Being a good neighbor: can students come to value homosexual persons?

Attitudes toward gay/lesbian persons and behavior were initially assessed among Christian college students. Students with either uniformly positive or negative attitudes toward homosexual persons and behavior were then exposed to psychological and spiritual interventions designed to help them see more clearly the value of homosexual persons. Attitudes toward homosexual persons and behavior were then reassessed immediately after the intervention and one month later. Generally, the intervention improved attitudes toward homosexual persons. The picture for attitudes toward homosexual behavior was more complicated. With students who were uniformly rejecting, the intervention made their attitudes toward homosexual behavior less rejecting. However, with students who were uniformly accepting, the intervention diminished their acceptance of gay/lesbian behavior.

**********

Christ was clear, when making the Good Samaritan (Luke 10:30-37) prototypic for being a "good neighbor," that people of faith should value everyone. In the Kingdom of God, there is no "colored section." Yet, in our culture, when it comes to the issue of sexual orientation, it can be argued that homonegativism and perhaps homophobia are pervasive (e.g., Kite & Whitley, 1996). Even within many Christian cultures, the issue of sexual orientation has become emotionally loaded and divisive (Olson & Cadge, 2002; Tapia, 1993). Often, Christians claim to make the distinction between valuing homosexual persons while rejecting homosexual behavior. Yet, some Christians are simply prejudiced when it comes to gay and lesbian persons (Bassett, et al., 2000). The label of "gay" or "lesbian" seems to blind them to the value of some persons. Attempting to be faithful to Christ's commands about "good neighbors," is it possible to encourage prejudiced Christians to be better neighbors toward gay and lesbian persons?

Becoming good neighbors for prejudiced individuals will require a change in attitude. One factor that may help to reduce prejudice toward gay and lesbian persons by heterosexuals is simple contact. Span and Vidal (2003) reported that among undergraduate students the number of homosexual friends was inversely related to levels of homophobia. Admittedly, as the 'friendship' measure and contact theory (Farley, 1982) suggest, simple contact may not be sufficient to reduce prejudice. The conditions of contact may require such things as equal status or working toward common goals for the reduction of prejudice (Lance, 2002).

Researchers have considered other factors that seem to reduce prejudice against gay and lesbian persons. Embracing biological explanations for homosexuality (Altemeyer, 2001; Landen & Innala, 2002), attending human sexuality classes (Cerny & Polyson, 1984; Serdahely & Ziemba, 1984), listening to a class guest speaker who is identified as homosexual (Pagtolun-An & Clair, 1986), exposure to a peer panel of gay and lesbian students (Nelson & Krieger, 1997), and watching films addressing issues of prejudice in general and homosexuality in particular (Goldberg, 1982) have all been shown to reduce prejudice against gay and lesbian persons. For example, the film intervention by Goldberg consisted of showing a film addressing prejudicial issues across a variety of groups, a film of a homosexual clergyman discussing issues related to homosexuality, or two films presenting sexually explicit behavior between same-sex couples. Goldberg reported that students who watched the general film and the film of the homosexual clergyman self-reported greater tolerance toward homosexuality than students who watched the sexually explicit films.

But, understanding and treating prejudice towards gay and lesbian persons may be more complex than the above discussion implies. One consideration in understanding the issue of prejudice against gay and lesbian persons is the issue of religiousness. Apparently, traditional or conservative Christian beliefs predict homonegativism or homophobia (see Hinrichs & Rosenberg, 2002; Plugge-Foust, 2000). Yet, some have argued that this type of finding may be misleading (Bassett, et al., 2000; Bassett, et al., 2002). Measures of prejudice may not take into consideration distinctions that are commonly made in the Christian community when considering the issue of homosexuality. Typically, Christians feel called to affirm the value of all persons. However, when people act in ways that violate perceived Christian values, a common position embraced by many Christians is to affirm the value of that person but reject the value of the behavior. In other words, the Christian will "love the sinner, but hate the sin." Measures of attitudes that gloss over this distinction between persons and behavior for homosexuality may distort how Christians really feel about gay and lesbian persons.

Adding another wrinkle to the issue of prejudice against gay and lesbian persons may be individuals' desire to control prejudicial reactions toward gay and lesbian persons. In the context of racial prejudice, Fazio and his colleagues (Fazio, Jackson, Dunton, & Williams, 1995) developed an instrument to measure this desire to control prejudicial reactions. Bassett, Angelov, Mack, Monfort, & Monroe (2003) have reported an effort to modify this scale to assess the desire to control prejudicial reactions toward gay and lesbian persons. The product of this effort was an instrument that seemed to tap three basic factors: (a) desire to control prejudicial reactions toward gay and lesbian persons, (b) intolerance for gay and lesbian persons, and (c) valuing gay and lesbian persons. It is obvious these three factors may moderate the impact of any efforts to modify prejudicial attitudes toward gay and lesbian persons.

In an effort to combine all of the above issues, we designed a pretest/posttest study. The pretest included a measure of attitudes toward gay and lesbian persons. This measure made it possible to assess attitudes toward homosexual persons separate from attitudes toward homosexual behavior. We used this measure to identify students who were accepting of homosexual persons and behavior and students who were rejecting of homosexual persons and behavior. In addition, we included in the pretest the Bassett et al. (2003) measure designed to assess motivation to control prejudicial attitudes toward gay and lesbian persons. We then introduced interventions designed to enhance attitudes toward gay and lesbian persons that were based upon psychological and spiritual concepts. We followed these interventions with the same measures that were used in the pretest (immediately following the intervention and about one month later). Our prediction was that the intervention would enhance attitudes toward homosexual persons but not homosexual behavior for those students who were initially rejecting of both.

METHOD

Participants

Students were recruited from a variety of psychology classes at a Christian liberal arts college. Those students who were invited to participate in the study were those who had indicated on an earlier questionnaire that they rejected homosexual persons and homosexual behavior (e.g., universally rejecting), or that they accepted homosexual persons and homosexual behavior (e.g., universally accepting). Of those students invited to participate, 56 signed-up for a one-hour session where they understood they would be watching video clips and reading scripture that addressed issues like sexual orientation. Of these students, 48 (86%) actually attended one of the sessions. Of the individuals that attended, 38 students were included in the final sample for the study. The reduction in sample size from 48 to 38 was a function of two things: (a) not all the students who signed-up for the study met the criterion of being universally rejecting or universally accepting (see below for the criteria), and (b) some students were randomly dropped from the sample to create an equal n for both groups.

The 38 students included in the final sample, had the following characteristics: (a) average age = 19.6 years; (b) 22 (57.9%) were female; and (c) 20 (52.6%) were first year students, 12 (31.6%) were sophomores, 5 (13.2%) were juniors, and one student did not indicate class status. In addition, in response to the statement, "I am a Christian," the average response was 7.8 on a 9-point scale (9 = "strongly agree").

Questionnaire

At the beginning of the semester, a questionnaire containing several randomly ordered surveys was given to students in a variety of psychology classes at a Christian liberal arts college. Students received a small amount of class extra credit for anonymously responding to the questionnaire. One of the surveys included within the questionnaire was the Sexual Orientation and Practice (SOAP) scale. The SOAP scale largely reflects the work of Rod Bassett and Christopher Rosik (see Bassett et al., 2003). The instrument contains three subscales: (a) attitudes toward single homosexuals who are sexually active (sexually active homosexuals), (b) attitudes toward single homosexuals who are voluntarily celibate (celibate homosexuals), and (c) attitudes toward single heterosexuals who are sexually active (sexually active heterosexuals). Each subscale contains 5 items that have participants evaluate the target person in a particular context. For example, "I would attend the performance of a person whom I knew was a sexually active homosexual." Participants respond to each item on a 9-point Likert-type scale (1 = "strongly disagree" and 9 = "strongly agree"). Bassett et al. reported respectable Cronbach's alphas for the three subscales ranging from .75 to .84. Comparison of the homosexual and sexually active subscale with the homosexual and celibate subscale makes it possible to differentiate between the valuing of homosexual persons and the valuing of homosexual behavior.

Another survey within the questionnaire was the Motivation to Control Attitudes scale developed by Bassett et al. (2003). This survey represents a modification of the Motivation to Control Prejudiced Reactions Scale (Dunton & Fazio, 1997). The original scale was designed to measure someone's desire to control prejudicial racial attitudes. The modification by Bassett et al. changed the focus of the scale to sexual orientation. The resulting modification produced a 19-item scale with a Likert-type 9-point response format (1 = "strongly disagree" and 9 = "strongly agree"). Principal components analysis with a Varimax rotation revealed the items tapped three basic factors: (a) desire to control prejudicial reactions toward gay and lesbian persons, (b) intolerance for gay and lesbian persons, and (c) valuing gay and lesbian persons.

Procedures

About two months after responding to the initial questionnaire, some students were invited to participate in the intervention study. The invitation to participate was a function of how students responded to the SOAP scale. A median-split was performed on the distribution of average scores for the subscale measuring attitudes toward single homosexual individuals who were sexually active and the subscale measuring attitudes toward single homosexual individuals who were voluntarily celibate. A group of students who were identified as universally rejecting had average scores that were above 3.2 on the sexually active subscale and above 4.6 on the celibate subscale (1 = greatest acceptance, 9 = greatest rejection). Another group of students were identified as universally accepting who have average values below the medians for both subscales. On the original questionnaire, students had also indicated their mother's maiden name and their own birthday. That information was included for both student groups on sign-up lists that were circulated through the classes that had responded to the original questionnaire. Those students identified by their mother's maiden name and their birthdays were then invited to sign-up for the intervention phase of the study. Participation in this phase produced a small amount of class extra credit.

Up to six students could sign-up for a particular one-hour time slot. On the sign-up sheet, students provided their first name and a phone number where they could be reached. Students were called the night before their scheduled participation to remind them of the time and location for the study. When students arrived for the study, they were greeted by one of four student researchers who were not aware of whether the students were universally rejecting or accepting of homosexuals (a mixture of students from both groups signed-up for each session). As students arrived, they were invited to take seats at a large conference table where it would be difficult for them to observe the written work of others.

When all of the students arrived for a particular session, the researcher began by thanking them for participating in the study. The researcher then explained that the first activity would involve watching four scenes from the movie, As Good as it Gets (Brooks & Andrus, 1997). The researcher would provide the background for each scene and then the students would simply watch that scene. The scenes included: (a) a gay man (Simon) being brutalized by burglars, (b) Simon talking with friends in the hospital as he realized the extent of his injuries, (c) Simon discussing with two friends a childhood experience that lead to estrangement from his father, and (d) Simon encouraging his roommate (Melvin) to develop a relationship with a female friend. These particular video segments were selected because they tended to emphasize the humanness of Simon (the gay character in the film). These judgments were made by the researchers (four undergraduate students, a social psychologist, and a counseling psychologist). The total time spent watching the video clip (along with the brief descriptions of the context for each video clip provided by the researcher) was about 10 minutes.

The second activity involved the students reading passages of scripture and then writing out their reactions to those passages. The specific instructions on the cover page were as follows:

Jesus showed love and great concern for people who thought and acted
differently. On the following pages are passages of scripture. After
reading each set of passages please write down your personal reflections
in the space provided. Please expand on your thoughts as much as
possible. In your reflections think about how the passages can be
applied to your actions and attitudes toward people who are gay or
lesbian.

The following three pages were randomly ordered for each student and consisted of a title/theme and passages of scripture (passages were taken from The Message [Peterson, 2002] translation of the Bible) at the top with room for written responses on the rest of the page (students were encouraged to continue their thoughts on the back of the page if needed). The three pages were: (a) Living Love (1 Peter 4:8; Matthew 5:46-47; Matthew 25:35, 37-38, 40; and John 13:34), (b) Living Acceptance (Matthew 9:10-12; John 4:7,9; John 8:3-4, 7, 10-11; and Romans 14:10), and (c) Living Peaceably (Titus 3:1-5).

The final activity involved students responding to the SOAP scale. With all of the written materials, the only identifying information was the student's mother's maiden name, the student's birthday, and a participant number. This information allowed us to connect information from each student while protecting the identity of the student.

The intervention phase of the study took a week to complete. About a month later, all students from the various psychology classes were invited to anonymously respond to the SOAP scale (the only identifying information on the survey was mother's maiden name and student's birthday along with basic demographic information like gender and year in school). This allowed us to determine the endurance of any effects from the intervention phase of the study. Thirty (15 from each group) of the original participants responded to this second post intervention measure.

RESULTS

Because only thirty of the 38 participants in the intervention process responded to the second post intervention measure, the attitude (SOAP) data was analyzed using a series of mixed model ANOVAs. For each attitude object (celibate homosexuals, sexually active homosexuals, and sexually active heterosexuals) two mixed model ANOVAs were performed. The 2 X 2 ANOVAs considered attitudes for universally accepting and universally rejecting students at the pretest and the initial posttest. The 2 X 3 ANOVAs added the second posttest to the analyses. The means for all these analyses are presented in Table 1.

Looking at attitudes toward celibate homosexuals, a 2 X 2 mixed model ANOVA (n = 19 for both groups of students) revealed a main effect for student group, F(1, 36) = 88.85, p [less than or equal to] .001. Universally rejecting students had more negative attitudes than universally accepting students (5.82 vs. 2.69). There was also a main effect for testing time, F(1, 36) = 15.28, p [less than or equal to] .001. Attitudes before the intervention were more negative than attitudes immediately following the intervention (4.67 vs. 3.85). However, both main effects were qualified by a significant interaction, F(1, 36) = 6.75, p [less than or equal to] .05. There was a tendency for all students to be more accepting of celibate homosexuals following the intervention. But, this trend was more dramatic for universally rejecting students than for universally accepting students (see Table 1).

Adding the second posttest to this data created a 2 X 3 ANOVA (n = 15 for both groups of students). Again, the main effect for student group was significant, F(1, 28) = 55.14, p [less than or equal to] .001. Universally rejecting students still had more negative attitudes than universally accepting students (5.60 vs. 2.83). There was also a main effect for testing time (Linear: F[1, 28] = 6.56, p [less than or equal to] .05; Quadratic: F[1, 28] = 11.12, p [less than or equal to] .01). Attitudes became more positive immediately after the intervention (4.78 vs. 3.87) but then shifted back toward the original position with the second post measure (3.99). Again, these main effects were qualified by a significant interaction (Quadratic: F[5.27] = 5.27, p [less than or equal to].05). This tendency, across measurement times, to become more positive and then somewhat drift back, was more dramatic for universally rejecting students than for universally accepting students (see Table 1).

Considering attitudes toward sexually active homosexuals, a 2 X 2 mixed model ANOVA (n = 19 for both groups of students) revealed a main effect for student group, F(1, 36) = 70.17, p [less than or equal to] .001. Universally rejecting students had more negative attitudes than universally accepting students (4.48 vs. 2.18). However, this main effect was qualified by a significant interaction, F(1, 36) = 6.24, p [less than or equal to] .05. Following the intervention, universally rejecting students showed increased acceptance of sexually active homosexuals while universally accepting students showed decreased acceptance (see Table 1).

Adding the second posttest to the design for this data produced a 2 X 3 mixed model ANOVA (n = 15 for both groups of students). Again, there was a significant main effect for student group, F(1, 28) = 28.00, p [less than or equal to] .001. Generally, universally rejecting students had more negative attitudes toward sexually active homosexuals than universally accepting students (4.30 vs. 2.51). There also was a significant interaction between student group and time of measurement (Linear: F[1, 28] = 9.21, p [less than or equal to] .005). Universally accepting students demonstrated diminished acceptance across the three measurement times. Universally rejecting showed diminished rejection following the intervention and then a return to greater rejection with the final measurement (see Table 1).

Looking at attitudes toward sexually active heterosexuals, a 2 X 2 mixed model ANOVA (n = 19 for both groups of students) only revealed a main effect for student group, F(1, 36) = 8.50, p [less than or equal to] .01. Universally rejecting students had more negative attitudes than universally accepting students (3.87 vs. 2.73). Expanding the design by adding the second posttest and creating a mixed model 2 X 3 ANOVA (n = 14 for the universally rejecting students and 15 for the universally accepting students), again produced only a significant main effect for group, F(1, 27) = 4.48, p [less than or equal to] .05. Universally rejecting students were less accepting of sexually active heterosexuals than universally accepting students (3.44 vs. 2.64).

Finally, the design of the study allowed for the consideration of the role of participants' motivation to control prejudicial attitudes toward gay and lesbian persons at three points in the intervention process (pretest, 1st posttest, and 2nd posttest). The scale designed to assess this motivation was included in the initial questionnaire and contained three subscales: (a) desire to control prejudicial reactions toward gay and lesbian persons (desire to control), (b) intolerance for gay and lesbian persons (intolerance), and (c) valuing gay and lesbian persons (valuing). Comparison of the universally rejecting and the universally accepting students on these three subscales revealed some interesting patterns. A one-way ANOVA revealed significant differences between the two groups of students (n = 18 for universally rejecting, n = 19 for universally accepting) for the subscale measuring desire to control prejudicial reactions toward gay and lesbian persons, F(1, 35) = 12.52, p [less than or equal to] .001. Universally rejecting students showed less of a desire to control than universally accepting students (means = 5.86 vs. 7.08 with greatest desire to control prejudicial reactions = "9"). In addition, a one-way ANOVA revealed another difference between the two groups of students (n = 19 for both groups) regarding intolerance for gay and lesbian persons, F(1, 36) = 28.27, p [less than or equal to] .001. Universally rejecting students showed greater intolerance than universally accepting students (means = 6.32 vs. 3.58 with greatest intolerance = "9"). However, there was no difference between the two groups of students regarding the valuing of gay and lesbian persons (means = 7.64 and 7.37).

Looking at the relationship between the subscales for the Motivation to Control Attitudes Scale and the SOAP scales the researchers calculated correlations across all three measurement opportunities. Those correlations are presented in Table 2. Considering this table, several general trends stand out. First, desire to control prejudicial reactions predicted diminished rejection for celibate homosexuals, sexually active homosexuals, and sexually active heterosexuals. Albeit, the trend seemed most clear for celibate homosexuals. Second, intolerance for gay and lesbian persons predicted increased rejection for both categories of homosexual individuals. Again, this trend was stronger for celibate homosexuals. Third, valuing gay and lesbian persons did not predict any of the SOAP measures. Finally, the trends for desire to control and intolerance seemed to be muted by the intervention (the correlations were smaller for the 1st posttest).

To consider the connection between the three Motivation to Control Attitudes subscales and the effectiveness of the intervention, change scores were calculated. For each student, the posttest attitude score toward celibate homosexuals was subtracted from the pretest attitude score. As a result, larger positive change scores indicated increased acceptance after the intervention. Larger negative change scores indicated increased rejection after the intervention. In the same manner, change scores were calculated for attitudes toward sexually active homosexuals. Collapsing together all the students who participated in the intervention, correlations were calculated between the three subscales and the two change scores. Between the subscales and the change scores, there were two significant correlations. There was a significant inverse relationship between desire to control prejudicial reactions and the change in attitude toward celibate homosexuals, r(36) = -.38, p [less than or equal to] .05. Higher levels of desire to control prejudicial reactions toward gay and lesbian persons predicted increased rejection of celibate homosexuals following the intervention. There was a significant direct relationship between intolerance for gay and lesbian persons and change in attitude toward celibate homosexuals, r(37) = .38, p [less than or equal to] .05. Higher levels of intolerance predicted greater acceptance of celibate homosexual individuals following the intervention. Looking within the two groups of students suggested that both these trends were more characteristic of the universally rejecting students (desire to control, r[17] = -.57, p [less than or equal to] .05; intolerance, r[18] = .37, n.s.) than the universally accepting students (desire to control, r[18] = .25, n.s.; intolerance, r[18] = .00, n.s.). There were no significant correlations between the three subscales and change in attitude toward sexually active homosexuals. However, when all the students were combined there was a significant direct relationship between the change scores for attitudes toward celibate homosexuals and the change scores for attitudes toward sexually active homosexuals, r(37) = .47, p [less than or equal to] .01.

DISCUSSION

Students who rejected homosexual persons and behavior and students who accepted homosexual persons and behavior participated in an intervention. The intervention consisted of: (a) exposing students, in a salutary fashion, to a gay man through the medium of film, and (b) having students contemplate passages of scripture that emphasized the value of persons regardless of those persons' behaviors. This intervention was then followed by two posttests that allowed students to differentiate between the value of homosexual persons and homosexual behavior.

Immediately following the intervention attitudes toward homosexual persons (celibate homosexuals) became more positive. This trend was most dramatic for those students who were initially universally rejecting. About a month later, there was somewhat of a trend for these attitudes to drift back to their original position, but for universally rejecting students this shift back was quite minimal.

Following the intervention, attitudes toward homosexual behavior (sexually active homosexuals) showed very different patterns for the two groups of students. Universally rejecting students showed increased acceptance of homosexual behavior while universally accepting students showed decreased acceptance of homosexual behavior. About a month later, this decreased acceptance for universally accepting students continued while the increased acceptance for universally rejecting students leveled off.

Apparently, the intervention worked. But, how it worked depended upon the group of students and the attitude object. With universally rejecting students, the intervention produced less negative attitudes toward homosexual persons and homosexual behavior. Albeit, the change in attitude toward homosexual persons was more than double the change for homosexual behavior. With universally accepting students, the intervention seemed to have more of an effect upon attitudes toward homosexual behavior than homosexual persons. The trend, across both post measurement times, was diminished acceptance of homosexual behavior. The students who attend the college where this study was conducted often come from conservative Christian backgrounds. It may have been that the scriptural intervention reoriented some of those students to a more traditional Christian view of homosexual behavior.

An evaluation of how well this intervention worked really hinges on two issues: (a) did the intervention have an effect?, and (b) was the effect in the desired direction? From the above conclusions, the answer to the first issue seems to be "yes." However, the answer to the second probably depends upon what is perceived as "ideal." If ideal means making a distinction between the value of homosexual persons and homosexual behavior, then the intervention seems to have worked well for both groups of students. But, if ideal is perceived as a greater openness to the value of homosexual persons and homosexual behavior, then the intervention worked less well (especially for the universally accepting students and homosexual behavior).

The trends for subscales from the Motivation to Control Attitudes Scale were also interesting. Generally, desire to control prejudicial attitudes toward gay and lesbian persons predicted less rejection of homosexual persons and behavior. Intolerance toward gay and lesbian persons generally predicted the exact opposite pattern. However, for both subscales these tendencies (especially toward the value of homosexual persons) were dampened immediately after the intervention. Apparently, the intervention moderated the predictive value of desire to control and intolerance in regards to attitudes toward homosexual persons.

The nature of the relationship between the two subscales (desire to control and intolerance) and the attitude change scores (pretest vs. initial posttest) is more complicated. The positive correlation between the change in attitude for sexually active homosexuals and the change in attitude for celibate homosexual makes sense if changes in attitude toward homosexual persons were related to changes in attitude toward homosexual behavior. And, the positive correlation between intolerance and change in attitude to homosexual persons makes sense if the intervention (which included contemplating scripture) was especially effective with students who were intolerant of homosexual persons because of religious reasons. However, what made less sense was the negative correlation between desire to control and change in attitudes toward homosexual persons. Why would students who initially reported a greater desire to control prejudicial reactions toward gay and lesbian persons demonstrate less improvement in their attitudes toward homosexual persons (more acceptance and/or less rejection) after the intervention? Perhaps the answer lies not in students who were high in desire to control but in those students who were low in desire to control. For those students who were initially high in desire to control, our intervention may have been akin to "preaching to the choir." However, for those students who were initially low in desire to control, our interventions may have opened their eyes to the value of homosexual persons. Thus, the significant negative correlation may have been more a function of people low in desire to control being affected by our intervention than people high in desire to control resisting our intervention. Consistent with this line of reasoning was the tendency for this pattern to be most robust for the universally rejecting students (the ones for whom we most hoped to open their eyes to the value of homosexual persons).

Another curious finding was the tendency on the pretest for sexually active homosexuals to be more accepted than celibate homosexuals. A repeated measures ANOVA across both groups of students revealed that this trend was statistically significant, F(1, 37) = 37.29, p [less than or equal to] .001. Both groups of students initially viewed sexually active homosexuals in a more positive light (mean = 3.40) than celibate homosexuals (mean = 4.67). This was somewhat confusing since the descriptions for these categories indicated that the decision for celibacy was anchored in ethical considerations. Presumably, people who anchored their decisions about sexual behavior in ethical considerations would have been perceived in a more positive light than other people of the same sexual orientation for whom ethical considerations were not mentioned. Yet, our choice of the term "celibate" may have had an unintended consequence for some of our students. In the last few years, the Catholic Church has struggled with accusations that some priests have violated their vows of celibacy by having same-sex relations with young parishioners. These accusations have received widespread publicity. Could it be that, for some of our students, using the term "celibate homosexuals" created an unintended connection with this priestly scandal within the Catholic Church? If so, then perhaps our intervention was even more powerful than we anticipated. Not only did it work for "average" individuals, but it also worked for individuals who have been stigmatized in the media.

When considering future research directions, several possibilities come to mind. First, our methodology left unresolved the question of what was the active ingredient that made the intervention work. Was it watching the film segments? Was it contemplating the passages of scripture? Or, was it some combination of the two? Second, our methodology could be easily exported. Would the intervention work with other age groups? Would the intervention work with other branches of the Christian family tree (our students were largely conservative Protestants)? Third, what effect would our intervention have upon people who were selectively accepting? Some individuals respond to the SOAP scale in such a way as to indicate relative rejection of homosexual behavior but relative acceptance of homosexual persons. Would their response to the intervention be like that of universally rejecting students, universally accepting students, or would they plot their own course in response to the intervention? And finally, our project focused upon the impact of our intervention upon explicit attitudes toward homosexual persons and behavior. However, the current popularity of dual-process models in social psychology (see Chaiken and Trope, 1999) raises the issue of implicit attitudes. In addition to affecting thoughtful conscious reactions toward homosexual persons and behavior, would our interventions also affect reactions that are more automatic and unconscious?

Table 1 Mean Attitude Scores of Participants Who Were Universally
Accepting and Universally Rejecting of Homosexual Individuals

             Attitudes Toward Celibate Homosexual Individuals
             Pretest          1st Posttest        2nd Posttest

Universally  6.51(6.56)       5.15(5.03)             (5.23)
 Rejecting
Universally  2.83(3.00)       2.56(2.72)             (2.76)
 Accepting

Mean values could range from 1 to 9 with 9 indicating greatest
rejection. The participants who responded to only the pretest and the
initial posttest produced bold values. Participants who responded to all
three measures produced values in parentheses.

             Attitudes Toward Sexually Active Homosexual Individuals
             Pretest          1st Posttest        2nd Posttest

Universally  4.77(4.77)       4.19(3.89)             (4.24)
 Rejecting
Universally  2.03(2.13)       2.33(2.47)             (2.92)
 Accepting

             Attitudes Toward Sexually Active Heterosexual Individuals
             Pretest          1st Posttest        2nd Posttest

Universally  4.00(3.79)       3.76(3.34)             (3.20)
 Rejecting
Universally  2.89(2.81)       2.57(2.61)             (2.47)
 Accepting

Table 2 Significant Correlations Between the Motivation to Control
Attitudes Subscales and the SOAP Subscales

SOAP Subscales                 Desire to Control  Intolerance  Valuing

                                    Pretest
Celibate Homosexuals           -.61***            .70***
Sexually Active Homosexuals    -.58***            .60***
Sexually Active Heterosexuals  -.42**

                                    1st Posttest
Celibate Homosexuals           -.45**             .57***
Sexually Active Homosexuals    -.46**             .46**
Sexually Active Heterosexuals  -.42**

                                    2nd Posttest
Celibate Homosexuals           -.57***            .62***
Sexually Active Homosexuals                       .38*
Sexually Active Heterosexuals

*Significant at p [less than or equal to] .05 **Significant at
p [less than or equal to] .01 ***Significant at
p [less than or equal to].001

REFERENCES

Altemeyer, B. (2001). Changes in attitudes toward homosexuals. Journal of Homosexuality, 42, 63-75.

Bassett, R. L., Angelov, A. B., Mack, W. J. A., Monfort, K., & Monroe, J. (June, 2003). Spontaneous and deliberative attitudes toward gay and lesbian persons among Christian college students. Paper presented at the annual meeting of the Christian Association for Psychological Studies, Anaheim, CA.

Bassett, R. L., Hodak, E., Allen, J., Bartos, D., Grastorf, J., Sittig, L., & Strong, J. (2000). Homonegative Christians: Loving the sinner but hating the sin? Journal of Psychology and Christianity, 19, 258-269.

Bassett, R. L., Baldwin, D., Tammaro, J., Mackmer, D., Mundig, C., Wareing, A., & Tschorke, D. (2002). Reconsidering intrinsic religion as a source of universal compassion. Journal of Psychology and Theology, 30, 131-143.

Brooks, J. L. (Producer/Director), & Andrus, M. (Writer). (1997). As Good as it Gets [Motion picture]. United States: Gracie Films & TriStar Pictures.

Cerny, J. A., & Polyson, J. (1984). Changing homonegative attitudes. Journal of Social and Clinical Psychology, 2, 366-371.

Chaiken, S., & Trope, Y. (Eds.). (1999). Dual-process theories in social psychology. New York: The Guilford Press.

Dunton, B. C., & Fazio, R. H. (1997). An individual difference measure of Motivation to control prejudiced reactions. Personality and Social Psychology Bulletin, 23, 316-326.

Farley, J. (1982). Majority-minority relations. Upper Saddle River, NJ: Prentice-Hall, Inc.

Fazio, R.H., Jackson, J.R., Dunton, B.C., & Williams, C.J. (1995). Variability in automatic activation as an unobtrusive measure of racial attitudes: A bona fide pipeline? Journal of Personality and Social Psychology, 69, 1013-1027.

Goldberg, R. (1982). Attitude change among college students toward homosexuality. Journal of American College Health, 30, 260-268.

Hinrichs, D. W., & Rosenberg, P. J. (2002). Attitudes toward gay, lesbian, and bisexual persons among heterosexual liberal arts college students. Journal of Homosexuality, 43, 61-84.

Kite, M. E., & Whitley, B. E. J. (1996). Sex differences in attitudes toward homosexual persons, behaviors and civil rights: A meta-analysis. Personality and Social Psychology Bulletin, 22, 336-352.

Lance, L. M. (2002). Heterosexism and homophobia among college students. College Student Journal, 36, 410-414.

Landen, M., & Innala, S. (2002). The effect of a biological explanation on attitudes towards homosexual persons: A Swedish national sample study. Nordic Journal of Psychiatry, 56, 181-186.

Nelson, E. S., Krieger, S. L. (1997). Changes in attitudes toward homosexuality in college students: Implementation of a gay men and lesbian peer panel. Journal of Homosexuality, 33, 63-81.

Olson, L. R., & Cadge, W. (2002). Talking about homosexuality: The views of mainline protestant clergy. Journal for the Scientific Study of Religion, 41, 153-167.

Pagtolun-An, I. G., & Clair, J. M. (1986). An experimental study of attitudes toward homosexuals. Deviant Behavior, 7, 121-135.

Peterson, E. H. (2002). The message: The bible in contemporary language. Colorado Springs, CO: Navpress.

Plugge-Foust, C. (2000). Homophobia, irrationality, and Christian ideology: Does a relationship exist? Journal of Sex Education & Therapy, 25, 240-244.

Serdahely, W., & Ziemba, G. J. (1984). Changing homophobic attitudes through college sexuality education. Journal of Homosexuality, 10, 109-116.

Span, S.A., & Vidal, L.A. (2003). Cross-cultural differences in female university students' attitudes toward homosexuals: A preliminary study. Psychological Reports, 92, 565-572.

Tapia, A. (1993). Homosexuality debate strains campus harmony: Homosexuals at Christian colleges press for acceptance. Christianity Today, November 22, 38-40.

RODNEY L. BASSETT, MARIKE VAN NIKKELEN-KUYPER, DEANNA JOHNSON, ASHLEY MILLER, ANNA CARTER, and JULIA P. GRIMM

Roberts Wesleyan College 

AUTHORS

BASSETT, RODNEY L.: Address: Roberts Wesleyan College, 2301 Westside Drive, Rochester, NY 14624. Title: Professor of Psychology. Degree: PhD, Ohio State University. Specializations: Social Psychology, Psychology of Religion, Integration of Psychology and Christianity.

VAN NIKKELEN-KUYPER, MARIKE. Address: Roberts Wesleyan College, 2301 Westside Drive, Rochester, NY 14624. Degrees: BA, Psychology; BS, Biology, Roberts Wesleyan College.

JOHNSON, DEANNA. Address: Roberts Wesleyan College, 2301 Westside Drive, Rochester, NY 14624. Degree: BA, Psychology, Roberts Wesleyan College.

MILLER, ASHLEY. Address: Roberts Wesleyan College, 2301 Westside Drive, Rochester, NY 14624. Degree: BA, Psychology, Roberts Wesleyan College.

CARTER, ANNA. Address: Roberts Wesleyan College, 2301 Westside Drive, Rochester, NY 14624. Degree: BA, Psychology, Roberts Wesleyan College.

GRIMM, JULIA. Address: Roberts Wesleyan College, 2301 Westside Drive, Rochester, NY 14624. Title: Professor of Psychology. Degree: PhD, Iowa State University. Specialization: Counseling Psychology.

Correspondence concerning this article should be addressed to Rodney L. Bassett, Roberts Wesleyan College, 2301 Westside Drive, Rochester, NY 14624. Email: [e-mail]