Perceptions of Positive Relationship Traits in
Gay and Lesbian Couples
Miki D. Skinner: McNair Scholar
Dr. Wind Goodfriend: Mentor
Psychology
The following study examined perceptions of positive traits in homosexual relationships. Students (n = 216) and
professional counselors (n = 96) read one of three variations of a transcript of a couple’s counseling session that
were identical in all aspects except for the names of the couple members and associated pronouns, implying sexual
orientation (either John and Amy, Amy and Jennifer, or John and David). Participants then rated the couple’s level
of commitment, satisfaction, investment, and closeness. Surprisingly, the student group perceived no differences
between the couples, but the counselor group perceived the gay and lesbian couples as having higher levels of the
positive relationship traits. Implications regarding counselor bias are discussed.
Introduction
In an ideal world, an individual or group of individuals would always be assessed based on their own
personal attributes and actions. Unfortunately, however, this is not always the case, especially for those individuals
belonging to an outgroup
11, 17, 21, 30
. This tendency affects all areas of life, but psychological research has tended to
focus on ingroup versus outgroup distinctions based on race, skin color, gender, religious affiliation, age, disability,
and sexual orientation.
The research regarding sexual orientation has grown steadily over the last century. Kinsey and associates
were certainly a catalyst in this field when they began to examine human sexuality in general in the middle of the
last century
14, 17
. Their findings that homosexual acts and thoughts were far more common than previously believed
rocked American society. Their research took the notion of sexual orientation from a dichotomous field with a
minute minority falling in the homosexual category, and expanded it to a prism where total heterosexuality and
homosexuality were the minorities and everyone else fell somewhere in between. Although much of the current
culture may still tend to believe in the dichotomous model, the work done by Kinsey opened up a new venue in
which researchers could begin to explore not just homosexuality, but society’s view of homosexuality
7, 18, 22
.
Prejudice based on sexual orientation is a growing area of study
10, 16, 24
. It is not surprising to find that
patterns of prejudice towards gays, lesbians, and bisexuals follow the general pattern of prejudice over all. Jellison
and colleagues found that in heterosexual male participants, there was a negative correlation between attitudes
toward heterosexuality and attitudes toward homosexuality, showing the common practice of ingroup versus
outgroup bias
16
. Prejudice relevant to sexual orientation is a function of both negative preconception and simple
ignorance. For example, in 1997, Eliason found that a large percentage of participants were generally ignorant when
it came to stereotypes and social concerns that both involved and affected bisexuals
10
.
The majority of this research has focused on prejudices and attitudes towards non-heterosexual orientations
in general. In order to better understand these overlying biases it is important to take a closer look at more specific
areas of perception affected by prejudice. The current political climate regarding gay and lesbian relationships
makes such research even more valuable. As social scientists, it is important to understand not only the true nature of
non-heterosexual relationships, but also the way in which the general populace perceives those relationships.
The current study examines individuals’ perception of commitment, satisfaction, investment, and closeness
in a non-heterosexual relationship as compared with a heterosexual relationship. Previous research has shown that,
in general, gay and lesbian relationships are no less satisfying, close, or well-adjusted than are heterosexual
relationships. In 1978, gay, lesbian, and heterosexual couples were compared on a measure of “marital” adjustment,
and no distinguishable differences were found between the groups
28
. Other research has found similar results, with
gay and lesbian relationships having no significant differences in relationship satisfaction, compared to
heterosexuals
8, 27
. When it comes to relationship commitment, there is less extant research, but that which has been
51
done does not suggest either more or less commitment in homosexual couples, and it is likely that emotional
qualities related to relationship commitment are similar in both heterosexual and homosexual couples
5, 12, 19, 26
.
However, this may not be the commonly held view among the general populace, and may also differ in regards to
perceptions of gay males as opposed to lesbians. In 2002, Hewitt and Moore found that there were different
constructs of homosexuality regarding lesbians and gay males, stating that, “It would seem prudent, then, that
researchers avoid conducting research in which homosexuality is used generically to refer to both lesbians and gay
men, as beliefs about lesbians and gay men may well differ” (p. 70)
13
.
In the current study it was expected that gay, lesbian, and heterosexual couples would be perceived to have
different levels of commitment, satisfaction, investments, and closeness, based only on the sexual orientation of the
couple. Because previous research has noted that gay males are perceived as being less emotionally involved in their
relationships
5, 20
, it was expected that the gay male couple would be perceived as having the least amount of
commitment, satisfaction, investments, and closeness. The lesbian couple was expected to be perceived slightly
more favorably, with the heterosexual couple being perceived as having the greatest amount of the aforementioned
qualities.
Commitment is the subjective state of dependence that individuals experience daily regarding their
relationship
2
. Commitment is hypothesized to be a function of three interrelated factors: (1) the level of satisfaction
with the relationship (or the degree to which a relationship is experienced as gratifying); (2) the quality of a
relationship’s alternatives (or the perceived desirability of the best available alternative to the current relationship),
and (3) the size of investments in the relationship. Investment size is defined as “the magnitude and importance of
the resources that are attached to a relationship—resources that would decline in value or be lost if the relationship
were to end”
29
. Closeness is defined as “people’s sense of interpersonal interconnectedness” (p. 597)
3
.
As with many psychological studies, the initial participant group in this research was made up of students
attending a metropolitan university. It was expected that some amount of bias against non-heterosexual couples
would be found in this population. However, certain subgroups within the American culture are expected to be less
biased in their views towards others. Because of specific ethics and sensitivity training, professional counselors and
therapists are expected to fall into this category
13, 15
. It is important that this expectation hold true because the nature
of the helping professions, such as counseling, demands a significant level of trust and understanding between the
client and the therapist
9
. Unfortunately, past studies have found that this is not always the case
23
. Although the
primary hypothesis of this study was that perception of commitment, satisfaction, investments, and closeness in non-
heterosexual couples would be lower than the levels perceived in heterosexual relationships, it was also expected
that professional counselors would perceive less of a difference between the control and experimental groups as
compared to the difference perceived by the student group.
One must also recognize that there are numerous other factors that may affect an individual’s response to
the relationships of others, especially different others. One such area is that person’s desire to be socially acceptable
6, 25
. Current social norms teach us that there are certain behaviors and beliefs that are beneficial for us to conform to
if we wish to participate and be accepted in society. At times an individual’s personal beliefs may contradict these
socially accepted beliefs, but that individual will attest to holding those beliefs in order to appear acceptable to
others. Sexual orientation tends to be a volatile and somewhat taboo subject in the American culture, while at the
same time gaining support in political circles to combat prejudice. Because of this duality, participants’ desire to
appear socially acceptable should be assessed.
Hypotheses
To summarize, three major hypotheses were made in the current study. Hypothesis 1: It was expected that
for perceived levels of commitment, satisfaction, investments, and closeness, college students would perceive the
gay couple as having the lowest levels of the qualities, the lesbian couple as having slightly more, and the
heterosexual couple as having the highest. Hypothesis 2: Conversely, no significant differences in the experimental
groups were expected to be found when testing professional counselors. Hypothesis 3: It was anticipated that there
would be a significant difference in the levels of commitment, satisfaction, investments, and closeness in the gay
male and lesbian group as perceived by the student population compared to that of the counselor population.
Method
Participants
This study included two separate participant groups. The initial participant group consisted of 216
undergraduate students attending a metropolitan research university in the Northwestern United States. The students
were enrolled in a general psychology course and were given the option to participate in ongoing research to fulfill
52
partial course credit. The demographic break down of the participants was as follows: 109 of the participants
identified as female, and 107 identified as male. The participants were asked to identify their race or ethnicity; 81%
identified as White, 6% as Hispanic, 1% as Black, 2% as Asian, and 11% as other or did not respond. The mean age
of the participants was 20.6 years (SD = 4.4) with a range from 18 years to 51 years. The participants’ current
romantic relationship status was assessed; 46% were currently single, 6% were dating more than one person, 30%
were dating one person, 4% were engaged, 7% were cohabiting with their romantic partner, 6% were married, and
1% declined to answer. In response to the extended sexual orientation scale, 65 of the male students identified as
exclusively heterosexual, 1 as exclusively homosexual, and 41 had mixed responses. For the female students, 64
identified as exclusively heterosexual, 3 as exclusively homosexual, and 42 had mixed responses.
The second participant group consisted of professional counselors and therapists from the Northwestern
United States. Participants were contacted through the name and address published in the public business phone
directory. Surveys were mailed to 1,000 counselors and therapists, of which 96 responded by returning the
completed survey, for a response rate of 9.6%. The demographic breakdown of this group is as follows: 61 of the
participants identified as female, and 35 identified as male. When identifying their race or ethnicity, 93% identified
as Caucasian, 1% identified as Native American, 1% identified as Hispanic, and 5% did not respond. The mean age
of the participants in the second group was 55 years (SD = 8.04) with a range from 33 years to 74 years. The
participants’ current romantic relationship status was assessed; 4% were currently single, 3% were dating one
person, 3% were engaged, 5% were cohabitating with their romantic partner, 67% were married, 17% were
divorced, separated, or widowed, and 1% declined to answer. For sexual orientation, 18 of the male counselors
identified as exclusively heterosexual, 17 had mixed responses, and none responded as exclusively homosexual. For
the female counselors, 21 identified as exclusively heterosexual, 3 as exclusively homosexual, and 37 had mixed
responses.
The second group was asked two additional demographic questions. They were asked to identify which
professional licenses they held; 9 held a LPC, 15 a LCPC, 9 a LMFT, and 37 a LCSW, 17 held a Ph.D. or Psy.D.,
and 11 of the respondents held other licenses. It should also be noted that 8.33% held multiple licenses. The
participants were also asked how long they had been in practice. The mean time was 20 years (SD = 8.67), ranging
from 3 to 40 years.
Materials
Independent variable. Participants were asked to read an excerpt from a fictional couple’s initial counseling
session. Participants randomly received one of three variations of the transcript that were identical in all aspects
except for the names of the couple members and associated pronouns. The control group (student group n = 72,
counselor group n = 27) read about a couple with the names Jennifer and David. The first experimental group
(student group n = 73, counselor group n = 36) read about a couple with the names John and David, and the second
experimental group (student group n = 71, counselor group n = 33) read about a couple with the names Jennifer and
Amy. The names used in the transcript were chosen from the most popular names for children born from 1974 to
1984 as listed by the Social Security Administration
31
.The four names used were also chosen because of their
gender specificity in order to accentuate the perception that the couple was either heterosexual, gay, or lesbian.
The transcript was brief, and in it the issues addressed by the couple include general, mild dissatisfaction
with the relationship. An example of a conflict experienced by the couple was given. The example involved one
partner prioritizing a career opportunity over a five-year anniversary celebration, and feeling that (s)he would be
accused of infidelity.
A pre-test asked four professional counselors whether the fictional transcript appeared realistic. Each of the
counselors who were asked to review the transcript was licensed to practice counseling and therapy in the state in
which the research was developed. They held varying degrees of higher learning; two held Ph.D.s, one held a
Psy.D., and one held a M.A. All four counselors attested to the transcript’s seeming authenticity.
After they read the transcript, all participants were asked to write a short response to two questions about
the couple’s relationship. First, they were asked to write about what they felt the major issue or issues were in the
relationship. Second they were asked what, if anything, the couple could do to overcome them.
Dependent variables. To measure perceived commitment, satisfaction, and investment, the participants
responded to a seventeen item assessment that was a modified version of the Rusbult, Martz, and Agnew scale
29
.
The original scale has all items in first person (e.g., “My relationship is close to ideal”); the modified version used in
the current study changed all items to refer to the couple from the transcript and asked the participants what they
believed the likelihood of the statements was (e.g., “How likely is it that this couple has a relationship that is close to
53
their ideal?”). Participants responded to the items on a 7-point Likert scale ranging from 1 (very unlikely) to 7 (very
likely). There were seven items assessing commitment, five items for satisfaction, and five items for investments.
Internal consistency for the composite measures was very good (alpha = .83 for commitment, .85 for satisfaction,
and .80 for investments).
The participants were next presented with the Inclusion of Other in Self Scale (IOS) pictograph
representing seven different levels of interconnectedness, or relationship closeness. The scale uses seven pictographs
of two circles that grow consecutively closer, so a rating of 1 (choosing the two circles that are barely touching)
indicates the least amount of closeness, and a rating of 7 (choosing the two circles that are nearly on top of one
another) indicates the greatest level of closeness. Participants were asked to circle the image they felt most
accurately described the relationship of the couple about whom they had read
3
. Demographic questions were then
asked. Participants responded to questions regarding their age, gender, race/ethnicity, and relationship status, and in
the counselor group licenses held, and length of time in practice.
Finally the participants responded to a forty-question social desirability scale
25
. Responses were on a 7-
point Likert scale ranging from 1 (not true) to 7 (very true). Examples of the questions asked include, “I have said
something bad about a friend behind his or her back” and, “I never regret my decisions.” Internal consistency for the
social desirability measures was good (alpha = .70 for self deception, and .79 for impression management).
Procedure
For the student participant group the survey occurred in a classroom setting. The participants were given 60
minutes to complete the full survey. Most participants completed the survey within 15 minutes. Following the
conclusion of the survey the participants were thanked for their time and debriefed regarding the nature of the study.
The participant group consisting of professional counselors was mailed the survey along with a cover letter
and standard consent information. The cover letter invited the recipient to participate in the enclosed survey to help
the researchers better understand perceptions of people in relationships. They were also informed that in order to
complete the research, the researchers require professional counselors to complete the survey. The letter directed the
recipient to contact the researchers if he or she wished to be debriefed on the full nature of the research. One
participant contacted the researchers and requested a report of the results of the research; this was provided.
Results
Hypothesis 1
The first hypothesis expected that for perceived levels of commitment, satisfaction, investments, and
closeness, student participants would perceive the gay couple as having the lowest levels of the qualities, the lesbian
couple perceived as having slightly more, and the heterosexual couple perceived as having the highest levels. A
series of one-way analyses of variance were conducted to test Hypothesis 1, one test for each of the four dependent
variables (commitment, satisfaction, investment, and closeness). No significant differences were found in
commitment, F(2, 214) = 1.03, p = 0.36, satisfaction, F(2, 215) = 0.12, p = 0.89, or investment, F(2, 214) = 0.03, p =
0.97. As previously mentioned, perception of closeness was measured using the Inclusion of the Other in Self Scale
(IOS). As with the other dependent variables in the student group, these means were not significantly different, F(2,
213) = 0.55, p = 0.58. Thus, Hypothesis 1 was not supported. For means and standard deviations on all four of the
dependent variables for both the counselor and student group, please refer to Table 1.
Hypothesis 2
In the second hypothesis, no significant differences were expected to be found in perceptions of the three
groups when testing professional counselors. As with the first hypothesis, a one-way analysis of variance was run for
each of the four dependent variables (commitment, satisfaction, investment, and closeness). Contrary to the
hypothesis, significant differences between the groups were found for satisfaction, F(2, 93) = 5.21, p < 0.01, and
investments, F(2, 92) = 7.95, p < 0.001, and marginal significance was found for commitment, F(2, 92) = 3.04, p = .
053. Interestingly, the gay couple was perceived as the most satisfied, the lesbian couple slightly less satisfied, and
the heterosexual couple as the least satisfied. The same pattern held true for investments. The gay couple was
perceived to be the most invested, followed by the lesbian couple, with the heterosexual couple perceived as
significantly less invested than either the gay or lesbian couple. For commitment, the lesbian couple was rated as the
54
most committed, followed by the gay couple, and the heterosexual couple rated with the lowest level of
commitment.
As with the student population, no significant differences were found among groups in perceived closeness.
Although the differences were not statistically significant, F(2,82) = 1.27, p = 0.27, the trend in means showed a
similar pattern as the other dependent variables, with the heterosexual couple being perceived to have the lowest
level of closeness (refer to Table 1).
Hypothesis 3
The third hypothesis conjectured that there would be a significant difference in the levels of perceived
commitment, satisfaction, investments, and closeness for the gay male and lesbian groups in the student sample,
compared to the counselor sample. Paired samples t-tests were completed for each dependent variable.
For the first experimental group (gay male couple), there was a significant difference in the level of
investments perceived by the student population and the counselor population, t(105) = -2.21, p = 0.03. The
difference in perceived commitment was marginally significant, t(105) = -1.87, p = 0.06. The difference in closeness
was also marginally significant, t(101) = -1.75, p = 0.08. No significant difference was found for perceived levels of
satisfaction. In sum, for investments, commitment, and closeness, the counselors perceived the gay couple to have at
least marginally higher levels of these variables than did the student sample.
Perceived closeness was the only significant difference in the second experimental group (lesbian couple), t
(95) = -2.65, p < 0.01. Again, the counselors perceived the lesbian couple to have a higher level of closeness than
did the student sample.
Table 1. Mean Responses of Commitment, Satisfaction, Investment, and Closeness
Student
Counselor
Independent
Variables
Mean
Standard
Deviation
Mean
Standard
Deviation
Commitment
Control
5.01
0.99
4.71
1.06
Gay
4.82
1.00
5.18
0.86
Lesbian
5.03
0.93
5.30
0.96
Satisfaction
Control
3.77
0.91
3.22
0.65
Gay
3.84
1.11
3.95
0.86
Lesbian
3.83
1.15
3.67
1.08
Investment
Control
4.89
1.06
4.49
0.88
Gay
4.85
1.09
5.32
0.86
Lesbian
4.88
0.98
5.23
0.82
Closeness
Control
3.67
1.56
4.12
1.67
Gay
3.90
1.60
4.48
1.38
Lesbian
3.90
1.48
4.78
1.42
Note. Scores range form 1 to 7 with the higher number representing a greater perception of
the given quality.
55