Content uploaded by Justin J. Lehmiller
Author content
All content in this area was uploaded by Justin J. Lehmiller on Dec 04, 2017
Content may be subject to copyright.
Content uploaded by Justin J. Lehmiller
Author content
All content in this area was uploaded by Justin J. Lehmiller on Dec 04, 2017
Content may be subject to copyright.
Commitment in Friends with Benefits Relationships 1
Running head: COMMITMENT IN FRIENDS WITH BENEFITS RELATIONSHIPS
Commitment in Friends with Benefits Relationships:
Implications for Relational and Safe-Sex Outcomes
Laura E. VanderDrift
Purdue University
Justin J. Lehmiller
Colorado State University
Janice R. Kelly
Purdue University
This is an Accepted Manuscript of an article published by Wiley in Personal Relationships,
available online: http://onlinelibrary.wiley.com/doi/10.1111/j.1475-6811.2010.01324.x/abstract
VanderDrift, L. E., Lehmiller, J. J., & Kelly J. R. (2012). Commitment in friends with benefits
relationships: Implications for relational and safer sex outcomes. Personal Relationships,
19, 1-13. DOI: 10.1111/j.1475-6811.2010.01324.x
Commitment in Friends with Benefits Relationships 2
Abstract
The present research sought to explore the antecedents and consequences of commitment in the
context of Friends with Benefits relationships (FWBRs). Data collected from an Internet sample
of participants (all of whom currently had a FWBR) revealed that these relationships do indeed
uniquely combine aspects of both sexual relationships and friendships. In addition, results
indicated that satisfaction, investment, and alternatives were predictive of commitment,
consistent with the antecedent factors specified by Rusbult’s (1980) investment model.
Moreover, high commitment was associated with desiring a transition into a more interdependent
relationship (i.e., a true romance), as well as reduced condom use during intercourse. These
findings have both theoretical and practical relevance to understanding relationship transitions
and safer sex practices across many different types of partnerships.
Commitment in Friends with Benefits Relationships 3
Commitment in Friends with Benefits Relationships:
Implications for Relational and Safe-Sex Outcomes
Friends with benefits relationships (FWBRs) have recently begun to attract the attention
of lay people and scholars alike. As a result, our understanding of these involvements is in a very
early stage. Based on the limited empirical information that exists, we know that they are novel
compared to most other relationship types in that they are a relational hybrid, combining aspects
of both a sexual relationship and a friendship without integrating the two components into a
traditional romantic relationship (Mongeau, Ramirez, & Vorell, 2003). This novelty makes
FWBRs interesting to study in their own right, but they also have many potential implications for
understanding relationship dynamics more broadly. Utilizing the unique context provided by
FWBRs, the current study aimed to achieve two goals.
First, we sought to measure the antecedents of commitment in the context of FWBRs.
Given the dual nature of these involvements, it is possible that commitment to the different
aspects of the relationship (i.e., friendship, sexual component) may vary independently from one
another. Examining how commitment to the distinct aspects of such relationships combine to
influence outcomes may provide valuable information about FWBRs, as well as the nature of
commitment more broadly.
Second, we examined two outcomes of commitment that have the potential to be
extremely important in the context of a FWBR. The first is relationship persistence, which is
most typically measured as a dichotomous end state of the relationship (for an exception, see
VanderDrift, Agnew, & Wilson, 2009). FWBRs provide a unique venue to study relationship
change, given that they can not only stay the same or become less interdependent (e.g., dissolve),
but they can also become more interdependent (e.g., turn into a romantic relationship). As such,
Commitment in Friends with Benefits Relationships 4
FWBRs provide a vehicle to study the transformation of a relationship into a different type,
rather than the more traditional end states of continuation or dissolution. Another outcome of
commitment that has important implications in FWBRs is that of safer sex practices, particularly
condom use. These relationships, despite the friendship at their core, are essentially casual sexual
relationships, in which risk for sexually transmitted infection (STI) acquisition is likely to be
high. We argue that FWBRs may be particularly dangerous in terms of STI acquisition, because
the friendship component may lead partners to erroneously believe that their partner is not a risk
to their health and subsequently forego using condoms, despite the fact that they are engaging in
a casual sexual relationship that is not necessarily exclusive (and indeed, many FWBRs lack
exclusivity; see Lehmiller, VanderDrift, & Kelly, in press).
Taken together, the goals of the current study are intended to not only provide
information about FWBRs, but also to utilize this unique context in order to gain information
about relationship processes more generally.
Friends with Benefits Relationships
Among American college students, FWBRs are relatively common and, of late, have
been the topic of much discussion throughout the popular media (Hughes, Morrison, & Asada,
2005). These relationships are characterized as combining the psychological intimacy of a
friendship with the physical intimacy of a romantic relationship, while avoiding the “romantic”
label. As alluded to by their title, both researchers and lay people alike tend to categorize these
relationships as friendships with an added component (i.e., sexual relations), rather than as
romantic relationships with missing components (e.g., the label “romantic,” exclusivity, romantic
love). A primary argument for such categorization, at least among scholars, is that the partners in
FWBRs are not romantically dedicated to one another (Bisson & Levine, 2009), which is a
Commitment in Friends with Benefits Relationships 5
feature that differentiates romantic relationships from friendships (Davis & Todd, 1982).
Interestingly, though, we know very little about the actual activities that occur within the context
of FWBRs, either in terms of their sexual activities or in terms of their friendship activities.
Additionally, we know little about the prototypical trajectory of a FWBR, including what type of
relationship (if any) the partners shared prior to beginning the FWBR or how the partners want
the relationship to change over time.
It has been shown that sexual encounters in FWBRs can further the sexual component of
the relationship, without necessarily leading toward romantic relationship commitment or
exclusivity (Williams, Shaw, Mongeau, Knight, & Ramirez, 2007). Evidence suggests that
individuals enter FWBRs to have the benefits of a sexual relationship within the relative safety
of a friendship, and without the “drama” associated with romantic relationships (Williams et al.,
2007). This “drama” refers to perceived negative aspects of romantic relationships, including
jealousy, heartbreak, and obligation. But how successful are individuals at attaining the ideal
FWBR, devoid of the potential negatives of romantic entanglements, but flush with the
positives? Existing research suggests that such success is limited, but that there is wide
variability regarding the outcomes derived from FWBRs (Williams et al., 2007). Whereas the
literature indicates that some people do achieve a sexual relationship without attachments, many
others experience the negatives they sought to avoid when beginning a FWBRs, causing the
subjective experience of the relationship to be not unlike the experience of an unrequited love for
one, if not both, of the partners (Baumeister, Wotman, & Stillwell, 1993).
It is logical that not all FWBRs yield comparable outcomes, given that relationships of
any type rarely yield identical outcomes for individuals, either between dyads or within dyads.
As a way of predicting what outcomes individuals will receive from their relationships in general
Commitment in Friends with Benefits Relationships 6
(e.g., romantic relationships, workplace associations), many researchers have found the notion of
relationship commitment important to consider.
The Causes of Relationship Commitment
While there is some divergence among theorists and theories about the exact nature of
commitment, most agree that it involves factors that cause individuals to persist in a relationship
(as cited in Arriaga & Agnew, 2001). The primary application of commitment, in terms of the
volume of research generated to date, is certainly to romantic relationships, but commitment has
been successfully measured in numerous other relationship types, some of which are
interpersonal (e.g., friendships, workplace associations; Lin & Rusbult, 1995; Rusbult & Farrell,
1983) and some of which are non-interpersonal (e.g., job, sport teams, college; Carpenter &
Scanlan, 1998; Hatcher, Kryter, Prus, & Fitzgeralt, 1992; Oliver, 1990).
One model that has been shown to be especially useful in predicting commitment across
many different relationship types, including romantic relationships and friendships, is the
investment model (Rusbult, 1980). This model holds that commitment to a target is fueled by
three independent, but related factors: 1) satisfaction level (i.e., the individual’s subjective
appraisal of the positivity and negativity that he or she experiences in a relationship), 2) quality
of alternatives (i.e., the most attractive option an individual perceives he or she would have if the
current relationship were to end), and 3) investment size (i.e., the resources attached to a
relationship that would be diminished in value or completely lost should the relationship
dissolve). Together, these three factors have been found to reliably predict commitment,
explaining 61% of the variance in commitment on average across diverse targets (Le & Agnew,
2003). Of importance, experimental manipulations of the investment model constructs support
Commitment in Friends with Benefits Relationships 7
the hypothesized causal direction (Agnew, Hoffman, Lehmiller, & Duncan, 2007; Hoffman,
Agnew, Lehmiller, & Duncan, 2009).
As stated previously, one goal of the current study is to explore the antecedents of
commitment in the present context. To accomplish this, we will test the utility of the investment
model’s theorized predictors of commitment among people who are currently in FWBRs. An
important modification to the model must be made before such an application is possible,
however. Specifically, because FWBRs are a hybrid relationship type, combining aspects of
friendships and sexual relationships without integrating the two into an overall romantic
relationship, it is possible that commitment to the two discrete aspects of the relationship may
vary independent of each other. As such, it is important to consider that commitment to the
friendship and commitment to the sexual component of the relationship may be distinct, and that
both may contribute to commitment to the global FWBR.
Outcomes of Commitment
The investment model specifies that the association between the three bases of
commitment (i.e., satisfaction, alternatives, and investments) and various outcome variables is
mediated by commitment. In other words, from a theoretical standpoint, commitment is a more
proximal precursor to the outcomes of interest than are the bases (Rusbult, 1983; VanderDrift et
al., 2009). Thus, whereas the three bases provide explanatory power toward understanding an
individual’s level of commitment, as well as information regarding the nature of that
commitment, it is commitment itself that is useful to the prediction of an individual’s behavior
and what outcomes he or she is able to derive from the relationship.
A frequently examined outcome in the relationship literature is stay-leave behavior (i.e.,
relationship persistence or termination). In the context of romantic relationships, the association
Commitment in Friends with Benefits Relationships 8
between commitment and stay-leave behavior is moderate (r = .47; Le & Agnew, 2003). This
context differs from FWBRs in an important way, however. Specifically, whereas in romantic
involvements, relationship termination necessarily means that the partners have selected a less
interdependent relationship type (e.g., friendship, no relationship at all), termination of a FWBR
could either mean the partners have selected a relationship characterized by less interdependence
(e.g., friendship), or one characterized by even more interdependence (e.g., romantic
relationship). Moreover, whereas commitment to the overall FWBR should be associated with
wanting the relationship type to remain intact (i.e., no change in interdependence), high
commitment to the different facets of the relationship may predict desiring that the relationship
will transition to one with greater interdependence (e.g., a romantic relationship).
Another outcome of commitment that has been studied extensively is that of condom use.
Individuals are less likely to use condoms with partners when involved in a long-term, steady
relationship than with partners in a casual relationship (Tucker, Elliott, Wenzel, &
Hambarsoomian, 2007). Research reliably finds that as an individual’s level of commitment to
their partner increases, their intentions to use condoms, as well as their actual condom use,
decrease. The primary explanations for this phenomenon implicate the cognitive shifts that occur
when commitment rises. Namely, individuals shift their focus from self-protection and self-
enhancement to relationship maintenance and relationship enhancement when commitment is
high (Misovich, Fisher, & Fisher, 1997). Also, as commitment increases, so does trust in one’s
partner (Wieselquist, Rusbult, Foster, & Agnew, 1999), and as such, condom use might
potentially decrease as a function of greater trust in a partner. Finally, individuals view their
partner as less of a risk to their health when commitment is high, which reliably leads to failure
to use protection (Gerrard, Gibbons, & Bushman, 1996).
Commitment in Friends with Benefits Relationships 9
Because FWBRs lack the exclusivity and steadiness of a true romance (Lehmiller et al.,
in press; Williams et al., 2007), it may seem intuitive to predict that individuals would be more
likely to use condoms within FWBRs. However, insofar as individuals are committed to their
FWBRs, particularly the friendship that is at their base, we theorize that they may actually be
less likely to use condoms with their FWBR partners, given that commitment often leads
partners to forgo protection. As a result, it is possible that, because of the friendship and trust
inherent in FWBRs, individuals in these relationships may place themselves at higher risk for
contracting STIs and experiencing unwanted pregnancies.
The Current Study
In the current study, we collected data relevant to the theorized antecedents and
consequences of commitment from individuals who were currently involved in FWBRs. With
regard to the antecedents of commitment, we examined whether the investment model’s
theorized bases of commitment did indeed predict commitment in this context. More specifically,
we assessed the importance of satisfaction level, investment size, and perceived quality of
alternatives in predicting commitment to the friendship aspect and to the sexual aspect of the
relationship separately, and then examined how commitment to each of these aspects predicts
commitment to the FWBR generally. Finally, with regard to the hypothesized consequences of
commitment, we examined how each type of commitment relates to partners’ desires for the
future of their relationships (i.e., transition into a romantic relationship, continue as a FWBR,
transition into a friendship), as well as frequency of condom use during intercourse.
Method
Participants and Design
Commitment in Friends with Benefits Relationships 10
We recruited participants to complete a survey study of “Friends with Benefits”
relationships from commonly used and recommended Internet websites for relationships research
such as Craig’s List and the Social Psychology Network (for a discussion of Internet participant
recruitment, see Lehmiller, 2008). We encouraged all people over the age of 18 to participate,
but indicated that we were especially interested in those people who were currently involved in a
“friends with benefits” relationship.
Participants were 181 women (73.6% of the sample) and 65 men (24.4% of the sample)
who indicated current involvement in a FWBR. If a participant indicated involvement in more
than one FWBR, they were told at the beginning of each page to consider only their ‘most
important’ FWBR partner when filling out the survey. The age range for the sample was 18 to 65
years old (M = 28.7, SD = 9.68). The majority of participants were White (70.7%, with 3.3%
Asian, 16.7% Black, 5.7% Hispanic, and 3.6% other) and reported being heterosexual (88.2%;
the remainder of the sample was 1.6% homosexual and 9.4% bisexual).
During the same timeframe, 678 individuals who were not currently involved in a FWBR
also began the survey and provided their age, sex, and sexual orientation. Although they were not
permitted to complete the full survey because they did not meet the selection criteria, we
compared their demographic features to the FWBR sample for exploratory purposes. The non-
FWBR sample was significantly younger (M = 26.26; t(922) = 3.74, p < .001) and was more
likely to be male than the FWBR sample (33.6%;
χ
2 = 4.39, p < .05). The two samples did not
differ with regard to sexual orientation (
χ
2 = 1.46, ns).
Measures
In line with our goal of assessing the antecedents of commitment, we included measures
of satisfaction with, investment in, alternatives to, and commitment toward the sexual and
Commitment in Friends with Benefits Relationships 11
friendship aspects of the FWBR. Specifically, we used a modified version of the Investment
Model Scale (Rusbult, Martz, & Agnew, 1998) and reduced the number of questions in each
subscale in order to enhance completion rates, given that participants were not compensated for
taking part in the study (a common practice in on-line research; e.g., Lehmiller & Agnew, 2006).
For each subscale, two items tapping each construct for each FWBR component were used, with
each item assessed on an 9-point scale ranging from 1 (strongly disagree) to 9 (strongly agree).
In selecting items for the shortened subscales, we chose the items with the highest item-total
averages based on Rusbult et al. (1998). We then modified each item to meet the goals of our
work. First, the general Investment Model Scale items were modified to be applicable
specifically to the friendship component of the FWBR (e.g., “I feel satisfied with our friendship,”
to tap satisfaction; “My other options for friends are desirable,” to tap alternatives; “I've put a lot
into developing this friendship,” to tap investment; and “I am committed to maintaining our
friendship” to tap commitment). All four scales evidenced acceptable reliability (satisfaction, r =
.81; investment, r = .86; alternatives, r = .65; commitment, r = .93). Second, we modified these
same Investment Model Scale items to be applicable specifically to the sexual aspect of the
FWBR (e.g., “Our sexual relationship is much better than others' sexual relationships,” to tap
satisfaction; “My needs could easily be fulfilled by other sex partners,” to tap alternatives; “I feel
very involved in this sexual relationship” to tap investment; and “I feel very attached to our
sexual relationship” to tap commitment). Again, the four scales evidenced acceptable reliability
(satisfaction, r = .90; investment, r = .85; alternatives, r = .76; commitment, r = .85). In addition
to the above measures, we administered three items that we modified from the investment
model’s commitment subscale to gauge participants’ feelings about their FWBRs in general
Commitment in Friends with Benefits Relationships 12
(e.g., “I want our ''friends with benefits'' relationship to last forever”). Together, these items
evidenced acceptable reliability (α = .79).
To meet our second goal, examining the outcomes of commitment in a FWBR, we
administered two additional measures. The first was intended to assess relationship change. We
asked each participant how he or she hopes his or her relationship changes over time (i.e.,
participants selected whether they wanted it to change into a romantic relationship, into a
friendship, discontinue altogether, or stay the same). Participants could only select one of these
four options. We also assessed the frequency of condom use by asking participants to indicate on
a scale from 1 (never) to 9 (every time) how frequently they use condoms during intercourse with
their FWBR partner.
Additionally, we collected measures of age, sex, sexual orientation, and race from
participants, as well as general demographics about the FWBR, including how the relationship
began (i.e., participants selected whether it began as a romantic relationship, friendship, or purely
sexual relationship), and the exclusivity of the relationship (i.e., participants indicated whether
other sexual partners or romantic relationship partners existed for themselves, and/or for their
current FWBR partner). To provide enhanced descriptive information for our sample,
participants were asked to indicate which of the following sexual acts they had performed with
their partner: kissing, sexual touching, mutual masturbation, oral sex, vaginal intercourse, and
anal intercourse. Each item was dichotomously coded with either a ‘1’ (have performed) or ‘0’
(have not performed). The sum of these items was taken to serve as a measure of the number of
sexual activities the partners have engaged in together. Likewise, to measure the number of
friendship activities the partners have shared, we asked participants to indicate which of the
following items, adapted from Reis, Sheldon, Gable, Roscoe, and Ryan (2000), they had engaged
Commitment in Friends with Benefits Relationships 13
in: talked about something meaningful, went out together with a larger group of friends or
acquaintances, hung out casually, gone on what could be considered a ‘date,’ tried something
new together, made each other feel insecure or self-conscious, and argued or had conflict. The
first five of these activities are those that Reis et al. (2000) found to be especially useful in
generating friendship, whereas the last two were found to hinder friendship. To create a measure
of how many friendship activities the partners shared, we reverse-coded the final two items, and
then took the sum of all seven.
Results
Descriptive Analysis of the Sample
Before delving into the nature and consequences of commitment in the context of a
FWBR, we first provide descriptive information regarding our sample, including how these
relationships began, the types of friendship and sexual activities engaged in, what the partners
desire for the future of the relationship, and the level of relationship exclusivity. Most FWBRs in
our sample began as friendships (n = 125, 47.89%), with the remainder developing from
romantic partners (n = 54, 20.7%), sexual partners (n = 64, 24.5%), or other types of
relationships (e.g., co-workers first; n = 18, 6.9%).
Participants reported engaging in an approximately equal amount of the 7 friendship and
6 sex activities we inquired about (Ms = 4.57 and 4.48, respectively; SDs = 1.37 and 1.02,
respectively), providing some empirical evidence that this type of relationship truly is a
friendship with a sexual component. See Table 1 for frequencies of each sex and friendship
activity. Over time, an approximately equal proportion of participants said they wanted their
FWBR to transition into a romantic relationship (n = 92, 35.4%) as said they wanted their FWBR
to stay the same (n = 89, 34.2%). The remaining participants wanted to become close friends
Commitment in Friends with Benefits Relationships 14
who do not have sex (n = 37, 14.2%), wanted to discontinue their relationship altogether (n = 15,
5.8%), or chose some other options (e.g., uncertain; n = 27, 10.4%).
Lastly, the majority of FWBRs in our sample were exclusive, meaning that the
participant was not involved in a romantic relationship or other FWBR (n = 146, 60.3%). Of the
96 participants in nonexclusive FWBRs, 60 had other FWBRs (23.2% of the total sample) and
36 had a romantic relationship partner (13.9% of the total sample). Given that FWBRs are not
always sexually exclusive, condom use is important to consider in this context. Overall,
participants reported using condoms during intercourse most of the time with their FWBR
partners, with the mean significantly above the midpoint of the scale (M = 5.74, SD = 3.42; t =
5.77, p < .001).
Antecedents of Commitment to a FWBR
The first goal of this study was to examine the applicability of the investment model to
FWBRs, which involved separating the relationship into its two facets: the friendship and the
sexual relationship. Participants reported moderate levels of commitment to both facets (sexual
aspect: M = 5.78, SD = 2.39; friendship M = 6.23, SD = 2.37). See Table 2 for complete
descriptive information about the investment model variables collected.
To examine the applicability of the investment model to FWBRs, we examined the
degree to which satisfaction, alternatives, and investment predicted commitment to each facet
using multiple regression. We then examined the extent to which commitment to the friendship
and commitment to the sexual relationship predicted overall commitment to the FWBR.
When considered concurrently, satisfaction and investment significantly and positively
predicted commitment to the friendship (respectively, β = .206, p < .01; β = .670, p < .01),
whereas alternatives significantly and negatively predicted friendship commitment (β = -.099, p
Commitment in Friends with Benefits Relationships 15
< .01). Together, these three variables explained 66.8% of the variance in friendship
commitment.
With regard to the sexual relationship, when considered concurrently, satisfaction and
investment both positively predicted commitment (respectively, β = .088, p = .06; β = .763, p <
.01), whereas alternatives did not (β = -.041, p = .25). Collectively the three predictors explained
67.9% of the variance in commitment to the sexual relationship. Post-hoc theorizing about the
nonsignificant association between alternatives and commitment led us to wonder whether the
fact that many FWBRs are not sexually exclusive could explain this finding. It may be the case
that when an individual feels free to have additional concurrent sexual relationships, alternatives
do not decrease commitment to one of the relationship partners (i.e., if one is already free to
explore other sexual options because the partners have agreed not to be exclusive, perceived
alternatives might not be consequential for commitment). Results support this rationale,
revealing that the impact of sexual alternatives was moderated by the exclusivity of the
relationship [F(1, 248) = 3.22, p < .05]. For participants who were in nonexclusive FWBRs (i.e.,
they either had other FWBRs or a romantic relationship; n = 95), sexual alternatives did not
significantly predict sexual commitment (β = .121, p = .24). For participants who were in
exclusive FWBRs (i.e., they had only one FWBR and no romantic relationships, n = 144), sexual
alternatives significantly predicted sexual commitment in the expected direction (β = -.190, p <
.05).
We next assessed the relative contribution of friendship commitment and sexual
commitment in a model predicting commitment to the overall FWBR. The two types of facet
commitment were significantly correlated (r = .64, p < .01), but the correlation was not so high
as to suggest they tap the same construct. Further, when considered concurrently, both friendship
Commitment in Friends with Benefits Relationships 16
commitment (β =.348, p < .01) and sexual commitment (β = .519, p < .01) predicted general
FWBR commitment, and together explained 61.0% of the variance. These results support the
notion that this is a hybrid relationship type, combining aspects of a friendship with those of a
sexual relationship without integrating the two into a cohesive, wholly interdependent
relationship.1
Outcomes of Commitment
The second goal of the current study was to examine how commitment to each aspect of
the FWBR is associated with two outcomes of interest (i.e., desired future state of the
relationship and condom use). To begin, we considered how commitment to the facets of a
FWBR is associated with participants’ desired future state of the relationship. Because the
FWBR is a type of relationship that has the potential to change to either a higher or lower state of
interdependence, we expanded the traditional stay/leave conceptualization of relationship
stability to reflect the broad possibilities of change in this relationship type. To do so, we used
multinomial logistic regression to examine whether friendship, sexual, and general commitment
to the FWBR independently predicted differential desired future relationship states that vary in
interdependence (i.e., become a romantic couple, stay the same, become close friends who do not
have sex, discontinue the relationship altogether). Results indicated that the levels of all three
types of commitment predicted differential desired future states of the relationship [friendship:
χ
2(3)= 45.2, p < .01; sex:
χ
2(3)= 53.0, p < .01; general FWBR:
χ
2(3)= 58.6, p < .01]. See Table 3
for the results of the post-hoc tests we conducted to probe these significant effects. In general,
this analysis indicates that commitment to both the friendship and to the sexual component of the
FWBR was positively associated with a desire for the relationship to transition to a more
interdependent relationship type (i.e., a romantic relationship). In comparison, commitment to
Commitment in Friends with Benefits Relationships 17
the FWBR more broadly was associated with an equal desire to either transition into a more
interdependent relationship type, or to maintain the FWBR.
The final outcome of commitment we considered in the current study was condom use.
We assessed whether commitment to the FWBR generally, or to either of the components
specifically, was associated with regularity of condom use during intercourse. In separate
regression analyses, general commitment to the FWBR, as well as commitment to the friendship
component of the relationship, were both negatively associated with regular condom use
(general: t = -2.14, p < .05; friendship: t = -2.21, p < .05). These effects were not moderated by
whether the participant had more than one sex partner, whether the participant’s partner had
more than one sex partner, how the relationship began, how the participant wanted the
relationship to change over time, the participant’s gender, or the duration of the sexual
relationship (all ts for the interaction terms < 1.00). Of interest, commitment to the sexual
component of the relationship had only a marginal negative association with regularly using a
condom during intercourse [t = -1.60, p > .10]. Taken together, the results regarding condom use
in FWBRs indicate that commitment to the friendship at the base of the FWBR may lead partners
to forego using condoms, more so than commitment to the sexual component of the relationship.
Discussion
The first goal of this work was to study the potential antecedents of commitment in the
context of FWBRs. Using the investment model, we examined whether satisfaction level,
perceived quality of alternatives, and investment size predicted commitment in FWBRs, as they
do in other relationship types (Rusbult et al., 1998). When considering the antecedents of
commitment to the friendship aspect of the FWBR, the investment model bases all predicted
commitment as expected. Interestingly, the investment-commitment association was quite strong
Commitment in Friends with Benefits Relationships 18
relative to the satisfaction-commitment association, which typically emerges as the strongest link
in the investment model (Le & Agnew, 2003). There are several possible explanations for this.
For instance, it is possible that, because most of our participants reported that they had
previously been friends or lovers with their FWBR partner, perhaps FWBRs have more
interpersonal history and past investment when they begin compared to other types of
relationships. It is also possible that because FWBRs are less common than romantic
relationships and potentially less accepted by society, it takes greater amounts of time and effort
to initiate one, which leads to a greater sense of investment. Future research is needed to explore
these possibilities in greater depth. Regardless of which association was strongest, the three
bases combined explained two-thirds of the variability in commitment, which is comparable to
previous work on the investment model in romantic relationships (see Le & Agnew, 2003).
When examining commitment to the sexual relationship, though, another notable difference from
previous investment model research emerged. Whereas satisfaction and investment both
predicted commitment, quality of alternatives failed to do so. In probing this nonsignificant
effect, we found that perceived quality of alternatives was a significant predictor of commitment
for partners in exclusive FWBRs, but not for partners whose involvements were non-exclusive.
Two important points can be made regarding these findings. First, these results
emphasize the importance of specificity with regard to the bases of commitment (i.e., which
aspects of the relationship—sexual or intimate—are partners most committed to?). Future
research is necessary to see whether this same pattern holds when examining romantic
relationships, but at least among partners in FWBRs, sexual exclusivity moderates the
importance of sexual alternatives on commitment, but not the role of friendship alternatives on
commitment. This suggests that circumstances can exist in which a potential threat to
Commitment in Friends with Benefits Relationships 19
relationship commitment arises, but fails to have an impact. In the case of FWBRs, that
circumstance arises when a sexual alternative tempts a partner in a sexually non-exclusive
FWBR. However, in other types of relationships, such a circumstance may have quite different
consequences and be interesting to study for both theoretical and practical reasons. Second, these
results suggest that the impact of alternatives on commitment may be mitigated, at least to some
degree, by the exclusivity of the relationship. Previous research on romantic relationships has
found that the association between general relationship alternatives and commitment does not
differ significantly depending on whether the relationship is “nonexclusively dating” or
“exclusively dating” (Le & Agnew, 2003), but specificity with regard to the type of alternatives
under consideration may be important to understanding the role of exclusivity more fully. Future
work aimed at determining the circumstances under which alternatives exert a heavy influence
on commitment (e.g., in sexually exclusive romantic relationships), and the circumstances in
which they do not (e.g., in sexually non-exclusive relationships), may shed light on the role of
alternatives, and provide a framework for determining when they will exert a maximal impact on
relationship commitment.
Our second goal in this work was to examine two commonly researched consequences of
commitment to determine whether commitment to a FWBR is predictive of the same outcomes.
The first, relationship change, is a widely studied outcome in relationship research (Le, Smoak,
Agnew, Korn, & Mutso, in press), but rarely is it studied in a context that allows for change to
not only a less interdependent relationship type (e.g., no relationship), but also to one with even
greater interdependence (e.g., a romantic relationship). Our results indicated that commitment to
all three aspects of the relationship (i.e., friendship, sexual, and general) was predictive of the
desired end state of the relationship, and that within each type of commitment, the same pattern
Commitment in Friends with Benefits Relationships 20
of desired future states was evidenced: those who wanted their FWBR to transition into a
romantic relationship had the highest commitment level, followed by those who desired their
FWBR remain intact, followed by those who desired their relationship transition into a
friendship, and finally those who desired no relationship at all. Notably, global FWBR
commitment was equally high among those who desired their relationship transition into a
romantic relationship and those who wanted their FWBR to remain intact. In that way, it is clear
that there are people for whom a FWBR is the desired end state of the relationship in and of itself
(i.e., those high in generalized commitment to the FWBR), but for most, being highly committed
to one aspect of the relationship was associated with a desire to form a romantic relationship with
their FWBR partner.
Finally, we believed that condom use would be especially important to study in the
context of a FWBR, given that sex outside of exclusive relationships is a known risk factor for
STI acquisition and unplanned pregnancy (e.g., Levinson, Jaccard, & Beamer, 1995). As
expected, commitment to both the FWBR and to the friendship aspect of the relationship were
associated with lower frequency of condom use. These effects were robust, and were not
moderated by myriad relationship variables (e.g., non-exclusivity, desired future state). Contrary
to expectations, though, commitment to the sexual aspect of the relationship was not
significantly related to condom use. Taken together, these results suggest that the friendship at
the base of a FWBR may be an important contributor to partners foregoing condom use. Again,
future research is necessary to determine if there is indeed a casual association here, but we
believe this may have interesting implications for the study of condom use more generally.
Potentially, interventions designed to bolster intentions to use condoms in any relationship type
may benefit from a focus on combating erroneous beliefs regarding the friendship element of the
Commitment in Friends with Benefits Relationships 21
relationship, which we theorize is what leads partners to put themselves at risk (e.g., by assuming
that an intimate partner is unlikely to harm them; Gerrard et al., 1996).
Limitations and Strengths
There are some strengths and limitations of this work that merit discussion. Of
importance, because we employed the Internet for data collection, our sample was
demographically more diverse than most past work on FWBRs. For example, our participants’
ages spanned 47 years, indicating that this type of relationship is not limited to young adults.
This stands in stark contrast to most previous studies of FWBRs, which have focused primarily
on college student samples. We should note, though, that Internet data collection is known to
yield samples that may not be representative of the overall population, particularly with regard to
education level and socio-economic status (Gosling, Vazire, Srivastava, & John, 2004). Thus,
despite being reasonably diverse, our sample may not necessarily reflect the full range and
variability of individuals involved in FWB relationships.
Additionally, we collected only minimal data from individuals who were not involved in
FWBRs, which limited our ability to compare the structure and functioning of FWBRs to other
types of partnerships, such as romances and friendships. We believe future work aimed at
determining how FWBRs differ from other types of relationships in terms of both the structure
(e.g., how they began, what the partners hope for the future of the relationship) and dynamics
(e.g., antecedents of commitment) would be beneficial to understanding not only FWBRs, but
also relationship processes in general.
One final limitation is that we did not study whether FWBRs actually transformed over
time. Instead, we studied participants’ desires for the future of their relationships. Future research
would benefit from studying how these relationships change over time, obtaining data from both
Commitment in Friends with Benefits Relationships 22
members of the partnership, and examining whether commitment longitudinally predicts the
transformations our results suggest.
Conclusion
FWBRs have the potential to provide information that is relevant both theoretically and
practically to many different types of involvements. Because the components of these
relationships are not cohesively integrated as they are in a typical romance, they provide fertile
ground to test predictions about the respective roles and impacts of friendship and sex within a
relationship. Much is yet to be learned about these relationships, though, and we hope the current
study provides initial results sufficient to incite others to explore these partnerships in their own
work.
Commitment in Friends with Benefits Relationships 23
References
Agnew, C. R., Hoffman, A. M., Lehmiller, J. J., & Duncan, N. T. (2007). From the interpersonal
to the international: Understanding commitment to the "War on terror.". Personality and
Social Psychology Bulletin, 33, 1559-1571.
Arriaga, X. B., & Agnew, C. R. (2001). Being committed: Affective, cognitive, and conative
components of relationship commitment. Personality and Social Psychology Bulletin, 27,
1190-1203.
Baumeister, R. F., Wotman, S. R., & Stillwell, A. M. (1993). Unrequited love: On heartbreak,
anger, guilt, scriptlessness, and humiliation. Journal of Personality and Social
Psychology, 64, 377-394.
Bisson, M. A., & Levine, T. R. (2009). Negotiating a friends with benefits relationship. Archives
of Sex Behavior, 38, 66-73.
Carpenter, P. J., & Scanlan, T. K. (1998). A longitudinal test of elite youth cricketers'
commitment. International Journal of Sport Psychology, 29, 195-365.
Davis, K. E., & Todd, M. J. (1982). Friendship and love relationships. Advances in Descriptive
Psychology, 2, 79-122.
Gerrard, M., Gibbons, F. X., & Bushman, B. J. (1996). Relation between perceived vulnerability
to HIV and precautionary sexual behavior. Psychological Bulletin, 119, 390-409.
Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust web-based
studies? American Psychologist, 59, 93-104.
Hatcher, L., Kryter, K., Prus, J. S., & Fitzgeralt, V. (1992). Predicting college student
satisfaction, commitment, and attrition from Investment Model constructs. Journal of
Applied Social Psychology, 22.
Commitment in Friends with Benefits Relationships 24
Hoffman, A. M., Agnew, C. R., Lehmiller, J. J., & Duncan, N. T. (2009). Satisfaction,
alternatives, investments, and the microfoundations of audience cost models.
International Interactions, 35, 365-389.
Hughes, M., Morrison, K., & Asada, K. J. K. (2005). What's love got to do with it? Exploring the
impact of maintenance rules, love attitudes, and network support on friends with benefits
relationships. Western Journal of Communication, 69, 49-66.
Le, B., & Agnew, C. R. (2003). Commitment and its theorized determinants: A meta-analysis of
the investment model. Personal Relationships, 10, 37-57.
Le, B., Smoak, N. D., Agnew, C. R., Korn, M. S., & Mutso, A. A. (in press). Predicting non-
marital romantic relationship dissolution: A meta-analytic synthesis. Personal
Relationships.
Lehmiller, J. J. (2008). Conducting relationships research online: A primer on getting started and
making the most of your resources. Relationship Research News, 7, 6-12.
Lehmiller, J. J. (2010). Differences in relationship investments between gay and heterosexual
men. Personal Relationships, 17, 81-96.
Lehmiller, J. J., & Agnew, C. R. (2006). Marginalized relationships: The impact of social
disapproval on romantic relationship commitment. [Original]. Personality and Social
Psychology Bulletin, 32, 40-51.
Lehmiller, J. J., VanderDrift, L. E., & Kelly, J. R. (in press). Sex differences in approaching
friends with benefits relationships. Journal of Sex Research.
Levinson, R. A., Jaccard, J., & Beamer, L. (1995). Older adolescents' engagement in casual sex:
Impact of risk perception and psychosocial motivations. Journal of Youth and
Adolescence, 24, 349-364.
Commitment in Friends with Benefits Relationships 25
Lin, Y. W., & Rusbult, C. E. (1995). Commitment to dating relationships and cross-sex
friendships in America and China. Journal of Social and Personal Relationships, 12, 7-
26.
Misovich, S. J., Fisher, J. D., & Fisher, W. A. (1997). Close relationship and elevated HIV risk
behavior: Evidence and possible underlying psychological processes. General
Psychology Review, 1, 72-107.
Oliver, N. (1990). Rewards, investments, alternatives and organizational commitment: Empirical
evidence and theoretical development. Journal of Occupational Psychology, 63, 19-31.
Reis, H. T., Sheldon, K. M., Gable, S. L., Roscoe, J., & Ryan, R. M. (2000). Daily well-being:
The role of autonomy, competence, and relatedness. Personality and Social Psychology
Bulletin, 26, 419-435.
Rusbult, C. E. (1980). Commitment and satisfaction in romantic associations: A test of the
investment model. Journal of Experimental Social Psychology, 16, 172-186.
Rusbult, C. E. (1983). A longitudinal test of the investment model: The development (and
deterioration) of satisfaction and commitment in heterosexual involvements. Journal of
Personality and Social Psychology, 45, 101-117.
Rusbult, C. E., & Farrell, D. (1983). A longitudinal test of the Investment Model: The impact on
job satisfaction, job commitment, and turnover of variations in rewards, costs,
alternatives, and investments. Journal of Applied Psychology, 68.
Rusbult, C. E., Martz, J. M., & Agnew, C. R. (1998). The Investment Model Scale: Measuring
commitment level, satisfaction level, quality of alternatives, and investment size.
Personal Relationships, 5, 357-391.
Commitment in Friends with Benefits Relationships 26
Tucker, J. S., Elliott, M. N., Wenzel, S. L., & Hambarsoomian, K. (2007). Relationship
commimtment and its implications for unprotected sex among impoverished women
living in shelters and low-income housing in Los Angeles County. Health Psychology,
26, 644-649.
VanderDrift, L. E., Agnew, C. R., & Wilson, J. E. (2009). Non-marital romantic relationship
commitment and leave behavior: The mediating role of dissolution consideration.
Personality and Social Psychology Bulletin, 35, 1220-1232.
Wieselquist, J., Rusbult, C. E., Foster, C. A., & Agnew, C. R. (1999). Commitment, pro-
relationship behavior, and trust in close relationships. Journal of Personality and Social
Psychology, 77, 942-966.
Williams, J., Shaw, C., Mongeau, P. A., Knight, K., & Ramirez, A. (2007). Peaches n’ cream to
rocky road: Five flavors of friends with benefits relationships. University of Arizona at
Tempe.
Commitment in Friends with Benefits Relationships 27
Footnote
1Past research on commitment to various targets has found that the structure of commitment does
not differ significantly by sex (see Le & Agnew, 2003 for a review). Thus, we did not proffer
any gender-related predictions in these analyses. For exploratory purposes, we did test whether
there were gender differences in the structure of commitment to the FWBR, and found that
neither the satisfaction-commitment association (friendship: Z = .23, p > .05; sexual relationship:
Z = .20, p > .05), nor the alternatives-commitment association (friendship: Z = -.78, p > .05;
sexual relationship: Z = .21, p > .05), nor the investment-commitment association (friendship: Z
= -1.16, p > .05; sexual relationship: Z = -.88, p > .05) differed significantly based on gender.
Thus, because the structure of commitment does not appear to differ as a function of gender, we
did not include any gender-based analyses in the remainder of this paper. We do not wish to
imply that gender does not potentially play an important role in this relationship type, however,
because it certainly does in some respects. For a full discussion of the role of gender in FWBRs,
see Lehmiller, VanderDrift, and Kelly (2010).
Commitment in Friends with Benefits Relationships 28
Table 1.
Frequencies of friendship and sexual activities engaged in by sample participants.
Activity
Frequency (% of sample)
Friendship
Talked about something meaningful
226 (91.9%)
Gone out together with a larger group of friends
144 (58.5%)
Hung out casually
207 (84.2%)
Made each other feel insecure or self-conscious
75 (30.5%)
Argued or had a conflict
126 (51.2%)
Gone on what could be considered a “date”
134 (54.5%)
Tried something new together
123 (50.0%)
Sexual Relationship
Kissing
228 (92.7%)
Sexual touching
236 (95.9%)
Mutual masturbation
134 (54.5%)
Oral sex
216 (87.8%)
Vaginal intercourse
241 (98.0%)
Anal intercourse
49 (19.9%)
Commitment in Friends with Benefits Relationships 29
Table 2.
Descriptive statistics of and correlations between the Investment Model Scale Variables
Target
Friendship Relationship
Sexual Relationship
FWBR
Scale
1. SAT
2. ALT
3. INV
4. COM
5. SAT
6. ALT
7. INV
8. COM
9. COM
Mean (SD)
5.66 (2.25)
5.30 (2.16)
5.87 (2.40)
6.44 (2.40)
6.45 (2.11)
5.38 (2.36)
5.82 (2.25)
5.75 (2.42)
4.95 (2.13)
Reliability
0.814
0.654
0.863
0.932
0.905
0.762
0.849
0.848
0.791
1
-
-
-
-
-
-
-
-
-
2
.12
-
-
-
-
-
-
-
-
3
.62***
.01
-
-
-
-
-
-
-
4
.63***
-.07
.81***
-
-
-
-
-
-
5
.53***
.08
.43***
.38***
-
-
-
-
-
6
-.04
.58***
-.15*
-.20**
-.03
-
-
-
-
7
.48***
.01
.72***
.60***
.64***
-.01
-
-
-
8
.48***
-.10
.61***
.63***
.58***
-.05
.83***
-
-
9
.58***
-.03
.61***
.67***
.43***
-.10
.63***
.73***
-
Note. SAT = Satisfaction, ALT = Alternatives, INV = Investment, COM = Commitment
* < .05, ** < .01, *** < .001
Commitment in Friends with Benefits Relationships 30
Table 3.
Mean levels of each type of commitment at each desired future relationship state
Desired Future State of Relationship
More interdependent ß------------------------------------------à Less interdependent
Commitment
Romantic
Relationship
Continue FWBR
Friendship Only
No Relationship
Friendship
7.41a
6.29b
6.13b
2.97c
Sexual Relationship
6.87a
5.90b
4.29c
3.03c
FWBR
5.70a
5.36a
3.54b
2.47c
Note. Subscripts denote statistically significant differences within commitment type (i.e., within rows), as determined by multinomial
logistic regresion.
