Journal of Family Psychology

© 2019 American Psychological Association 2020, Vol. 34, No. 1, 90 –100

ISSN: 0893-3200 http://dx.doi.org/10.1037/fam0000614

An Online Intervention to Promote Predictors of Supportive Parenting for

Sexual Minority Youth
Joshua A. Goodman and Tania Israel
University of California, Santa Barbara

This article presents the Parent Resource for Increasing Sexual Minority Support (PRISMS), an
interactive online intervention designed by the authors to promote parental self-efficacy and behavioral
intentions for supporting a sexual minority child. The intervention was developed based upon psycho-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

logical literature about parent support and feedback from parents of sexual minority youth and psychol-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ogists, and contains 5 interactive online modules: normalizing parent experiences, psychoeducation,
reflection upon existing support, rehearsal of support, and affirmation. We assessed the feasibility and
acceptability of PRISMS and collected pilot data to assess its efficacy. One hundred eighty-four U.S.
parents of sexual minority youth Ages 13 to 18 were recruited from Amazon Mechanical Turk and were
randomly assigned to complete either PRISMS (n ⫽ 92) or a control: an existing resource with
psychoeducation about sexual orientation (n ⫽ 92). Analyses revealed that PRISMS was highly
acceptable to participants and established the noninferiority of the intervention relative to the control in
terms of credibility and affect at posttest. Two analyses of covariance did not establish significant
associations between PRISMS and parental self-efficacy or behavioral intentions for supportive parent-
ing, although we found small effect sizes (d ⫽ .31 and .21). Post hoc exploratory analyses offered
preliminary support for the efficacy of PRISMS with parents who were highly distressed about their
child’s sexual orientation, suggesting that this might be an important population for future research. We
discuss implications for research and practice, including the importance of interventions for parents of
sexual minority youth.

Keywords: parents of LGB youth, parent support, online intervention, parental self-efficacy

Sexual minority (lesbian, gay, bisexual, queer, questioning, or Diaz, & Sanchez, 2010); suicidality (Liu & Mustanski, 2012; Ryan
otherwise nonheterosexual) youth face notable challenges, includ- et al., 2010); self-esteem (Ryan et al., 2010; Snapp, Watson,
ing higher rates of mental health concerns than their heterosexual Russell, Diaz, & Ryan, 2015); alcohol, tobacco, and substance use
peers (National Institutes of Health, 2016; Russell & Fish, 2016). (Newcomb, Heinz, Birkett, & Mustanski, 2014; Newcomb, Heinz,
An important predictor of mental health outcomes for sexual & Mustanski, 2012; Ryan et al., 2010); and overall mental distress
minority youth is their perception of emotional support and pres- (Shilo & Savaya, 2011). Whereas a lack of family support can lead
ence from parents and other family members. An extensive body to negative outcomes, high levels of family support can serve as a
of research has found significant associations between sexual protective factor against psychological distress, including suicid-
minority youths’ levels of perceived family support and such ality (Ryan et al., 2010).
psychological outcomes as depression (Ryan, Russell, Huebner, Parents can be supportive of their sexual minority children in
various ways. General support includes conveying warmth and
love, doing enjoyable activities with one’s child, and fostering a
relationship in which the parent and child regularly communicate
This article was published Online First December 2, 2019.
(Needham & Austin, 2010; Thoma & Huebner, 2014). Direct
X Joshua A. Goodman and Tania Israel, Department of Counseling,
support for the child’s sexual orientation includes a willingness to
Clinical, and School Psychology, University of California, Santa Barbara.
This research was supported by the American Psychological Association discuss sexual orientation, gender expression, and identity explo-
of Graduate Students Psychological Science Research Grant and the Hos- ration; acknowledging the presence and impact of heterosexism;
ford Memorial Research Grant. This study is based upon Joshua A. and connecting one’s child with sexual minority media or com-
Goodman’s doctoral dissertation. Select findings were presented at the 11th munity events such as a youth group (Harkness & Israel, 2018;
Biennial National Multicultural Conference and Summit in Denver, Col- Ryan et al., 2010). Family members can also demonstrate support
orado in January 2019, and at the 127th Annual Convention of the Amer- for their child’s sexual minority relationships, including validating
ican Psychological Association in Chicago, Illinois in August 2019. We a child’s romantic relationships (to the extent that one would for
thank Miya Barnett and Erika Felix for their feedback regarding the
heterosexual relationships) and welcoming a child’s sexual minor-
development of this study.
Correspondence concerning this article should be addressed to Joshua A. ity friends into one’s home (D’amico, Julien, Tremblay, & Char-
Goodman, who is now at the Department of Psychology, Bates College, 4 trand, 2015; Ryan et al., 2010).
Andrews Road, Lewiston, ME 04240. E-mail: joshua.goodman.phd@ Parents can also demonstrate support outside of their direct
gmail.com interactions with their child. Parents can create more supportive

90
 PARENT SUPPORT FOR SEXUAL MINORITY YOUTH 91

environments for their child by encouraging family members and minority youth and suggestions for demonstrating support, and
friends to respect their child’s sexual orientation, standing up to found that this increased parents’ self-efficacy for supporting their
heterosexist comments, advocating on behalf of one’s child when child (Huebner et al., 2013). More broadly, online interventions
appropriate (e.g., in response to bullying at school), and taking have successfully targeted parents across a variety of domains,
steps to promote sexual minority inclusivity within their religious including advocating for a child with a disability (Glang,
community (D’amico et al., 2015; Ryan et al., 2010). Additionally, McLaughlin, & Schroeder, 2007), and communication with an
parents can seek support for their own adjustment and education, adolescent, including discussing sexuality (Villarruel, Loveland-
including reading books, using online resources, and finding a Cherry, & Ronis, 2010). Online interventions have several advan-
confidential space—away from their child—to process their expe- tages for serving parents of sexual minority youth, including
rience as a parent of a sexual minority child (D’amico et al., 2015; anonymity for parents concerned about revealing that they have a
Huebner, Rullo, Thoma, McGarrity, & Mackenzie, 2013). sexual minority child, low cost, scalability, and accessibility in
In contrast, parenting behaviors perceived as unsupportive in- communities in which there is limited access to in-person lesbian,
clude both overt rejection of a child’s sexual orientation, such as gay, bisexual, transgender, and queer (LGBTQ) affirming re-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

verbal rejection or sexual orientation change efforts (e.g., conver- sources (e.g., rural areas).
This document is copyrighted by the American Psychological Association or one of its allied publishers.

sion therapy), and covert or unintentional disapproval, such as The theory of planned behavior (Ajzen, 1991) posits that atti-
microaggressions (Nadal et al., 2011). Microaggressions can in- tudes, norms, and perceived behavioral control (self-efficacy) pre-
clude suggesting that sexual minority identities are pathological or dict behavioral intentions, which, in turn, predict enacted behavior.
“just a phase,” discouraging a child from coming out, and avoiding Applying this theory to parents of sexual minority youth, a parent
discussions related to sexual orientation (D’amico et al., 2015; who, for example, is confident that they can support their child’s
Harkness & Israel, 2018). same-sex relationship, has the attitude that it is good to support a
Although sexual minority youth can benefit from family support child’s same-sex relationships, and perceives social norms to be
as they navigate adolescence and emerging adulthood, parents may supportive of teenagers’ same-sex relationships is more likely than
experience barriers to providing support. Parents of sexual minor- another parent to intend to support their child’s same-sex relation-
ity youth may lack information about sexual orientation, know few ship. This behavioral intention is predictive of actions that convey
sexual minority individuals, and hold prejudices toward sexual support for their child’s relationship. Certain predictors of behav-
minorities (Conley, 2011a; D’amico et al., 2015; Pew Research ior in the theory of planned behavior may be more malleable than
Center, 2019). Notably, whereas children from other minority others—for example, whereas a brief intervention may shift self-
groups (e.g., racial or religious minorities) generally receive pa- efficacy and behavioral intentions, it is unlikely to change unsup-
rental affirmation of their minority status and support based upon portive social norms in a parent’s community, though it may alter
shared group experiences, sexual minority youth are typically the perceived norms (see Tankard & Paluck, 2016).
offspring of heterosexual parents, who may be uncertain of how to
support a nonheterosexual child. Finally, adjusting to a role as a Current Study
parent of a sexual minority child includes gaining knowledge,
processing what it means to have a sexual minority child, navi- To address calls for interventions to facilitate parent support for
gating concerns (e.g., about a child’s safety, spirituality, or hap- sexual minority youth, the authors developed an interactive online
piness), and, for some parents, a sense of loss about what they psychological intervention—the Parent Resource for Increasing
believed to be their child’s heterosexuality (Chrisler, 2017). Al- Sexual Minority Support (PRISMS). We conducted a pilot test to
though parents typically become more supportive of their child’s assess the feasibility and acceptability of this intervention, and
sexual orientation over time (see Phillips & Ancis, 2008), sexual collected preliminary outcome data to assess whether PRISMS
minority youth may benefit from parental support earlier. would be associated with higher parental self-efficacy and behav-
ioral intentions for sexual-minority-supportive parenting than a
control group at posttest. Although measurement of parent behav-
Increasing Parent Support ior was beyond the scope of this study, higher parental self-
Although the importance of parental support for sexual minority efficacy and behavioral intentions are theoretically associated with
youth is well established, and researchers have identified numer- enacted behaviors in the theory of planned behavior (Ajzen, 1991);
ous supportive and unsupportive parenting practices, there are few supportive parenting behaviors, in turn, are associated with better
psychological resources available to increase parent support. There mental health outcomes for sexual minority youth. We hypothesize
is scarce research about family therapy for increasing parent sup- that participants will find the PRISMS intervention to be accept-
port (but see Diamond et al., 2012; Willoughby & Doty, 2010). able and will have higher self-efficacy and behavioral intentions
Support groups such as PFLAG involve peer support among for sexual-minority-supportive parenting at posttest relative to a
parents of sexual minority youth; however, PFLAG groups pre- control group.
dominantly serve White parents in urban areas and may not ade-
quately reach racial minority and rural populations (Conley, Method
2011b). Researchers have called for more resources to help parents
of sexual minority youth support their children (Needham & PRISMS Intervention
Austin, 2010; Newcomb et al., 2019; Russell & Fish, 2016).
Online psychological interventions are a promising strategy for The development of the PRISMS intervention was informed by
increasing parent support. One such intervention asked parents to intervention mapping (Kok, Harterink, Vriens, de Zwart, & Hos-
watch a video featuring narratives of family members of sexual pers, 2006), including, among other steps, review of the literature,
 92 GOODMAN AND ISRAEL

needs assessment, seeking feedback during the design of the in- esteem, and unprotected sex. This module sought to increase
tervention, and developing an evaluation plan. An outline of participants’ motivation for applying the forthcoming information
PRISMS was initially drafted based upon parent training literature, (Stajkovic & Luthans, 2002).
theoretical underpinnings (described later in this section), and best Module 3: Reflection upon existing support. In this module,
practices for online interventions. Joshua A. Goodman then inter- participants considered their use of six types of supportive parent-
viewed five parents of sexual minority individuals, recruited from ing practices identified in the psychological literature. An example
a PFLAG chapter and personal connections, to provide feedback item is “When my child experiences bullying or mistreatment due
on the initial outline of PRISMS. Participants were queried regard- to their sexual orientation, I consider it most important to,” with
ing aspects of the intervention they perceived as helpful and response options including providing emotional support, suggest-
unhelpful, engaging and unengaging, and what could be added to ing behavioral changes to reduce the likelihood of future bullying,
the intervention to make it more helpful. After a draft of the and encouraging their child to “move on.” Participants were then
PRISMS intervention was completed, Joshua A. Goodman sought provided with feedback based upon their responses and sugges-
feedback about the intervention materials from 12 experts, four of tions for additional ways to convey each type of support. After-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

whom provided feedback; experts were selected based upon re- ward, participants watched an animated video in which a child
This document is copyrighted by the American Psychological Association or one of its allied publishers.

search and clinical experience in relevant areas (e.g., LGBTQ discusses an incident of homophobic bullying with their parent;
youth mental health, parental self-efficacy). Next, Joshua A. participants were asked to identify the parent’s supportive and
Goodman conducted usability testing with a different sample of unsupportive parenting behaviors. This module sought to (a) help
three parents of sexual minority individuals recruited from a parent participants to recognize the ways they were already being sup-
support organization and personal connections to assess their ex- portive, (b) increase self-efficacy by equipping participants with
perience with the intervention materials in real time (e.g., clarity of knowledge about how to demonstrate support (Morawska et al.,
materials, fatigue). All interviews with parents of sexual minority 2015), and (c) increase self-efficacy through vicarious experience
individuals were completed under an institutional review board and performance accomplishment when participants viewed and
approved protocol. Feedback from each step was integrated into identified supportive and unsupportive parenting in the video
the current version of the PRISMS intervention. (Bandura, 1977).
The PRISMS intervention consists of five modules, each sup- Module 4: Rehearsal of support. Participants were presented
ported by psychological research about increasing self-efficacy with a vignette in which a fellow parent shares that their child
and behavioral intentions. Three of the four mechanisms to in- recently came out as nonheterosexual and asks for ideas about how
crease self-efficacy outlined by Bandura (1977) were integrated to be supportive. Participants were provided with a textbox to write
into the intervention: verbal persuasion, performance accomplish- suggestions to this parent about how to best support their child.
ment, and vicarious experience. In addition, the intervention draws This activity sought to increase self-efficacy through performance
upon findings that knowledge about how to communicate about accomplishment (Bandura, 1977) and increase behavioral inten-
sexuality-related topics increases self-efficacy for sexuality-related tions through self-justification theory (Leahy, 2000).
communication with adolescents (Morawska, Walsh, Grabski, & Module 5: Affirmation. Participants were provided with
Fletcher, 2015). The intervention also drew upon applications of quotes about positive aspects of being a parent of a sexual minority
social– cognitive theory suggesting that high self-efficacy alone is child from Gonzalez, Rostosky, Odom, and Riggle (2013). Fol-
not sufficient for behavioral change; rather, it must be paired with lowing this, participants watched a slideshow featuring uplifting
motivation to change (Stajkovic & Luthans, 2002). Intervention music and images of sexual minority youth and their families. This
materials were therefore designed to motivate parents to support module sought to reemphasize motivation for increasing support-
their sexual minority child. All materials, including videos, were ive parenting practices and end the intervention on an uplifting
developed by the researchers. A brief description of each module note.
follows.
Module 1: Normalize parent experiences. Participants watched
Participants
a video with quotes from parents about their reactions to their child
coming out that encompassed a range of emotional reactions (e.g., Participants were recruited from April to May 2018 through
denial, sadness, joy, relief) and were provided with a framework of Amazon Mechanical Turk (MTurk), a crowdsourcing website in
parent adjustment to a child coming out (Phillips & Ancis, 2008). which participants can earn small amounts of money for complet-
This module sought to establish rapport with participants by con- ing online tasks. A notable benefit of MTurk is the ability to reach
necting with a range of emotional reactions and increase recep- a diverse sample: With parents, specifically, MTurk may yield
tiveness to the forthcoming modules by emphasizing the ways in more diversity in terms of ethnicity, income, and gender than
which parents adapt to and ultimately grow from their child samples recruited through listservs (Dworkin, Hessel, Gliske, &
coming out. Rudi, 2016) and may reach parents who are not connected to
Module 2: Psychoeducation. Participants were provided with support networks for parents of LGBTQ youth. Steps were taken
information about common experiences of sexual minority youth, to gain reliable data through MTurk, including using the affiliated
completed multiple choice items about the role of parent support website TurkPrime to restrict recruitment to individuals who had
for sexual minority youth, and were provided with feedback about previously identified as a parent of a child Age 9 to 19 who lives
their answers. A sample multiple choice item is “LGB youth who with them, and by only making the study available to participants
perceive their parents as supportive have higher levels of with an established reputation on MTurk (Peer, Vosgerau, &
__________ compared to LGB youth who perceive their parents as Acquisti, 2014). Recruitment was also attempted via online adver-
unsupportive,” with the following response options: anxiety, self- tisements, social media, and community organizations but yielded
 PARENT SUPPORT FOR SEXUAL MINORITY YOUTH 93

few participants; as such, only results from MTurk are analyzed in ther the PRISMS intervention described above or the control
this article. Due to technical errors with how the survey website condition using a function from the host website. The control
coded recruitment sources, it is possible that a small number of condition involved psychoeducation about sexual orientation using
participants from other recruitment sources (up to five) are in- a common online resource (American Psychological Association,
cluded in the MTurk sample. 2008), and multiple-choice and true–false questions to assess for
Participants were 184 U.S. parents, legal guardians, and primary comprehension. As this resource could be easily found through a
caretakers of youth Ages 13 to 18 who identified as nonhetero- web search, the control condition was conceptualized as Internet
sexual or believed their child to be nonheterosexual. Sample size treatment as usual. The control materials made a brief mention of
was selected based upon a power analysis that used effect sizes the importance of parent support. Following the intervention or
from similar studies. The average age of participants was 42 years, control, participants completed measures of parental self-efficacy
with a range from 30 to 60 years (SD ⫽ 6.70). A majority of and behavioral intentions for sexual-minority-supportive parent-
participants identified as White (84%; n ⫽ 155) in a “check all that ing, a measure of affect, acceptability and credibility questions,
apply” format, and slightly more than one in five (22%; n ⫽ 39) validity check items, and additional demographic items. Figure 1
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

identified with a racial or ethnic minority group, including Black/ provides information about participant flow and attrition.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

African American (13%; n ⫽ 23), Latino/a (6%; n ⫽ 11), Asian/

Pacific Islander (2%; n ⫽ 3), and Native American/Alaska Native Measures
(2%; n ⫽ 3). Most participants identified as women (73%; n ⫽
134), and two thirds identified as Christian (67%; n ⫽ 123). The Credibility. The credibility of materials was assessed with
modal participant perceived themselves as middle class on the seven items adapted from the nine-item Credibility Questionnaire
MacArthur Scale of Subjective Social Status (Adler & Stewart, (Cred-Q; Cartreine, Locke, Buckey, Sandoval, & Hegel, 2012);
2007), which asks participants to rate their socioeconomic status two items that were not relevant to the current study were not
on a ladder ranging from the first (lowest) to the tenth (highest) retained, and certain items were revised to refer to supporting LGB
step (M ⫽ 5.96, SD ⫽ 1.67). About half of participants reported youth. Items were measured on a 10-point scale ranging from not
living in a suburban location (56%; n ⫽ 103), with the remaining at all to completely. The Cred-Q was developed to assess the
participants living in rural (25%; n ⫽ 45) and urban (17%; n ⫽ 32) credibility of an online intervention for depression, and includes
locations.
Participants also provided demographic information about their
sexual minority child. The mean age of participants’ children was
15 years. Nearly three fifths of participants’ children identified as
a girl (58%; n ⫽ 106), 37% identified as a boy (n ⫽ 68), and 5%
identified as nonbinary (n ⫽ 10). A plurality of participants’
children identified as lesbian or gay (39%; n ⫽ 72), and one third
identified as bisexual, pansexual, or queer (33%; n ⫽ 61); about
one in four parents either knew their child to be questioning their
sexual orientation or were unsure of their child’s specific sexual
orientation (27%; n ⫽ 49). Nearly half of participants’ children
came out to their parent within the past year (46%; n ⫽ 85), and
one third came out more than 1 year ago (33%; n ⫽ 60); about one
in five (21%; n ⫽ 39) had a child whom they believed to be a
sexual minority but had not come out.

Procedure
Participants viewed a recruitment page on MTurk inviting par-
ents of sexual minority youth to use a resource for a research study,
and provided informed consent anonymously through the survey
website (Qualtrics) by clicking that they agreed to participate, in
accordance with our approved Human Subjects Committee proto-
col at the University of California, Santa Barbara. Participants then
completed an eligibility questionnaire to confirm that they were a
parent or primary caretaker of a child Age 13 to 18 who lived with
them, that they knew or believed their child to be nonheterosexual,
and that they lived in the United States; participants also com-
pleted a demographic survey. Next, participants completed mea-
Figure 1. CONSORT diagram with participant flow and attrition. Valid-
sures of attitudes toward sexual minorities and non-LGB parenting ity checks included (a) an attention check item, (b) comparison of partic-
behaviors and confidence involving sexuality, which were pre- ipant report of child’s age at pretest with age at which parent first believed
dicted to be covariates of the outcome variables of parental self- their child to be nonheterosexual at posttest, and (c) participant self-report
efficacy and behavioral intentions for sexual-minority-supportive for how seriously they took the study. PRISMS ⫽ Parent Resource for
parenting. Afterward, participants were randomly assigned to ei- Increasing Sexual Minority Support.
 94 GOODMAN AND ISRAEL

items about the trustworthiness and perceived bias of materials. A consistency of ␣ ⫽ .91 and ␣ ⫽ .89 for parenting confidence and
sample item is “How much do you believe what these materials tell behaviors.
you?” The adapted seven-item version of the Cred-Q had excellent Distress about child’s sexual orientation. Parental distress
internal consistency of ␣ ⫽ .93. about their child’s sexual orientation at pretest was assessed with
Positive and negative affect. Affect at posttest was measured the following item: “To what extent do you feel distressed about
with the Positive and Negative Affect Schedule (PANAS; Watson, your child’s sexual orientation?” The item was measured on a
Clark, & Tellegen, 1988), a 20-item survey consisting of two 5-point Likert scale ranging from not at all distressed (1) to
10-item subscales measuring positive (e.g., excited) and negative extremely distressed (5).
(e.g., ashamed) affect. Participants rated their present affect on a Parental self-efficacy. There was no existing measure of
Likert scale ranging from very slightly or not at all (1) to extremely parental self-efficacy to engage in sexual-minority-supportive
(5). With the current sample, the PANAS had strong internal parenting practices; therefore, the researchers used an existing
consistency for both the Positive Affect (␣ ⫽ .91) and Negative scale format for measuring therapist self-efficacy for working
Affect (␣ ⫽ .89) subscales. with sexual minority clients (Dillon & Worthington, 2003),
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Feasibility and acceptability. The feasibility and acceptabil- with items specific to sexual-minority-supportive parenting.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ity of the intervention was additionally assessed with questions Items and instructions were informed by recommendations for
about participants’ reactions to the intervention, satisfaction with developing self-efficacy scales from Bandura (2006). An initial
the intervention, and the helpfulness of the intervention. A major- pool of items was reviewed by three psychologists, and a final
ity of items are from two feasibility and acceptability studies set of 18 items was selected by Joshua A. Goodman; items were
(Hightow-Weidman et al., 2012; Mustanski, Garofalo, Monahan, based upon specific examples of sexual-minority-supportive
Gratzer, & Andrews, 2013) that served sexual minority commu- parenting practices in the literature. A sample item is “I can ask
nities. Feasibility and acceptability items, all on a 5-point Likert about crushes or attractions using language consistent with the
scale, unless otherwise specified, included (a) reactions to the gender(s) my child is attracted to.” Items were rated on a
intervention on scales ranging from terrible to wonderful, difficult 6-point Likert scale from not at all confident (1) to highly
to easy, frustrating to satisfying, and dull to stimulating; (b) the confident (6). An exploratory factor analysis was performed to
helpfulness of each activity within the PRISMS intervention; (c) assess whether items loaded together as a consistent construct;
items assessing acceptability for oneself and for other parents of all items loaded on a factor with acceptable to strong loadings
sexual minority youth; (d) satisfaction with the PRISMS interven- of .40 or higher. The parental self-efficacy measure had internal
tion as a whole; (e) motivation to use the information in the consistency of ␣ ⫽ .93.
intervention and self-reported change in motivation; and (f) satis- Behavioral intentions. Because no measure existed to assess
faction with the length and amount of information on a 5-point behavioral intentions to engage in sexual-minority-supportive par-
scale, with 3 being just right, and 1 and 5 representing too little or enting behaviors, the researchers developed a measure to assess for
too much length and information. Control participants completed this construct. The directions and Likert scale anchors were based
the same items for comparison purposes, with the exception of upon suggestions for measuring behavioral intentions outlined in
answering items about the helpfulness of specific activities in the Ajzen (2006). As with the parental self-efficacy measure, an initial
control condition. pool of items was reviewed by three psychologists, and Joshua A.
Attitudes. The Lesbian, Gay, and Bisexual Knowledge and Goodman selected a set of 22 items. Both supportive (e.g., “I plan
Attitudes Scale for Heterosexuals (LGB-KASH; Worthington, Dil- to watch a TV show or movie about LGB topics with my child”)
lon, & Becker-Schutte, 2005) was used as a pretest covariate to and unsupportive (e.g., “I plan to be upfront and honest with my
assess attitudes toward sexual minority populations. The LGB- child about my discomfort with their lesbian, gay, or bisexual
KASH contains 28 self-report items on a 7-point Likert scale identity”) intentions were included in the measure. Items were
anchored by very uncharacteristic of me or my views (1) and very measured on a 7-point Likert scale ranging from highly unlikely (1)
characteristic of me and my views (7). It contains five subscales: to highly likely (7) to engage in each parenting behavior within the
Hate; Knowledge of LGB History, Symbols, and Community; next 3 months. Unsupportive items were reverse-scored. As with
LGB Civil Rights; Religious Conflict; and Internalized Affirma- the parental self-efficacy measure, an exploratory factor analysis
tiveness. With the current sample, the LGB-KASH had strong was performed to assess whether items loaded together; one item
internal consistency of ␣ ⫽ .88, with reliability on subscales did not load with any other items and was removed, leaving a final
ranging from acceptable to strong (␣ ⫽ .73 to .89). set of 21 items. The behavioral intentions measure had a strong
Sexuality-related parenting practices. A questionnaire de- internal consistency of ␣ ⫽ .88.
veloped by Morawska et al. (2015) was used to examine partici-
pants’ sexuality-related parenting confidence and practices as pre-
Results
test covariates. The questionnaire asks parents to rate their
confidence and frequency with which they used 17 parenting Although this trial was not preregistered, analyses reported here
strategies. Confidence was measured on a 10-point Likert scale are consistent with what was planned within the study proposal.
ranging from certain I can’t do it (1) to certain I can do it (10), and Data analysis was performed using SPSS, and multiple imputation
behaviors were assessed on a 4-point Likert scale ranging from not was performed for missing items on scales (Schlomer, Bauman, &
true of me at all (1) to true of me very much or most of the time (4). Card, 2010). Between-groups differences for parental self-efficacy
Items are not sexual minority specific. Sample items include and credibility were significant; the intervention and control par-
“encouraged my child to ask questions about sexuality” and “ex- ticipants did not differ significantly with regard to any other pretest
pressed comfort with my own sexuality.” The measure had internal covariate or outcome measures (see Table 1). Correlations of
 PARENT SUPPORT FOR SEXUAL MINORITY YOUTH 95

Table 1
Descriptive Statistics for Measures

PRISMS Control
Measure Scale range Observed range M SD M SD t

Parental self-efficacyⴱ 18–108 44–108 101.02 9.38 97.43 13.75 2.07

Behavioral intentions 21–147 51–147 131.71 13.24 128.20 19.06 1.45
Sexuality: Parenting practices 17–68 28–68 54.44 9.02 54.94 9.14 .37
Sexuality: Parenting confidence 17–170 51–170 145.44 21.46 142.22 23.98 .96
LGB-KASH 28–196 79–196 138.54 26.31 138.20 26.98 .11
PANAS Positive 10–50 11–50 37.91 8.75 36.22 9.02 1.29
PANAS Negative 10–50 10–34 13.71 5.56 13.92 5.54 .28
Cred-Qⴱ 7–70 21–70 61.83 8.66 58.09 11.38 2.66
Note. PRISMS ⫽ Parent Resource for Increasing Sexual Minority Support; LGB-KASH ⫽ Lesbian, Gay, and
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Bisexual Knowledge and Attitudes Scale for Heterosexuals; PANAS ⫽ Positive and Negative Affect Schedule;
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Cred-Q ⫽ Credibility Questionnaire.

ⴱ
p ⬍ .05.

variables used to assess the intervention’s efficacy can be found in Third, the helpfulness of intervention materials was evaluated
Table 2. on Likert scales ranging from not helpful (1) to very helpful (5) for
each of seven aspects of the PRISMS intervention. Mean helpful-
Feasibility and Acceptability ness ratings were 4 or above for each aspect of the intervention and
ranged from 4.05 to 4.59. Participants in the control condition
Several strategies were used to test the hypothesis that PRISMS answered six items about the activities they completed using the
would be feasible and acceptable. First, to determine whether the same Likert scale, with means ranging from 3.81 to 4.16.
intervention was sufficiently credible, participant ratings of cred-
Fourth, participant satisfaction was examined. Intervention and
ibility were compared with the control condition. Participants in
control participants reported comparable levels of overall satisfac-
both conditions rated the materials they used to be highly credible
tion (M ⫽ 4.57 and 4.38, respectively, on a 5-point Likert scale
on the adapted Cred-Q (see Table 1). Noninferiority testing was
ranging from not satisfied [1] to highly satisfied [5]). Reactions as
used to assess whether PRISMS was no less credible than treat-
measured on 5-point Likert scales, ranging from terrible to won-
ment as usual because a similar level of credibility as a well-
derful, difficult to easy, frustrating to satisfying, and dull to stim-
regarded psychological resource would satisfy the hypothesis re-
ulating, also indicated similar satisfaction among intervention
garding acceptability (Walker & Nowacki, 2011). A comparison of
(M ⫽ 17.37) and control (M ⫽ 16.67) participants. Participants
confidence intervals (CIs) established the noninferiority of
also evaluated acceptability for self and others by indicating the
PRISMS compared with the control condition (95% CI [60.25,
extent to which the resource they used was a good way to “learn
63.86] and 95% CI [55.73, 60.44], respectively).
Second, affect at posttest was examined to ensure that the more about how to support my child” and “help other parents with
intervention did not lead to harm in the form of higher negative an LGB child” on a 5-point Likert scale from strongly disagree (1)
affect or lower positive affect compared with the control group. to strongly agree (5). Means for these items were comparable
Levels of positive affect at posttest were similar in the intervention between the intervention (4.61 and 4.74) and control (4.30 and
and control conditions (see Table 1). A comparison of CIs for 4.47) conditions.
positive affect established the noninferiority of PRISMS relative to Fifth, items about motivation were examined due to its role in
the control (95% CI [36.10, 39.72] and 95% CI [34.36, 38.09], behavioral change (Stajkovic & Luthans, 2002). The first item,
respectively). Levels of negative affect at posttest were compara- motivation to use the information from the study, was measured on
ble for the intervention and control conditions (see Table 1), and a 5-point Likert scale from strongly disagree (1) to strongly agree
noninferiority for the intervention was established (95% CI [12.67, (5); means were similar in the intervention (4.67) and control
14.74] and 95% CI [12.77, 15.07], respectively). (4.64) conditions. The second item was “the activities I completed
made me less (1)/more (5) motivated to support my child as an
LGB person,” with means of 4.62 and 4.20 for the intervention and
Table 2 control conditions, respectively. An independent samples t test
Correlation Matrix for Variables Used in Efficacy Tests established that this difference was significant, t(181) ⫽ 8.28, p ⫽
.001.
Measure 1 2 3 4 5 The final step to examine the feasibility and acceptability of the
intervention was to evaluate the time it took participants to com-
1. Parental self-efficacy —
2. Behavioral intentions .86 — plete the study. Mean time to complete PRISMS, as measured by
3. Attitudes about sexual orientation .57 .59 — Qualtrics, was 47 min (Mdn ⫽ 39), and 90% of responses took 24
4. Sexuality: Parenting confidence .58 .45 .46 — to 69 min—this time includes screening, preliminary, and posttest
5. Sexuality: Parenting practices .44 .29 .36 .75 — measures in addition to the intervention. Participants were asked,
Note. All correlations are significant at p ⬍ .001. on a 5-point scale, to indicate whether they felt that the interven-
 96 GOODMAN AND ISRAEL

tion was too short (1), just right (3), or too long (5). Two thirds of As a post hoc assessment of whether the PRISMS intervention
participants stated that the length was just right (n ⫽ 62), 1% may be associated with predictors of supportive parenting with this
wished it were longer (n ⫽ 1), and nearly one third wished it were population, the two ANCOVAs were reconducted with partici-
shorter (n ⫽ 28). About three quarters of participants reported that pants who reported high distress about their child’s sexual orien-
the amount of information in the intervention was just right (n ⫽ tation. Our analyses yielded significant associations between
71), while fewer participants reported wanting more information PRISMS and parental self-efficacy when controlling for confi-
(n ⫽ 9) or less information (n ⫽ 12). dence and parenting practices regarding sexuality in general, F(1,
16) ⫽ 6.28, p ⫽ .023, as well as behavioral intentions when
Preliminary Efficacy Analyses controlling for attitudes about sexual orientation, F(1, 17) ⫽ 6.80,
p ⫽ .018. A visual inspection of demographic features of parents
We conducted two one-way analyses of covariance (ANCOVAs) who reported high distress about their child’s sexual orientation
as a preliminary assessment of the efficacy of the PRISMS inter- found that this group was comparable with the sample at large with
vention. To assess parental self-efficacy, we selected confidence the exception of parent ethnicity and when or if a child came out,
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

with and use of sexuality parenting practices as covariates because which were selected for analysis. Thirty percent of high-distress
This document is copyrighted by the American Psychological Association or one of its allied publishers.

self-efficacy in parenting about sexuality may generalize to self- parents identified as Black/African American (n ⫽ 6) compared
efficacy about parenting with sexual orientation, specifically, and with 10% of low-distress participants (n ⫽ 17); a chi-square test of
because performance accomplishment in parenting with sexuality independence found that this difference was significant (␹2 ⫽
generally was hypothesized to enhance self-efficacy for parenting 6.21, p ⫽ .013). Most highly distressed participants had a child
about sexual orientation (see Bandura, 1977). When accounting for who came out in the last year (55%; n ⫽ 11) or who[m] they
these covariates, the association between PRISMS and parental believed to be nonheterosexual but had not come out (40%; n ⫽ 8).
self-efficacy was not significant, F(1, 180) ⫽ 3.54, p ⫽ .061. A A chi-square test of independence established that when, or if, a
small effect size was present (Cohen’s d ⫽ .31). To assess whether child came out was associated with participant distress (␹2 ⫽ 9.37,
PRISMS was associated with behavioral intentions to support p ⫽ .009); a post hoc evaluation of residuals indicated that par-
one’s child, we conducted a one-way ANCOVA, with attitudes, an ticipants whose child came out more than 1 year ago were less
antecedent of behavioral intentions in the theory of planned be- likely to be in the high-distress group than the low-distress group
havior (Ajzen, 1991), as a covariate. Results did not indicate that (z ⫽ ⫺2.80, p ⫽ .005).
PRISMS had a significant effect on behavioral intentions for
sexual-minority-supportive parenting, F(1, 181) ⫽ 2.97, p ⫽ .086,
although a small effect size was found (d ⫽ .21). Discussion
One potential explanation for our nonsignificant findings was We developed an innovative, interactive online intervention for
the presence of ceiling effects on the outcome measures used. Five parents of sexual minority youth and found that it was highly
sixths of participants had an average score of at least 5 out of 6 for acceptable in terms of participant satisfaction, credibility, engage-
parental self-efficacy (n ⫽ 154; 84%), and nearly three quarters of ment, and helpfulness. Although tests of the efficacy of the inter-
participants had an average score of at least 6 out of 7 for vention did not establish significant associations between PRISMS
behavioral intentions (n ⫽ 135; 73%). This may both speak to the and parental self-efficacy and behavioral intentions with our sam-
items used to measure parental self-efficacy and behavioral inten- ple at large when controlling for pretest covariates, each had a
tions, and the fact that participants who were already supportive of small effect size. We also found preliminary evidence that
their sexual minority children may have been more likely to PRISMS was associated with parental self-efficacy and behavioral
participate than other parents. Ceiling effects were not present intentions among parents who reported high distress about their
among a key group of interest: parents who reported feeling a high child’s sexual orientation. Additionally, results conveyed that in-
level of distress about their child’s sexual orientation (4 or 5 on a tervention participants self-reported a higher degree of gain in
5-point Likert scale ranging from not at all distressed [1] to motivation to support their child relative to control participants.
extremely distressed [5]; n ⫽ 20; see Table 3). Our preliminary tests of the efficacy of PRISMS, though non-
significant, suggest that the effect size of PRISMS for parental
self-efficacy (d ⫽ .31) is comparable with that of other online
Table 3 psychological interventions with parents. Although this effect size
Mean Levels of Outcome Variables Based on Parent Distress is slightly lower than the average effect size of .36 found in a
About Child’s Sexual Orientation meta-analysis of online parenting interventions (Nieuwboer, Fuk-
kink, & Hermanns, 2013), it is remarkable when considering (a)
Low distress High distress that PRISMS is a single-session intervention, whereas most studies
PRISMS Control PRISMS Control in the meta-analysis involved multiple sessions or prolonged ac-
Measure (n ⫽ 81) (n ⫽ 83) (n ⫽ 11) (n ⫽ 9) cess to a resource; and (b) the fact that parents of teenagers are less
confident in their parenting than parents of younger children (Glatz
Parental self-efficacy 102.27 100.42 91.45 69.89
Behavioral intentions 133.13 132.27 120.27 90.67
& Buchanan, 2015), who were the target population of a majority
of studies in the meta-analysis. In contrast, the effect size for
Note. Parent distress about child’s sexual orientation was measured on a behavioral intentions (d ⫽ .21) was smaller, even as PRISMS
scale from not at all distressed (1) to extremely distressed (5). A score of
3 or lower was considered low distress, and a score of 4 or 5 was
appeared to motivate participants to use the information they
considered high distress. PRISMS ⫽ Parent Resource for Increasing Sex- learned about supportive parenting. One explanation is that
ual Minority Support. PRISMS addressed one predictor of behavioral intentions in the
 PARENT SUPPORT FOR SEXUAL MINORITY YOUTH 97

theory of planned behavior—self-efficacy— but not two others— More broadly, little is known about parents who feel highly dis-
attitudes and social norms (Ajzen, 1991). A second explanation is tressed about their child’s sexual orientation, and further research
that it may take some period of time for the intervention materials about the reasons parents feel distressed may inform future inter-
to fully impact behavioral intentions. vention research. One challenge in research with this population is
Our exploratory finding that PRISMS was associated with pa- recruitment: Parents who hold nonaffirming attitudes about their
rental self-efficacy and behavioral intentions for parents who were child’s sexual orientation may be less likely than others to choose
distressed about their child’s sexual orientation offers preliminary to participate in a study for parents of sexual minority youth.
support for the efficacy of the intervention with this population. Because parents who are highly distressed about their child’s
Parents who feel distressed about their child’s sexual orientation sexual orientation are especially likely to have a child who is not
are an important group, both because of the need to address the out or recently came out, these parents may also be underrepre-
distress that these parents feel and because parents who feel sented in LGBTQ parent support or community organizations,
distressed about their child’s sexual orientation may be particularly from which recruitment often occurs. Thus, online interventions
likely to engage in unsupportive parenting practices (e.g., convey may be a particularly promising delivery format with this popula-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

verbal disapproval, enroll their child in conversion therapy). We tion. Research about strategies for recruiting parents who feel
This document is copyrighted by the American Psychological Association or one of its allied publishers.

found that parents reporting high distress were particularly likely distressed about their child’s sexual orientation would be a wel-
to have a child who had not been out for at least 1 year, consistent come contribution to the field.
with past findings that parents become more comfortable with their Second, PRISMS can be adapted with additional modules, tai-
child’s sexual orientation over time (Chrisler, 2017; Phillips & loring, and follow-up assessment. Because we did not find a
Ancis, 2008). Highly distressed parents were also more likely than significant association between PRISMS and behavioral inten-
others to be Black, which is consistent with past research about tions, the addition of modules to address attitudes and perceived
parental attitudes toward sexual minorities among different racial social norms—the two predictors of behavioral intentions besides
groups (Calzo & Ward, 2009) and may also relate to awareness of self-efficacy in the theory of planned behavior (Ajzen, 1991)—
a social environment in which intersecting prejudices affect sexual may be promising. PRISMS can also be tailored toward specific
minority youth of color (Sutter & Perrin, 2016). Given the adverse populations (e.g., cultural, religious, or gender groups) to speak to
effect of parental rejection of sexual orientation on the mental the specific experiences and needs of members of these popula-
health of sexual minority youth (see Liu & Mustanski, 2012; Ryan tions. Finally, collecting follow-up data would allow for assess-
et al., 2010), preliminary evidence for the efficacy of PRISMS ment of how PRISMS affects supportive parenting over time.
with parents who were distressed about their child’s sexual orien- Given the established relationship between parent support and
tation is encouraging. sexual minority youth mental health, longitudinal research also
offers an opportunity to assess the impact of PRISMS on the
mental health of participants’ children.
Implications for Research
If established as efficacious, research can focus on dissemina-
Whereas past psychological research surrounding sexual minor- tion and implementation of PRISMS. Such research can identify
ity youth has predominantly focused on the youth themselves, this people or organizations (e.g., health care providers, religious lead-
study is one of a handful to address parent support— one of the ers, school personnel) that can effectively distribute PRISMS to
most influential predictors of sexual minority youth mental health. parents of sexual minority youth and evaluate the intervention’s
In taking a systems-based approach, this study acknowledges that external validity. Furthermore, although our MTurk sample was
parents of sexual minority youth have their own needs that can be more diverse in terms of ethnicity and gender than samples re-
served related to their experience as a parent of a sexual minority cruited from parent support organizations (e.g., Conley, 2011b;
child. In particular, this study highlights the potential for future Gonzalez et al., 2013), ethnic minorities and fathers are nonethe-
research and clinical interventions to address parent support less underrepresented. There is a need for further research about
through online resources. Our study joins Huebner et al.’s (2013) recruitment for interventions with underserved groups of parents
Lead with Love video as, to our knowledge, the second online of sexual minority youth. If a modified version of PRISMS is
psychological intervention for parents of sexual minority youth tailored toward a specific group (e.g., a religious group, fathers),
and the first to use interactive multimedia to address predictors of targeting recruitment specifically toward that group is one strategy
parent support. Advantages of this study’s online medium included for recruiting a diverse sample.
interactivity (e.g., tailored feedback, a writing exercise), the low
marginal cost of providing the intervention to each participant, and
Implications for Practice
reaching individuals who may lack access to in-person support
resources due to distance (e.g., in rural communities) or cost. In In a society in which sexual minority youth face mental health
particular, the interactive nature of PRISMS may explain its high disparities due to their social environment (Needham & Austin,
helpfulness ratings and acceptability and suggests promise in the 2010), there is a need for clinical interventions to address the
use of interactive multimedia in future interventions for parents of systems in which these youth live. Although there is a dearth of
sexual minority youth. therapy research about increasing family support for sexual mi-
Future research can build from the current study in several key nority youth, this study suggests that efforts to promote family
ways. First, PRISMS can be further tested with parents who report support may help to prepare parents who are distressed about their
high distress about their child’s sexual orientation. Replication of child’s sexual orientation to support their child. Several compo-
the exploratory analyses with a larger sample would help to nents of PRISMS can be integrated into counseling with parents of
establish the efficacy of PRISMS with this specific population. LGB youth, including reflection upon ways one already demon-
 98 GOODMAN AND ISRAEL

strates support, rehearsal of supporting one’s child, and psychoe- ioral intentions with our sample at large, we did find preliminary
ducation. An important component of PRISMS is to connect with evidence that it was associated with these outcomes for parents
parents’ emotional experiences (e.g., using quotes from parents reporting high distress about their child’s sexual orientation. This
about their reaction to their child coming out) while emphasizing study is part of a small but growing body of research that seeks to
the significance of parents’ actions toward their child; therapeutic facilitate parent support for sexual minority youth and, more
interventions may likewise seek to find this balance. broadly, create more supportive spaces for these youth. Whereas
most research to date has focused on the experiences of sexual
minority youth, and this continues to be of critical importance, the
Limitations
current study highlights the value of both serving the needs of
This study’s findings should be interpreted with consideration of parents of sexual minority youth and targeting psychological in-
its limitations. First, this study, like other studies with parents of terventions toward members of the family systems in which sexual
sexual minority youth, relied upon self-report measures. Self- minority youth live.
report questionnaires about parenting can be affected by difficul-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ties with estimating parenting behavior, social desirability effects, References

This document is copyrighted by the American Psychological Association or one of its allied publishers.

and a tendency to overstate help provided toward one’s child Adler, N., & Stewart, J. (2007). The MacArthur Scale of Subjective Social
(Mandemakers & Dykstra, 2008; Morsbach & Prinz, 2006). In Status. MacArthur Research Network on SES & Health. Retrieved from
particular, it may be difficult to accurately assess self-efficacy and https://macses.ucsf.edu/research/psychosocial/subjective.php
behavioral intentions, which are more abstract than assessing how Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior
frequently one uses a specific parenting practice. Social desirabil- and Human Decision Processes, 50, 179 –211. http://dx.doi.org/10.1016/
ity effects may also lead participants to underreport bias (Hof- 0749-5978(91)90020-T
mann, Gawronski, Gschwendner, Le, & Schmitt, 2005) or overre- Ajzen, I. (2006). Constructing a theory of planned behavior questionnaire.
port acceptability. A second limitation is that ceiling effects on Retrieved from http://people.umass.edu/~aizen/pdf/tpb.measurement
outcome measures may limit our understanding of the effect of the .pdf
American Psychological Association. (2008). Answers to your questions:
intervention. In addition, although this study, like others assessing
For a better understanding of sexual orientation and homosexuality.
self-efficacy and behavioral intentions, adapted existing frame- Retrieved from http://www.apa.org/topics/lgbt/orientation.pdf
works to the specific content of the study, the measures may Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral
benefit from further validation. change. Psychological Review, 84, 191–215. http://dx.doi.org/10.1037/
Third, findings from this study only apply immediately at post- 0033-295X.84.2.191
test and further research is needed to assess for efficacy and Bandura, A. (2006). Guide for constructing self-efficacy scales. In F.
acceptability over time or whether PRISMS may be associated Pajares & T. Urdan (Eds.), Self-efficacy beliefs of adolescents (Vol. 5,
with the use of supportive parenting behaviors. Additionally, pp. 307–337). Greenwich, CT: Information Age Publishing.
whereas our design was informed by the theory of planned behav- Calzo, J. P., & Ward, L. M. (2009). Contributions of parents, peers, and
ior (Ajzen, 1991), we did not assess for attitudes and social norms media to attitudes toward homosexuality: Investigating sex and ethnic
differences. Journal of Homosexuality, 56, 1101–1116. http://dx.doi.org/
at posttest despite the potential that the intervention affected these
10.1080/00918360903279338
factors. This may limit our understanding of the results and the Cartreine, J. A., Locke, S. E., Buckey, J. C., Sandoval, L., & Hegel, M. T.
interaction between variables. (2012). Electronic problem-solving treatment: Description and pilot
Regarding our exploratory analyses with parents reporting high study of an interactive media treatment for depression. JMIR Research
distress about their child’s sexual orientation, it is important to Protocols, 1, e11. http://dx.doi.org/10.2196/resprot.1925
note that this involves a small subset of participants, and results Chrisler, A. J. (2017). Understanding parent reactions to coming out as
should be interpreted with caution. More broadly, it is not possible lesbian, gay, or bisexual: A theoretical framework. Journal of Family
to rule out hypothesis guessing in our results, as PRISMS provided Theory & Review, 9, 165–181. http://dx.doi.org/10.1111/jftr.12194
specific information about supportive parenting practices that was Conley, C. L. (2011a). The development and initial validation of the
assessed in the outcome measures; the fact that the control condi- COPLAG scale: Measuring the concerns of parents of lesbians and gays.
Journal of Gay & Lesbian Social Services: The Quarterly Journal of
tion also contained affirmative information about sexual orienta-
Community & Clinical Practice, 23, 30 –52. http://dx.doi.org/10.1080/
tion suggests that a similar, albeit potentially lower, risk existed for 10538720.2010.532083
control participants, too. Likewise, we did not include a formal Conley, C. L. (2011b). Learning about a child’s gay or lesbian sexual
manipulation check to determine whether participants knew the orientation: Parental concerns about societal rejection, loss of loved
condition they were assigned to; however, the fact that both ones, and child well being. Journal of Homosexuality, 58, 1022–1040.
conditions received materials that they considered relevant and http://dx.doi.org/10.1080/00918369.2011.598409
helpful reduces the likelihood that this affected the results. D’amico, E., Julien, D., Tremblay, N., & Chartrand, E. (2015). Gay,
lesbian, and bisexual youths coming out to their parents: Parental reac-
tions and youths’ outcomes. Journal of GLBT Family Studies, 11,
Conclusion 411– 437. http://dx.doi.org/10.1080/1550428X.2014.981627
Diamond, G. M., Diamond, G. S., Levy, S., Closs, C., Ladipo, T., &
The current study pilot tested a brief, online intervention for
Siqueland, L. (2012). Attachment-based family therapy for suicidal
parents of sexual minority youth and found that this intervention lesbian, gay, and bisexual adolescents: A treatment development study
was highly acceptable to participants at a level similar to the and open trial with preliminary findings. Psychotherapy, 49, 62–71.
acceptability of an existing psychological resource used by parents http://dx.doi.org/10.1037/a0026247
of sexual minority youth. Although we did not establish that the Dillon, F., & Worthington, R. L. (2003). The Lesbian, Gay, and Bisexual
intervention was associated with parental self-efficacy and behav- Affirmative Counseling Self-Efficacy Inventory (LGB-CSI): Develop-
 PARENT SUPPORT FOR SEXUAL MINORITY YOUTH 99

ment, validation, and training implications. Journal of Counseling Psy- mechanisms for lesbian, gay, and bisexual individuals. Journal of LGBT
chology, 50, 235–251. http://dx.doi.org/10.1037/0022-0167.50.2.235 Issues in Counseling, 5, 21– 46. http://dx.doi.org/10.1080/15538605
Dworkin, J., Hessel, H., Gliske, K., & Rudi, J. H. (2016). A comparison of .2011.554606
three online recruitment strategies for engaging parents. Family Rela- National Institutes of Health. (2016). Sexual and gender minorities for-
tions: An Interdisciplinary Journal of Applied Family Studies, 65, 550 – mally designated as a health disparity population for research purposes.
561. http://dx.doi.org/10.1111/fare.12206 Retrieved from https://www.nimhd.nih.gov/about/directors-corner/
Glang, A., McLaughlin, K., & Schroeder, S. (2007). Using interactive messages/message_10-06-16.html
multimedia to teach parent advocacy skills: An exploratory study. The Needham, B. L., & Austin, E. L. (2010). Sexual orientation, parental
Journal of Head Trauma Rehabilitation, 22, 198 –205. http://dx.doi.org/ support, and health during the transition to young adulthood. Journal of
10.1097/01.HTR.0000271121.42523.3a Youth and Adolescence, 39, 1189 –1198. http://dx.doi.org/10.1007/
Glatz, T., & Buchanan, C. M. (2015). Change and predictors of change in s10964-010-9533-6
parental self-efficacy from early to middle adolescence. Developmental Newcomb, M. E., Heinz, A. J., Birkett, M., & Mustanski, B. (2014). A
Psychology, 51, 1367–1379. http://dx.doi.org/10.1037/dev0000035 longitudinal examination of risk and protective factors for cigarette
Gonzalez, K. A., Rostosky, S. S., Odom, R. D., & Riggle, E. D. (2013). smoking among lesbian, gay, bisexual, and transgender youth. Journal
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

The positive aspects of being the parent of an LGBTQ Child. Family of Adolescent Health, 54, 558 –564. http://dx.doi.org/10.1016/j
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Process, 52, 325–337. http://dx.doi.org/10.1111/famp.12009 .jadohealth.2013.10.208

Harkness, A., & Israel, T. (2018). A thematic framework of observed Newcomb, M. E., Heinz, A. J., & Mustanski, B. (2012). Examining risk
mothers’ socialization messages regarding sexual orientation. Psychol- and protective factors for alcohol use in lesbian, gay, bisexual, and
ogy of Sexual Orientation and Gender Diversity, 5, 260 –272. http://dx transgender youth: A longitudinal multilevel analysis. Journal of Studies
.doi.org/10.1037/sgd0000268 on Alcohol and Drugs, 73, 783–793. http://dx.doi.org/10.15288/jsad
Hightow-Weidman, L. B., Pike, E., Fowler, B., Matthews, D. M., Kibe, J., .2012.73.783
McCoy, R., & Adimora, A. A. (2012). HealthMpowerment.org: Feasi- Newcomb, M. E., LaSala, M. C., Bouris, A., Mustanski, B., Prado, G.,
bility and acceptability of delivering an internet intervention to young Schrager, S. M., & Huebner, D. M. (2019). The influence of families on
Black men who have sex with men. AIDS Care, 24, 910 –920. http://dx LGBTQ youth health: A call to action for innovation in research and
.doi.org/10.1080/09540121.2011.647677
intervention development. LGBT Health, 6, 139 –145. http://dx.doi.org/
Hofmann, W., Gawronski, B., Gschwendner, T., Le, H., & Schmitt, M.
10.1089/lgbt.2018.0157
(2005). A meta-analysis on the correlation between the implicit associ-
Nieuwboer, C. C., Fukkink, R. G., & Hermanns, J. M. (2013). Online
ation test and explicit self-report measures. Personality and Social
programs as tools to improve parenting: A meta-analytic review. Chil-
Psychology Bulletin, 31, 1369 –1385. http://dx.doi.org/10.1177/
dren and Youth Services Review, 35, 1823–1829. http://dx.doi.org/10
0146167205275613
.1016/j.childyouth.2013.08.008
Huebner, D. M., Rullo, J. E., Thoma, B. C., McGarrity, L. A., & Mack-
Peer, E., Vosgerau, J., & Acquisti, A. (2014). Reputation as a sufficient
enzie, J. (2013). Piloting Lead with Love: A film-based intervention to
condition for data quality on Amazon Mechanical Turk. Behavior Re-
improve parents’ responses to their lesbian, gay, and bisexual children.
search Methods, 46, 1023–1031. http://dx.doi.org/10.3758/s13428-013-
The Journal of Primary Prevention, 34, 359 –369. http://dx.doi.org/10
0434-y
.1007/s10935-013-0319-y
Pew Research Center. (2019). Attitudes on same-sex marriage. Retrieved
Kok, G., Harterink, P., Vriens, P., de Zwart, O., & Hospers, H. J. (2006).
from http://www.pewforum.org/fact-sheet/changing-attitudes-on-gay-
The gay cruise: Developing a theory-and evidence-based Internet HIV-
marriage/
prevention intervention. Sexuality Research & Social Policy: A Journal
Phillips, M. J., & Ancis, J. R. (2008). The process of identity development
of the NSRC, 3, 52– 67. http://dx.doi.org/10.1525/srsp.2006.3.2.52
Leahy, R. L. (2000). Sunk costs and resistance to change. Journal of as the parent of a lesbian or gay male. Journal of LGBT Issues in
Cognitive Psychotherapy, 14, 355–371. http://dx.doi.org/10.1891/0889- Counseling, 2, 126 –158. http://dx.doi.org/10.1080/15538600802125605
8391.14.4.355 Russell, S. T., & Fish, J. N. (2016). Mental health in lesbian, gay, bisexual,
Liu, R. T., & Mustanski, B. (2012). Suicidal ideation and self-harm in and transgender (LGBT) youth. Annual Review of Clinical Psychology,
lesbian, gay, bisexual, and transgender youth. American Journal of 12, 465– 487. http://dx.doi.org/10.1146/annurev-clinpsy-021815-093153
Preventive Medicine, 42, 221–228. http://dx.doi.org/10.1016/j.amepre Ryan, C., Russell, S. T., Huebner, D., Diaz, R., & Sanchez, J. (2010).
.2011.10.023 Family acceptance in adolescence and the health of LGBT young adults.
Mandemakers, J. J., & Dykstra, P. A. (2008). Discrepancies in parent’s and Journal of Child and Adolescent Psychiatric Nursing, 23, 205–213.
adult child’s reports of support and contact. Journal of Marriage and http://dx.doi.org/10.1111/j.1744-6171.2010.00246.x
Family, 70, 495–506. http://dx.doi.org/10.1111/j.1741-3737.2008 Schlomer, G. L., Bauman, S., & Card, N. A. (2010). Best practices for
.00496.x missing data management in counseling psychology. Journal of Coun-
Morawska, A., Walsh, A., Grabski, M., & Fletcher, R. (2015). Parental seling Psychology, 57, 1–10. http://dx.doi.org/10.1037/a0018082
confidence and preferences for communicating with their child about Shilo, G., & Savaya, R. (2011). Effects of family and friend support on
sexuality. Sex Education, 15, 235–248. http://dx.doi.org/10.1080/ LGB youths’ mental health and sexual orientation milestones. Family
14681811.2014.996213 Relations: An Interdisciplinary Journal of Applied Family Studies, 60,
Morsbach, S. K., & Prinz, R. J. (2006). Understanding and improving the 318 –330. http://dx.doi.org/10.1111/j.1741-3729.2011.00648.x
validity of self-report of parenting. Clinical Child and Family Psychol- Snapp, S. D., Watson, R. J., Russell, S. T., Diaz, R. M., & Ryan, C. (2015).
ogy Review, 9, 1–21. http://dx.doi.org/10.1007/s10567-006-0001-5 Social support networks for LGBT young adults: Low cost strategies for
Mustanski, B., Garofalo, R., Monahan, C., Gratzer, B., & Andrews, R. positive adjustment. Family Relations: An Interdisciplinary Journal of
(2013). Feasibility, acceptability, and preliminary efficacy of an online Applied Family Studies, 64, 420 – 430. http://dx.doi.org/10.1111/fare
HIV prevention program for diverse young men who have sex with men: .12124
The keep it up! intervention. AIDS and Behavior, 17, 2999 –3012. Stajkovic, A. D., & Luthans, F. (2002). Social cognitive theory and
http://dx.doi.org/10.1007/s10461-013-0507-z self-efficacy: Implications for motivation theory and practice. In L. W.
Nadal, K. L., Wong, Y., Issa, M. A., Meterko, V., Leon, J., & Wideman, Porter, G. A. Bigley, & R. M. Steers (Eds.), Motivation and work
M. (2011). Sexual orientation microaggressions: Processes and coping behavior (7th ed., pp. 126 –140). New York, NY: McGraw-Hill.
 100 GOODMAN AND ISRAEL

Sutter, M., & Perrin, P. B. (2016). Discrimination, mental health, and Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
suicidal ideation among LGBTQ people of color. Journal of Counseling dation of brief measures of positive and negative affect: The PANAS
Psychology, 63, 98 –105. http://dx.doi.org/10.1037/cou0000126 scales. Journal of Personality and Social Psychology, 54, 1063–1070.
Tankard, M. E., & Paluck, E. L. (2016). Norm perception as a vehicle for http://dx.doi.org/10.1037/0022-3514.54.6.1063
social change. Social Issues and Policy Review, 10, 181–211. http://dx Willoughby, B. L., & Doty, N. D. (2010). Brief cognitive behavioral family
.doi.org/10.1111/sipr.12022 therapy following a child’s coming out: A case report. Cognitive and
Thoma, B. C., & Huebner, D. M. (2014). Parental monitoring, parent- Behavioral Practice, 17, 37– 44. http://dx.doi.org/10.1016/j.cbpra.2009
adolescent communication about sex, and sexual risk among young men .04.006
who have sex with men. AIDS and Behavior, 18, 1604 –1614. http://dx Worthington, R. L., Dillon, F. R., & Becker-Schutte, A. M. (2005).
.doi.org/10.1007/s10461-014-0717-z Development, reliability, and validity of the Lesbian, Gay, and Bisexual
Villarruel, A. M., Loveland-Cherry, C. J., & Ronis, D. L. (2010). Testing Knowledge and Attitudes Scale for Heterosexuals (LGB-KASH). Jour-
the efficacy of a computer-based parent-adolescent sexual communica- nal of Counseling Psychology, 52, 104 –118. http://dx.doi.org/10.1037/
tion intervention for Latino parents. Family Relations: An Interdisciplin- 0022-0167.52.1.104
ary Journal of Applied Family Studies, 59, 533–543. http://dx.doi.org/
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

10.1111/j.1741-3729.2010.00621.x
Walker, E., & Nowacki, A. S. (2011). Understanding equivalence and Received March 15, 2019
This document is copyrighted by the American Psychological Association or one of its allied publishers.

noninferiority testing. Journal of General Internal Medicine, 26, 192– Revision received October 23, 2019
196. http://dx.doi.org/10.1007/s11606-010-1513-8 Accepted October 24, 2019 䡲

Correction to Roddy et al. (2019)

In the article “Mechanisms of Change in a Brief, Online Relationship Intervention,” by McKenzie
K. Roddy, Caitlin A. Stamatis, Karen Rothman, and Brian D. Doss (Journal of Family Psychology,
advance online publication, August 5, 2019, http://dx.doi.org/10.1037/fam0000569), the effect sizes
listed in the “Mech. slope ¡ CSI slope” and “Indirect effect of treatment” columns of Table 2 were
calculated incorrectly and mistakenly reported as Cohen’s d values. These errors do not change any
of the conclusions of the study. However, the correct values, column heads, and associated footnote
are as follows:

Mech. slope ¡ CSI slope Indirect effect of treatment

B ab/sya

X X
⫺.037 .002
1.140 .129
⫺.381 .021
1.666 .085
⫺1.588 .048
⫺1.304 .057
1.456 .034

⫺.307 .017
.050 .006
2.027 ⫺.023
.033 .000
a
Difference in relationship-satisfaction SD units attributable to indirect effect.

http://dx.doi.org/10.1037/fam0000607