three items.
Approximated incremental validity was good, as the
AIS-12 correlated only weakly with each subscale and
the total score of the Klein scale.
4. Discriminant validity. The Childhood Trauma Question-
naire (CTQ; Bernstein et al., 1994; Bernstein & Fink,
1998) was included to ensure that endorsement of items
that may be construed to indicate asexuality are not,
instead, an indicator of negative sexual experiences. The
CTQ provides a brief, reliable, and valid assessment of a
broad range of potentially traumatic experiences experi-
enced in childhood. Subscales on the CTQ include: Emo-
tional Abuse (EA), Physical Abuse (PA), Emotional Ne-
glect (EN), Sexual Abuse (SA), and Physical Neglect
(PN). Higher scores on the CTQ indicate higher rates of
abuse. We predicted that Total scores on the AIS-12
would not be strongly correlated with CTQ scores. Fur-
ther, interpersonal relationships were measured using the
Short-Form Inventory of Interpersonal Problems-
Circumplex scales (IIP-SC; Soldz, Budman, Demby, &
Merry, 1995), a 32-item questionnaire that assesses var-
ious interpersonal problems and ways of relating to oth-
ers (e.g., socially avoidant, vindictive, exploitative). The
Big-Five Inventory (BFI; John, Donahue, & Kentle,
1991; John, Naumann, & Soto, 2008; John & Srivastava,
1999) was also included to assess personality traits. The
BFI is a 44-item self-report questionnaire designed to
measure the Big Five personality traits; openness, con-
scientiousness, extraversion, agreeableness, and neuroti-
cism. Both the IIP-SC and the BFI were included to
determine whether our measure identified asexuality
above and beyond basic interpersonal and personality
traits.
Discriminant validity was as expected, as scores on the
AIS-12 were not significantly correlated with the Child-
hood Trauma Questionnaire for asexual individuals. For
sexual participants, the CTQ scores were weakly corre-
lated with the AIS-12 total score. Similarly, there were no
significant moderate or strong correlations with the
AIS-12 and the IIP-SC, or the Big-Five Inventory for
asexual or sexual individuals (see Table 4).
Cut-off scores for the AIS-12. A score of 40/60 on the
AIS-12 was found to capture 93% of individuals who self-
identified as asexual. That is, 93% of asexual participants scored at
or above 40 on the AIS-12, while 95% of self-identified sexual
participants scored below 40.
Discussion
The objective of the present study was to develop a valid and
reliable sex- and gender-neutral self-report measure of asexuality.
The AIS-12 (hereafter referred to as the AIS
2
) was designed to be
used as a valid and reliable tool to distinguish asexual individuals
from sexual individuals for research purposes, and is considered
necessary for researchers to obtain more representative samples
than those that have been studied to date, most of whom had been
recruited through self-identification as asexual (Hinderliter, 2009).
The AIS was developed in a series of stages, including devel-
opment and administration of open-ended questions to sexual and
asexual individuals, development of initial multiple choice items,
and analysis of these items to facilitate selection of final items.
Based on discriminant analytic methods, 12 items were retained in
the final AIS. Sexual and asexual participants significantly differed
in their AIS total score with high statistical significance. Further,
a cut-off score of 40/60 was found to identify 93% of self-
identified asexual individuals, while excluding 95% of sexual
individuals. This suggests that the AIS is a useful tool for identi-
fying asexuality, and could be used in future research to identify
individuals with a lack of sexual attraction. We believe that the
AIS does not depend on one’s self-identification as asexual; rather,
it would also capture the individual who scores high on the AIS but
has not yet identified him/herself as asexual.
Psychometric validation of the 12-item AIS was conducted for
construct validity. Specifically, the total score of the AIS showed
excellent ability to distinguish between asexual and sexual subjects
2
AIS is a most appropriate name given that asexuals euphemistically
refer to themselves as aces.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
8YULE, BROTTO, AND GORZALKA
(discriminant validity), while showing only moderate correlations
with an adaptation of a previously established measure of sexual
orientation (approximated incremental validity). When compared
to an existing measure of sexual desire, the AIS was found to
correlate highly with scores on the Dyadic subscale of the
Sexual Desire Inventory, but not with Solitary sexual desire
scores. This finding demonstrates the previously recognized
and considerable overlap between the constructs of sexual at-
traction and sexual desire. The strong correlation between the
SDI Dyadic subscale and the AIS also confirms previous re-
search showing that asexual individuals score significantly
lower on sexual desire measures than sexual samples (Brotto et
al., 2010), but do not differ from others on solitary sexual desire
(Brotto & Yule, 2011).
Because of the difficulty disentangling sexual attraction and
desire, and the assumption of some level of sexual attraction
implied by many items on the SDI, we would expect low scores on
the SDI (indicating low sexual desire) to be correlated with high
scores on the AIS (indicating asexuality) if the AIS is indeed
assessing a lack of sexual attraction. To what extent does the AIS
add value as a measure of asexuality above and beyond that
captured by the SDI (or other measures of low desire for that
matter)? The items on the SDI focus on desire for sexual behavior,
rather than on level of sexual attraction. This may make the AIS
more accessible than the SDI for individuals who lack sexual
attraction. Further, items on the SDI are worded such that they
presume an existing sexual attraction (i.e., “When you first see an
attractive person [. . .]”). Also, a number of items address solitary
sexual desire, which is not relevant to assessing asexuality because
of the finding that patterns of solitary sex do not differ between
asexual and nonasexual individuals (Brotto et al., 2010; Prause &
Graham, 2007).
Table 4
Asexuality Identification Scale-12 Validity
Asexual participants
only Sexual participants
only
Known groups validity
AIS-Total score, Mean (SD)
ⴱⴱⴱ
51.45 (6.59) 21.14 (8.83)
Convergent validity (Pearson correlations)
Sexual Desire Inventory:
Solitary ⫺.190
ⴱ
⫺.308
ⴱⴱ
Dyadic ⫺.567
ⴱⴱ
⫺.757
ⴱⴱ
Approximated incremental validity
(Pearson correlations)
Klein Scale subscales:
Attraction ⫺.185
ⴱ
⫺.034
Behavior ⫺.253
ⴱⴱ
⫺.116
ⴱⴱ
Fantasy ⫺.237
ⴱⴱ
⫺.053
Emotional ⫺.056 ⫺.046
Social .122 ⫺.065
Lifestyle ⫺.133 ⫺.159
ⴱⴱ
Identity ⫺.163
ⴱ
⫺.161
ⴱⴱ
Political ⫺.169
ⴱ
⫺.168
ⴱⴱ
Total Score ⫺.240
ⴱⴱ
⫺.131
ⴱⴱ
Discriminant validity (Pearson correlations)
Childhood Trauma Questionnaire:
Emotional abuse ⫺.032 .147
ⴱⴱ
Physical abuse ⫺.078 .135
ⴱⴱ
Emotional neglect ⫺.125 .205
ⴱⴱ
Sexual abuse ⫺.155 .189
ⴱⴱ
Physical neglect ⫺.016 .143
ⴱⴱ
Total score ⫺.100 .213
ⴱⴱ
Short-Form Inventory of Interpersonal
Problems-Circumplex Version:
Domineering .003 .120
ⴱⴱ
Vindictive .142 .195
ⴱⴱ
Cold .185
ⴱ
.280
ⴱⴱ
Socially avoidant .186
ⴱ
.203
ⴱⴱ
Exploitive .098 .135
ⴱⴱ
Nonassertive .163
ⴱ
.120
ⴱⴱ
Overly nurturant ⫺.001 .159
ⴱⴱ
Intrusive ⫺.067 .096
ⴱ
Total score .176
ⴱ
.245
ⴱⴱ
Big Five Inventory:
Extraversion ⫺.219
ⴱⴱ
⫺.172
ⴱⴱ
Agreeableness ⫺.061 ⫺.165
ⴱⴱ
Consciousness .100 ⫺.072
Openness .018 ⫺.155
ⴱⴱ
Neuroticism .004 .171
ⴱⴱ
ⴱ
p⬍.05.
ⴱⴱ
p⬍.01.
ⴱⴱⴱ
p⬍.001.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
9
ASEXUALITY IDENTIFICATION SCALE
There has been previous speculation that asexual individuals
may not experience sexual attraction due to a history of childhood
trauma or sexual abuse (Brotto et al., 2010; Jay, 2008). Our finding
that AIS total scores were not correlated with scores on the
Childhood Trauma Questionnaire provides evidence that the AIS
is not tapping into an experience of childhood trauma.
This questionnaire was developed with the sole intent of differ-
entiating asexual individuals from sexual individuals, and not to
provide any information about the phenomenon of asexuality
itself. One inherent challenge in developing such a questionnaire is
the need to avoid assuming a strict gender binary (Hinderliter,
2009). In fact, asexual individuals may reject identifying as either
male or female in favor of more ambiguous (and inclusive) terms
such as agendered,genderqueer,orpan-asexual gender-free to
label their gender (Brotto et al., 2010). We believe that the result-
ing 12-item measure is, indeed, sex- and gender-identity neutral,
and thus will be desirable for use. Furthermore, items are worded
clearly and the AIS is simple to administer and score. While we
view (a)sexuality as likely occurring on a dimensional spectrum, it
was necessary to create a cut-off score that allowed researchers to
categorize participants as “asexual” or “sexual,” and this catego-
rization may be in conflict with the participant’s identity. The
purpose of this questionnaire is not to question an individual’s
identity, but allow the recruitment of representative samples of
individuals who lack sexual attraction despite how they might
identify.
Limitations
There is a possibility that some participants in this study who
identified as “sexual” actually lack sexual attraction, and would
thus be better categorized as “asexual” in research. Given the
relative recency with which the term asexuality has been available,
it maybe that an asexual is misclassifying him/herself as sexual
despite having a high AIS score. Instead, such an individual
might select heterosexual, homosexual (gay or lesbian), or bi-
sexual (rather than asexual) in response to a query about their
sexual orientation. This is problematic, as it means that our
asexual category does not include those participants who have not
heard of asexuality, which could influence our understanding of
asexuality. In fact, as can be seen in Figure 1, there is a subgroup
of sexual participants in this study (4%) who received a score
greater than 40 on the AIS, and approximately 1% who scored
above 50. While this is a limitation of our study, it could be
argued that this confirms that the AIS is functioning as it was
designed and is identifying those participants who might be
better classified as asexual instead of sexual. Further, because
this questionnaire was developed on a sample of well-defined
self-identified asexual individuals, who are potentially highly
asexual, it is possible that the AIS may have greater sensitivity at
higher levels of asexuality, and decreased sensitivity at lower
levels. Follow-up research will focus on further development of
this scale in large population-based studies to ensure sensitivity
across a broader range of individuals who lack sexual attraction.
Future research will also attempt to determine sensitivity and
specificity of individual items on the AIS.
Another important limitation that should be considered is that
the concept of asexuality is relatively new, and thus the construct
is still poorly understood (Brotto et al., 2010). While the current
and most widely used definition of asexuality is, as noted through-
out this manuscript, an individual who does not experience sexual
attraction, there appears to be a range of experiences within this
group. For example, a sizable number of self-identifying asexual
individuals report experiencing low levels of sexual attraction,
rather than no sexual attraction at all (Carrigan, 2011). This is
reflective of the large amount of diversity present among the
asexual community, and is common enough that the term
“Gray-A” is used to refer to those who fall in the “gray area”
between sexual and asexual. While AVEN describes an asexual
individual as being “someone who does not experience sexual
attraction,” this definition is seen as a blanket term among asexual
individuals themselves (Carrigan, 2011). Several asexual individ-
uals in Scherrer’s (2008) study based their identification as asexual
not on a lack of sexual attraction, but on a lack of intent to engage
in sexual behavior. As Scherrer (2008, p. 626) points out, “as with
other sexual identities, the meaning of an asexual identity varies.”
The participants in this study who self-identified as asexual likely
represent this variability.
The psychometric analyses used to assess validity in Stage III
were conducted on the items that made up the AIS-12 as com-
pleted by the participants in Stage II (i.e., those who completed the
AIS-111), as the validity questionnaires were not included in the
final stage of the study. While we believe this to be a reasonable
0
2
4
6
8
10
12
1 3 5 7 9 11131517192123252729313335373941434547495153555759
Percentage of respondents who received this score
AIS-12 Score
Asexual (%)
Sexual (%)
Figure 1. Distribution of scores on the AIS-12 for asexual and sexual participants.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
10 YULE, BROTTO, AND GORZALKA
approximation of the AIS’ validity, these analyses should be con-
ducted on a new sample using only the items on the final AIS.
Conclusion
The AIS, a 12-item questionnaire, has been developed as a brief,
valid, and reliable self-report instrument for assessing asexuality.
It is psychometrically sound, easy to administer, and has demon-
strated ability to discriminate between sexual and asexual individ-
uals. It should prove useful to allow researchers to cast a wider net
when recruiting individuals who lack sexual attraction, and may
therefore lead to more representative samples of the asexual pop-
ulation, allowing us to further increase our understanding of asex-
uality. We hope that researchers interested in the construct of
asexuality consider use of the AIS.
References
Asexuality Visibility and Education Network Wiki. (2013). Attraction.
Retrieved from http://www.asexuality.org/wiki/index.php?title⫽
Romantic_attraction#Romantic_Attraction
Berkey, B. R., Perelman-Hall, T., & Kurdek, L. A. (1990). The multidi-
mensional scale of sexuality. Journal of Homosexuality, 19, 67–88.
http://dx.doi.org/10.1300/J082v19n04_05
Bernstein, D., & Fink, L. (1998). Childhood Trauma Questionnaire: A
retrospective self-report. San Antonio, TX: The Psychological Corpo-
ration.
Bernstein, D. P., Fink, L., Handelsman, L., Foote, J., Lovejoy, M., Wenzel,
K.,...Ruggiero, J. (1994). Initial reliability and validity of a new
retrospective measure of child abuse and neglect. The American Journal
of Psychotherapy, 151, 1132–1136.
Bogaert, A. F. (2004). Asexuality: Prevalence and associated factors in a
national probability sample. Journal of Sex Research, 41, 279–287.
http://dx.doi.org/10.1080/00224490409552235
Bogaert, A. F. (2006). Toward a conceptual understanding of asexuality.
Review of General Psychology, 10, 241–250. http://dx.doi.org/10.1037/
1089-2680.10.3.241
Brotto, L. A. (2013, April). Lack of desire or lack of attraction? Asexuality,
sexual desire disorder, and the spaces between. Paper presented at the
annual meeting of the Society for Sex Therapy and Research, Baltimore,
MD.
Brotto, L. A., Knudson, G., Inskip, J., Rhodes, K., & Erskine, Y. (2010).
Asexuality: A mixed-methods approach. Archives of Sexual Behavior,
39, 599–618. http://dx.doi.org/10.1007/s10508-008-9434-x
Brotto, L. A., & Yule, M. A. (2009). Reply to Hinderliter (2009). Archives
of Sexual Behavior, 38, 622–623. http://dx.doi.org/10.1007/s10508-009-
9514-6
Brotto, L. A., & Yule, M. A. (2011). Physiological and subjective sexual
arousal in self-identified asexual women. Archives of Sexual Behavior,
40, 699–712. http://dx.doi.org/10.1007/s10508-010-9671-7
Carrigan, M. (2011). There’s more to life than sex? Difference and com-
monality within the asexual community. Sexualities, 14, 462–478. http://
dx.doi.org/10.1177/1363460711406462
Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory
factor analysis: Four recommendations for getting the most from your
analysis. Practical Assessment, Research & Evaluation, 10(7). Retrieved
from http://pareonline.net/getvn.asp?v⫽10&n⫽7
Hinderliter, A. C. (2009). Methodological issues for studying asexuality.
Archives of Sexual Behavior, 38, 619–621. http://dx.doi.org/10.1007/
s10508-009-9502-x
Jay, D. (2008). Asexuality Visibility and Education Network. Retrieved
from http://www. asexuality.org/home/overview.html
John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five
Inventory–Versions 4a and 54. Berkeley, CA: University of California,
Berkeley, Institute of Personality and Social Research.
John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift to the
integrative Big-Five trait taxonomy: History, measurement, and concep-
tual issues. In O. P. John, R. W. Robins, & L. A. Pervin (Eds.),
Handbook of personality: Theory and research (3rd ed., pp. 114–158).
New York, NY: Guilford Press.
John, O. P., & Srivastava, S. (1999). The Big Five trait taxonomy: History,
measurement, and theoretical perspectives. In L. Pervin & O. P. John
(Eds.), Handbook of personality: Theory and research (2nd ed., pp.
102–138). New York, NY: Guilford Press.
Kinsey, A. C., Pomeroy, W. R., & Martin, C. E. (1948). Sexual behavior
in the human male. Philadelphia, PA: W. B. Saunders.
Klein, F., Sepekoff, B., & Wolf, T. J. (1985). Sexual orientation: A
multi-variable dynamic process. Journal of Homosexuality, 11, 35–49.
http://dx.doi.org/10.1300/J082v11n01_04
Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002).
To parcel or not to parcel: Exploring the question, weighing the merits.
Structural Equation Modeling, 9, 151–173. http://dx.doi.org/10.1207/
S15328007SEM0902_1
Lucassen, M. F. G., Merry, S. N., Robinson, E. M., Denny, S., Clark, T.,
Ameratunga, S.,...Rossen, F. V. (2011). Sexual attraction, depression,
self-harm, suicidality and help-seeking behaviour in New Zealand sec-
ondary school students. Australian & New Zealand Journal of Psychi-
atry, 45, 376–383. http://dx.doi.org/10.3109/00048674.2011.559635
MacCallum, R. C., Widaman, K. F., Zhang, S., & Hong, S. (1999). Sample
size in factor analysis. Psychological Methods, 4, 84–99. http://dx.doi
.org/10.1037/1082-989X.4.1.84
MacInnis, C. C., & Hodson, G. (2012). Intergroup bias toward “Group X”:
Evidence of prejudice, dehumanization, avoidance, and discrimination
against asexuals. Group Processes & Intergroup Relations, 15, 725–743.
http://dx.doi.org/10.1177/1368430212442419
McClave, C. H. (2013). Asexuality as a spectrum: A national probability
sample comparison to the sexual community in the U. K. (Master’s
thesis). Retrieved from http://academiccommons.columbia.edu/item/ac:
162382
McDowell, C. (1996). Measuring health: A guide to rating scales and
questionnaires (2nd ed.). Oxford, UK: Oxford University Press.
Meston, C. M., Trapnell, P. D., & Gorzalka, B. B. (1996). Ethnic and
gender differences in sexuality: Variations in sexual behavior between
Asian and non-Asian university students. Archives of Sexual Behavior,
25, 33–72. http://dx.doi.org/10.1007/BF02437906
Neth. (2011, June 8). Ableism and asexuality...anote. [Blog]. Retrieved
from http://sexualprivilegechecklist.tumblr.com/
NextStepCake. (2011, May 13). I’m an aromantic asexual. [Blog] Re-
trieved from http://nextstepcake.tumblr.com/post/5473267595/
southpawscopic-buildmeatower-the-whole-thing
Nurius, P. S. (1983). Mental health implications of sexual orientation.
Journal of Sex Research, 19, 119–136. http://dx.doi.org/10.1080/
00224498309551174
Poston, D. L., Jr., & Baumle, A. K. (2010). Patterns of asexuality in the
United States. Demographic Research, 23, 509–530. http://dx.doi.org/
10.4054/DemRes.2010.23.18
Prause, N., & Graham, C. A. (2007). Asexuality: Classification and char-
acterization. Archives of Sexual Behavior, 36, 341–356. http://dx.doi
.org/10.1007/s10508-006-9142-3
Radloff, K. (2008). Beyond the disorder: Change of discourse on asexuality
[Abstract]. Sexologies, 17(Suppl. 1), S156–S157. http://dx.doi.org/
10.1016/S1158-1360(08)72952-9
Rothblum, E. D., & Brehony, K. A. (1993). Boston marriages: Romantic
but asexual relationships among contemporary lesbians. Amherst, MA:
University of Massachusetts Press.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
11
ASEXUALITY IDENTIFICATION SCALE
Scherrer, K. S. (2008). Coming to an asexual identity: Negotiating identity,
negotiating desire. Sexualities, 11, 621–641. http://dx.doi.org/10.1177/
1363460708094269
Sloan, D. (Director). (2006). 20/20, March 24. [Television series episode].
In D. Sloan (Executive Producer), Cincinnati, OH: ABC News Produc-
tions.
Soldz, S., Budman, S., Demby, A., & Merry, J. (1995). A short form of the
Inventory of Interpersonal Problems Circumplex scales. Assessment, 2,
53–63. http://dx.doi.org/10.1177/1073191195002001006
Spector, I. P., Carey, M. P., & Steinberg, L. (1996). The Sexual Desire
Inventory: Development, factor structure, and evidence of reliability.
Journal of Sex & Marital Therapy, 22, 175–190. http://dx.doi.org/
10.1080/00926239608414655
Storms, M. D. (1979). Sexual orientation and self-perception. In P. Pliner,
K. R. Blanstein, I. M. Spigel, T. Alloway, & L. Krames (Eds.), Advances
in the study of communication and affect: Vol 5. Perception of emotion
in self and others (pp. 165–180). New York, NY: Plenum Press.
Tabachnick, B. G., & Fidell, L. A. (2001). Using multivariate statistics (4th
ed.). Needham Heights, MA: Allyn & Bacon.
Walters, B., & Geddie, B. (Creators). (2006). Secrets of Asexuals. [Tele-
vision series episode]. In B. Walters & B. Wolff (Executive producer),
The View. United States: Barwell Productions.
Yule, M. A., Brotto, L. A., & Gorzalka, B. B. (2013). Mental health and
interpersonal functioning in self-identified asexual men and women.
Psychology and Sexuality, 4, 136–151. http://dx.doi.org/10.1080/
19419899.2013.774162
Yule, M. A., Brotto, L. A., & Gorzalka, B. B. (2014a). Biological markers
of asexuality: Handedness, birth order, and finger length ratios in self-
identified asexual men and women. Archives of Sexual Behavior, 43,
299–310. http://dx.doi.org/10.1007/s10508-013-0175-0
Yule, M. A., Brotto, L. A., & Gorzalka, B. B. (2014b). Sexual fantasy and
masturbation among asexual individuals. Canadian Journal of Human
Sexuality, 23, 89–95. http://dx.doi.org/10.3138/cjhs.2409
Appendix
Asexuality Identification Scale (AIS)
(Yule, M. A., Brotto, L. A. & Gorzalka, B. B., 2014)
These questions ask about your experiences over your life-
time, rather than during a short period of time such as the past
few weeks or months. Please answer the questions as honestly
and as clearly as possible while keeping this in mind. In
answering these questions, keep in mind a definition of sex or
sexual activity that may include intercourse/penetration, caress-
ing, and/or foreplay.
What is your sexual orientation? _____________________
1. I experience sexual attraction toward other people
1⫽Completely True
2⫽Somewhat True
3⫽Neither True nor False
4⫽Somewhat False
5⫽Completely False
2. I lack interest in sexual activity
1⫽Completely False
2⫽Somewhat False
3⫽Neither True nor False
4⫽Somewhat True
5⫽Completely True
3. I don’t feel that that I fit the conventional categories of sexual
orientation such as heterosexual, homosexual (gay or lesbian), or
bisexual
1⫽Completely False
2⫽Somewhat False
3⫽Neither True nor False
4⫽Somewhat True
5⫽Completely True
4. The thought of sexual activity repulses me
1⫽Completely False
2⫽Somewhat False
3⫽Neither True nor False
4⫽Somewhat True
5⫽Completely True
5. I find myself experiencing sexual attraction toward another
person
1⫽Always
2⫽Often
3⫽Sometimes
4⫽Rarely
5⫽Never
(Appendix continues)
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
12 YULE, BROTTO, AND GORZALKA
6. I am confused by how much interest and time other people put
into sexual relationships
1⫽Completely False
2⫽Somewhat False
3⫽Neither True nor False
4⫽Somewhat True
5⫽Completely True
7. The term “nonsexual” would be an accurate description of my
sexuality
1⫽Completely False
2⫽Somewhat False
3⫽Neither True nor False
4⫽Somewhat True
5⫽Completely True
8. I would be content if I never had sex again
1⫽Completely False
2⫽Somewhat False
3⫽Neither True nor False
4⫽Somewhat True
5⫽Completely True
9. I would be relieved if I was told that I never had to engage in
any sort of sexual activity again
1⫽Completely False
2⫽Somewhat False
3⫽Neither True nor False
4⫽Somewhat True
5⫽Completely True
10. I go to great lengths to avoid situations where sex might be
expected of me
1⫽Completely False
2⫽Somewhat False
3⫽Neither True nor False
4⫽Somewhat True
5⫽Completely True
11. My ideal relationship would not involve sexual activity
1⫽Completely False
2⫽Somewhat False
3⫽Neither True nor False
4⫽Somewhat True
5⫽Completely True
12. Sex has no place in my life
1⫽Completely False
2⫽Somewhat False
3⫽Neither True nor False
4⫽Somewhat True
5⫽Completely True
Which of the following best describes you?
Heterosexual
Bisexual
Homosexual (Lesbian or Gay)
Asexual
Scoring
Total AIS scores are calculated by summing responses from all
12 questions. Higher scores indicate greater tendency to endorse
traits that may indicate asexuality. A cut-off score of 40/60 has
been proposed, such that those participants who score at or above
40 on the AIS are likely to experience a lack sexual attraction. The
final item (“Which of the following best describes you?”) is
unscored.
Received October 12, 2012
Revision received August 11, 2014
Accepted August 12, 2014 䡲
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
13
ASEXUALITY IDENTIFICATION SCALE
