Development and Validation of Videotaped Scenarios

Abstract

Researchers using scenarios often neglect to validate perceived content and salience of embedded stimuli specifically with intended participants, even when such meaning is integral to the study. For example, sex and aggression stimuli are heavily influenced by culture, so participants may not perceive what researchers intended in sexual aggression scenarios. Using four studies, the authors describe the method of scenario validation to produce two videos assessing alcohol-related sexual aggression. Both videos are identical except for the presence in one video of antiforce cues that are extremely salient to the young heterosexual men. Focus groups and questionnaires validate these men's perceptions that (a) the woman was sexually interested, (b) the sexual cues were salient, (c) the antiforce cues were salient (antiaggression video only), and (e) these antiforce cues inhibited acceptance of forced sex. Results show the value of carefully selecting and validating content when assessing socially volatile variables and provide a useful template for developing culturally valid scenarios.

The advantage of studying social behavior in the laboratory is that it allows one to isolate and systematically vary specific factors that are often embedded, infrequent, illegal, unethical, and/or unpredictable in the natural social environment. Of course, such “laboratory analogues” also raise the question: How well does behavior in a laboratory setting predict behavior in the natural environment? This dilemma has spawned several types of laboratory methods intended to make the study of social situations as externally valid as possible while preserving laboratory control. In studying sexual aggression, specifically acquaintance rape (Abbey, 2002), one frequently employed method is the use of scenarios. Typically, the participant reads a short description of a situation and is asked to respond to a series of questions. The independent variable(s) within the story can be manipulated through embedded cues, and the use of a between-respondents design provides an indication of the variables’ effects. Recently, technological advances have expanded scenario methods to include auditory presentation, video presentation, and even computerized presentation that can vary with the participant's responses (e.g., Maisto, Carey, Carey, Gordon, & Schum, 2004). A video presentation, for example, may contribute significantly to verisimilitude, with the increase in generalizability offsetting the additional expense of development (cf. Johnson, Noel, & Sutter-Hernandez, 2000).

However, the problem of validation with the scenario method cannot be resolved by technology. Researchers still need to demonstrate that the situations presented in the stories are realistic, salient, and meaningful to the specific participant who is expected to respond realistically to the scenario. The researcher may present the participant with a situation that perfectly reflects reality, but if the participant “doesn't buy it,” his or her in vitro reaction will not be a good analogy to an in vivo reaction. Thus, the need is not to produce generalizable scenarios but instead scenarios that are specifically suited to the population being studied.

This problem is significant in an area of study such as sexual aggression. Sexual aggression is a particularly difficult topic to study as a laboratory analogue for ethical and practical reasons. Sexual aggression, when it occurs in the everyday environment, is private, socially proscribed, and determined and defined by culture, gender, and age cohort. Yet in many studies using written, taped, or videotaped scenarios of sexual aggression, researchers often write scenarios without validating the perceived content and the salience of the embedded cues specifically with the intended study population (i.e., those who are asked to make judgments of the situation). Most scenarios in these studies appear to have been developed based on what the researchers believed represented a common acquaintance rape incident, but the scenario may not be perceived by the specific population under study as what the researcher believes it represents.

In this article, we present a series of four studies illustrating our development and validation of two videotaped scenarios for assessing young men's judgment of sexual aggression situations. This series is presented in the spirit of suggesting a framework that other researchers could use in developing scenarios specific to the populations they wish to study. Our method is based on a three-step series of studies successfully used by Maisto, Carey, Carey, and Gordon (2002), Maisto, Carey, Carey, Gordon, and Schum (2004), and Maisto, Carey, Carey, Gordon, Schum, and Lynch (2004), to develop and validate videotaped scenarios to research the effects of alcohol on condom use in heterosexual situations. Maisto and his colleagues used this method to develop one set of scenarios for women and a different set of scenarios for men.

Background of the Example Alcohol and Sexual Aggression Study

We developed the scenarios described in this article exclusively for use in a laboratory study of the effects of alcohol intoxication on young men's judgments of sexual aggression. A brief explanation of the laboratory study follows. (A more detailed description of that study has been presented by Noel et al., 2003). A common assumption is that intoxication contributes to sexual aggression by impairing a man's normal prosocial inhibition to use force to have sex (e.g., Barbaree & Marshall, 1991), but mechanisms involved in the process are widely debated. We have proposed an “Alcohol myopia” (Steele & Josephs, 1988) explanation, hypothesizing that an intoxicated male in a heterosexual interaction will focus on the most salient cues in the situation (usually stimuli they interpret as sexual cues) and decrease attention to other cues (stimuli they interpret as inhibiting the use of force to have sex). To test this hypothesis, our laboratory conducts a double-blind randomized dose alcohol administration study in which male participants (aged 21 through 30) consume a beverage and then view one of two videotaped scenarios depicting a man and a woman on a first date. In one scenario, we intend that only sexually explicit cues are salient. In the second scenario, we intend these cues to be accompanied by salient antiforce cues. Alcohol myopia versus disinhibition explanations generates differential predictions about participants’ judgments of the man's behavior. Most important to this article is the point that the young male participants need to perceive the sexual and antiforce cues as we intended, regardless of the real or ecological validity of the cues. In future research, we expect to examine women's judgments and will use this same method to generate video scenarios specific to their perceptions.

The validation study series had to produce two videos that met our study criteria. We drew material exclusively from the identical population that would be used in the alcohol administration study: heterosexual male non-problem drinkers in the target age range (21 through 30). The validation issues considered were that they perceived (a) the woman in both videos as flirting and showing sexual interest, (b) the sexual cues as salient in both videos, (c) the antiforce cues as salient in the antiaggression video, and (d) the antiforce cues as truly inhibitory to acceptance of forced sex.

Study 1: Eliciting Salient Cues for Sexual Interest from the Target Population

Method

Participants

Males, 21 to 30 years old (M age = 23.2, SD = 1.2), responded to advertisements placed in the local community for male heterosexual regular drinkers to participate in a focus group in which they would be asked about heterosexual dating and alcohol. All participants were light, moderate, or heavy drinkers (no abstainers or infrequent drinkers) as self-reported on a modified Quantity-Frequency Index (QFI; adapted from Cahalan, Cisin, & Crossley, 1969). In addition, we screened out those with alcohol problems through the use of the Rutgers Alcohol Problem Index (RAPI; White & Labouvie, 1989) and those with severe psychiatric symptoms reported on the Symptom Checklist-90-Revised (SCL-90-R; Derogatis, 1977). Twenty-five men were screened and defined as eligible to participate in the research, but only 20 (19 Caucasian, 1 Latino) participated in focus groups. We do not have further information on the 5 people who did not participate.

Procedure

A male graduate student conducted the focus groups, assisted by a male undergraduate who audio-recorded and took notes on the group proceedings. At the beginning of each session, participants were breath-tested (Alco-Sensor III from Intoximeters) to verify that all blood alcohol concentrations (BACs) were 0. The participants (Group 1, n = 4; Group 2, n = 5; Group 3, n = 5; Group 4, n = 6) had not been acquainted before the focus group meetings. All signed a consent form before the meeting, stating that all information obtained during the focus groups would be kept confidential. They were also asked to refrain from using names during the discussions, which would be recorded. Recordings were made and used by the primary author only to verify the accuracy of the notes taken by the assistant in the room. Recordings were erased at the end of the study.

Participants were asked to discuss the following questions for at least 20 minutes each:

• Question 1: Under what circumstances would a man expect that a woman was likely to have sex with him on a first or early date?

• Question 2: What are the signals a woman would use to make it clear that she was interested in sex with a man on a first or early date (what would she say or do)?

• Question 3: Under what circumstances would it be clear that the woman was interested in the man, but did not want to have sex on the first date? (What would she say or do to convey this?).

Focus group sessions were held in a comfortable lounge-type room with snacks and soft drinks available. In all four groups, after an initial brief ice-breaking period, discussion was lively (and sometimes raucous). The moderator introduced each question and helped keep the discussion going. He also ensured that each participant had time to address each topic.

Each participant was paid $15 at the end of the group session.

Results

Participants agreed that it was easy for them to answer the first two questions (“Under what circumstances would a man expect that a woman was likely to have sex with him on a first or early date?” and “What are the signals a woman would use to make it clear that she was interested in sex with a man on a first or early date?”). Lists were compiled of elements that were discussed and agreed on in all four groups by at least 50% of the group's members. The consolidated answers for the first two questions included:

• — Woman is dressed to “look good” but not too sexy.

• — They are in her apartment (important, because if they are in his apartment, she can go home; but if she invites him into her place, they believed she was inviting him in for sex).

• — They were at a party, bar, or some other group activity, and she singled him out, asked him home to her place so they could be alone together.

• — She can't resist touching him, plays with his hair, puts her hand on his arm or on his chest.

• — She makes frequent eye contact.

• — She asks him for a backrub or other physical contact.

• — She initiates kissing with him, and both he and she appear to enjoy it.

• — She initiates discussion of sex, but in a subtle kind of way, as in “testing the waters.”

• — Some physical contact involving “playing with force” takes place. For example, she steals his wallet to “look at the pictures” and he tries to grab it back. They wrestle on the sofa but end up kissing and groping each other.

• — She tells him she likes men who know what they want; men who are in control.

Question 3 (“Under what circumstances would it be clear that the woman was interested in the man, but did not want to have sex on the first date?”) was difficult for participants to answer in a positive way. Most participants felt that the way she would convey her desire to wait would be through the absence of most of the elements above. Just saying “no” would apparently be too strong and not convey the message that the potential for sex at a later time still existed.

An alcohol myopia effect would only be seen in ambiguous situations—those in which the person had to weigh the consequences of several competing actions—so our dilemma was to develop a situation in which the participants perceived that there was a potential for sexual interaction (she might want it) while also perceiving equally salient cues suggesting that using force to have sex would be unacceptable. Previous research in our laboratory (Johnson et al., 2000) showed that the woman's flat statement, “I don't want to have sex” would not have been useful as a cue, because there would be no ambiguity about sex at that point. To develop and incorporate some tangible, salient antiforce cues, we turned to the literature (e.g., Johnson, Benson, Teasdale, Simmons, & Reed, 1997; Johnson et al., 2000; Johnson & Russ, 1989) and suggested some cues that might be construed as mitigating against force but not necessarily against sex. These included the following:

• — She discusses a “women's issues” course or some similar topic.

• — She has posters that promote women's issues or women's rights (e.g., a poster for the National Organization for Women).

• — She is wearing a T-shirt that reads “Rape Crisis Center Volunteer” in letters big enough to be noticeable immediately.

Most participants agreed that these cues were very salient antiforced-sex cues. However, because these cues were not generated by the focus groups, we decided that they required further testing with the target audience to determine if they were perceived as inhibitory to aggression (see Study 4, below).

Discussion of Study 1

This study clearly shows the value of focus group content generation. The participants were often very specific about what they perceived as sexual cues and innuendos. Other populations (e.g., women and men of other age groups) might not have been able to predict the reactions of young men. For example, note that participants distinguished between the man's versus the woman's residence as the setting. They believed that if the woman went to the man's apartment, her sexual intentions were unclear. “She could still decide to leave and go home,” said one participant. In contrast, if she invited him to her place, they felt she was signaling a strong sexual desire. Furthermore, participants appeared unable to articulate signals from the woman that could indicate both a strong interest in the man but a desire to wait in terms of sexual intercourse. This seems to be a likely source of potential miscommunication between men and women in early dating.

For our next step, we engaged a professional video company to write scripts incorporating the content generated by the focus groups. The primary author and the two male graduate student research assistants met with the scriptwriter several times as the scenarios were being developed. Notably, the research assistants and the scriptwriter were all males in the target age range. When the scripts were completed, we began Study 2.

Study 2: Assessment of Scripts for Salience of Cues to Target Audience

Method

Participants

Twenty-two men (aged 21 through 30; M = 23.2; SD = 2.5; 21 Caucasian, 1 Latino) responded to the same advertisements and were screened for alcohol and psychiatric problems with the RAPI and SCL-90-R. None had participated in Study 1. All were regular alcohol drinkers as reported on the QFI. Unfortunately, the Latino man had to be excluded because of difficulty reading the vernacular English, leaving 21 participants.

Procedure

Each participant was tested individually. When he entered the laboratory, a male research assistant breath-tested him to assure that his BAC was 0 (Alco Sensor III) and then read him the following instructions:

In this laboratory, we are studying people's opinions about dating and sex. In an upcoming study, we want to show a short video to our participants and then ask them to make ratings about the behavior of the people in the video. We have a script for the video, but before we actually shoot the video, we want to make sure that people who see it will remember its contents. That is why we have asked for your participation.

What we will be asking you to do is to read the script of the video carefully, as though you were watching the finished story. After you have completed your reading, we will ask you some questions about the script: first, what you remember, and then, your opinions about the people in the story. The story is about two young people, Kip and Jenn, having a social encounter. Please take the time to read it carefully.

The research assistant then left the room so the participant could read in private. He monitored the participant through a one-way window and returned when the participant was finished reading.

Scripts

The participant read one of two scripts, randomly assigned. In both cases, the scripts depicted a young couple (Jenn and Kip) entering her apartment after she had singled him out from a crowd of friends at a bar. Her apartment is decorated with posters, candles, framed photos, and feminine knick-knacks. She invites him in and they discuss again the fact that she singled him out at the bar. Jenn cuddles next to Kip on the sofa. Among other topics, they discuss classes she is taking at the university. Jenn requests that Kip get her a glass of wine while she changes into “something more comfortable.” She leaves the bedroom door open just enough that he can glimpse her changing clothes. She returns wearing a T-shirt and very short shorts. Jenn asks for a backrub, then progresses to a little wrestling match on the sofa to try to grab his wallet (she says she wants to look at his driver's license photo). He pushes her down on the sofa, kisses her, and begins to try to have sex with her. As he tries to pull down her shorts, she says to stop. He continues, pushing her hands away, and as the camera fades away, he is clearly overpowering her while she sounds progressively more distressed.

In the Antiforce Cues script (n = 11), three antiforce stimuli were embedded: (a) the posters on her walls were feminist in theme (e.g., a National Organization for Women convention poster); (b) the class Jenn and Kip discussed was for her women's studies major; and (c) the T-shirt into which she changed was red with “Rape Crisis Center” emblazoned on the back. In the contrasting No Cues script (n = 10), the corresponding stimuli were (a) posters of kittens and a rock band; (b) a psychology class in animal behavior; and (c) a plain, white T-shirt.

Responses to the scripts

The research assistant returned after the participant signaled that he had finished reading (usually about 10 to 15 minutes) and asked the following questions to assess recall of the embedded cues:

1. What posters were displayed in Jenn's apartment?

2. What class was Jenn taking at the university?

3. What was Jenn wearing in the second half of the script?

Answers were copied verbatim and later were scored for accuracy. The research assistant remained impassive so as to provide no feedback to the participant on the accuracy of his recall.

Next, to conduct a pilot assessment of whether cues in both scripts were equally effective in conveying Jenn's sexual interest and whether both were perceived as equally realistic and believable, despite the different antiforcedsex cues, the research assistant gave the participant a questionnaire and again left the room so that the participant could fill it out in private. Four of the items were statements, each followed by a 7-point Likert-type scale anchored with 1 = strongly disagree and 7 = strongly agree.

• Item 1. Jenn was flirting with and teasing Kip.

• Item 2. In this situation, Kip should try to have sex with Jenn.

• Item 3. This script was realistic.

• Item 4. This script was believable.

Finally, embedded in the questionnaire were these two items, intended to assess the participant's judgment of fault or responsibility for the sexual aggression:

• Item 5. Assume that Jenn and Kip had sexual intercourse; to what extent is this Jenn's responsibility?

• Item 6. Assume that Jenn and Kip had sexual intercourse; to what extent is this Kip's responsibility?

For these items, participants placed an X on a continuous 6-inch line with none as the anchor point on the left side and 100% as the anchor point on the right side. These were scored by calculating the distance, in inches, from the beginning of the line (none) to the X.

After the participant had completed these items, he was debriefed by the research assistant (a male graduate student) who was supervised by the primary author (a doctoral clinical psychologist). Debriefing included a statement and discussion of the fact that forced sex is wrong, both legally and morally. Participants were thanked for contributing to a study that might help prevent forced sex from occurring in the future. Each participant was paid $15.

Results

Responses to the scripts were examined with the intention of determining whether they were ready for videotaping. As noted earlier, our goal was that the sexual interest cues be salient in both scenarios, with no difference in these cues between conditions; and in the second scenario only, we wanted the antiforce cues to be salient. We also wanted both scripts to be rated highly believable and realistic, again with no difference between scripts on these ratings.

Evaluation of sexual interest cues

Participants’ responses to the overall scenarios are shown in Table 1. All responses to these questions were on a 1 = strongly disagree to 7 = strongly agree scale. For both scripts, the mean rating for flirting was 6 and the minimum was 4, so the participants appeared to agree that sexual interest cues were salient. Furthermore, some participants in each condition gave a positive rating to the question, should Kip try to have sex with Jenn. Believability and realism were also highly rated, with a mean between 5 and 6.

Table 1.

Ratings of the Scripts in Study 2^a

	Antiforce Cues				No Antiforce Cues
Item	M rating	SD	Min	Max	M rating	SD	Min	Max
1. Jenn was flirting	6.0	0.9	4.0	7.0	6.0	1.1	4.0	7.0
2. Kip should have sex	2.7	2.1	1.0	7.0	2.4	1.6	1.0	5.0
3. Story was realistic	5.4	1.2	3.0	7.0	6.1	1.1	4.0	7.0
4. Story was believable	5.8	1.2	3.0	7.0	6.3	1.0	4.0	7.0
5. Kip's responsibility	4.1	1.2	2.3	5.7	4.1	1.2	2.5	5.9
6. Jenn's responsibility	2.1	1.1	0.4	3.4	2.5	1.5	0.4	5.9

^aOn items: 1 = strongly disagree to 7 = strongly agree; on responsibility, 0 = none to 6 = 100%.

In the result, t tests showed no significant differences in any of these ratings between conditions (all ps nonsignificant). Most participants felt that both scripts depicted believable and realistic situations in which the woman was flirting and the man might be justified in continuing efforts to have sexual intercourse with her.

Salience of antiforce cues

The 11 participants who read the Antiforce Cues script recalled the class and the red “Rape Crisis Center” T-shirt with 100% accuracy and no hesitation. However, the women's rights posters were forgotten by about half the participants. The 10 participants in the No Cues script condition also had trouble recalling some of their posters, indicating that the posters were not salient in either script. We felt that because the posters were only mentioned at the beginning of the script as part of the room description, they would need more emphasis in both scenarios. The script was changed so that the camera would focus on the posters several times and Jenn and Kip would actually discuss them at one point. Because they are mostly visual cues, we believed that the increased camera time and discussion would make them more salient in both scenarios. In the no cues condition, the white T-shirt and the animal behavior classes were recalled with 100% accuracy.

Attribution of responsibility

Recall that levels of responsibility were measured separately for Jenn versus Kip on a continuous scale on 0 (no responsibility) to 6 (100% responsibility). These data also appear in Table 1. Participants appeared to be consistent in assigning more responsibility to Kip than Jenn, but again, t tests between conditions showed no significant differences between the scripts on these two ratings (ps nonsignificant).

Discussion of Study 2

The important finding in Study 2 is that the cues intended to inhibit forced sex were salient, even though they were embedded in the script and they were easily distinguished from those in the no cues script. In addition, participants indicated that the sexual interest cues were salient equally and at a high level in both scripts. Furthermore, they considered both scripts believable and realistic.

Pilot data on the levels of responsibility attributed to the man versus the woman were also informative. Participants did not, in most cases, give responses that added up to 100%, so separating the two questions (Jenn's responsibility vs. Kip's responsibility) appeared to be the most valid method of assessing responsibility attribution. Furthermore, it is important that most participants were willing to attribute some level of responsibility to Jenn. When we use the videos to assess for an alcohol myopia effect, we will test the hypothesis that participants in the cues condition, when given the higher alcohol dose, will be more aware of the antiforce cues than those in the no alcohol condition and will attribute even less responsibility to Jenn. If the ratings of Jenn's responsibility in this pilot, no-alcohol study were at 0, a floor effect would obscure the results and provide an incomplete test of the alcohol myopia hypothesis.

Of course, these scenarios were created to be presented audiovisually in the experimental study, so we conducted a parallel validation study with the two videos once they were completed. Adjustments were made in the scripts (including more emphasis on Jenn's posters). The video company shot the scenarios with professional actors and crew with the primary author as an on-set consultant. The videos, about 10 minutes each, are identical, except for the scenes depicting the cues.

Study 3: Assessment of Videos for Salience of Cues

Method

Participants

Nineteen men (M age = 24.1, SD = 2.1; 17 Caucasian, 1 African American, 1 Asian American) who passed the same screening criteria as described earlier participated individually in the video validation study. Again, none of them had participated in the previous studies.

Procedure

To assess whether the cues were both salient and memorable in the audiovisual productions, a slight variation was used in this study. The male experimenter told each participant that the two videos were a test of social memory. The participants were asked to observe carefully and try to remember the details. The experimenter left the room after handing the participant two envelopes. Participants watched both videos, with the order randomly selected. Ten participants saw the Antiforce Cues video first and 9 saw the No Cues video first. At the end of each video, the participant opened an envelope. In the first envelope, were the six statements as in Study 2 (e.g., “Jenn was flirting with and teasing Kip.”) and the same response scales. In the second envelope was the same questionnaire along with a question: “What were the differences between the two videos? Please describe these as specifically as possible.” Participants were given as much time as they wanted to write their answers. When they finished, participants signaled the experimenter. The experimenter then debriefed each participant and paid him $15.

Results

In response to the final question, all participants noted the differences accurately. Two wrote that the posters were different but did not specify what that meant. Otherwise, all participants noted the cue differences in detail.

Responses to the videos

In contrast to Study 2, all 19 participants in Study 3 viewed both videos and made ratings of sexual perceptions and believability after each. However, we expected a pattern of results similar to those from Study 2 (the script study). One-way analyses of variance with repeated measures showed no differences between these videos on any of these ratings (ps nonsignificant). Ratings are shown in Table 2.

Table 2.

Ratings of the Scripts in Study 3^a

	Antiforce Cues				No Antiforce Cues
Item	M rating	SD	Min	Max	M rating	SD	Min	Max
1. Jenn was flirting	6.2	0.7	5.0	7.0	5.9	1.1	5.0	7.0
2. Kip should have sex	3.3	2.1	1.0	6.0	3.5	2.0	1.0	5.0
3. Story was realistic	5.5	2.2	4.0	7.0	5.0	1.6	4.0	7.0
4. Story was believable	4.7	1.2	3.0	7.0	4.8	1.0	3.0	7.0
5. Kip's responsibility	114	24	66	140	111	26	70	140
6. Jenn's responsibility	35	31	0	65	42	34	0	70

^aOn items: 1 = strongly disagree to 7 = strongly agree; on responsibility, 0 = none to 6 = 100%.

Attribution of responsibility

We changed the scale of the attribution of responsibility only by measuring in millimeters rather than inches. The line was similar (140 mm in length) and was scored according to how far the X was placed from the left side (none). One-way analyses of variance with repeated measures found no differences between the videos on attributions of Kip's responsibility or Jenn's responsibility (ps nonsignificant).

Discussion of Study 3

Most important, the results of Study 3 again provided evidence that the antiforce cues were quite salient and memorable. Every one of the 19 participants accurately noticed and remembered the crucial differences between the videos. In addition, although the participants clearly perceived content differences between the two videos, they did not react differentially to the sexual cues. Jenn was perceived as flirting, and there were equal positive ratings on the “Kip should try to have sex with her” question. Both videos were rated as relatively believable and realistic with no difference between videos on these scales.

One final question remained in regard to the three antiforce cues and was assessed in Study 4.

Study 4: Are Antiforce Cues Inhibitory?

This study was designed to provide empirical support that the three cues exclusive to the antiforce video would indeed be perceived by young male participants as inhibitors to the use of force to have sex.

Method

Participants

Twenty men (aged 21 through 30, M age = 24.1; SD = 2.3; 17 Caucasian, 2 African American, 1 Latino), screened as described earlier, volunteered for this study. None were in the previous studies. All were breath-tested for alcohol at the beginning of the session to confirm that their BACs were 0.

Procedure

Each participant was tested individually. Participants signed an informed consent and then viewed the Antiforce Cues video while sitting individually in the laboratory. The video lasted approximately 10 minutes. At the conclusion of the video, the experimenter returned and asked the participant to write responses to three questions. The participant completed this questionnaire and placed it in a manila envelope to ensure his confidentiality.

The three questions were as follows:

1. What do the posters do to the chance (or probability) that Kip will try to force Jenn to have sex?

2. What does the Rape Crisis Center T-shirt do to the chance (or probability) that Kip will try to force Jenn to have sex?

3. What does the book on women's studies do to the chance (or probability) that Kip will try to force Jenn to have sex?

Each question was followed by a 7-point Likert-type scale (–3, –2, –1, 0, +1, +2, +3) with these three anchors (–3 = decreases; 0 = doesn't affect; +3 = increases). Each participant was debriefed and paid $15 at the end of the session.

Results

The first question, “What do the posters do to the chance (or probability) that Kip will try to force Jenn to have sex?” resulted in a negative mean response (M = –1.23, SD = 1.15). The second question, “What does the Rape Crisis Center T-shirt do to the chance (or probability) that Kip will try to force Jenn to have sex?” also resulted in a negative mean score (M = –1.95, SD = 1.15). On the final question, “What does the book on women's studies do to the chance (or probability) that Kip will try to force Jenn to have sex?” the mean response was slightly negative (M = –0.8, SD = 1.15), but in no case did any participant say that any of these cues increased the probability of forced sex (i.e., the highest response from any participant was 0), and all said that at least one of the three cues decreased the probability.

Discussion of Study 4

The results of Study 4 suggest that all nonforce cues in the video were perceived as inhibitors to forced sex in the scenario. No responses were rated higher than a 0 on the Likert-type scales, suggesting that the cues act primarily as antagonist to forced sex. Accepting these results as validation that the nonforce cues in the video act as inhibitory stimuli indicates that the videos can be used in the alcohol administration study and not reduce its internal validity.

General Discussion

This article described a series of four studies intended to produce and validate two videotaped scenarios meeting specific criteria required to test an alcohol myopia model of sexual aggression exclusively with young male participants. To test the model, the two videos needed to present stimuli that elicited the men's sexual interest at an identical level in both videos, while also meeting the requirement that one video also had to present very salient antiforced sex cues. Furthermore, we needed to show that the specified cues were relevant to the target participant population (males aged 21 through 30) and were not just the researchers’ sense of what the target group ought to be perceiving.

The four studies generated the content, tested the salience of the cues, first in a written and then in an audiovisual presentation, and then assessed if the cues were performing as intended (i.e., were the antiforced sex cues really inhibitory for forced sex?). In all four studies, the participants were drawn from the population to be used later in the experimental study, and their sobriety at the beginning of each research session was ascertained with a breath test. Furthermore, contact with researchers was minimized to limit socially biased answers. All research assistants who had contact with the participants were males in the same age group.

We did not involve women in creating these scenarios, because our intention was to focus strictly on the male perspective. However, as we develop the companion study for women, using the same procedure to create their scenarios, a contrast between the two viewpoints is likely to be evident that might provide some insight into male and female miscommunication in regard to sexual intentions.

This article shows the value of carefully selecting and validating content to assess socially volatile variables. Its importance lies in allowing our experience to provide a template for researchers to develop specific scenarios as vehicles for testing hypotheses with specific populations. Using this same procedure, Maisto et al.'s laboratory has developed sets of realistic and believable scenarios of unsafe sexual behavior, one specifically for use with heterosexual women, one for use with heterosexual men, and one for men who have sex with men (Maisto et al., 2002; Maisto, Carey, Carey, Gordon, & Schum, 2004; Maisto, Carey, Carey, Gordon, Schum, & Lynch, 2004).

The alcohol administration study based on the scenarios described in this article is in progress. Informal anecdotal evidence supports the value of these scenarios with this population. In one experimental session, for example, an intoxicated participant was overheard talking to and shouting at the characters in the video. In addition, continued formal assessment conducted during debriefing supports the contention that the videos are perceived as intended. Men viewing the videos are rating them as believable and realistic and spontaneously inform the interviewer doing the debriefing that “this happens more often than you'd think.” We will continue to assess the utility of the scenarios throughout the alcohol administration study to provide a follow-up report. In addition, using this same procedure, we are developing separate scenarios for a women's study.

Biography

Nora E. Noel is a professor of psychology at the University of North Carolina–Wilmington. She received her PhD in clinical psychology from SUNY–Binghamton in 1983. Her research and clinical interests are on the involvement of substance use and substance use disorders in friendship and couples’ relationships.

Stephen A. Maisto is a professor of psychology at Syracuse University. He received his PhD in experimental psychology in 1975 from the University of Wisconsin–Milwaukee and completed a postdoctoral respecialization in clinical psychology in 1985 at George Peabody College of Vanderbilt University. His main research and clinical interests are the assessment and treatment of the substance use disorders and substance use and sexual risk.

James D. Johnson is a social psychologist and a professor in the Department of Psychology at the University of North Carolina–Wilmington. He received his PhD from the Indiana University in 1984. His program of research involves the media, stereotypes, and the role of stereotypes in healthcare decision making.

Lee A. Jackson, Jr. is a professor of psychology at the University of North Carolina–Wilmington. He received his PhD in social and community psychology from the University of Florida in 1972. His current research centers on alcohol use and sexual aggression as well as stereotyping and stereotype threat.

Christopher D. Goings is currently a graduate student in psychology at the University of North Carolina–Wilmington, where he received his BA in Psychology in 2002 and his MA in 2007. His current research is on the assessment of alcohol use by college students.

Brett T. Hagman is currently a doctoral candidate in the University of Medicine and Dentistry of New Jersey School of Public Health. He received his MA in psychology from the University of North Carolina–Wilmington in 2004. His research is in assessment of substance use and consequences, particularly with college students.