The following is an excerpt of the intervention methodology. For more information, please see the full text of the article on the publisher's website or through your institution's library.

Design
This study adopted a 2 (E-contact Condition: E-contact vs. Control) x 2 (Interaction Partner Sex: Male vs. Female) x 2 (Participant Sex: Male vs. Female) between-subjects, fixed effects design. The E-contact condition (intergroup contact) involved an interaction between a heterosexual participant and a bisexual interaction partner. The control condition (intragroup contact) involved an interaction between a heterosexual participant and a heterosexual interaction partner. Participants were randomly allocated to either the E-contact or control condition with either a male or female interaction partner. The outcome variables were intergroup anxiety, tolerance toward bisexual individuals, and the perceived stability of bisexuality.[...]

Procedure[...]

In the introductory stage, participants were asked by the moderator to introduce themselves to their chat partner by giving their name, occupation, hobbies, “something interesting that not many people know about them,” and what they liked about living in Australia. [...] Sexual orientation was disclosed in the introductory phase. [...]

This was followed by the cooperative stage, which included a collaborative task to prompt cooperation and a common goal, with the support of the chat moderator (authority figure), to support Allport’s [...] conditions for optimum contact. The moderator then invited both the interaction partner and participant to each give contribute ideas that could help Australians achieve better work-life balance. Once both participants gave their contributions, the moderator thanked the participants for their contributions and directed participants back to the survey. Depending on participant response time, the chat interaction lasted approximately 10 to 15 minutes. [...]

Besides the absence of sexual orientation disclosure in the control condition, and sex-typical information (e.g., pre-programmed interaction partner name Sally vs. Joseph), all interactions were identical between experimental conditions (both intergroup and intragroup interactions). To increase chat realism, the interaction included emoticons, an indication of typing time, varied response times, and grammatical errors. Further, both the moderator and the interaction partner addressed the participant by name during the chat (e.g., Hey [Participant]! I’m Sally ☺). [...]

To ensure valid engagement, programming guaranteed participants could not progress until the pre-scripted interaction had been completed. Once completed, participants were redirected to Qualtrics and presented with manipulation check items. To reinforce the idea that participants were taking part in two unrelated studies [...] participants were then told the second study aimed to investigate perceptions of several social groups. Participants completed a brief filler task and then were given a fake random selection item, where they were led to believe choosing one of five items (labeled Group 1–5) would allocate them to a different social group – about which they would complete the remainder of “second” study. In reality, after selection of an item, without their knowledge, all participants were directed to complete the same outcome measures regarding bisexual individuals. Finally, participants were asked to provide demographic information and a concluding open-ended question was given to probe for suspicion regarding the chat interaction or true aims of the study. Participants were then debriefed about the true nature of the study and community participants given the opportunity to enter a prize draw. [...]

Measures [...]

Intergroup Anxiety [...]

Participants were asked to rate the extent to which they would be happy, awkward, confident, self-conscious, defensive, and relaxed in a scenario where they were the only person in a group made up entirely of bisexual individuals, as a measure of outgroup anxiety. An identical set of items was presented regarding heterosexual individuals as a measure of ingroup anxiety. Participants rated all 12 items on a 7-point scale, with scores ranging from Strongly Disagree (1) to Strongly Agree (7). [...]

Bisexual Prejudice [...]
Bisexual prejudice was measured using the Attitudes Regarding Bisexuality Scale (ARBS) as a self-report instrument ([...]. It contains two subscales reflecting two distinct factors. The tolerance subscale contained six items that assessed the degree to which bisexuality was considered a tolerable, moral sexual orientation (e.g., “Bisexuality is immoral”). The stability subscale contained six items that assessed the degree to which bisexuality was perceived as a legitimate, stable sexual orientation (e.g., “Most people who identify as bisexual have not yet discovered their true sexual orientation”). The 12 items were presented in an integrated manner. [...]

Participants rated the extent to which they agreed with each item on a 5-point scale, with scores ranging from Strongly Disagree (1) to Strongly Agree (5).