Arch Sex Behav (2010) 39:1105–1119

DOI 10.1007/s10508-008-9448-4

ORIGINAL PAPER

Predictors and Consequences of Sexual ‘‘Hookups’’

Among College Students: A Short-term Prospective Study
Robyn L. Fielder Æ Michael P. Carey

Received: 14 July 2008 / Revised: 29 October 2008 / Accepted: 29 October 2008 / Published online: 9 January 2009
Ó Springer Science+Business Media, LLC 2009

Abstract ‘‘Hookups,’’ sexual interactions between part- Introduction

ners who do not expect a romantic commitment, are believed
to be common among adolescents and young adults. Most ‘‘Hookup’’ is a catch-all term used by adolescents and young
existing research is cross-sectional and has not investigated adults to describe a sexual interaction between two partners
the antecedents or consequences of hookups. To our knowl- who expect no romantic commitment. Hookups are believed
edge, this study provides the first prospective investigation of to be very common on college campuses, with estimates
the hypothesized predictors of penetrative sex hookups (i.e., ranging as high as 81% of students reporting at least one
oral, vaginal, and anal sex) and the first exploration of the hookup experience (Bisson & Levine, in press; Lambert,
short-term mental health consequences of hookups. A total of Kahn, & Apple, 2003; Owen, Rhoades, Stanley, & Fincham,
140 first-semester college students (109 females, 31 males) in press; Paul & Hayes, 2002; Paul, McManus, & Hayes,
completed an anonymous survey early in their first semester; 2000). Popular books (e.g., Bogle, 2008; Seaman, 2005;
the survey assessed 18 potential predictors of hooking up that Stepp, 2007) suggest that hookups have negative outcomes,
were identified from theory and past research. At the end of but research on the predictors and consequences of hookups
their first semester, students again responded to a survey and has been scarce.
provided data on their oral and vaginal sex hookup behavior
(occurrence and number of partners), distress, and self-es- Predictors of Sexual Hookups
teem. Baseline and follow-up data were linked using unique
codes that protected participants’ anonymity. Pre-college Potential predictors of sexual hookups have been suggested
hookup patterns, peak intoxication level, and situational by both theory and research. For example, the Theory of
triggers for hookups were consistent predictors of oral and Interpersonal Behavior (Triandis, 1977, 1980) has been used
vaginal sex hookup behavior (and number of hookup part- to explain casual sex intentions and behavior (Apostolopo-
ners) in the first semester of college. Penetrative sex hookups ulos, Sönmez, & Yu, 2002; Maticka-Tyndale, Herold, &
increased psychological distress for females, but not for Mewhinney, 1998). Triandis suggested that attitudes and
males. Implications for education and intervention as well as norms influence behavioral intentions, which—along with
suggestions for future research are discussed. situational factors and prior experience with a behavior—
determine whether an individual will engage in a future
Keywords Sexual behavior
Hookup
Casual sex
(sexual) behavior.
Mental health
Adolescents
College students Evolutionary theories (e.g., Symons, 1987) may also help
to increase understanding of the origins of hookup behavior.
This perspective suggests that gender should be an important
predictor of hookup behavior; that is, because males accrue
advantages from having multiple partners, they should be
R. L. Fielder
M. P. Carey (&)
more likely to engage in hookups. In contrast, females would
Center for Health and Behavior, Syracuse University,
430 Huntington Hall, Syracuse, NY 13244-2340, USA be expected to eschew sexual encounters devoid of emotional
e-mail: [email protected] intimacy in order to find a mate who invests more in the

123
1106 Arch Sex Behav (2010) 39:1105–1119

relationship. Sociocultural perspectives might challenge this provide sexual interaction with interesting or attractive men
view, and suggest that, because gender roles are socially without compromising their freedom or independence.
constructed, differences between men and women should Several parental factors may influence hookups. For
diminish as social norms change to be more egalitarian. example, there is some evidence that the marital status of
Social-cognitive theory (Bandura, 1977) provides a more parents may influence adolescents’ attitudes about relation-
general framework for understanding how the social envi- ships and sexual behavior (Jennings, Salts, & Smith, 1991;
ronment shapes behavior. This approach recognizes the Newcomer & Udry, 1987). Numerous studies have found an
powerful role that modeling and vicarious learning play in the effect of parental attitudes about sex on the sexual attitudes
formation of behavior patterns. For many young people, the and behavior of adolescents (Daugherty & Burger, 1984;
most salient model of an intimate relationship is their parents’ Dittus & Jaccard, 2000; Maguen & Armistead, 2006). Fur-
relationship. In addition, the immediate social environment thermore, parental discouragement of relationships may
of the college campus (represented by social norms) and the motivate some young people to avoid serious relationships
larger cultural context, transmitted through mass media, due to parental messages suggesting that relationships might
would also be expected to shape sexual behavior. reduce their opportunities or distract them from academic and
In addition to these theoretical frameworks, prior quali- career goals (Glenn & Marquardt, 2001; Stepp, 2007).
tative and quantitative research has identified other potential There are other situational factors that influence college
predictors of hookups, comprising person characteristics, students’ beliefs and hookup practices. Alcohol use, social
parental influences, and situational (social–cultural) factors. norms, and exposure to media messages about sexuality
At least four person variables might be expected to influ- emerge as likely influences on hookups. Alcohol use (and
ence hookup behavior: religiosity, gender, career-minded- intoxication) before hookups is common; for example, 65%
ness, and the desire to be carefree. The popularity of of Grello, Welsh, and Harper’s (2006) sample reported
‘‘virginity pledges’’ suggests that religiosity might affect drinking before their most recent episode of casual sex. Paul
sexual behavior and, indeed, some research suggests that et al. (2000) found that frequency of alcohol intoxication was
religiosity affects sexual behavior choices (e.g., Rostosky, lowest among individuals who had never hooked up, was
Wilcox, Wright, & Randall, 2004; Thornton & Camburn, higher among those who had a history of hookups without
1989; Zaleski & Schiaffino, 2000). Religious feelings and sexual intercourse, and was highest among those who had a
attendance at religious services were related to number of history of hookups with sexual intercourse. Moreover, Owen
hookups and frequency of intercourse during hookups in a et al. (in press) found that an average of quantity and fre-
recent study (Penhollow, Young, & Bailey, 2007). quency of alcohol use was a multivariate predictor of hooking
As noted earlier, gender is likely to influence hookup up in a recent cross-sectional study.
behavior. Compared to women, men have more sexual part- Social norms have also been implicated as possible
ners, are more sexually permissive (Oliver & Hyde, 1993), determinants of sexual hookups. College students overesti-
and are more likely to engage in sex without emotional mate the frequency of their peers’ sexual behavior and
involvement (Maticka-Tyndale et al., 1998; Townsend, number of sexual partners (Martens et al., 2006; Scholly,
1995). Despite this, prior studies of hooking up have found no Katz, Gascoigne, & Holck, 2005) as well as acceptance of
gender differences in hookup experience (Flack et al., 2007; casual sex (Cohen & Shotland, 1996). Norm misperceptions
Paul & Hayes, 2002). The lack of gender differences has led to are associated with increased sexual activity and multiple
speculation about a possible ‘‘change’’ in gender roles. partners (Page, Hammermeister, & Scanlan, 2000). Students
Based on interviews with high-achieving female college also overestimate the percentage of their peers with hookup
and high school students, Stepp (2007) suggests that, for experience (actual: 70% vs. estimated: 85%; Paul & Hayes,
some females, career-mindedness (i.e., academic and career 2002). The perception that ‘‘everyone’s doing it’’ may
goals) may take priority over personal relationships; for such encourage some students to hook up themselves.
females, the time commitment needed for a long-term rela- There are two types of social norms (Carey, Borsari,
tionship may limit their ability to seek out educational or Carey, & Maisto, 2006). Descriptive norms refer to an indi-
career opportunities that would benefit them personally, vidual’s perception of the prevalence of a certain behavior,
making them more willing to engage in hookup behavior in whereas injunctive norms refer to perceptions of peer ap-
lieu of committed relationships. Similar to the qualitative proval of a behavior. Both descriptive and injunctive norms
findings from Stepp (2007), Glenn and Marquardt (2001) tend to be overestimated for risky behaviors, and research
suggested that the desire to be carefree in college may suggests that the greater the discrepancy between a student’s
motivate hookup behavior, especially for high-achieving personal behaviors and attitudes, and the behaviors and
adolescent females. They argued that young women who attitudes of their peers (i.e., self-other differences [SODs]),
strive for self-sufficiency and independence prefer hookups the greater the pressure that student will feel to conform to the
to traditional committed relationships because hookups perceived norms.

123
Arch Sex Behav (2010) 39:1105–1119 1107

Finally, mass media (e.g., music videos, magazines, had engaged in sex with only romantic partners. Distress
the internet) are rife with sexual content (Escobar-Chaves increased for women, but not for men, as the number of
et al., 2005; Greenberg & Hofschire, 2000). Cross-sectional partners increased. The temporal sequence of the hookups
(L’Engle, Brown, & Kenneavy, 2006) and longitudinal and distress remains unclear due to the use of a cross-sec-
studies (Collins et al., 2004) demonstrate the effects that tional design. Another recent cross-sectional study revealed
media depictions of sex can have on adolescent sexual that male and female college students have different emo-
behavior. tional reactions to hookups (Owen et al., in press). Females
In summary, based on several social-cognitive theories, as were more likely than males to report a negative reaction to
well as empirical evidence, we identified a large number of hookups over the past year, and females were less likely than
plausible predictors of hookup behavior, from the individual males to report a positive reaction.
to the sociocultural level. At the time of its initiation, this Engaging in hookups may also affect other mental health
study was the first prospective study of predictors of hooking outcomes, such as an individual’s self-esteem. In the only
up. Therefore, we explored the utility of a wide variety of study to examine self-esteem related to hooking up in college
hypothesized predictors of hookup behaviors in college stu- students, Paul et al. (2000) found that both males and females
dents. The purpose of this exploratory study was to conduct who had ever hooked up had lower self-esteem than those
an initial evaluation of the strength of these hypothesized who had not; however, this study used a cross-sectional de-
predictors in order to improve our conceptual understanding sign, precluding causal inference.
of, and future research on, hookup behavior.
Study Objectives
Consequences of Hooking Up
The purposes of this study were: (1) to explore a range of
possible predictors of sexual hookups as suggested by pre-
Sexual behavior may involve risk for physical and mental
vious theory and research and (2) to investigate the short-term
health. Physical health consequences include unintended
psychological consequences of hooking up in college stu-
pregnancy, sexually transmitted infections (STIs), and sexual
dents. Unlike previous cross-sectional studies (e.g., Grello
assault. Unintended pregnancies are relatively rare on most
et al., 2006; Owen et al., in press; Paul et al., 2000; Paul
college campuses (Scholly et al., 2005) due to the wide-
& Hayes, 2002), we used a prospective design. Thus,
spread availability of hormonal contraceptives. STIs (Rim-
when students arrived on campus (T1), we assessed pre-
sza, 2005) and sexual assaults (Messman-Moore, Ward, &
college hookup behavior and hypothesized predictors of
Brown, in press) appear to be more common, but still have
future hookups; at the end of their first semester (T2), we
received relatively limited investigation, especially in the
assessed collegiate hookup behavior and psychological
hookup context. Challenges limiting such research include
consequences. We focused on penetrative sex (i.e., oral,
the relatively low base rates of some outcomes, the high cost
vaginal, and anal sex) hookups because of their public health
and perceived invasiveness of biological testing, and the
importance.
stigma and legal issues associated with sexual assault.
This prospective design permitted evaluation of two sets
Hookups might also be expected to have mental health
of hypotheses:
consequences, which are easier to investigate. According to
traditional ‘‘sociocultural expectations’’ (Paul, 2006, p. 146), 1. We predicted that penetrative sex hookup behavior at
men are lauded for sexual prowess and experience, whereas T2 would be more likely for students with the follow-
women are shamed for these. This ‘‘double standard’’ means ing characteristics at study entry: male gender, hav-
that women are more likely to feel guilty or anxious if they ing divorced parents, having engaged in pre-college
engage in casual sex (Herold & Mewhinney, 1993; Lottes, hookup behavior/having more pre-college hookup part-
1993). As a result, hooking up may lead to short-term psy- ners, greater self-other differences in injunctive norms,
chological distress for women. On the other hand, a pro- greater hookup prevalence overestimations, stronger
spective study of younger adolescents suggested that distress intentions to engage in hookups, less religiosity, more
may lead to hookups, not vice versa (Grello, Welsh, Harper, permissive parental attitudes toward hooking up, more
& Dickson, 2003); however, this relationship has not been situational triggers, higher peak intoxication levels,
examined prospectively in college students. In a cross-sec- greater media exposure, and more permissive media
tional study of college students, Grello et al. (2006) found that messages about hooking up. We also predicted interac-
women who had engaged in casual sex reported more distress tions between gender and five predictors, such that the
than virgins or women who had engaged in sex with only likelihood of T2 hookup behavior would be increased for
romantic partners. In contrast, men who had engaged in ca- females (but not males) who reported greater distress,
sual sex had lower levels of distress than virgins or men who lower self-esteem, greater career-mindedness, greater

123
1108 Arch Sex Behav (2010) 39:1105–1119

desire to be carefree in college, and greater parental dis- items. Only gender was used as a predictor. Sexual history
couragement of relationships. (i.e., number of lifetime oral, vaginal, and anal sex partners)
2. We predicted that females who transitioned from no was assessed with three items.
previous penetrative sex hookups at study entry to a To assess pre-college oral (vaginal) sex hookup behavior
penetrative sex hookup by the end of their first semester (yes/no), students were given the following definition3 of a
would report increased distress and decreased self- casual partner: ‘‘someone whom you were not dating or in a
esteem. romantic relationship with, and at the time of the sexual
interaction, you understood that there was no mutual expec-
tation of a romantic commitment.’’ Students were asked
with how many casual partners they had engaged in oral
Method
(vaginal) sex before arriving on campus. These responses
also provided the pre-college number of oral (vaginal) sex
Participants
hookup partners.
Religiosity was measured using one item that asked stu-
Participants were 140 first-semester college students, 18 to
dents to indicate the intensity of their religious beliefs
19 years old (M = 18.03 years, SD = 0.18).1 Most were
(0 = not at all intense to 20 = very intense). This item was
female (78%) and Caucasian (69%); other racial/ethnic
reported by Mahoney (1980) to have a high (r = .88) cor-
identities included Asian (13%), Hispanic (10%), African
relation with Rohrbaugh and Jessor’s (1975) eight-item
American (5%), and other (3%). The sample was represen-
religiosity scale that assesses four dimensions of religion.
tative of the typical psychology class from which they were
Self-esteem was measured with the Rosenberg (1965)
recruited.2 Most (61%) females were single or uninvolved at
scale. The 10-item scale is internally consistent (a = .86),
study entry, whereas 18% were in a committed relationship,
has high test–retest reliability (r = .82), and has demon-
19% were dating one person, and 1% were dating more than
strated convergent and discriminant validity (Blascovich &
one person. Most (77%) males were single or uninvolved at
Tomaka, 1991). Self-esteem was used as a predictor of T2
study entry, whereas 13% were in a committed relationship,
hookups and as a consequence of T1 hookup transition.
6% were dating one person, and 3% were dating more than
Distress was assessed with the 9-item Center for Epide-
one person. On average, females reported 2.5 lifetime oral
miological Studies-Depression scale (CES-D; Santor &
sex partners (SD = 3.3, median = 2) and 1.5 lifetime vagi-
Coyne, 1997). The 9-item CES-D correlates highly with the
nal sex partners (SD = 1.9, median = 1), and males reported
20-item version (r = .93). Higher scores indicate greater
2.8 lifetime oral sex partners (SD = 4.1, median = 1) and
distress (a = .78). Distress was used as a predictor of T2
1.6 lifetime vaginal sex partners (SD = 3.0, median = 1).
hookups and as a consequence of T1 hookup transition.
However, 25% of participants reported that they had not yet
Intentions were assessed by asking students to rate
had either oral or vaginal sex.
(1 = strongly disagree to 7 = strongly agree) two state-
ments: (1) in the coming semester, I plan to have oral sex with
Measures a casual partner, and (2) in the coming semester, I plan to have
vaginal sex with a casual partner.
Descriptive information (i.e., gender, age, ethnicity, sexual Social norms were assessed following published proce-
orientation, relationship status, weight) was assessed with six dures (Carey et al., 2006). Therefore, to assess general
injunctive norms, students rated (1 = strongly disagree to
7 = strongly agree) four statements: (1) hooking up is a part
of the college experience, (2) college students are expected to
hook up, (3) freshmen look forward to being able to hook up at
1
A total of 158 students enrolled in the study; for the T2 assessment, college, and (4) hooking up is important to my social life (cf.
144 participants returned, for a retention rate of 91%. There were no
differences between attriters and completers on any of the predictor
variables or number of oral or vaginal sex hookup partners at study entry.
Of the 144 students providing data at both T1 and T2, two students were
excluded due to sexual orientation (because the hookups of homosexual 3
Pilot testing of our survey items with 10 individuals confirmed that
individuals may differ from those of heterosexual individuals), and two participants understood the definition provided in the survey. All pilot
were excluded because they provided contradictory answers on their T1 participants completed the survey and were asked for feedback
surveys (undermining data quality). Thus, the final sample size at T2 was regarding the survey. None of the pilot participants expressed confusion
140. over or suggested clarification regarding any of the relationship or
2
Enrollment in the introductory psychology course is typically at least sexual behavior terms used in the survey. Furthermore, none of the 158
65% female. At the university overall, 62% of first-year students that participants asked questions about the meaning of terms or definitions
year were Caucasian. used in the survey.

123
Arch Sex Behav (2010) 39:1105–1119 1109

Rimal & Real, 2005). Students also rated the extent to which a higher scores indicate a greater effect of the situation on the
typical male and female freshman would agree with those participant.4
statements. To assess self-other differences (SOD) for the Perceived parental attitudes toward hooking up were
general injunctive norm (GIN-SOD), the student’s average assessed with items adapted from Daugherty and Burger
score was subtracted from the average score that the student (1984). Participants rated (1 = strongly disagree to 7 =
provided for the typical same-sex freshman. A positive value strongly agree) the extent to which their parents would agree
indicates that the typical same-sex student is perceived to be with four statements: (1) hooking up is bad or wrong, (2)
more permissive than the participant. hooking up is pleasurable or fun, (3) there are problems
Next, to assess hookup limits, students were asked connected with hooking up (pregnancy, loss of respect,
‘‘Which statement best captures how far (1) you, (2) the emotional difficulties), and (4) hooking up is okay. After
typical male freshman at this school, and (3) the typical fe- reverse scoring, responses were averaged (a = .79); higher
male freshman at this school, would think it is okay to go scores indicate more permissive attitudes.
sexually with a casual partner?’’ Response options were: no Parental marital status was assessed by asking students if
sexual activity of any kind, kissing, touching each other’s their biological parents were currently married. Parental
bodies, oral sex, and vaginal sex. To obtain a hookup limit discouragement of relationships was assessed with two items
injunctive norm SOD (HLIN-SOD), the rank of the statement designed to capture this construct: (a) my parents encourage
endorsed by the student was subtracted from the rank of the me to avoid getting too serious in romantic relationships
typical same-sex freshman. while I am young, and (b) my parents would be disappointed
To assess descriptive norms at T1, students estimated the if I got engaged or married while I was still in college. Par-
percentage of male and female freshmen who had engaged in ticipants rated (1 = strongly disagree to 7 = strongly agree)
oral and vaginal sex with a casual partner before college. each statement. Responses were averaged (a = .59), with
Prevalence accuracy was formed by subtracting the actual T1 higher scores indicating greater parental discouragement.
prevalence rate of oral (vaginal) sex with a casual partner Career-mindedness was assessed with four items: (a) I
among the student’s gender from the estimated T1 prevalence have educational and career goals that I want to accomplish
rate of oral (vaginal) sex with a casual partner among the before I settle down in a serious relationship, (b) I am too
participant’s gender. A positive value indicates that the stu- focused on succeeding in school to invest my time in a serious
dent overestimated the prevalence of hooking up. relationship right now, (c) With all my school, work, and/or
Peak intoxication level was assessed using peak blood social activities, I don’t have time for a serious relationship
alcohol content (BAC) in the past month. Participants indi- right now, and (d) I want to go to graduate school and/or
cated the number of standard drinks (i.e., a 10–12 oz. can or establish my career before I commit to a serious relationship.
bottle of 4–5%-alcohol beer, a 4-oz. glass of 12%-alcohol Participants rated (1 = strongly disagree to 7 = strongly
table wine, a 12-oz. bottle or can of wine cooler, or a 1.25-oz. agree) each statement. Responses were averaged (a = .83);
shot of 80-proof liquor either straight or in a mixed drink; higher scores indicate a greater degree of career-mindedness.
Dufour, 1999) they had on their heaviest drinking day in the Desire to be carefree in college was assessed with seven
past month and how many hours passed from the beginning of items (e.g., Being involved in a committed relationship
the first drink to the finishing of the last drink (Carey et al., would prevent me from enjoying my time in college to the
2006). Peak intoxication level was calculated using the for- fullest; I don’t want to be ‘‘tied down’’ with a committed
mula BAC = [(drinks/2) * (GC/weight)] - (.016*hours), relationship while I am in college). Participants rated
where (1) drinks = number of standard drinks consumed, (2) (1 = strongly disagree to 7 = strongly agree) the extent to
GC = gender constant (9.0 for females and 7.5 for males), which they agreed with each statement. Responses were
(3) weight = weight in pounds, and (4) hours = number of averaged (a = .87); higher scores indicate a greater desire
hours over which the drinks were consumed (Matthews & to be carefree.
Miller, 1979). Media exposure was assessed using seven items asking
Situational triggers for oral (vaginal) sex hookups were participants how many hours they spend in a typical week (1)
assessed with three items (adapted from Apostolopoulos watching television, (2) listening to music, (3) watching
et al., 2002; Herold, Maticka-Tyndale, & Mewhinney, 1998; music videos, (4) reading popular magazines, (5) watching
Maticka-Tyndale et al., 1998). Participants rated (1 = not at movies, (6) using social networking websites, and (7) reading
all likely to 7 = extremely likely) if they would engage in campus newspapers. The number of hours spent using all
oral (vaginal) sex with a casual partner in three situations: (1)
4
when you meet someone at a bar or party, (2) when someone The situational triggers predictors were not confounded with alcohol
use. Peak intoxication level and situational triggers for oral sex hookups
attractive wants to hook up with you, and (3) when it seems
did not correlate highly, r = .22, p = .005, and peak intoxication level
like everyone else is hooking up. Responses to these items and situational triggers for vaginal sex hookups were not correlated,
were averaged (oral sex a = .89, vaginal sex a = .88); r = .07, p = .42.

123
1110 Arch Sex Behav (2010) 39:1105–1119

seven media types were summed to create a composite media know. This nine-digit code was designed such that (1) it
exposure score.5 would protect participants’ anonymity (i.e., it could not be
Permissiveness of media messages about hooking up was linked to participants’ identities by the researchers) and (2) its
assessed with seven items adapted from a study by L’Engle contents could not be forgotten by the participants (i.e.,
et al. (2006). Participants rated their agreement (1 = strongly all digits were unlikely to change over the course of the
disagree to 7 = strongly agree) with items such as ‘‘The semester). The code comprised the first two digits of the
messages that college students get from television shows are participants’ university identification number, their day of
that it’s okay for people our age to hook up.’’ The other items birth (e.g., 05 for the 5th), the second two digits of their
substituted songs lyrics, music videos, magazines, movies, university identification number, their month of birth (e.g., 07
social networking websites, and campus newspapers for for July), and the first letter of the city in which they were
television shows. Responses were averaged (a = .88), with born. The code allowed the T1 and T2 data to be linked but
higher scores indicating more permissive perceived media also to remain anonymous.
messages about hooking up.
The dichotomous outcome variable T2 oral (vaginal) sex Data Analysis
hookup behavior (i.e., engaged in oral [vaginal] sex with a
casual partner in the first semester: yes/no) was determined Prior to running any statistical tests, the data were examined
based on the students’ responses to these questions: Since you for outliers and checked for univariate and multivariate
arrived on campus, with how many casual partners have you normality. The relatively few outliers were replaced with the
had oral (vaginal) sex? Students who reported zero partners unstandardized score for which z = 3. Continuous predictors
were coded as ‘‘no,’’ and participants who reported one or were inspected for multicollinearity and were centered at
more partners were coded as ‘‘yes.’’ Responses to this ques- their means prior to analysis (Tabachnick & Fidell, 2007). All
tion also indicated T2 number of oral (vaginal) sex hookup p values were two-tailed unless otherwise indicated, with
partners. alpha set at .05. When a directional effect was hypothesized
(i.e., for a priori contrasts), one-tailed tests were used.
Procedure Stepwise regression was used for model development due
to the large number of predictors included, the lack of pre-
Students enrolled in an Introductory Psychology course re- vious research on predictors of hooking up, and the absence
sponded to electronically posted invitations to participate in a of theoretical rationale for entering certain predictors first.
study of the health behaviors and interpersonal relationships The first step in the model development process was uni-
of young adults. After receiving an overview of the study, variate analyses to determine which predictors to test in the
students provided written consent and completed a self- exploratory multivariate model. All predictors with p \ .25
administered, anonymous survey in small groups with ample in univariate analyses were retained for multivariate analyses
privacy. The initial survey was administered in mid-to-late (Hosmer & Lemeshow, 2000). Stepwise logistic regression
September (T1); the follow-up survey was administered was conducted using an entry probability of .15 and a removal
10 weeks later, during the last 2 weeks of the semester (T2). probability of .20 (Hosmer & Lemeshow, 2000). For the two
Predictor variables were assessed at T1, and criterion vari- logistic regression models, the continuous variables selected
ables were assessed at T2. Each survey took 30 min to by the stepwise regression procedure were checked for lin-
complete. Upon completion, students received course credit. earity in the logit; there were no violations of this assumption.
All procedures were approved by the Institutional Review Next, all potential interactions between variables selected in
Board. the stepwise regression were tested. Significant interactions
Participants’ responses on the T1 and T2 surveys were and the predictors were then entered simultaneously into
linked using a unique identification code that only they would another regression model. If any predictors had p [ .05 in
these models, the models were re-run without those predic-
5
To minimize participant burden, these questions measure general tors, and the models with and without the predictor were
media exposure rather than exposure to sexual media content in compared to determine if the predictor should be retained.
particular. A recent review confirmed that sexual content is pervasive on Models were also run with and without any interactions to
television, in song lyrics and music videos, in movies, in magazines, and
determine if the interactions should be retained in the final
on the internet (Escobar-Chaves et al., 2005); thus, we assumed that
participants with greater media exposure in general would have greater model.
exposure to sexual media content. Social networking internet websites Paired samples t-tests were used for the specific tests of
were included due to their popularity with college students in particular, hypothesis two. Two separate repeated-measures analyses
their high traffic volume, and the high amount of photo-sharing (e.g.,
of variance (ANOVAs) were conducted on distress and
Facebook: Statistics, n.d.). Campus newspapers were included because
the campus newspaper had featured articles on hooking up within the self-esteem for further analysis of the effect of hookup tran-
past year (e.g., Tousigant, 2007). sition group among women; the within-groups independent

123
Arch Sex Behav (2010) 39:1105–1119 1111

Table 1 Univariate relationships between predictors and sexual risk behaviors

Predictors (measured at Time 1) Criterion variables (measured at Time 2)
Oral sex hookup Oral sex Vaginal sex Vaginal sex
behavior no. of partners hookup behavior no. of partners

Self-esteem .17
Psychological distress .12 .12
Intentions to hook up in first semester \.0001 \.0001 .003 .005
Career-mindedness .06
Desire to be carefree in college .002 \.0001 .06 .06
Parental attitudes toward hooking up .23
Parental discouragement of relationships .14
Situational triggers for hookups \.0001 \.0001 .003 .002
Peak intoxication level \.0001 \.0001 .0003 .001
Media exposure .24
Permissiveness of media messages about hooking up .04 .04 .23
a
Prevalence accuracy .11 .02 .15 .19
Injunctive norm SOD, limit .01 .006 .0008 .005
Injunctive norm SOD, general .01 .002 .02 .006
Gender .23 .18
Pre-college hookup behavior or number of partners .0001 \.0001 \.0001 .0003
Parental marital status
Religiosity .15
Gender 9 self-esteem
Gender 9 psychological distress .22
Gender 9 desire to be carefree
Gender 9 parental discouragement of relationships .18
Gender 9 career-mindedness
Note: Unless otherwise noted, N = 138 for oral sex hookup behavior and number of hookup partners analyses, and N = 140 for vaginal sex hookup
behavior and number of partners analyses. The absence of a value in any cell indicates that p [ .25. SOD = self-other difference
a
N = 137 for oral sex analyses and 139 for vaginal sex analyses

variable was time, and the between-groups independent partners (count). These four outcomes were selected because
variable was hookup transition group.6 There were three of their public health importance (i.e., greater likelihood of
groups: inexperienced—no transition (i.e., never hooked up generating mental and physical health consequences).7
at T1 or T2), transition-to-hooking-up (i.e., had not hooked
up at T1 but hooked up at T2), and experienced (i.e., had Oral Sex Hookup Behavior
hooked up at T1 and may or may not have hooked up at T2).
The dependent variable (DV) in this model was T2 oral
sex hookup behavior (yes/no). During their first semester
Results of college, 33% of the sample reported an oral sex hook-
up. Table 1 displays predictors with p-values \ .25 in the
Predictors of Hooking Up univariate logistic regression analyses. In the stepwise
regression procedure, three predictors entered the model:
Using a prospective design and psychometrically valid situational triggers for oral sex hookups, peak intoxication
measures of hypothesized predictors, we sought to develop level, and pre-college oral sex hookup behavior. There were
models of four outcomes: oral sex hookup behavior (yes/no), no significant interactions. In the final model, situational
vaginal sex hookup behavior (yes/no), number of oral sex triggers for oral sex hookups, peak intoxication level, and pre-
hookup partners (count), and number of vaginal sex hookup college oral sex hookup behavior predicted T2 oral sex

6 7
The effect of hookup transition group could not be meaningfully No participants reported engaging in anal sex during a hookup;
examined in males due to small cell sizes. therefore, we report only on oral and vaginal sex hookups.

123
1112 Arch Sex Behav (2010) 39:1105–1119

Table 2 Final model: multivariate predictors of Time 2 oral sex hookup behavior
Predictor (measured at Time 1) B SE Wald v2 (df = 1) p Adjusted odds ratio 95% CI

Intercept -1.44 0.33 19.40 \.0001

Situational triggers for oral sex hookups 0.46 0.13 12.47 .0004 1.58 1.23–2.04
Peak intoxication levela 6.09 2.40 6.43 .01 1.84 1.15–2.94
b
Pre-college oral sex hookup behavior 1.06 0.43 5.90 .02 2.88 1.23–6.74
Note: N = 140. B coefficients represent logits. SE = standard error; CI = confidence interval
a
For the odds ratio, the unit for BAC is .10, instead of 1
b
Reference group is no pre-college oral sex hookup behavior

hookups, Likelihood Ratio (LR) v2(df = 3) = 41.87, p \ The six predictors were entered as predictors for the
.0001. Parameter estimates and odds ratios for the final model multivariate model. The model was re-run with five, four,
are displayed in Table 2. and three predictors when distress, gender, and situational
Given a prior probability of 50%, the model correctly triggers for vaginal sex hookups, respectively, had p-val-
identified (ID) 76% of cases. Sensitivity, or the proportion of ues [ .05. The different models were compared on AIC,
participants who had an oral sex hookup in the first semester deviance, correct ID, number of predictors with p [ .05,
of college and were correctly predicted as doing so by the and number of predictors. The three-predictor model, LR
model, was 50%. Specificity, or the proportion of participants v2(df = 3) = 44.54, p \ .0001, was retained as the final
who did not have an oral sex hookup and were correctly model. The four-, five-, and six-predictor models produced
predicted as doing so by the model, was 88%. For this model, lower deviances and AICs than the three-predictor model;
c, which is equivalent to the area under the receiver operating however, they each included at least one non-significant
characteristic curve and ranges from .5 to 1 (Tabachnick & predictor. The difference in deviance between the three- and
Fidell, 2007), was .81. Chance prediction is represented by a c four-predictor models was not significant, v2(df = 1) = 3.2,
of .5 and perfect prediction is represented by a c of 1; c of .8–.9 p = .07; therefore, the model with fewer variables was fa-
is considered excellent discrimination (Hosmer & Leme- vored to avoid over-fitting the model (Tabachnick & Fidell,
show, 2000). The Hosmer and Lemeshow Goodness-of-Fit 2007). The three-predictor model also resulted in a higher
test was not significant, v2(df = 8) = 4.97, p = .76, indi- percentage of correct ID and specificity than the four- and
cating good model fit (Tabachnick & Fidell, 2007). The five-predictor models. Parameter estimates and odds ratios
Akaike Information Criterion (AIC) for the final model was for the final model appear in Table 3.
143.4 and the deviance was 135.4. The performance of the Given a prior probability of 50%, the final model correctly
final model improved on the fit of a model with pre-college identified 77% of cases. Sensitivity was 49%, and specificity
oral sex hookup behavior as the only predictor (AIC = 165.3, was 88%. The Hosmer and Lemeshow Goodness-of-Fit test
deviance = 161.3, correct ID = 44%) and a model with pre- was not significant, v2(8) = 3.54, p = .90, indicating good
college oral sex hookup behavior and peak intoxication level model fit (Tabachnick & Fidell, 2007), and c was .83. The
as the only predictors (AIC = 155.1, deviance = 149.1, final model’s AIC was 129.1, and deviance was 121.1. The
correct ID = 70%). The difference in deviance between the final model improved on the fit of a model with pre-college
final model and the two-predictor model was significant, LR vaginal sex hookup behavior as the only predictor (AIC =
v2(df = 1) = 13.7, p = .0002. 147.8, deviance = 143.8, correct ID = 76%) and a model
with pre-college vaginal sex hookup behavior and peak
Vaginal Sex Hookup Behavior intoxication level as the only predictors (AIC = 134.7,
deviance = 128.7, correct ID = 76%). The difference in
The DV in this model was T2 vaginal sex hookup behavior deviance between the final model and the two-predictor
(yes/no). During their first semester of college, 28% of model was significant, LR v2(df = 1) = 7.6, p = .006.
the sample reported a vaginal sex hookup. Table 1 displays
predictors with p-values \ .25 in the univariate logistic Number of Oral Sex Hookup Partners
regression analyses. In the stepwise regression procedure,
six predictors entered the model: pre-college vaginal sex The DV in this model was T2 number of oral sex hookup
hookup behavior, peak intoxication level, HLIN-SOD, situ- partners. Table 1 displays predictors with p-values \ .25 in
ational triggers for vaginal sex hookups, gender, and distress. the univariate regression analyses. In the stepwise regression
None of the interactions between these six predictors was procedure, four predictors entered the model: situational
significant. triggers for oral sex hookups, number of oral sex hookup

123
Arch Sex Behav (2010) 39:1105–1119 1113

Table 3 Final model: multivariate predictors of Time 2 vaginal sex hookup behavior
Predictor (measured at Time 1) B SE Wald v2 (df = 1) p Adjusted odds ratio 95% CI

Intercept -1.91 0.33 33.40 \.0001

Pre-college vaginal sex hookup behaviora 1.88 0.47 15.78 \.0001 6.57 2.59–16.61
Peak intoxication levelb 9.28 2.58 12.96 .0003 2.53 1.53–4.19
Same-sex hookup limit injunctive norm SOD –0.55 0.21 6.73 .01 0.58 0.38–0.87
Note: N = 140. B coefficients represent logits. SOD = self-other difference; SE = standard error; CI = confidence interval
a
Reference group is no pre-college vaginal sex hookup behavior
b
For the odds ratio, the unit for BAC is .10, instead of 1

Table 4 Final model: multivariate predictors of Time 2 number of oral sex hookup partners
Predictor (measured at Time 1) B SE b t (df = 1) p

Intercept 0.45 0.08 0 5.98 \.0001

a
Peak intoxication level 0.30 0.08 .02 3.55 .0005
Peak intoxication levela 9 number of oral sex hookup partners 0.14 0.05 .03 2.84 .005
Peak intoxication levela 9 situational triggers for oral sex hookups 0.14 0.06 .02 2.26 .03
Situational triggers for oral sex hookups 0.22 0.05 .32 4.55 \.0001
Parental discouragement of relationships 9 situational triggers for oral sex hookups 0.11 0.03 .26 4.22 \.0001
Parental discouragement of relationships 0.10 0.04 .15 2.50 .01
Number of oral sex hookup partners 0.03 0.05 .05 \1 ns
Note: N = 138. SE = standard error
a
Unit for BAC is .10, instead of 1

partners, peak intoxication level, and parental discourage- the univariate regression analyses. In the stepwise regression
ment of relationships. All six potential interactions between procedure, five predictors entered the model: situational
the four predictors were tested with a Bonferroni-corrected triggers for vaginal sex hookups, peak intoxication level,
a = .008; three were significant: parental discouragement number of vaginal sex hookup partners, gender, and parental
and situational triggers for oral sex hookups, F(1, 132) = attitudes. The interaction between situational triggers for
13.03, p = .0004; peak intoxication level and situational vaginal sex hookups and gender was significant, F(1, 133) =
triggers for oral sex hookups, F(1, 132) = 8.19, p = .005; 9.94, p = .002.
and peak intoxication level and number of oral sex hookup The five predictors and one interaction were entered as
partners, F(1, 132) = 12.68, p = .0005. The three interac- predictors of T2 number of vaginal sex hookup partners. The
tions were included with the four predictors in the final model was run without parental attitudes (p = .16), and the
multivariate model, F(7, 130) = 20.97, p \ .0001, R2 = two models were compared, DR2 = .02, F(1, 133) = 3.85,
.53, adjusted R2 = .51. The interactions were retained in the p = .052; the more parsimonious five-predictor model was
final model because their addition resulted in a significant retained to avoid over-fitting the model (Tabachnick & Fi-
increase in variance explained by the model, DR2 = .13, F(3, dell, 2007). Inclusion of the interaction of situational triggers
134) = 11.81, p \ .0001. Parameter estimates for the final for vaginal sex hookups and gender resulted in a significant
model are displayed in Table 4. The final model explained increase in variance explained, DR2 = .06, F(1, 134) =
more variance than a model with number of oral sex hookup 10.61, p = .001. Thus, the final model included the following
partners as the only predictor, R2 = .21, and a model with predictors: situational triggers for vaginal sex hookups, peak
number of oral sex hookup partners and peak intoxication intoxication level, gender, number of vaginal sex hookup
level as the only predictors, R2 = .26, DR2 = .27, F(5, partners, and the interaction of situational triggers for vaginal
132) = 16.31, p \ .0001. sex hookups and gender. R2 for the final model was .29, and
adjusted R2 was .27. Parameter estimates for the final model
Number of Vaginal Sex Hookup Partners are displayed in Table 5. The final model explained more
variance than a model with number of vaginal sex hookup
The DV in this model was T2 number of vaginal sex hookup partners as the only predictor, R2 = .09, and a model with
partners. Table 1 displays predictors with p-values \ .25 in number of vaginal sex hookup partners and peak intoxication

123
1114 Arch Sex Behav (2010) 39:1105–1119

Table 5 Final model: multivariate predictors of Time 2 number of vaginal sex hookup partners
Predictor (measured at Time 1) B SE b t (df = 1) p

Intercept 0.67 0.08 0 8.15 \.0001

Situational triggers for vaginal sex hookups 0.34 0.07 0.52 4.73 \.0001
Gender 9 situational triggers for vaginal sex hookups -0.36 0.11 -0.37 -3.25 .002
a
Peak intoxication level 0.24 0.08 0.02 3.11 .002
Number of vaginal sex hookup partners 0.17 0.06 0.23 3.05 .003
Genderb -0.36 0.20 -0.15 -1.78 .08
Note: N = 140. SE = standard error
a
Unit for BAC is .10, instead of 1
b
Coded as female = 0, male = 1

level as the only predictors, R2 = .16, DR2 = .13, F(3, interaction were significant, F(2, 106) = 1.14. However,
134) = 8.18, p \ .0001. a priori contrasts confirmed that (a) the inexperienced group
and the transition-to-hooking-up groups (M = 6.45) re-
Consequences of Hooking Up ported less distress at T1 than the experienced group (M =
8.36), F(1, 106) = 3.64, p = .03, one-tailed; and (b) the
Table 6 summarizes the mental health consequences (i.e., inexperienced group (M = 6.86) reported less distress than
psychological distress and self-esteem) of hooking up, by the transition-to-hooking-up group and the experienced
gender, for each of the three hookup transition groups. groups (M = 8.40) at T2, F(1, 106) = 2.79, p = .05, one-
tailed.
Psychological Distress An additional analysis tested whether increases in distress
for females were restricted to hookups in which penetrative
As summarized in Table 6, the mean CES-D score for fe- sex occurred. The effect of the transition to non-penetrative
males in the transition-to-hooking-up group increased from sex hookups and the effect of the transition to penetrative sex
6.91 at T1 to 8.91 at T2; despite a large effect size (d = .45), hookups could not be compared because only two females
this change was not statistically significant, t(10) = -1.44, made the former transition. Instead, a priori contrasts were
p = .09, one-tailed. conducted separately at T1 and T2. Females who had hooked
The effect of hookup transition group on distress was up before college but did not engage in penetrative sex during
examined in females. Neither the between-subjects effect, their hookups (M = 6.2, SD = 3.7, n = 38) reported lower
F(2, 106) = 2.42, p = .09, nor the time-by-transition group distress at T1 than those who had engaged in penetrative sex

Table 6 Psychological distress and self-esteem over the first semester of college by gender and hookup transition group
Gender Hookup transition group n T1 T2
M (SD) Range M (SD) Range

Psychological distress
Females Inexperienced, no transition 44 6.34 (3.75) 0–16 6.86 (4.16) 0–17
Transition to hooking up 11 6.91 (3.73) 2–13 8.91 (4.93) 4–21
Experienced 54 8.36 (4.59) 0–20 8.30 (4.96) 1–19
Males Inexperienced, no transition 12 7.00 (5.19) 0–17 6.92 (4.74) 0–15
Transition to hooking up 4 3.25 (2.22) 1–6 4.25 (2.06) 2–7
Experienced 15 5.89 (2.92) 1–12 5.33 (2.41) 0–10
Self-esteem
Females Inexperienced, no transition 44 33.05 (4.63) 20–40 33.05 (4.42) 25–40
Transition to hooking up 11 31.55 (4.95) 23–37 31.68 (4.61) 23.5–37
Experienced 54 34.02 (4.75) 19.5–40 33.91 (4.48) 24–40
Males Inexperienced, no transition 12 34.17 (4.99) 26–40 34.17 (4.53) 25–40
Transition to hooking up 4 36.75 (0.96) 36–38 35.50 (3.11) 33–40
Experienced 15 35.47 (3.11) 30–40 35.47 (3.04) 29–40

123
Arch Sex Behav (2010) 39:1105–1119 1115

hookups prior to college (M = 8.1, SD = 4.5, n = 60), F(1, behavior predicted future hookups supports the Theory of
115) = 4.76, p = .02, one-tailed. In addition, females who Interpersonal Behavior (Triandis, 1977, 1980) and corrobo-
hooked up in their first semester but did not engage in pen- rates past research on casual sex (e.g., Apostolopoulos et al.,
etrative sex during those hookups (M = 6.0, SD = 4.3, 2002; Maticka-Tyndale et al., 1998). The link between heavy
n = 33) reported lower distress at T2 than females who en- drinking and hookup behavior also corroborated previous
gaged in penetrative sex hookups in the first semester of research that has found an association between alcohol use
college, (M = 8.9, SD = 4.6, n = 39), F(1, 106) = 7.07, and hookup behavior (e.g., Grello et al., 2006; Owen et al., in
p = .005, one-tailed. press; Paul et al., 2000). Alcohol may facilitate hookups by
lowering emerging adults’ inhibitions, increasing their con-
Self-esteem fidence to approach potential partners, or increasing their
susceptibility to real or perceived social pressures to hook up.
A paired samples t-test conducted on T1 and T2 self-esteem In addition, alcohol use may serve an ‘‘anticipatory excuse
in females who transitioned to hooking up revealed no change function’’ (Paul, 2006, p. 151), allowing students to attribute
in self-esteem over the first semester, t(10) \ 1, one-tailed, their behavior to alcohol.
d = -.03. The effect of hookup transition group on self-es- The social environment does not appear to overwhelm
teem was examined in females. Neither the between-subjects students’ pre-existing personal characteristics or family-of-
effect of hookup transition group, F(2, 106) = 1.53, nor the origin influence. For example, social norms predicted vaginal
time-by-transition group interaction was significant, F(2, sex hookups, but not as we expected; that is, as hookup limit
106) \ 1. Similarly, a priori contrasts of group differences at self-other differences increased, the likelihood of vaginal sex
T1 and T2 indicated that the three groups did not differ (all hookups decreased, contrary to our hypothesis and past
ps [ .10). alcohol research (cf. Carey et al., 2006; Prentice & Miller,
An additional analysis tested whether changes in self-es- 1993). The vast majority (94%) of students who indicated a
teem for females were restricted to hookups in which pene- stringent hookup limit at study entry did not engage in a
trative sex occurred. Females who had hooked up prior to vaginal sex hookup during their first semester. Thus, those
college but did not engage in penetrative sex during their students who accounted for the large, positive self-other
hookups (M = 33.7, SD = 4.6, n = 38) and those who had differences remained firm in their beliefs, despite feeling that
engaged in penetrative sex hookups prior to college (M = their peers did believe oral and vaginal sex hookups were
34.1, SD = 4.7, n = 60) did not differ in self-esteem at T1 acceptable. Future research might investigate moderators of
(p [ .10). However, females who hooked up in the first such findings, such as religiosity and alcohol use, to deter-
semester of college but did not engage in penetrative sex mine when such beliefs are protective and when they result in
during those hookups (M = 35.5, SD = 3.6, n = 33) re- pressure to conform to the perceived norm.
ported higher self-esteem at T2 than females who engaged in One family-of-origin influence emerged as an important
penetrative sex hookups in the first semester of college, predictor of number of oral sex partners, namely, parental
(M = 32.9, SD = 4.9, n = 39), F(1, 106) = 6.39, p = .005, discouragement of relationships. As expected, the more stu-
one-tailed. dents perceived parental discouragement of relationships, the
more oral sex partners they reported. Some students may find
benefits from heeding their parents’ preference that they re-
Discussion main unattached while in college (e.g., greater indepen-
dence); at the same time, they may not want to forego sexual
This study provides the first prospective exploration of the intimacy. Oral sex hookups may be an acceptable compro-
hypothesized predictors of penetrative sex hookups and the mise in this situation.
short-term mental health effects of hooking up for females. Many hypothesized predictors assessed at T1 were not
associated with any of the criterion variables. Possible
Predictors of Hooking Up explanations for the lack of hypothesized relationships
include restriction of range (e.g., self-esteem), imprecise
We identified three consistent predictors for sexual hookup measurement (e.g., media exposure), and study design.
outcomes, namely: prior hookup behavior/number of hookup Regarding the latter, a single semester may be too brief
partners, peak intoxication level, and situational triggers for for exaggerated descriptive norms to influence students’
sexual hookups. Prior hookup behavior/number of hookup behavior, and data collection during the first semester may be
partners and peak intoxication level (or their interaction) too soon to observe the influence of career-mindedness.
were significant predictors in all four models. Situational Conceptual notions may need revision, such as parental
triggers for hookups was a significant predictor in three of the variables (e.g., may be too distal to participants’ first semester
four models. The finding that situational triggers and past in college; cf. Owen et al., in press) and gender (e.g., cultural

123
1116 Arch Sex Behav (2010) 39:1105–1119

norms regarding female sexuality may have changed). Be- preceded hookups, not vice versa. Developmental differ-
cause no single study is definitive, the hypothesized rela- ences between adolescents and college students may explain
tionships should be examined again with larger samples. the disparate results. Continued investigation of the mental
Other variables emerged as univariate, but not multivari- health—hookup relationship is needed. Because students
ate, predictors of oral or vaginal sex hookups and number of undergo several life transitions during their first year of col-
partners. Intentions to hook up predicted all four outcome lege (e.g., leaving home, social and academic adjustment),
measures in the univariate context, but not in the multivariate and these transitions may also affect their mental health
context. Although intentions to hook up likely vary as a (Friedlander, Reid, Shupak, & Cribbie, 2007; Lucas & Ber-
function of numerous situational factors (e.g., availability of kel, 2005), we cannot be sure that the increase in distress
attractive partners, mood, alcohol use), we measured inten- we observed in some women is due exclusively to hookup
tions only once at T1. Accordingly, situational triggers and experiences. To best characterize the consequences of hook-
peak intoxication level may have cancelled out or suppressed ups, future research will be most informative if investigators
intentions in the multivariate models because early-semester also assess these co-occurring experiences and transitions.
intentions are more distal than situational triggers, which were Self-esteem was also examined as a second indicator of
also measured only once at T1 but reflect a more general mental health. Participants showed few changes from T1 to
pattern of behavior. Desire to be carefree, hookup limit norms, T2 in self-esteem, regardless of hookup transition group. In
and general injunctive norms were consistent univariate lieu of the global measure of self-esteem that was used, a
predictors, but not multivariate predictors. Thus, these vari- measure of sexual self-esteem may be more appropriate to
ables in particular need to be assessed in future studies to detect related changes in this construct. Alternatively, the 10-
determine their utility in predicting hookup behavior. week follow-up interval may not have been long enough for
changes to occur in self-esteem.
Mental Health Consequences of Hooking Up
Public Health Implications
Our findings suggest that penetrative sex hookups may lead to
an increase in distress for females. This result emerged de- Our findings suggest that engaging in penetrative sex hook-
spite the relatively small number of participants in the tran- ups may lead to an increase in distress for young women. If
sition-to-hooking-up group, and corroborates results from a replicated, these findings should be communicated to young
cross-sectional study (Grello et al., 2006). At study entry, adults, especially females, so they can be informed about the
females with prior hookup experience reported higher dis- emotional risks of hooking up. Females are less likely than
tress than females in the two inexperienced groups. At T2, males to experience positive emotions and more likely than
females in the experienced group still reported relatively high males to experience negative emotions following hookups
levels of distress, and females in the inexperienced group still (Owen et al., in press). Townsend (1995) found that even
reported relatively low levels of distress; however, females in women who entered sexual relationships without intending
the transition group, who had their first penetrative sex to become emotionally involved experienced emotional
hookup in the first semester of college, reported a level of vulnerability and concern over their partners’ investment in
distress similar to the experienced group. Thus, the pattern of the relationship. Post-hookup reasons for regret among wo-
means for the three groups at T1 and T2 is what would be men and men appear to differ, with women focusing more on
expected if penetrative sex hookups were indeed detrimental emotional factors (e.g., feeling ‘‘used’’) and men focusing
to the short-term mental health of females. more on physical factors (e.g., partner was unattractive; Paul
In contrast to the pattern seen in females, males in the & Hayes, 2002). Thus, despite hooking up as much as men,
inexperienced group reported the highest levels of distress at women are not affected by hookups in the same way. Young
T1 and T2. For males, having a high number of sexual part- women may benefit from personal reflection and group dis-
ners is associated with higher self-esteem (Walsh, 1991) and cussion about gender differences in how hookups are expe-
masculinity (Pleck, Sonenstein, & Ku, 1993; Walsh, 1995); rienced. Another potential focus for intervention is the link
accordingly, those who are ‘‘missing out’’ on hooking up between alcohol use and hookups. Interventionists may
could be expected to show poorer mental health than those educate college students on the connection between alcohol
who are hooking up. The relationship between hooking up use and hooking up and offer suggestions for monitoring
and mental health in males warrants further investigation alcohol use (Sugarman & Carey, 2007).
with larger samples. Education related to hookups should begin in middle or
The temporal order of poorer mental health and hookups is high school. Younger adolescents hook up (Manning, Giord-
unclear. Distress was not a significant predictor of oral or ano, & Longmore, 2006; Manning, Longmore, & Giordano,
vaginal sex hookups in our sample. However, in a study of 2005), and those who do so prior to college are likely to
younger adolescents, Grello et al. (2003) found that distress continue hooking up upon entering college. Indeed, the

123
Arch Sex Behav (2010) 39:1105–1119 1117

strongest predictors of penetrative sex hookup behavior and that the predictors and consequences we identified are spe-
number of hookup partners in the first semester of college cific to hookups or are related to sex in general or to new
were pre-college hookup behavior and number of hookup sexual relationships in general. Research might compare the
partners. predictors of oral and vaginal sex during these two relation-
ship contexts to determine if (and how) romantic and hookup
relationships differ.
Limitations and Future Research Sixth, our data raise many interesting questions. For
example, we cannot determine the level of hookup intimacy
These findings must be interpreted mindful of study limita- required to trigger negative mental health effects. Our find-
tions, which can also guide future research. First, we used a ings suggest that non-penetrative hookups may be benign, but
convenience sample of heterosexual, primarily female and hookups that include penetrative sex may lead to negative
Caucasian college students; therefore, generalization to other health consequences. Perhaps moderate hookup involvement
student sub-groups is premature. To broaden our under- allows young women to participate in the dominant social
standing of hookup behavior and its consequences, future scene on campus (Bogle, 2008) and experience physical
research should sample more broadly, including more males; pleasure, without having to worry about the consequences of
ethnic minority students; gay, lesbian, and bisexual students; penetrative sex hookups (e.g., feeling as if they were ‘‘used’’
upperclassmen; and members of fraternities and sororities. for sex). We did not assess participants’ motives for, inter-
Second, we measured only two mental health outcomes: pretation of, or satisfaction with their hookups; any of these
distress and self-esteem. Future research can extend our exam- experiential variables may affect the potential mental health
ination of mental health consequences by including a broader effects of hookups. Research might investigate the differ-
range of health indicators, such as positive and negative ences in emotional risk between hookups that do and do not
affect, perceived stress, and relationship and sexual satis- progress to penetrative sex, and explore the individual and
faction. Longer-term follow-up intervals will provide infor- situational factors that influence the transition from non-
mation on the durability of these effects and allow more time penetrative to penetrative sex hookups.
for delayed effects to emerge. Investigation of the positive (as Seventh, we included a wide array of hypothesized
well as negative) consequences of hooking up can help in- predictors, but there are other person variables that should be
crease understanding of the full range of sexual, social, and explored as potential predictors of hookup behavior (cf. Gute
psychological effects associated with sexual hookups. In & Eshbaugh, 2008). For example, sensation-seeking is pos-
addition, research might explore how students interpret their itively correlated with number of sexual partners (Kraft
hookup experiences and the mechanism(s) by which hooking & Rise, 1994; Walsh, 1995). In addition, conscientiousness,
up influences mental health. extraversion, gregariousness, and impulsivity might also
Third, although our sample included 140 students, only a be explored as potential risk or protective factors in relation
small number transitioned to penetrative sex hookups in the to hooking up. For females, acceptance of traditional gen-
first semester of college; as a result, the mental health analyses der roles may impact willingness to engage in hookups.
were underpowered. In addition, the sample size prevented In summary, this study expands the existing knowledge
cross-validation of the four models that were developed. about the hookup phenomenon. The strongest predictors of
Future research should recruit larger samples to address these first-semester penetrative sex hookups are prior hookup
limitations. behavior, peak intoxication level, and situational factors that
Fourth, we relied upon self-report data; participants may encourage hookups. Our prospective assessment of the effect
have had inaccurate memories or may have responded in a of hooking up on participants’ mental health suggests that
socially desirable way. To minimize memory difficulties, we penetrative sex hookups may lead to an increase in distress
used measures with short-term recall intervals. To limit so- among females, but not males. To extend knowledge, we
cial desirability responding, our survey was anonymous and encourage replication with larger and more diverse samples,
self-administered. To improve overall measurement preci- longer follow-up intervals, and a broader array of health
sion, most of the measures we used were already established, outcomes and predictor variables.
psychometrically validated measures. When new scales were
necessary, we conducted factor analyses and internal con- Acknowledgments We thank the students for their participation,
sistency analyses prior to model development to confirm their Kate B. Carey, and two anonymous reviewers for their comments on
appropriateness; however, we did not assess test–retest reli- an earlier draft of this article, and Hillary L. Bishop for her assis-
tance with data collection and data entry. Supported by Grant R01-
ability of the new scales. MH54929 from the National Institute of Mental Health to Michael
Fifth, we did not compare sex during hookups to sex in the P. Carey and by a Graduate Research Grant from Psi Chi to Robyn
context of romantic relationships. Thus, we cannot be sure L. Fielder.

123
1118 Arch Sex Behav (2010) 39:1105–1119

References Herold, E. S., Maticka-Tyndale, E., & Mewhinney, D. (1998).

Predicting intentions to engage in casual sex. Journal of Social
Apostolopoulos, Y., Sönmez, S., & Yu, C. H. (2002). HIV-risk and Personal Relationships, 15, 502–516.
behaviours of American spring break vacationers: A case of Herold, E. S., & Mewhinney, D. K. (1993). Gender differences in casual
situational disinhibition. International Journal of STD and AIDS, sex and AIDS prevention: A survey of dating bars. Journal of Sex
13, 733–743. Research, 30, 36–42.
Bandura, A. (1977). Social learning theory. Oxford, England: Prentice- Hosmer, D. W., & Lemeshow, S. (2000). Applied logistic regression
Hall. (2nd ed.). New York: Wiley.
Bisson, M. A., & Levine, T. R. (in press). Negotiating a friends with Jennings, A. M., Salts, C. J., & Smith, T. A. (1991). Attitudes toward
benefits relationship. Archives of Sexual Behavior. marriage: Effects of parent conflict, family structure, and gender.
Blascovich, J., & Tomaka, J. (1991). The Self-Esteem scale. In J. P. Journal of Divorce & Remarriage, 17, 67–79.
Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of Kraft, P., & Rise, J. (1994). The relationship between sensation
personality and social psychological attitudes: Volume 1 of seeking and smoking, alcohol consumption and sexual behavior
measures of social psychological attitudes (pp. 121–123). San among Norwegian adolescents. Health Education Research, 9,
Diego, CA: Academic Press, Inc. 193–200.
Bogle, K. A. (2008). Hooking up: Sex, dating, and relationships on Lambert, T. A., Kahn, A. S., & Apple, K. J. (2003). Pluralistic ignorance
campus. New York: New York University Press. and hooking up. Journal of Sex Research, 40, 129–133.
Carey, K. B., Borsari, B., Carey, M. P., & Maisto, S. A. (2006). Patterns L’Engle, K. L., Brown, J. D., & Kenneavy, K. (2006). The mass media
and importance of self-other differences in college drinking norms. are an important context for adolescents’ sexual behavior. Journal
Psychology of Addictive Behaviors, 20, 385–393. of Adolescent Health, 38, 186–192.
Cohen, L. L., & Shotland, R. L. (1996). Timing of first sexual intercourse Lottes, I. L. (1993). Nontraditional gender roles and the sexual experi-
in a relationship: Expectations, experiences, and perceptions of ences of heterosexual college students. Sex Roles, 29, 645–670.
others. Journal of Sex Research, 33, 291–299. Lucas, M. S., & Berkel, L. A. (2005). Counseling needs of students who
Collins, R. L., Elliott, M. N., Berry, S. H., Kanouse, D. E., Kunkel, D., seek help at a university counseling center: A closer look at gender
Hunter, S. B., et al. (2004). Watching sex on television predicts and multicultural issues. Journal of College Student Development,
adolescent initiation of sexual behavior. Pediatrics, 114, 280–289. 46, 251–266.
Daugherty, L. R., & Burger, J. M. (1984). The influence of parents, Maguen, S., & Armistead, L. (2006). Abstinence among female
church, and peers on the sexual attitudes and behaviors of college adolescents: Do parents matter above and beyond the influence
students. Archives of Sexual Behavior, 13, 351–359. of peers? American Journal of Orthopsychiatry, 76, 260–264.
Dittus, P. J., & Jaccard, J. (2000). Adolescents’ perceptions of maternal Mahoney, E. R. (1980). Religiosity and sexual behavior among
disapproval of sex: Relationship to sexual outcomes. Journal of heterosexual college students. Journal of Sex Research, 16, 97–113.
Adolescent Health, 26, 268–278. Manning, W. D., Giordano, P. C., & Longmore, M. A. (2006). Hooking
Dufour, M. C. (1999). What is moderate drinking: Defining ‘‘drinks’’ up: The relationship contexts of ‘‘nonrelationship’’ sex. Journal of
and drinking. Alcohol Research and Health, 23, 5–14. Adolescent Research, 21, 459–483.
Escobar-Chaves, S. L., Tortolero, S. R., Markham, C. M., Low, B. J., Manning, W. D., Longmore, M. A., & Giordano, P. C. (2005).
Eitel, P., & Thickstun, P. (2005). Impact of the media on adolescent Adolescents’ involvement in non-romantic sexual activity. Social
sexual attitudes and behaviors. Pediatrics, 116, 303–326. Science Research, 34, 384–407.
Facebook: Statistics. (n.d.). Retrieved August 1, 2007 from http:// Martens, M. P., Page, J. C., Mowry, E. S., Damann, K. M., Taylor, K. K.,
static.ak.facebook.com/press/facebook_statistics.pdf?12:51442& & Cimini, M. D. (2006). Differences between actual and perceived
pwstdfy=95f3f0c257be3cfedb96b2df6269a271. student norms: An examination of alcohol use, drug use, and sexual
Flack, W. F., Daubman, K. A., Caron, M. L., Asadorian, J. A., D’Aureli, behavior. Journal of American College Health, 54, 295–300.
N. R., Gigliotti, S. N., et al. (2007). Risk factors and consequences Maticka-Tyndale, E., Herold, E. S., & Mewhinney, D. (1998). Casual
of unwanted sex among university students: Hooking up, alcohol, sex on spring break: Intentions and behaviors of Canadian students.
and stress response. Journal of Interpersonal Violence, 22, 139– Journal of Sex Research, 35, 254–264.
157. Matthews, D. B., & Miller, W. R. (1979). Estimating blood alcohol
Friedlander, L. J., Reid, G. J., Shupak, N., & Cribbie, R. (2007). Social concentration: Two computer programs and their applications in
support, self-esteem, and stress as predictors of adjustment to therapy and research. Addictive Behaviors, 4, 55–60.
university among first-year undergraduates. Journal of College Messman-Moore, T. L., Ward, R. M., & Brown, A. L. (in press).
Student Development, 48, 259–274. Substance use and PTSD symptoms impact the likelihood of rape
Glenn, N., & Marquardt, E. (2001). Hooking up, hanging out, and and revictimization in college women. Journal of Interpersonal
hoping for Mr. Right: College women on dating and mating today. Violence.
New York: Institute for American Values. Newcomer, S., & Udry, J. R. (1987). Parental marital status effects on
Greenberg, B. S., & Hofschire, L. (2000). Sex on entertainment adolescent sexual behavior. Journal of Marriage and the Family,
television. In D. Zillmann & P. Vorderer (Eds.), Media entertain- 49, 235–240.
ment: The psychology of its appeal (pp. 93–112). Mahwah, NJ: Oliver, M. B., & Hyde, J. S. (1993). Gender differences in sexuality: A
Lawrence Erlbaum Associates. meta-analysis. Psychological Bulletin, 114, 29–51.
Grello, C. M., Welsh, D. P., & Harper, M. S. (2006). No strings attached: Owen, J. J., Rhoades, G. K., Stanley, S. M., & Fincham, F. D. (in press).
The nature of casual sex in college students. Journal of Sex ‘‘Hooking up’’ among college students: Demographic and psy-
Research, 43, 255–267. chosocial correlates. Archives of Sexual Behavior.
Grello, C. M., Welsh, D. P., Harper, M. S., & Dickson, J. W. (2003). Page, R. M., Hammermeister, J. J., & Scanlan, A. (2000). Everybody’s
Dating and sexual relationship trajectories and adolescent func- not doing it: Misperceptions of college students’ sexual activity.
tioning. Adolescent & Family Health, 3(3), 103–112. American Journal of Health Behavior, 24, 387–394.
Gute, G., & Eshbaugh, E. M. (2008). Personality as a predictor of Paul, E. L. (2006). Beer goggles, catching feelings, and the walk of
hooking up among college students. Journal of Community Health shame: The myths and realities of the hookup experience. In D. C.
Nursing, 25, 26–43. Kirkpatrick, S. Duck, & M. K. Foley (Eds.), Relating difficulty: The

123
Arch Sex Behav (2010) 39:1105–1119 1119

process of constructing and managing difficult interaction (pp. Seaman, B. (2005). Binge: What your college student won’t tell you.
141–160). Mahwah, NJ: Lawrence Erlbaum Associates. Hoboken, NJ: Wiley.
Paul, E. L., & Hayes, K. A. (2002). The causalities of ‘‘casual’’ sex: A Stepp, L. S. (2007). Unhooked: How young women pursue sex, delay
qualitative exploration of the phenomenology of college students’ love, and lose at both. New York: Riverhead Books.
hookups. Journal of Social and Personal Relationships, 19, 639– Sugarman, D. E., & Carey, K. B. (2007). The relationship between
661. drinking control strategies and college student alcohol use.
Paul, E. L., McManus, B., & Hayes, A. (2000). ‘‘Hookups’’: Character- Psychology of Addictive Behaviors, 21, 338–345.
istics and correlates of college students’ spontaneous and anony- Symons, D. (1987). An evolutionary approach: Can Darwin’s view of
mous sexual experiences. Journal of Sex Research, 37, 76–88. life shed light on human sexuality? In J. H. Geer & W. T.
Penhollow, T., Young, M., & Bailey, W. (2007). Relationship between O’Donohue (Eds.), Theories of human sexuality (pp. 91–126). New
religiosity and ‘‘hooking up’’ behavior. American Journal of York: Plenum Press.
Health Education, 38, 338–345. Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics
Pleck, J. H., Sonenstein, F. L., & Ku, L. C. (1993). Masculinity ideology: (5th ed.). Boston: Pearson.
Its impact on adolescent males’ heterosexual relationships. Thornton, A., & Camburn, D. (1989). Religious participation and
Journal of Social Issues, 49, 11–29. adolescent sexual behavior and attitudes. Journal of Marriage and
Prentice, D. A., & Miller, D. T. (1993). Pluralistic ignorance and alcohol the Family, 51, 641–653.
use on campus: Some consequences of misperceiving the social Tousigant, C. R. (2007, April 26). Summer lovin’: The rules of hooking
norm. Journal of Personality and Social Psychology, 64, 243–256. up in the heat. The Daily Orange: (315) Life in Syracuse. Retrieved
Rimal, R. N., & Real, K. (2005). How behaviors are influenced by August 1, 2007 from http://media.www.dailyorange315.com/
perceived norms: A test of the theory of normative social behavior. media/storage/paper1204/news/2007/04/26/Explicit/
Communication Research, 32, 389–414. Summer.Lovin-2877810.shtml.
Rimsza, M. E. (2005). Sexually transmitted infections: New guidelines Townsend, J. M. (1995). Sex without emotional involvement: An
for an old problem on the college campus. Pediatric Clinics of evolutionary interpretation of sex differences. Archives of Sexual
North America, 52, 217–228. Behavior, 24, 173–206.
Rohrbaugh, J., & Jessor, R. (1975). Religiosity in youth: A personal and Triandis, H. C. (1977). Interpersonal behavior. Monterey, CA: Brooks/
social control against deviant behavior. Journal of Personality, 43, Cole.
136–155. Triandis, H. C. (1980). Values, attitudes, and interpersonal behavior. In
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, H. E. Howe & M. M. Page (Eds.), Nebraska symposium on
NJ: Princeton University Press. motivation 1979 (pp. 195–259). Lincoln, NE: University of
Rostosky, S. S., Wilcox, B. L., Wright, M. L. C., & Randall, B. A. (2004). Nebraska Press.
The impact of religiosity on adolescent sexual behavior: A review Walsh, A. (1991). Self-esteem and sexual behavior: Exploring gender
of the evidence. Journal of Adolescent Research, 19, 677–697. differences. Sex Roles, 25, 441–450.
Santor, D. A., & Coyne, J. C. (1997). Shortening the CES-D to improve Walsh, A. (1995). Parental attachment, drug use, and facultative sexual
its ability to detect cases of depression. Psychological Assessment, strategies. Social Biology, 42, 95–107.
9, 233–243. Zaleski, E. H., & Schiaffino, K. M. (2000). Religiosity and sexual risk-
Scholly, K., Katz, A. R., Gascoigne, J., & Holck, P. S. (2005). Using taking behavior during the transition to college. Journal of
social norms theory to explain perceptions and sexual health Adolescence, 23, 223–227.
behaviors of undergraduate college students: An exploratory study.
Journal of American College Health, 53, 159–166.

123