University of Southern California Dissertations and Theses

Factors and correlates of sexual behaviors among young adults from continuation high schools

(USC Thesis Other)

Factors and correlates of sexual behaviors among young adults from continuation high schools

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

FACTORS AND CORRELATES OF SEXUAL BEHAVIORS
AMONG YOUNG ADULTS FROM CONTINUATION HIGH SCHOOLS

by

Lilia Espinoza

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PREVENTIVE MEDICINE)

August 2009

Copyright 2009 Lilia Espinoza

ii
ACKNOWLEDGEMENTS

I sincerely thank the members of my dissertation committee, Dr. Jean L. Richardson
(Chair), Dr. Lourdes Baezconde-Garbanati, Dr. Chih-Ping Chou, Dr. Kristin M. Ferguson,
and Dr. Alan W. Stacy, for their assistance with development of the dissertation study and
this manuscript. My deepest appreciation and thanks goes to Drs. Richardson and
Baezconde-Garbanati for their guidance, support, encouragement, and extreme patience
throughout my doctoral training. Dr. Baezconde-Garbanati provided me with countless
research and evaluation opportunities. Dr. Richardson challenged me in a way that I never
could have imagined and made me a better researcher, thinker, and person in the process.
Special thanks go to the data collection team (Alexandra Q. Anderson, Lee Bette-
Prudhomme, Lauren Brown, Andrew Evans, Brett Mendenhall, and Callan Wampler) for
their hard work and dedication; Hee-Sung Shin for help with the TPRC study; Mufti Jantan
for creating the data entry screens; Dr. Jerry Gates for granting time off; and Marny
Barovich, IPR faculty and staff, and USC PAETC faculty and staff for guidance and support.
I also thank my cohort, Drs. Selena Nguyen-Rodriguez and Enrique Ortega, my friends
(especially Mary, Allison, and Julie), and colleagues for the years of support and friendship
that sustained me through this process.
I further thank my family, both immediate and acquired, who continually supported
and encouraged me to continue, especially my nieces who grew quickly to understand why I
could not spend weekends with them. I especially want to thank my younger sister, Claudia
Espinoza Wilson, who encouraged me to not drop out of the program back in July 2006.
Finally, I thank the three most inspirational people in my life: my mother, Leoveldina
Espinoza Muro; my father, Miguel Espinoza Rodriguez; and my fiancé, Daniel Scott Perry.
My mother instilled in me the importance of education and learning, and through her

iii
struggles, she demonstrated the importance of health, health education, and strength of
character in life. My father instills a strong work ethic and perseverance in me that I often
find challenging to fulfill. Lastly Danny’s love, support, and patience as well as his ability to
truly believe in me when I did not believe in myself provided me with the strength and
additional passion to complete this chapter of my life.
iv
TABLE OF CONTENTS

Acknowledgements ii
List of Tables v
List of Figures vi
Abstract vii
Chapter 1: Introduction 1
Chapter 2: Effects of Substance Use and Sensation-Seeking in Adolescence on
Age at Sexual Initiation and Number of Sexual Partners

7
Chapter 3: The Role of Depressive Symptomatology on Sexual Behaviors 32
Chapter 4: Sociocognitive Correlates of Condom Use among Young Adults
from Continuation High Schools

55
Chapter 5: Conclusion 83
Bibliography 91

v
LIST OF TABLES

Table 1. Characteristics of the Sample 23
Table 2. Associations with Sexual Maturation and Sexual Onset Experience 24
Table 3. Associations with Substance Use and Sensation-seeking 25
Table 4. Associations with Age at Sexual Onset and Number of Sexual
Partners

26
Table 5. Characteristics of the Sample 44
Table 6. Associations with Depressive Symptomatology 45
Table 7. Associations with Sexual Behaviors 47
Table 8. Associations between Depressive Symptomatology and Sexual
Behaviors

48
Table 9. Associations between Depressive Symptomatology and Number of
Sexual Partners by Gender

50
Table 10. Characteristics of the Sample 73
Table 11. Correlations among Sociocognitive Variables by Gender 75
Table 12. Correlations among Sociocognitive Variables by Ethnicity 75
Table 13. Regression Analyses 76

vi
LIST OF FIGURES

Figure 1. Theoretical Model 6
Figure 2. SEM Mediation Model 28

vii
ABSTRACT

With HIV increasingly impacting youth, the importance of studying an understudied
at-risk group of youth was evident. National epidemiologic data reveal that youth in
continuation high schools (CHS) exhibit greater HIV sexual risk behaviors than youth in
regular high schools yet these data are aged and cross-sectional in nature. In the present
study, sociocognitive factors and correlates of sexual behaviors among CHS youth were
explored. Baseline data from the Transdisciplinary Drug Abuse Prevention Research Center
as well as four-year follow-up data that were newly collected were used to conduct
descriptive and inferential statistics. Follow-up data from 111 young adults (M = 20.6 years,
S = 1.2 years) who attended CHS in southern California were matched to their baseline data.
Compared to Latinos, non-Latinos reported greater substance use and had higher sensation-
seeking scores at baseline and an earlier age at sexual initiation. The relationships between
baseline substance use and the numbers of lifetime and recent sexual partners reported at
follow-up were fully mediated by age at sexual onset. CHS youth with greater substance use
initiated sexual activity at an earlier age and reported multiple sexual partners. Sensation-
seeking had no effect on sexual behaviors. Females exhibited greater depressive
symptomatology than males. A pattern of greater depressive symptomatology at both time
periods and an earlier age at sexual onset was observed. Depressive symptoms at follow-up
were significantly associated with multiple lifetime sexual partners. Depressive
symptomatology was not related with frequency of condom use. Greater frequency of sexual
safety communication with a parent either during adolescence or recently was significantly
associated with more consistent lifetime and recent condom use. Greater perceived
susceptibility to HIV, peer norms that support safer sex, and greater condom use self-
efficacy were also significant correlates of greater lifetime condom use among non-Latinos.

viii
None of the sociocognitive constructs were significantly associated with condom use among
Latinos. Peer norms and condom use self-efficacy were significant correlates for lifetime
condom use among females. Findings provide starting points where additional research is
needed to determine what culturally and gender-appropriate variables are important to
include when developing HIV prevention interventions among CHS youth.

1
CHAPTER ONE: INTRODUCTION

Unprotected sexual intercourse is a known risk factor for infection with the human
immunodeficiency virus (HIV). Since the start of the documented HIV epidemic in the early
1980’s, changes in HIV infection rates and the number of acquired immune deficiency
syndrome (AIDS) cases are apparent. Specifically, what was an epidemic that primarily
affected White males is increasingly affecting females and disproportionately affecting
African-Americans and Latinos.
The HIV/AIDS epidemic is also affecting adolescents. Early in the epidemic,
mother-to-child (vertical) transmission and blood transfusion were the predominant modes
of exposure for children and adolescents living with HIV/AIDS. The risk of HIV infection
among adolescents by these exposures is now low due to advances and rigor in HIV testing,
highly active antiretroviral therapy to reduce mother-to-child transmission, and screening of
blood and blood products. Among adolescents and adults, the primary mode of HIV
transmission is through unprotected sexual behaviors.
Additionally, unprotected sexual behaviors are risk factors for sexually transmitted
diseases (STDs) and unintended pregnancy. The HIV/AIDS epidemic along with a high
prevalence of STDs and unintended pregnancy as surrogate markers of unprotected sex are
major health concerns associated with high-risk sexual activity that affect adolescents.
Sexual behavior is complex and requires special consideration of a complex set of variables.
Closer examination of intrapersonal and interpersonal factors of sexual behaviors may
uncover paths by which individual and interpersonal risk and protective factors affect sexual
risk-taking. Addressing these issues has significant relevance to public health and health
research because of the risks of HIV/AIDS, other STDs, and unintended pregnancy among
adolescents that may affect one’s current and future quality of life.

2
Adolescence is also an important time period in which to examine and intervene
because of the unique developmental elements associated with it. Adolescence is a time
when persons are transitioning from childhood to adulthood, and adolescents often engage in
precocious behaviors such as risky sexual behavior as efforts to assume adult
responsibilities. Some adolescents may engage in multiple risky behaviors (e.g., sexual
behaviors, substance use) that may constitute problem behaviors (Jessor, 1984). Health
behaviors are habituated during adolescence, making prevention efforts pivotal during this
time.
Key concepts that need to be considered when discussing sexual risk-taking include
intrapersonal and interpersonal factors of sexual behaviors. Individuals may engage in
certain behaviors that place them at increased sexual risk. For example, substance use is not
only a significant risk factor for health complications (e.g., addiction) but is also a risk factor
for HIV/AIDS. Injection drug use is the secondary HIV exposure category among
adolescents (Centers for Disease Control and Prevention [CDC], 2005). Non-injection drug
use also facilitates HIV transmission by affecting one’s level of coherence that predisposes
an individual to high-risk sexual behaviors such as unprotected sex and multiple sexual
partners.
However intrapersonal factors alone do not provide an adequate description of
sexual risk. Consistent with Social Cognitive Theory, an individual’s cognitions and
behaviors are affected by the social environment (Baranowski, Perry, & Parcel, 1997).
Interpersonal factors influence individual sexual behaviors through modeling and
reinforcements. Studying the effects of intrapersonal and interpersonal factors on sexual
behaviors, especially among at-risk adolescents, is warranted when the deleterious health
implications from high-risk sexual behaviors are considered. Identifying risk and protective

3
factors and entries for intervention may contribute to the primary prevention of HIV, other
STDs, and unintended pregnancy.
Several intrapersonal and interpersonal factors of unprotected sexual behaviors and
multiple sexual partners have been identified among adolescents. However extant research
on these factors or how they vary by select characteristics among youth in continuation high
schools (CHS) is limited. CHS youth are an important vulnerable subpopulation of youth.
They are not only at risk of dropping out of school and becoming available for high-risk
behaviors, but they are also exposed to a social environment with a high prevalence of health
risk behaviors. Substance use is at least twice as high among CHS youth as other high school
youth. CHS youth are also more likely to have a higher proportion of unprotected sex and a
greater number of lifetime sexual partners.
The current study explored factors related to sexual behaviors among young adults
who were in continuation high schools. Researchers at the University of Southern
California’s Institute for Health Promotion and Disease Prevention Research conducted drug
prevention research with CHS youth in Los Angeles County through the Transdisciplinary
Drug Abuse Prevention Research Center (TPRC). By building on this research, sexual risk
behaviors among CHS youth were explored. Baseline data for the TPRC study were
collected between February and November 2004. Select baseline data from the TPRC study
was employed in the current study.
Information on HIV sociocognitive factors, sexuality characteristics, and sexual
behaviors were collected in a supplemental study. Young adults who were at least 18 years
old were selected. Participant data was used to address the following aims: (1) to assess if
lifetime substance use and sensation-seeking at Time 1 (TPRC baseline) was predictive of
the numbers of lifetime and recent sexual partners by Time 2 (four-year follow-up) among
CHS youth; (2) to assess if the relationships of substance use and sensation-seeking with

4
number of sexual partners is partially mediated by age at sexual onset; (3) to assess if
depressive symptomatology at any time point was associated with the number of sexual
partners and frequency of condom use during vaginal and/or anal intercourse among CHS
youth; (4) to assess if HIV sociocognitive factors are associated with the frequency of
condom use among CHS youth; and (5) to investigate the moderating effects of gender and
race/ethnicity on the relationships between sociocognitive factors and condom use
frequency. Results from this study clarified factors to explore in more detail or consider
when creating and tailoring HIV prevention programs for at-risk youth.
Questions posed and covariates included for the present study are based on a review
of the literature. The following hypotheses were tested:
(1a) CHS young adults who reported alcohol, marijuana, and/or cigarette use at Time 1 are
more likely to report a higher number of lifetime and recent sexual partners by Time 2.
(1b) The relationship between substance use and number of sexual partners will be at least
partially mediated by age at sexual onset where earlier sexual initiators will have
greater substance use and a higher number of sexual partners.
(1c) CHS young adults who scored higher on sensation-seeking at Time 1 are more likely to
report a higher number of lifetime and recent sexual partners by Time 2.
(1d) The relationship between sensation-seeking and number of sexual partners will be at
least partially mediated by age at sexual onset where earlier sexual initiators will have
higher sensation-seeking and a higher number of sexual partners.
(2) CHS young adults who report depressive symptomatology at any time period are more
likely to engage in a lower frequency of condom use during vaginal and/or anal
intercourse and report a greater number of lifetime sexual partners.

5
(3a) CHS young adults who communicated more frequently on sexual safety with a parent
or parental figure during adolescence or within the past 6 months are more likely to
engage in a greater frequency of condom use.
(3b) CHS young adults who perceive themselves as susceptible to HIV are more likely to
engage in a greater frequency of condom use.
(3c) CHS young adults with higher peer norms for safer sex are more likely to engage in a
greater frequency of condom use.
(3d) CHS young adults with higher perceived condom use self-efficacy are more likely to
engage in a greater frequency of condom use.
(3e) The relationships between the sociocognitive construct and frequency of condom use
will vary by gender and race/ethnicity. Positive associations are expected for males and
non-Latinos. Each potential moderator will be tested separately.
The theoretical model is shown in Figure 1. In addition to gender and race/ethnicity,
covariates include age, age at sexual initiation, age differential with first sexual partner,
pubertal age, and acculturation.

6
Figure 1. Theoretical Model.

Number of
Sexual
Partners
Gender
Race/Ethnicity

Frequency of
Condom Use

Age
Age Differential with
First Sexual Partner

Pubertal Age

Parent-
adolescent
Communication

Peer Norms on
Safer Sex
Perceived
Susceptibility
to HIV
Covariates Independent Variables Mediator/Moderators Dependent Variables
Condom Use
Self-Efficacy

Substance
Use
Age at
Sexual Onset
Sensation-
Seeking
Depressive
Symptoms

Acculturation

7
CHAPTER TWO: EFFECTS OF SUBSTANCE USE AND SENSATION-
SEEKING IN ADOLESCENCE ON AGE AT SEXUAL INITIATION AND
NUMBER OF SEXUAL PARTNERS

Sexual transmission is the primary exposure category among adolescents in both the
United States and California followed by injection drug use (California State Office of
AIDS, 2005a; California State Office of AIDS, 2005b; CDC, 2005). Although not a direct
influence on HIV infection, non-injection substance use can facilitate exposure to HIV via
behavioral disinhibition and subsequent at-risk sexual behaviors. Temporal changes in the
proportions affected by the HIV epidemic in age, gender, and race/ethnicity have occurred
whereby adolescents, specifically women and ethnic minorities, have increasingly become
affected.
The relationship between substance use, either injected or non-injected, and sexual
risk behaviors has been well-documented among several adolescent samples (Brown,
DiClemente, & Beausoleil, 1992; Buzi, Tortolero, Roberts, Ross, Addy, & Markham, 2003;
Centers for Disease Control and Prevention [CDC], 1999; CDC, 2000; CDC, 2006a; Deas-
Nesmith, Brady, White, & Campbell, 1998; Devieux, Malow, Stein, Jennings, Lucenko, et
al., 2002; DiClemente, Lodico, Grinstead, Harper, Rickman, Evans, et al., 1996; Ellickson,
Collins, Bogart, Klein, & Taylor, 2005; Hallfors, Waller, Bauer, Ford, & Halpern, 2005;
Hallfors, Waller, Ford, Halpern, Brodish, & Iritani, 2004; Jemmott, Jemmott, & Fong, 1998;
Kowaleski-Jones & Mott, 1998; Krantz, Lynch, & Russell, 2002; Leigh & Stall, 1993;
O’Hara et al., 1996; O’Hara et al., 1998; Otto-Salaj, Gore-Felton, McGavry, & Canterbury,
2002; Sikkema, Brondino, Anderson, Gore-Felton, Kelly, Winett, et al., 2004; Whaley,
1999). Being under the influence of alcohol and/or illicit substances leads to behavioral

8
disinhibition whereby one’s decision-making abilities to practice safer sex are compromised
(e.g., consistent condom use, reduced number of sexual partners). Cigarette smoking has also
been shown to be positively associated with high-risk sexual behaviors among adolescents
(Sussman, 2005).
Although several studies have focused on traditionally at-risk adolescent samples
(e.g., homeless youth; Bailey, Gao, & Clark, 2006), extant research on the relationships
between substance use and sexual risk-taking among adolescents in continuation high
schools (CHS) is limited. Youth are transferred to alternative or continuation high schools
when they are considered at risk of dropping out of high school for various reasons and in
order to fulfill an educational mandate (Sussman, 1996). Not only are they at risk of
dropping out of school, but they are also exposed to a social environment where HIV-related
risk behaviors such as substance use and unprotected sex are high (Sussman, Dent, Simon,
Stacy, Galaif, et al., 1995a; Sussman, 1996; CDC, 1999).
The national Youth Risk Behavior Survey (YRBS) was adapted in 1998 to assess
behaviors among youth in alternative or continuation high schools (Alternative High School
Youth Risk Behavior Survey [ALT-YRBS]), shedding light on the prevalence of sexual risk
behaviors and substance use among this group (CDC 1999). Compared to the 1999 YRBS
sample, greater proportions of ALT-YRBS youth engaged in sexual intercourse, had
multiple sexual partners, and initiated sexual intercourse by 13 years of age (CDC, 1999;
CDC, 2000). CHS youth also reported less condom use at last sexual intercourse and greater
substance use prior to last sexual intercourse. While trend data are available for high school
youth and show decreasing sexual risk-taking (e.g., delayed sexual initiation, smaller
proportion of students with multiple sexual partners), trend data are not available for CHS
youth since ALT-YRBS has only been conducted in 1998. Regardless of the academic
setting, males and African-American youth are more likely to have an earlier age at sexual

9
debut and have multiple sexual partners. Gender differences in sexual activity vary by school
type as high school males and CHS females are more likely to have sex than their
counterparts.
Findings from O’Hara, Parris, Fichtner, & Oster (1998) support these national
results. Among sexually active youth in a dropout prevention program from the early 1990’s,
half reported not using a condom at their last sexual encounter, and 55% reported multiple
sexual partners in the previous three months. Males, African-Americans, and Latinos were
more likely to report sexual risk behaviors regardless of the academic setting (CDC, 1999;
CDC, 2000; CDC, 2006b; O’Hara et al., 1998).
Other research has shown that youth who have dropped out of school are more likely
to have an early onset of sexual initiation, lack contraceptive (including condoms) use, and
have higher pregnancy rates (Brewster, Cooksey, Guilkey, & Rindfuss, 1998; Darroch,
Landry, & Oslak, 1999; Manlove, 1998; Mauldon & Luker, 1996). Youth who are not in
school or are in a dropout prevention program also have higher substance abuse rates and are
at increased risk of HIV and other sexually transmitted diseases (CDC, 1994a; CDC, 1994b;
Grunbaum & Basen-Engquist, 1993).
Research that has been conducted among CHS youth has focused mostly on
substance use as the variable of interest (Newcomb, Maddahian, Skager, & Bentler, 1987;
Sussman et al., 1995a; Sussman, Stacy, Dent, Simon, Galaif, et al., 1995b). Substance use
research conducted among CHS youth has elicited knowledge on the prevalence of
substance use and factors associated with it. Lifetime and current cigarette smoking, illicit
drug use, and alcohol use are high among high school youth but are disproportionately
higher among CHS youth (CDC, 1999; CDC, 2000; Newcomb et al., 1987; Stacy, Ames,
Sussman, & Dent, 1996; Sussman et al., 1995a; Sussman et al., 1995b). The majority of
CHS youth from the ALT-YRBS sample reported drinking alcohol (92%), smoking a

10
cigarette (91%), and/or using marijuana (85%) at least once in their lifetime (CDC, 1999).
Unlike the ALT-YRBS sample, the national YRBS sample of youth in traditional high
schools (1999) had different substance using histories. Although the majority reported
drinking alcohol (81%) and smoking a cigarette (70%) at least once in their lifetime, less
than half had used marijuana (CDC, 2000). In similar time periods, the prevalence of
substance use was substantially higher among the ALT-YRBS sample although unknown if
statistically significant. Regardless of the academic setting, males were most likely to use
illicit substances than females with the exception of alcohol where similar usage was seen
(CDC, 1999; CDC, 2000). Both Whites and Latinos had comparable proportions of
substance use.
Like substance use, early onset of sexual initiation is a significant predictor of sexual
risk behaviors whereby youth who have an earlier age at sexual onset often engage in
multiple high-risk sexual behaviors (e.g., multiple sexual partners). Research has provided
sufficient evidence of early age at sexual onset as a predictor or correlate of HIV risk
behaviors among several adolescent samples (CDC, 1999; CDC, 2006a; Harwell, Trino,
Rudy, Yorkman, & Gollub, 1999; Hutchinson, Jemmott, Sweet Jemmott, Braverman, &
Fong, 2003; Krantz, Lynch, & Russell, 2002; Miller, Clark, & Moore, 1997; National
Research Council, 1987; O’Donnell, O’Donnell, & Stueve, 2001; O’Donnell, Stueve, San
Doval, Duran, & Haber, 1999; O’Hara, Parris, Fichtner, & Oster, 1998; Sikkema et al.,
2004; Stock, Bell, Boyer, & Connell, 1997). Among a sample of Latina and African-
American female adolescents, early sexual initiation was significantly correlated with an
older sexual partner at sexual onset but not with subsequent number of sexual partners
(Miller, Clark, and Moore, 1997).
Compared to the 1999 YRBS sample, the ALT-YRBS sample was almost three
times more likely to report an earlier age at sexual onset (CDC, 1999; CDC, 2000). Almost

11
one-fourth (22%) of ALT-YRBS youth reported having first had sexual intercourse before
13 years of age compared to 8.3% of YRBS youth. CHS youth reported higher levels than
the YRBS sample for males (29.6%) and females (12.8%) as well as Latino (19.6%), and
White youth (18.1%).
Substance use often co-occurs with risky sexual behaviors and has been associated
with an earlier age at sexual initiation. In fact, there is evidence that early use of substances
precedes sexual initiation (Rosenbaum & Kandel, 1990). Among most of these studies, a
significant relationship existed between early substance use and early sexual initiation
among males (Mott, Fondell, Hu, Kowaleski-Jones, & Menaghan, 1996; Smith, 1997;
Walter, Vaughan, Armstrong, Krakoff, Maldonado, Tiezze, et al., 1995), females (Kinsman,
Romer, Furstenberg, & Schwarz, 1998; Stueve & O’Donnell, 2005), and African-American
youth (Kinsman et al., 1998; Mott et al., 1996; Smith, 1997; Stueve & O’Donnell, 2005;
Walter et al., 1995). However, there was no effect of early substance use on early sexual
initiation among one sample of females (Bachanas, Morris, Lewis-Gess, Sarett-Cuasay,
Flores, Sirl, et al., 2002a) or among one predominantly White sample (Staton, Leukefeld,
Logan, Zimmerman, Lynam, Milich, et al., 1999).
Like substance use, sensation-seeking may produce some disinhibition that impacts
decision-making and impacts sexual risk-taking (Robbins & Bryan, 2004; Sussman, 2005).
Sensation-seeking refers to an individual’s need for “complex sensations and experiences”
and the willingness to take risk in order to fulfill them (Greene, Krcmar, Walters, Rubin,
Hale, & Hale, 2000). According to Greene et al. (2000), sensation-seeking impacts both
risky sexual behaviors and drug-using behaviors in conjunction with the belief that one is
unique and invincible regardless of one’s behaviors. Youth may engage in risky behavior
because of the positive emotional outcomes that are anticipated and may emerge.

12
The relationship between sensation-seeking and sexual risk-taking has been explored
among adult populations (Kalichman, Heckman, & Kelly, 1996; Leigh & Stall, 1993;
Valdiserri, Lyter, Leviton, Callahan, Kinglsey, Rinaldo, et al., 1988). When the relationship
has been explored in adolescent samples, it is often in conjunction with drug-using behaviors
(Bailey, Gao, & Clark, 2006; Crockett, Raffaelli, & Shen, 2006; Li, Stanton, Cottrell, Burns,
Pack, & Kaljee, 2000).
The effects of sensation-seeking on sexual behaviors have been discrepant. Among a
sample of urban community-based African-American adolescents, sensation-seeking scores
were high among adolescents who used drugs (either alcohol or illicit) and initiated sex over
a 4-year period (Li et al., 2000). Sensation-seeking scores were low for adolescents who
only reported initiating sex during the time period. Additionally, teens who initiated
substance use and sex during the study period were more likely to report multiple sexual
partners at the end of the study period. Adolescents with higher sensation-seeking scores
were more likely be younger, male, and to have initiated substance use prior to sex. Unlike
Li et al. (2000), Crocket, Raffaelli, & Shen (2006) found that the relationship between
sensation-seeking (termed “risk proneness”) and sexual risk-taking was fully mediated by
substance use among a national sample of youth whereas Bailey, Gao, & Clark, (2006)
found no correlation between sensation-seeking and high-risk sexual behaviors among a
sample of adolescents with substance use disorders.
As is evident, there is considerable literature on the relationships among substance
use, an earlier age at sexual onset, and sexual behavioral outcomes with variance in
populations studied and methodologies. Several studies have been conducted with CHS
youth specifically that reveal a higher prevalence of substance use and sexual behaviors with
varying results by gender and ethnicity. However extant literature on behaviors among
Latino CHS youth, current research, and research on the longitudinal effects of substance use

13
and sensation-seeking during adolescence on sexual risk behaviors through young adulthood
among a sample of CHS youth is limited. CHS youth are at particular risk for HIV given
their propensity to not attend classes regularly and subsequent availability for involvement in
risk behaviors. Although continuation high schools provide a structured environment and
specialized skills for youth, CHS youth are exposed to peers with higher risk behaviors.
In an effort to further elucidate the literature among youth in continuation high
schools, a study to identify sexual behaviors among young adults who were in continuation
high schools was conducted. The effects of substance use and sensation-seeking during
adolescence on subsequent sexual behaviors were of special interest. It is expected that CHS
young adults who reported greater substance use (i.e., alcohol, cigarette, and/or marijuana
use) or who scored higher on sensation-seeking at adolescence are more likely to report a
higher number of lifetime and recent sexual partners at four-year follow-up. Furthermore,
these relationships will be at least partially mediated by age at sexual initiation where an
earlier age at sexual initiation is associated with greater substance use, higher sensation-
seeking, and a higher number of sexual partners.
Methods
Sample
The present study uses baseline and four-year follow-up data from a sample of 111
young adults originally recruited from fourteen continuation high schools in southern
California. For the final model, three cases were omitted (further detailed in the Results
section) yielding a sample size of 108 subjects for statistical modeling. Young adults were in
ninth through twelfth grades, participating in a larger intervention study (Project Towards
No Drug Abuse [TND]) on the roles of associative memory and implicit cognition on
substance abuse among at-risk youth. Surveys assessed demographic factors and included

14
psychosocial and behavioral measures such as substance use, sensation-seeking, and sexual
behaviors.
School selection
Continuation high schools were selected from 29 southern California school districts
that contain continuation high schools. Schools were randomly sampled from the sample of
continuation high schools not receiving Project TND through Project TND3. Within the
schools, half of the classes were randomly selected. Participants were then randomly
assigned to the different assessment conditions.
Student recruitment
Baseline
After receiving approval from school administrators, invited classrooms were
selected by each school. All students listed in the classroom’s roster were invited to
participate in the study. Both parental consent forms and student assent forms were collected
for students under the age of 18 years. Students and parents were informed that the study
related to health behaviors, some of which may be considered sensitive and/or illegal.
Consenting students were randomized on the day of data collection. All materials and
procedures received approval from the Institutional Review Board (IRB) at the University of
Southern California’s Health Sciences Campus (USC HSC).
Four-year follow-up
All materials and procedures from the current study received IRB approval from the
USC HSC in May 2007. A letter signed by the TPRC Principal Investigator was sent to all
eligible subjects notifying them of the telephone-based follow-up study. Subjects were given
the ability to opt-out of being contacted by study staff by returning a postage-paid postcard
by a designated date indicating their decision to not participate. Those subjects who did not
actively refuse to be contacted were eligible for the study.

15
Student tracking
Baseline
TPRC participants provided contact information and received reminder slips with
the scheduling of future TPRC assessments. The participant’s contact information includes:
(1) participant’s name, (2) current address, (3) phone number, (4) contact person’s name and
phone number (friend or relative who knows how to reach the participant), (5) participant’s
parent(s) name, address, and phone number, (6) participant’s social security number, (7)
participant’s driver’s license number, (8) e-mail address, (9) dates and status of last
attempted contact (i.e., contacted, measured, refused), and (10) an extensive notes section
where all attempts to reach participants are recorded.
Four-year follow-up
For the current study, sampling was limited to students who were at least 18 years
old at the time of sampling and self-identified as Latino/a, White, or African-American
(n=747). The sample size was further limited to those individuals who had some reliable
contact information in the TPRC participant database (i.e., phone number or address) and the
name of at least one parent or guardian with whom an online search could be made for
additional information. Given the additional criteria, the final base sample was 391 CHS
youth.
The online search databases of White Pages, Net Detective, and PeopleFinders were
used to update addresses and phone numbers. Searches on each site were made using the
names of the parent(s) or guardian(s) and the student. Reverse searches by address and
phone numbers listed in the TPRC participant database were also conducted in the White
Pages database. If available, phone calls to contacts listed in the TPRC participant database
were made. A final search of the White Pages database was conducted in September 2008 to
update any information.

16
TPRC participants were considered “lost to follow-up” if the address and phone
number listed in the TPRC participant database were no longer valid, and updated
information was not available for both the parent or guardian and TPRC participant. TPRC
participants were considered “unable to reach” if phone calls were not returned; there was
never any answer; or no phone number was listed publicly (e.g., address listed but phone
number unlisted). A minimum of ten phone call attempts were made to each participant, but
messages were left only at every third call or after a few weeks of leaving a last message.
Procedure
Baseline
Self-administered paper-and-pencil surveys were completed in the classrooms at the
respective CHS. Since classes were taught in English, it was assumed that the student had
the ability to read and complete a survey in English. The survey took approximately 30 to 45
minutes to complete. Baseline data on 894 students were collected between February and
November 2004.
Four-year follow-up
Between February and April 2008, letters were sent to all individuals in the final
base sample (n=391) with contact information. Random sampling by gender and race (Latino
and White only) was conducted with ultimate sampling of all TPRC participants. The data
collection team consisted of the PI and five undergraduate students doing directed research
under the aegis of the Chair of the PI’s Doctoral Committee. Each undergraduate research
assistant participated in 5 hours of training prior to contacting TPRC participants. The
training was conducted by the PI and consisted of a 2-hour lecture on the study proposal; 1-
hour training on the study materials and protocol; and 1-hour training of conducting mock
interviews with one another. Once phone calls to TPRC participants started, the PI worked
alongside each research assistant for the first few days for quality assurance purposes.

17
Once the participant was reached by telephone, an initial screener with verbal
consent was read giving basic information on the purpose of the assessment. The survey was
administered by a trained interviewer once the participant consented verbally to participate.
The survey took approximately 15 minutes to complete. Subjects were given the same
unique identification number so that the first and second waves of data could be linked. To
minimize response biases and prior to asking the sexual behaviors questions, participants
were reminded that their name is not written on the survey, and they can refuse to answer
any question. A $15 gift card was mailed to each participant for any out-of-pocket expenses
related to completing a survey. Follow-up data were collected between March and
September 2008. Raw data were entered by the PI into an Access database which was then
imported into SAS.
Measures
Dependent variables. Lifetime and recent (i.e., past 3 months) numbers of sexual
partners were assessed as separate single-item outcome variables. These items were adapted
from the YRBS and collected during the current study (4-year follow-up). Sexual partners
were defined as any partner with whom the participant engaged in anal, receptive oral,
and/or vaginal intercourse in one’s lifetime and in the past 3 months (i.e., recent). For
participants who have had sex but not in the past 3 months, the number of recent partners
was coded as ‘0.’ The number of sexual partners was treated as a continuous variable.
Independent variables.
Substance use. Information on lifetime use of alcohol, marijuana, and cigarettes was
collected at baseline. An 11-point scale was used to assess the number of times each drug
was used ranging from 0 to 91 and more times (Graham, Flay, Johnson, Hansen, Grossman,
& Sobel, 1984). The possible range of raw scores was 1 to 11 with a higher number
indicating greater substance use. The Graham et al. (1984) substance use scale has been used

18
with continuation high school youth and has yielded high reliability ratings. For the full
TPRC baseline sample with available data (n=873), the Cronbach alpha for alcohol,
marijuana, and cigarette use was α = 0.79. Substance use by type at baseline was assessed as
a continuous variable. In descriptive statistics and tests of association, alcohol, marijuana,
and cigarettes were assessed separately. For structural equation modeling, substance use was
treated as a latent factor. Exploratory factor analysis was performed to select reliable items
with a minimum factor loading of 0.40 to retain in the final scale. The final substance use
scale (Cronbach α = 0.82) retained all three items.
Sensation-seeking. Assessment of sensation-seeking was conducted at baseline. The
sensation-seeking scale from the Zuckerman-Kuhlman Personality Questionnaire consisted
of five items embedded within 19 statements asked in a True/False format (Zuckerman,
Kuhlman, Thornquist, & Kiers, 1991). The sensation-seeking scale has been used in studies
on continuation high school and has been found to be reliable (Cronbach α = 0.75; Sussman,
Simon, Dent, Steinberg, & Stacy, 1999). The sensation-seeking items pertained to enjoyment
of “new and exciting experiences and sensations” regardless of fear; enjoy doing things for
thrill or doing things that can be frightening; doing “crazy things for fun;” and preferring
unpredictable friends. Items were recoded from a ‘2’ (True)/‘1’ (False) to ‘1/0’ coding
scheme, respectively. In exploratory factor analysis, the first four items loaded onto one
factor whereas the last item (preferring unpredictable friends) loaded onto a separate factor.
Therefore, the final sensation-seeking scale (Cronbach α = 0.67) consisted of the weighted
sum of the values of the first four variables based on factor loadings. The possible range of
raw scores was 0 to 4 with a higher number indicating higher sensation-seeking.
Mediator. Age at sexual onset was treated as a mediator between substance use and
sensation-seeking and numbers of lifetime and recent sexual partners. Age at sexual onset
was a single-item open-ended indicator that was treated as a continuous variable.

19
Covariates. Several variables were assessed for descriptive purposes and tests of
association.
Age differential of first sexual partner. Questions on if the first sexual partner was
the same age, older, or younger and by how many years were asked in order to determine the
age differential at sexual onset. A dichotomous composite variable (Cronbach α = 0.78) was
created from the two items where an older partner (“Older, 3 or more years”) was coded as
‘1’ and a partner that is younger, same age, or no more than 2 years older was treated as the
referent category (coded as ‘0’).
Sexual maturation. The Adolescence Scale-Imperial College School of Medicine
(AS-ICSM), a brief retrospective assessment of pubertal milestones, was used to assess
sexual maturation (Kaiser & Gruzelier, 1999). The AS-ICSM consists of one common item
(growth spurt); one female-specific item (menarche); and three male-specific items (voice
break, regular shaving, and first nocturnal emission). The AS-ICSM was normed among
young adults (mean age = 20.7 years) and shown to be a reliable retrospective assessment for
both females (Cronbach α = 0.87) and males (Cronbach α = 0.83; Kaiser & Gruzelier, 1999).
In the current study, the mean age at sexual maturation for females was computed
using both growth spurt and menarche (Cronbach α = 0.31). Almost one-fourth of males
(23%) either did not remember or had not experienced a nocturnal emission, and thus, this
item was omitted. For males, the mean age at sexual maturation was computed using ages at
growth spurt, voice break, and regular shaving (Cronbach α = 0.61).
Gender and race/ethnicity. Information on self-reported gender and race/ethnicity
were collected during the TPRC baseline assessment. Gender and race/ethnicity were
dichotomously coded. Females were coded as ‘0,’ and males were coded as ‘1.’ Latinos,
coded with a value of ‘1,’ were compared to non-Latinos (coded as ‘0’).

20
Age. Using birth date and follow-up interview date, age in years at the time of the
four-year follow-up assessment was computed.
Sexual behavior orientation. Sexual behavior orientation was defined using four
variables: self-identified sexual orientation; subject’s gender; and reported gender(s) of
lifetime and recent sexual partners. The three sexual behavior orientation categories were
men who had sex with both men and women (‘1’), women who had sex with men and/or
women (‘2’), and men who had sex with women (‘3’). If a male self-identified as
“heterosexual” but engaged in sex with both males and females, then he was coded as
behaviorally having had sex with both men and women (‘1’).
Acculturation. Acculturation is the process by which the values, norms, attitudes,
and behaviors of an individual of a minority culture are fused with those from the majority
culture and incorporated into one’s existence (Marín, Sabogal, Marín, Otero-Sabogal, &
Perez, Stable, 1987). Marín et al. (1987) created an acculturation scale that focused on the
extent of which languages were spoken (i.e., English, Spanish) in particular settings.
Although the Marín et al. (1987) linguistic acculturation scale was normed with Latino
adults (Cronbach α = 0.90), its subsequent use among Latino youth has yielded high
reliability (e.g., Cronbach α = 0.90 in Ford & Norris, 1993).
Both TPRC and the current study used the Marín et al. (1987) short acculturation
scale (Cronbach α = 0.88, both periods). The four items assessing linguistic acculturation
were language(s) most often read and spoken; language(s) spoken at home; language(s)
spoken with friends; and language(s) to which movies, television, and radio shows are
watched and listened. The response categories were: (1) Only English; (2) English more than
another language; (3) English and another language equally; (4) Another language more than
English; and (5) Only another language (not English). The final acculturation item was the

21
mean value of the four items with a possible range of 1 (“Only English”) to 5 (“Only another
language [not English]”).
Data analysis
Since baseline data were nested within schools, intraclass correlation (ICC)
coefficients were computed on each dependent variable to address the potential cluster effect
of schools. ICC coefficients provide an estimate of the degree of commonality in
observations within a given unit (e.g., school). School-level ICCs that are less than 0.03
indicate low degree of dependence in observations (Murray and Blitstein, 2003). ICCs were
calculated for variables in the model, and all were found to be less than 0.03. Also four-year
follow-up data were collected when the participant was no longer in continuation high
school, providing further support that controlling for school may not be pivotal in the present
analyses. Therefore, structural equation modeling was performed without accounting for
school.
Descriptive statistics were computed for all demographic variables. Independent t-
tests were used to test associations in sexual maturation, substance use, sensation-seeking,
age at sexual onset, and the number of sexual partners by gender, ethnicity, age at sexual
maturation, and sexual onset characteristics. Chi-square tests were used to test associations
in ever having had sexual intercourse and age differential with first sexual partner by gender
and ethnicity. Exploratory factor analyses were used to determine the relationships between
the observed variables and their underlying construct. The criterion for statistical
significance was set to a 0.05 level. Measures of reliability, descriptive statistics, tests of
association, and exploratory factor analyses were computed using SAS version 9.1.
The relationships of alcohol, marijuana, and cigarette use to its respective latent
factor of substance use were empirically assessed through confirmatory factor analyses.
Empirical data were used to statistically test the hypothesized model in order to confirm the

22
adequacy of the indicator variables used to represent the proposed latent factor. Factor
loadings of indicator variables were expected to have high loadings on their respective
factors and to be significant in order to show evidence of convergent validity. Following
inspection of confirmatory factor analysis results to verify the presence of a distinct
substance use construct, causal pathways were inserted to delineate the relationships among
substance use, sensation-seeking, age at sexual onset, and number of lifetime and recent
sexual partners.
Assessment of model fit was performed using the goodness-of-fit Chi-square test
statistic as well as the comparative fit index (CFI) and the root mean squared error of
approximation (RMSEA). A CFI of 0.90 or more and an RMSEA of less than 0.05 is
considered to be a good fit (Browne & Cudeck, 1993; Kline, 1998). The maximum-
likelihood estimation procedure was employed as a global test of the model (Bentler, 1990).
The Lagrange Modifier (LM) test was requested to identify parameters that would improve
model fit. Confirmatory factor analysis, assessment of model fit, and the maximum-
likelihood estimation procedure were computed using EQS version 6.1.
Results
A total of 111 young adults (14.9% of eligible sample) completed a follow-up
survey (Table 1). The sample was 51% female and 56.8% Latino with a mean age of 20.6
years. The mean acculturation score was 1.8 with participants speaking English more than
another language. The majority of the sample (98.2%) reported having had sexual
intercourse and a mean age of 14.8 years at sexual initiation. Data from three subjects were
omitted from analyses pertaining to sexual behaviors. Two individuals reported never having
sex, and one individual was an outlier. The majority either self-reported as or was
behaviorally heterosexual (96.4%).

23
Table 1. Characteristics of the sample
Variable Overall
N=111
Female
N=56 (50.5%)
Male
N=55 (49.5%)
Latino
N=63 (56.8%)
Non-Latino
N=48 (43.2%)

N (%) N (%) N (%) N (%) N (%)
Ever had sex 109 (98.2) 55 (98.2) 54 (98.2) 61 (96.8) 48 (100.0)
Older (3+ years) sexual partner at
sexual onset (n=109)
17 (15.6) 14 (25.5) 3 (5.6) 9 (14.8) 8 (16.7)
Sexual behavior orientation
Men who have sex with men 4 (3.6) --- --- 4 (7.3) 3 (4.8) 1 (2.1)
Women who have sex with men 56 (50.5) 56 (100.0) --- --- 33 (52.4) 23 (47.9)
Men who have sex with women 51 (45.9) --- --- 51 (92.7) 27 (42.9) 24 (50.0)

Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Range
Age (years)

20.6 (1.2) 20.3 (1.3) 20.9 (1.1) 20.6 (1.2) 20.5 (1.3) 18-23
Acculturation (score) 1.8 (0.9) 1.9 (0.9) 1.8 (0.8) 2.4 (0.7) 1.0 (0.1) 1-5
Sexual maturation (years) 13.5 (1.1) 13.1 (0.9) 14.0 (1.1) 13.5 (1.1) 13.5 (1.1) 11-18
Age at sexual onset (years; n=96) 14.8 (1.5) 14.9 (1.7) 14.7 (1.4) 15.1 (1.6) 14.5 (1.3) 11-17
Substance use
Alcohol 5.7 (3.8) 5.2 (3.7) 6.3 (3.8) 4.7 (3.7) 7.1 (3.5) 1-11
Cigarettes 5.0 (4.1) 4.9 (4.0) 5.2 (4.2) 3.1 (2.7) 7.7 (4.2) 1-11
Marijuana 5.7 (4.4) 4.9 (4.1) 6.4 (4.6) 4.0 (3.9) 7.9 (4.0) 1-11
Sensation-seeking 3.6 (1.4) 3.7 (1.3) 3.6 (1.4) 3.3 (1.4) 4.0 (1.2) 0-4

Median (Q1,Q3) Median (Q1,Q3) Median (Q1,Q3) Median (Q1,Q3) Median (Q1,Q3) Range
Number of sexual partners
Lifetime (n=108) 6 (3, 8) 4 (3, 8) 7 (5, 10) 5 (3, 7) 7 (5, 10) 1-25
Past 3 months (n=103) 1 (1, 2) 1 (1, 1) 1 (1, 2) 1 (1, 1) 1 (1, 2) 1-5

24
Select gender differences but no ethnic differences were evident in sexual maturation and
sexual experience (Table 2). Females had reached sexual maturity at an earlier age (13.1
years) than males (14 years; t = -5.06, p < 0.001). At sexual debut, females (25.5%) were
significantly more likely than males (5.6%) to have an older sexual partner (χ
2
= 8.2, p =
0.004). There were no gender differences in the proportion of persons who had ever engaged
in sexual intercourse.
Table 2. Associations with sexual maturation and sexual onset experience
Sexual maturation Ever had sexual
intercourse
Age differential with first
sexual partner, 3+ years
Mean
years
t % χ
2
% χ
2

Gender
Male (n=55) 14.0 -5.06* 98.2 0.0002 5.6 8.2*
Female (n=56) 13.1 98.2 25.5

Ethnicity
Latino (n=63) 13.5 0.02 96.8 1.55 14.8 0.07
Non-Latino (n=48) 13.5 100.0 16.7
*
p < 0.01

Substance use during adolescence was significantly associated with ethnicity and
age at sexual onset and demonstrated a trend-level association with gender (Table 3).
Overall, alcohol and marijuana were the most widely used substances (M = 5.7) followed by
cigarettes (M = 5.0). Compared to Latinos, non-Latino youth reported greater alcohol (M =
7.1 v. 4.7, t = 6.65, p = 0.01), cigarette (M = 7.6 v. 3.9, t = 5.16, p < 0.0001), and marijuana
use (M = 7.9 v. 4.0, t = 6.35, p < 0.0001). Persons who initiated sex by 15 years of age
reported higher alcohol (M = 6.5 v. 4.8, t = -2.26, p = 0.03), cigarette (M = 6.7 v 4.4, t = -2.7,
p = 0.008), and marijuana use (M = 5.8 v 4.3, t = -1.81, p = 0.07) than youth who delayed
sexual onset. Females reported less marijuana use (M = 4.9) than males (M = 6.5; t = -1.91, p
= 0.06), but there were no differences in alcohol and cigarette use.
Ethnic and age at sexual debut differences were also evident in adolescent sensation-
seeking. Non-Latino youth (M = 4.0 v. 3.3; t = 2.68, p = 0.008) had higher mean sensation-

25
Table 3. Associations with substance use and sensation-seeking
Substance use
a

Alcohol Cigarettes Marijuana Sensation-seeking
b

n
Mean
score
t n
Mean
score
t

n
Mean
score
t

n
Mean
score
t
Gender
Male 53 6.28 -1.54 52 5.17 -0.37 54 6.48 -1.91 53 3.57 0.36
Female 56 5.16 56 4.88 56 4.89 56 3.66

Ethnicity
Latino 62 4.66 3.44
**
62 3.94 5.16
**
62 3.95 6.35
**
61 3.31 2.68
**

Non-Latino 47 7.09 48 7.63 46 7.9 48 4.0

Sexual maturation
c

≤13.5 years old 66 5.58 0.44 65 4.72 0.92 66 5.30 1.07 66 3.55 0.65
13.6+ years old 43 5.91 43 5.47 44 6.23 43 3.72

Age at Sexual Onset
c,d

≤15 years 61 6.52 -2.26
*
62 6.74 -2.72
**
60 5.77 -1.81
†
62 3.79 -1.73
†

16+ years 44 4.84 44 4.43 44 4.30 43 3.33

Age Differential with
First Sexual Partner
c,d

3+ years 17 6.29 -0.60 17 5.06 0.04 17 5.47 0.29 17 3.47 0.48
<3 years 90 5.69 89 5.10 89 5.81 90 3.64
†
p < 0.10,
*
p < 0.05,
**
p < 0.01
Note: some frequencies do not sum to n due to missing values
a
Range: 0 (never) to 11 (91+ times)
b
Unstandardized, range 0 (low) – 4 (high)

c
Dichotomized at median
d
Limited to those who have ever had sexual intercourse (n=109)

26
seeking scores than Latino youth. A trend towards greater sensation-seeking was also
demonstrated among early sexual initiators (M = 3.8 v. 3.3; t = -1.73, p = 0.08). There were
no gender differences in sensation-seeking with both males and females exhibiting similar
sensation-seeking scores (M = 3.6 and 3.7, respectively).
Among this sample, non-Latino youth were more likely to have an earlier age at
sexual onset than Latino youth (M = 14.5 v. 15.1; t = -2.14, p = 0.03; Table 4). Persons who
had an older partner at sexual initiation were also more likely to have an earlier age at sexual
onset than persons with a peer-aged sexual partner (M = 13.7 v. 15.0; t = 3.38, p = 0.001).
No gender differences existed.
Table 4. Associations with age at sexual onset and number of sexual partners

Age at sexual
onset

Lifetime number of
sexual partners

Number of sexual
partners,
past 3 months

Mean
age
t
Mean
number
t
Mean
number
t
All 14.8 --- 7.39 --- 1.49 ---

Gender
Male 14.7 0.83 9.18 -2.47
*
1.73 -2.93
**

Female 14.9 5.61 1.25

Ethnicity
Latino 15.1 -2.14
*
6.53 1.40 1.37 1.66
†

Non-Latino 14.5 8.49 1.64

Sexual maturation
a

≤13.5 years old 14.8 0.34 7.30 0.18 1.33 2.21
*

13.6+ years old 14.9 7.54 1.74

Age at Sexual Onset
b,c

≤15 years old --- --- 9.47 -4.02
**
1.68 -3.02
**

16+ years old --- --- 4.43 1.22

Age Differential with
First Sexual Partner
b,c

3+ years 13.7 3.38
*
9.53 -1.29 1.71 -1.16
<3 years 15.0 6.96 1.45
†
p < 0.10,
*
p < 0.05,
**
p < 0.01
a
Dichotomized at mean
b
Limited to those who have ever had sexual intercourse (n=109)
c
Dichotomized at median

27
The numbers of sexual partners varied significantly by gender and age at sexual
onset characteristics. Females reported fewer lifetime (M = 5.6 v. 9.2; t = -2.47, p = 0.02)
and recent sexual partners (M = 1.3 v. 1.7; t = -2.93, p = 0.004) than males. Persons who
initiated sex by 15 years of age reported more lifetime (M = 9.5 v. 4.4; t = -4.02, p = 0.0001)
and recent sexual partners at follow-up (M = 1.7 v. 1.2; t = -3.02, p = 0.004) than individuals
who delayed sexual onset. Although there were no ethnic differences in the lifetime number
of sexual partners, ethnic differences in the number of recent sexual partners approached
significance. Non-Latinos were slightly more likely to report a higher number of recent
sexual partners than Latinos (M = 1.6 v. 1.4, t = 1.66, p = 0.10).
The measurement model was tested using confirmatory factor analyses. Alcohol,
marijuana, and cigarette use loaded significantly onto the substance use factor providing
evidence of construct and convergent validity (Figure 2). The basic theoretical model yielded
a statistically satisfactory fit to the data (χ
2
= 6.49, df = 9, p = 0.69) and an overall good fit
based on the fit indices (CFI = 1.0; RMSEA = 0.000). Therefore, model modification guided
by the LM test was not necessary.
The relationships between substance use and number of sexual partners were fully
mediated by age at sexual onset. Young adults who reported greater substance use at
baseline reported an earlier age at sexual onset (β = -0.23; p < 0.05) which in turn was
associated with a greater number of lifetime (β = -0.44; p < 0.05) and recent (β = -0.28; p <
0.05) sexual partners. Substance use did not have a significant direct effect on either lifetime
or recent sexual partners. Sensation-seeking did not have any direct or indirect effects on age
at sexual onset and numbers of sexual partners.

28
E6
E7
V6
Lifetime
number of
sexual partners
V7
Number of
sexual partners,
past 3 months

V5
Age at
sexual onset
E5
V4
Sensation-
seeking

F1
Substance
Use
V3 Cigarette Use
USE
V2 Marijuana Use

V1 Alcohol Use
E2
E3
E1
χ
2
(9) = 6.49, p = 0.69
0.67*
Figure 2. SEM Mediation Model.
Parameters and factor loadings are presented. *p < 0.05
0.52*
0.69*
0.75*
0.85*
0.73*
0.97*
0.04
-0.44*
-0.04 -0.23*
-0.06
-0.28*
-0.003
-0.11
0.89*
0.96*
0.65*

29
Discussion
It was hypothesized that higher substance use and sensation-seeking during
adolescence would be predictive of a greater number of lifetime and recent sexual partners at
four-year follow-up, and these relationships would be at least partially mediated by an earlier
age at sexual onset. Inconsistent with the literature, there was no direct effect of substance
use during adolescence on the number of sexual partners through young adulthood. The
relationship between substance use and number of sexual partners was fully mediated by age
at sexual initiation. CHS youth who reported greater substance use during adolescence were
more likely to initiate sex at an earlier age. Early initiation of both activities was predictive
of the number of lifetime and recent sexual partners by young adulthood. Supporting Bailey,
Gao, & Clarke’s (2006) findings, sensation-seeking was not predictive of the number of
sexual partners either directly or via age at sexual initiation.
Although direct effects of substance use and sensation-seeking were not significant
in a structural equation model, significant bivariate relationships by key demographic
variables existed. Unlike the ALT-YRBS sample from 1998 that showed no differences in
substance use between Latino and White youth, alcohol, cigarette, and marijuana use were
significantly higher among non-Latino than Latino youth in the current study.
In bivariate analyses, the relationship between sensation-seeking and age at sexual
initiation approached statistical significance. CHS youth with higher sensation-seeking
scores may be more likely to initiate sex at a younger age than lower sensation-seekers. Like
substance use, sensation-seeking may impact age at sexual initiation given a larger sample
size. Apart from age at sexual initiation, sensation-seeking may indirectly impact the number
of sexual partners through other factors not previously assessed or collected (e.g., decision-
making).

30
Males and early sexual initiators had greater numbers of lifetime and recent sexual
partners. Other factors may contribute to more sexual partners among males and reinforce
their sexual practices. Early initiation to sexual activity provides a longer period of time for
sexual behaviors to occur including having sexual relationships with multiple partners. Non-
Latinos had slightly more recent sexual partners. Unlike Miller, Clark, & Moore’s (1997)
findings, no differences in the number of sexual partners by age differential with a first
sexual partner existed. The differences may be due to the sample whereby youth in and who
graduated from continuation high schools have greater sexual risk behaviors than other
adolescent samples. Like Miller, Clark, & Moore (1997), an earlier age at sexual initiation
was associated with an older first sexual partner.
Despite the study findings, several limitations need to be considered. The ability to
find true significant predictive relationships and associations is limited by the sample size.
The small sample size does not provide sufficient power to detect significant findings.
Attrition was high with only 15% of the baseline having completed a four-year follow-up
survey. Although efforts were taken to contact a larger sample size, tracking participants
proved to be difficult, many of whom were lost to follow-up. However, the final sample size
is representative of a CHS sample that is about 50% Latino and 50% female. Given the high
attrition, baseline participants who were not interviewed in the most recent follow-up (85%
of the sample) may have sexual behaviors that differ from the current sample thereby
limiting the generalizability of the findings.
Validity of the data could also be affected by the self-report nature of the study,
especially with sensitive questions on sexual behaviors. Participants were assured of the
confidentiality of all data, and measures were taken prior to commencing and during the
survey to display this confidentiality to all participants. Furthermore, measures that were

31
used had been validated previously in adolescent samples. Therefore, there is no reason to
believe that participants were differentially dishonest in their answers.
Despite these limitations, results from this study support the effect of early sexual
onset on subsequent sexual risk-taking. Results highlight an important pathway between
substance use and age at sexual initiation during adolescence on the number of sexual
partners by young adulthood. Further research is warranted to understand the pathways by
which sensation-seeking and sexual risk-taking occurs among youth who were in
continuation high schools. These conclusions bear potential implications for the prevention
of sexual risk-taking that place an individual at risk of HIV and other sexually transmitted
diseases (STD). Interventions aimed at youth in continuation high schools could benefit by
incorporating comprehensive sexual education into substance abuse prevention curriculum.
Sexual education should not only focus on delaying sexual onset but also on limiting the
amount of sexual partners as a way of reducing one’s risk of HIV and other STD. Future
research is needed to determine whether this type of intervention can reduce sexual risk-
taking among this at-risk population.

32
CHAPTER THREE: THE ROLE OF DEPRESSIVE SYMPTOMATOLOGY
ON SEXUAL BEHAVIORS

The estimated numbers of HIV/AIDS cases among adolescents and young adults
(ages 13 to 24 years) increased by 37% between 2004 and 2007 (CDC, 2009). HIV risk
factors include having multiple sexual partners and engaging in unprotected sex. National
epidemiologic data reveal higher sexual risk behaviors among youth in continuation high
schools (CHS) than youth in regular high schools. According to the 1999 national Youth
Risk Behavior Survey, 49.9% of youth in high schools have had sexual intercourse; 16.2%
have had four or more sexual partners; and 58% used a condom at last sexual intercourse
(CDC, 2000). Larger proportions of CHS youth engaged in these sexual risk behaviors with
87.7% having had sexual intercourse; 50.3% having had four or more sexual partners; and
42.1% having used a condom at last sexual intercourse (CDC, 2000). CHS youth are at
particular risk for HIV given their propensity to not attend classes regularly and subsequent
availability for involvement in risk behaviors. Although continuation high schools provide a
structured environment and specialized skills for youth, CHS youth are exposed to peers
with higher risk behaviors.
Previous research has suggested that mental health problems may play an important
role in the development and maintenance of sexual risk behaviors. Depressive
symptomatology has been significantly associated with an earlier age at sexual onset
(Bachanas et al., 2002; Ethier, Kershaw, Lewis, Milan, Niccolai, & Ickovics, 2006;
Kowaleski-Jones & Mott, 1998; Smith, 1997; Rink, Tricker, & Harvey, 2007) and a history
of infection with a sexually transmitted disease (Mazzaferro, Murray, Ness, Bass, Tyus, &
Cook, 2006; Shrier et al., 2001).

33
The cross-sectional relationship between depressive symptomatology and number of
sexual partners has also been explored, but discrepant findings exist. In a cross-sectional
study of African-American adolescent and young adult females, 31% of the sample exhibited
depressive symptoms (Mazzaferro, Murray, Ness, Bass, Tyus, & Cook, 2006). Females with
high levels of depression were more likely to report multiple sexual partners than non-
depressed females. This relationship was more frequent among younger (14 to 19 years old)
than older (20 to 24 years old) females. Unlike Mazzaferro et al. (2006), Tubman, Windle, &
Windle (1996) and Bachanas et al. (2002) found no correlation between depressive
symptoms and number of sexual partners among their samples of White and African-
American female adolescents, respectively.
The predictive effect of depressive symptomatology on subsequent number of sexual
partners has been investigated with varying results. Among a sample of African-American
and Latina adolescent females, emotional distress consisting of depression, anxiety and
hostility at baseline was predictive of multiple sexual partners by 6-month follow-up (Ethier
et al., 2006). Emotional distress consisted of depression, anxiety and hostility, and therefore,
the effect of depression alone cannot be determined. Unlike Ethier et al. (2006), Hallfors et
al. (2005) did not find a predictive effect of baseline depression on the number of sexual
partners in a national sample of predominantly White middle and high school students.
However, the presence of depressive symptoms increased the odds among substance-using
females. For males, there continued to be no relationship between depression and number of
sexual partners.
The role of depressive symptomatology on the frequency of condom use has also
been explored. In Mazzaferro et al.’s (2006) sample of females, females with high levels of
depression were less likely to report consistent condom use than non-depressed females. In a
national sample of predominantly White adolescents and young adults, depressed females

34
were less likely to use contraceptives (including condoms) than non-depressed females
(Kowaleski-Jones & Mott, 1998). A similar trend towards significance was seen among
males. Using a different national sample of predominantly White youth, Shrier et al. (2001)
found a different relationship between depressive symptomatology and frequency of condom
use by gender. A dose-response relationship between depressive symptomatology and sexual
behaviors was observed where increasing depressive levels were associated with increasing
risk of not using a condom at last sexual intercourse among males. No effect was observed
among females. Bachanas et al. (2002) observed no correlation between depression and
condom use frequency, and Ethier et al. (2006) found that baseline depressive
symptomatology was not predictive of subsequent unprotected sex.
Limited research on the role of depressive symptomatology on health risk behaviors
has been done among youth in continuation high schools, but this research has not focused
on sexual risk behaviors. Depressive symptomatology has been shown to be significantly
correlated with suicidal ideation among females and Latinos whereas it was significantly
correlated with hard substance use among males and White CHS youth (Galaif, Chou,
Sussman, & Dent, 1998). Among this sample of CHS youth, 14% exhibited depressive
symptomatology. Depression has also been shown to be predictive of more perceived stress
but not anger coping, seeking social support, or substance use with a stronger relationship
exhibited among females and males (Galaif, Sussman, Chou, & Wills, 2002). No ethnic
differences were observed.
Individual and interpersonal health behavior theories (e.g., Theory of Planned
Behavior and Social Cognitive Theory, respectively) provide possible mechanisms by which
depressive symptomatology can impact sexual risk-taking. Psychological maladjustments
(e.g., depressive symptomatology) may affect proximal and distal factors (e.g., motivation to
comply, self-efficacy) which can ultimately impact sexual behaviors. Understanding the

35
effects of depressive symptomatology on sexual risk behaviors among an understudied at-
risk sample of youth is important when creating and tailoring effective interventions.
In an effort to further elucidate the literature among youth in continuation high
schools, a study to identify sexual behaviors among young adults who were in continuation
high schools was conducted. The effects of depressive symptomatology during adolescence
and young adulthood on sexual behaviors were of special interest. It is expected that CHS
young adults who reported greater depressive symptomatology at either time period will
have significantly higher numbers of lifetime and recent sexual partners and lower
frequencies of lifetime and recent condom use at four-year follow-up.
Methods
Sample
The present study uses baseline and four-year follow-up data from a sample of 111
young adults originally recruited from fourteen continuation high schools in southern
California. Young adults were in ninth through twelfth grades, participating in a larger
intervention study (Project Towards No Drug Abuse [TND]) of the roles of associative
memory and implicit cognition on substance abuse among at-risk youth. Surveys assessed
demographic factors and included psychosocial and behavioral measures such as substance
use, sensation-seeking, and sexual behaviors.
School selection
Continuation high schools were selected from 29 southern California school districts
that contain continuation high schools. Schools were randomly sampled from the sample of
continuation high schools not receiving Project TND through Project TND3. Within the
schools, half of the classes were randomly selected. Participants were then randomly
assigned to the different assessment conditions.

36
Student recruitment
Baseline
After receiving approval from school administrators, classrooms were selected by
each school. All students listed in the classroom’s roster were invited to participate in the
study. Both parental consent forms and student assent forms were collected for students
under the age of 18 years. Students and parents were informed that the study related to health
behaviors, some of which may be considered sensitive and/or illegal. Consenting students
were randomized on the day of data collection. All materials and procedures received
approval from the Institutional Review Board (IRB) at the University of Southern
California’s Health Sciences Campus (USC HSC).
Four-year follow-up
All materials and procedures from the current study received IRB approval from the
USC HSC in May 2007. A letter signed by the TPRC Principal Investigator was sent to all
eligible subjects notifying them of the telephone-based follow-up study. Subjects were given
the ability to opt-out of being contacted by study staff by returning a postage-paid postcard
by a designated date indicating their decision to not participate. Those subjects who did not
actively refuse to be contacted were eligible for the study.
Student tracking
Baseline
TPRC participants provided contact information and received reminder slips with
the scheduling of future TPRC assessments. The participant’s contact information includes:
(1) participant’s name, (2) current address, (3) phone number, (4) contact person’s name and
phone number (friend or relative who knows how to reach the participant), (5) participant’s
parent(s) name, address, and phone number, (6) participant’s social security number, (7)
participant’s driver’s license number, (8) e-mail address, (9) dates and status of last

37
attempted contact (i.e., contacted, measured, refused), and (10) an extensive notes section
where all attempts to reach participants are recorded.
Four-year follow-up
For the current study, sampling was limited to students who were at least 18 years
old at the time of sampling and self-identified as Latino/a, White, or African-American
(n=747). The sample size was further limited to those individuals who had some reliable
contact information in the TPRC participant database (i.e., phone number or address) and the
name of at least one parent or guardian with whom an online search could be made for
additional information. Given the additional criteria, the final base sample was 391 CHS
youth.
The online search databases of White Pages, Net Detective, and PeopleFinders were
used to update addresses and phone numbers. Searches on each site were made using the
names of the parent(s) or guardian(s) and the student. Reverse searches by address and
phone numbers listed in the TPRC participant database were also conducted in the White
Pages database. If available, phone calls to contacts listed in the TPRC participant database
were made. A final search of the White Pages database was conducted in September 2008 to
update any information.
TPRC participants were considered “lost to follow-up” if the address and phone
number listed in the TPRC participant database were no longer valid, and updated
information was not available for both the parent or guardian and TPRC participant. TPRC
participants were considered “unable to reach” if phone calls were not returned; there was
never any answer; or no phone number was listed publicly (e.g., address listed but phone
number unlisted). A minimum of ten phone call attempts were made to each participant, but
messages were left only at every third call or after a few weeks of leaving a last message.

38
Procedure
Baseline
Self-administered paper-and-pencil surveys were completed in the classrooms at the
respective CHS. Since classes were taught in English, it was assumed that the student had
the ability to read and complete a survey in English. The survey took approximately 30 to 45
minutes to complete. Baseline data on 894 students were collected between February and
November 2004.
Four-year follow-up
Between February and April 2008, letters were sent to all individuals in the final
base sample (n=391) with contact information. Random sampling by gender and race (Latino
and White only) was conducted with ultimate sampling of all TPRC participants. The data
collection team consisted of the PI and five undergraduate students doing directed research
under the aegis of the Chair of the PI’s Doctoral Committee. Each undergraduate research
assistant participated in 5 hours of training prior to contacting TPRC participants. The
training was conducted by the PI and consisted of a 2-hour lecture on the study proposal; 1-
hour training on the study materials and protocol; and 1-hour training of conducting mock
interviews with one another. Once phone calls to TPRC participants started, the PI worked
alongside each research assistant for the first few days for quality assurance purposes.
Once the participant was reached by telephone, an initial screener with verbal
consent was read giving basic information on the purpose of the assessment. The survey was
administered by a trained interviewer once the participant consented verbally to participate.
The survey took approximately 15 minutes to complete. Subjects were given the same
unique identification number so that the first and second waves of data could be linked. To
minimize response biases and prior to asking the sexual behaviors questions, participants
were reminded that their name is not written on the survey, and they can refuse to answer

39
any question. A $15 gift card was mailed to each participant for any out-of-pocket expenses
related to completing a survey. Follow-up data were collected between March and
September 2008. Raw data were entered by the PI into an Access database which was then
imported into SAS.
Measures
Dependent variables. Frequency of condom use during anal and/or vaginal
intercourse and the lifetime and recent (i.e. past 3 months) number of sexual partners were
assessed as separate outcome variables. Information on sexual behaviors was collected at
four-year follow-up.
Frequency of condom use. Two single-item indicator variables measured the
frequency of condom usage during anal and/or vaginal sex over one’s lifetime and within the
past 3 months. These questions were adapted from the National Institute of Mental Health
Multisite HIV Prevention Trial (Fishbein & Coutinho, 1997). The response format was a 5-
point Likert-type scale ranging from “All of the time” (1) to “Never” (5). The item was
reverse-coded so that a higher rating indicated greater frequency in condom use.
Numbers of sexual partners. Lifetime and recent (i.e., past 3 months) numbers of
sexual partners were assessed as separate single-item outcome variables. These items were
adapted from the YRBS and collected during the current study (4-year follow-up). Sexual
partners were defined as any partner with whom the participant engaged in anal, receptive
oral, and/or vaginal intercourse in one’s lifetime and in the past 3 months (i.e., recent). For
participants who have had sex but not in the past 3 months, the number of recent partners
was coded as ‘0.’ The number of sexual partners was treated as a continuous variable.
Independent variable. A short form of the Center for Epidemiologic Studies
Depression (CES-D) scale consisting of five items has been shown to be a reliable measure
of depressive symptomatology among youth in continuation high schools (Cronbach α =

40
0.81; Galaif et al. , 2003). The 5-item short form was used at both baseline and four-year
follow-up. The depressive symptomatology scale consisted of five items on a 4-point scale
assessing the number of days in the last week when one felt depressed, lonely, and sad;
thought that his/her life had been a failure; and felt that s/he “could not shake the blues”
(Andresen, Malmgren, Carter, & Patrick, 1994). The Cronbach alphas were α = 0.84 at
baseline and α = 0.89 at four-year follow-up. The possible range of raw scores is from 0 to
15. The cutpoint used in the 20-item CES-D (i.e., cutpoint of 16) was adjusted by
multiplying it by n/20 where n is the number of items used in the short form (Shrout &
Yager, 1989). Therefore, cumulative scores of 4 or greater indicate depressive
symptomatology. Depressive symptomatology was treated as a dichotomous variable and
assessed as: baseline only; follow-up only; depressive at either time point; and depressed at
both baseline and four-year follow-up.
Covariates. Several variables were assessed for descriptive purposes and tests of
association.
Age at sexual onset. Age at sexual onset was dichotomized at the mean age of 15
years where sexual initiation by age 15 years was coded as ‘1.’
Age differential of first sexual partner. Questions on if the first sexual partner was
the same age, older, or younger and by how many years were asked in order to determine the
age differential at sexual onset. A dichotomous composite variable (Cronbach α = 0.78) was
created where an older partner (“Older, 3 or more years”) was coded as ‘1’ and a partner that
is younger, same age, or no more than 2 years older was treated as the referent category
(coded as ‘0’).
Sexual maturation. The Adolescence Scale-Imperial College School of Medicine
(AS-ICSM), a brief retrospective assessment of pubertal milestones, was used to assess
sexual maturation (Kaiser & Gruzelier, 1999). The AS-ICSM consists of one common item

41
(growth spurt); one female-specific item (menarche); and three male-specific items (voice
break, regular shaving, and first nocturnal emission). The AS-ICSM was normed among
young adults (mean age = 20.7 years) and shown to be a reliable retrospective assessment for
both females (Cronbach α = 0.87) and males (Cronbach α = 0.83; Kaiser & Gruzelier, 1999).
In the current study, the mean age at sexual maturation for females was computed
using both growth spurt and menarche (Cronbach α = 0.31). Almost one-fourth of males
(23%) either did not remember or had not experienced a nocturnal emission, and thus, this
item was omitted. For males, the mean age at sexual maturation was computed using ages at
growth spurt, voice break, and regular shaving (Cronbach α = 0.61).
Gender and race/ethnicity. Information on self-reported gender and race/ethnicity
were collected during the TPRC baseline assessment. Gender and race/ethnicity were
dichotomously coded. Females were coded as ‘0,’ and males were coded as ‘1.’ Latinos,
coded with a value of ‘1,’ were compared to non-Latinos (coded as ‘0’).
Age. Using birth date and follow-up interview date, age in years at the time of the
four-year follow-up assessment was computed.
Sexual behavior orientation. Sexual behavior orientation was defined using four
variables: self-identified sexual orientation; subject’s gender; and reported gender(s) of
lifetime and recent sexual partners. The three sexual behavior orientation categories were
men who had sex with both men and women (‘1’), women who had sex with men and/or
women (‘2’), and men who had sex with women (‘3’). If a male self-identified as
“heterosexual” but engaged in sex with both males and females, then he was coded as
behaviorally having had sex with both men and women (‘1’).
Acculturation. Acculturation is the process by which the values, norms, attitudes,
and behaviors of an individual of a minority culture are fused with those from the majority
culture and incorporated into one’s existence (Marín, Sabogal, Marín, Otero-Sabogal, &

42
Perez, Stable, 1987). Marín et al. (1987) created an acculturation scale that focused on the
extent of which languages were spoken (i.e., English, Spanish) in particular settings.
Although the Marín et al. (1987) linguistic acculturation scale was normed with Latino
adults (Cronbach α = 0.90), its subsequent use among Latino youth has yielded high
reliability (e.g., Cronbach α = 0.90 in Ford & Norris, 1993).
Both TPRC and the current study used the Marín et al. (1987) short acculturation
scale (Cronbach α = 0.88, both periods). The four items assessing linguistic acculturation
were: language(s) most often read and spoken; language(s) spoken at home; language(s)
spoken with friends; and language(s) to which movies, television, and radio shows are
watched and listened. The response categories were: (1) Only English; (2) English more than
another language; (3) English and another language equally; (4) Another language more than
English; and (5) Only another language (not English). The final acculturation item was the
mean value of the four items with a possible range of 1 (“Only English”) to 5 (“Only another
language [not English]”).
Data analysis
Descriptive statistics were computed for all demographic variables. Chi-square tests
were used to test for associations between depressive symptomatology and gender, ethnicity,
and sexual onset characteristics. Independent t-tests were used to test associations in
frequency of lifetime condom use during vaginal and/or anal intercourse and the number of
lifetime sexual partners and gender, ethnicity, sexual onset characteristics, and depressive
symptomatology. Nonparametric Wilcoxon signed rank tests were used to test associations
in frequency of condom use in the past 3 months during vaginal and/or anal intercourse and
the number of sexual partners in the past 3 months and gender, ethnicity, sexual onset
characteristics, and depressive symptomatology. The criterion for statistical significance was
set to a 0.05 level. All analyses were conducted using SAS version 9.1.

43
Results
Demographic characteristics of the study sample are displayed in Table 5. The
sample was mostly Latino (56.8%) and heterosexual (96.4%) and almost equally represented
by gender (female, 51%). The mean age at four-year follow-up was 20.6 years. The majority
of the sample had already had consensual sex (98.2%) with 58.9% initiating sex by 15 years
of age. Larger proportions of females and non-Latinos had an early sexual onset. At first
sexual experience, 15.6% of the sample had an older (3+ years) sexual partner. Overall,
participants had a median number of six lifetime sexual partners and one sexual partner in
the past three months. Forty percent of the baseline sample and thirty percent of the four-
year follow-up sample exhibited depressive symptomatology. The majority (57.7%)
exhibited depressive symptoms at either time period, and 14.4% exhibited depressive
symptomatology at both time periods.
Percentages of participants with depressive symptomatology at different levels of
the demographic variables are summarized in Table 6. Regardless of time point, females
exhibited more depressive symptoms than males. Close to two-thirds of females compared to
20% of males exhibited depressive symptomatology at baseline (χ
2
= 19.5, p < 0.01). At
four-year follow-up, over one-third of females compared to 22% of males exhibited
depressive symptomatology (χ
2
= 3.27, p < 0.10). The majority (75.5%) of females exhibited
depressive symptoms at either time period, and over one-fourth exhibited symptoms at both
time periods.
Compared to persons who had delayed sexual initiation, persons who initiated sex
by 15 years of age exhibited significantly higher depressive symptomatology at follow-up
(38.1% v. 18.2%; χ
2
= 4.9, p < 0.05), and among them, depressive symptomatology at both

44
Table 5. Characteristics of the sample
Variable Overall
N=111
Female
N=56 (50.5%)
Male
N=55 (49.5%)
Latino
N=63 (56.8%)
Non-Latino
N=48 (43.2%)

N (%) N (%) N (%) N (%) N (%)
Age at sexual onset, ≤15 years
(n=107)
63 (58.9) 34 (64.2) 29 (53.7) 28 (47.5) 35 (72.9)
Older (3+ years) sexual partner at
sexual onset (n=109)
17 (15.6) 14 (25.5) 3 (5.6) 9 (14.8) 8 (16.7)
Sexual orientation
Men who have sex with men 4 (3.6) --- --- 4 (7.3) 3 (4.8) 1 (2.1)
Women who have sex with men 56 (50.5) 56 (100.0) --- --- 33 (52.4) 23 (47.9)
Men who have sex with women 51 (45.9) --- --- 51 (92.7) 27 (42.9) 24 (50.0)
Ever had sex
Lifetime 109 (98.2) 55 (98.2) 54 (98.2) 61 (96.8) 48 (100.0)
Past 3 months 103 (92.8) 52 (92.9) 51 (92.7) 51 (81.0) 43 (89.6)
Presence of depressive
symptomatology

Baseline (n=104) 43 (41.4) 33 (62.3) 10 (19.6) 24 (42.1) 19 (40.4)
Follow-up (n=111) 33 (29.7) 21 (37.5) 12 (21.8) 17 (27.0) 16 (33.3)
Either time period (n=104) 60 (57.7) 40 (75.5) 20 (39.2) 34 (59.7) 26 (55.3)
Both time periods (n=104) 15 (14.4) 14 (26.4) 1 (2.0) 6 (10.5) 9 (19.2)

Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Range
Age (years)

20.6 (1.2) 20.3 (1.3) 20.9 (1.1) 20.6 (1.2) 20.5 (1.3) 18-23
Acculturation (score) 1.8 (0.9) 1.9 (0.9) 1.8 (0.8) 2.4 (0.7) 1.0 (0.1) 1-5
Sexual maturation (years) 13.5 (1.1) 13.1 (0.9) 14.0 (1.1) 13.5 (1.1) 13.5 (1.1) 11-18

Median (Q1,Q3) Median (Q1,Q3) Median (Q1,Q3) Median (Q1,Q3) Median (Q1,Q3) Range
Number of sexual partners
(n=109)

Lifetime 6 (4, 10) 5 (2, 8) 6 (5, 10) 5 (3, 7) 7 (5, 10) 1-25
Past 3 months 1 (1, 2) 1 (1, 1) 1 (1, 2) 1 (1, 1) 1 (1, 2) 1-5
Frequency of condom use
Lifetime (n=109) 4 (3, 4) 3 (3, 4) 4 (3, 4) 3 (3, 4) 4 (3, 4) 1-5
Past 3 months (n=103) 2 (1, 3) 2 (1, 3) 2 (1, 4) 2 (1, 3) 2 (1, 3) 1-5

45

Table 6. Associations with depressive symptomatology
Presence of depressive symptomatology
Baseline
(n=104)
Follow-up
(n=111)
Either time period
(n=104)
Both time periods
(n=104)
n % χ
2
n % χ
2
n % χ
2
n % χ
2

Gender
Male 51 19.6 19.5
**
55 21.8 3.27
†
51 39.2 14.0
**
51 2.0 12.6
**

Female 53 62.3 56 37.5 53 75.5 53 26.4

Ethnicity
Latino 57 42.1 0.03 63 27.0 0.53 57 59.7 0.20 57 10.5 1.55
Non-Latino 47 40.4 48 33.3 47 55.3 47 19.2

Age at Sexual Onset
a

≤15 years 60 36.7 0.69 63 38.1 4.90
*
60 55.0 0.06 60 20.0 2.94
†

16+ years 40 45.0 44 18.2 40 57.5 40 7.5

Age Differential with
First Sexual Partner
a

3+ years 17 47.1 0.40 17 47.1 2.69 17 70.6 1.57 17 23.5 1.27
<3 years 85 38.8 92 27.2 85 54.1 85 12.9
†
p < 0.10,
*
p < 0.05,
**
p < 0.01
Note: some frequencies do not sum to n due to missing values
a
Dichotomized at median

46
time points demonstrated a trend-level effect (20% v. 7.5%; χ
2
= 2.94, p < 0.10). There was
no association between age at sexual onset and depressive symptomatology at baseline only
or at either time period. Ethnicity and age differential with first sexual partner were not
associated with depressive symptomatology at point of assessment.
Associations between different levels of the demographic variables and number of
sexual partners and condom usage were tested. As evidenced in Table 7, number of lifetime
partners was associated with select demographics. Compared to their counterparts,
significantly more lifetime sexual partners were reported by males (M = 8.06 v. 5.55; t =
-2.87, p < 0.01); non-Latinos (M = 8.38 v. 5.49; t = 3.33, p < 0.01); and early sexual
initiators (M = 8.44 v. 4.48; t = -4.91, p < 0.01). Males and early sexual initiators also
reported a higher median number of recent (i.e., past 3 months) sexual partners than females
and late sexual initiators, respectively. The median number of recent sexual partners was not
associated with either ethnicity or age differential with first sexual partner. Frequency of
lifetime and recent condom use during anal and/or vaginal intercourse was not associated
with any of the demographic variables.
Associations between number of sexual partners and condom usage and depressive
symptomatology at various time periods were tested and are displayed in Table 8. Persons
who exhibited depressive symptomatology at baseline reported a fewer number of lifetime
sexual partners at four-year follow-up than persons who did not exhibit depressive
symptoms at baseline (M = 5.55 v. 7.47; t = 2.10, p < 0.05). A different relationship at four-
year follow-up is observed. Persons who exhibited depressive symptomatology at follow-up
reported a higher number of lifetime sexual partners than persons who did not exhibit
depressive symptoms at baseline (M = 8.58 v. 5.99; t = -2.72, p < 0.01). A trend-level
association is observed between depressive symptomatology at follow-up and the median
number of recent sexual partners where persons with depressive symptoms reported a higher

47
Table 7. Associations with sexual behaviors

Lifetime number of
sexual partners

Number of sexual partners,
past 3 months

Frequency of condom use
with vaginal/anal intercourse,
lifetime
a

Frequency of condom use with
vaginal/anal intercourse,
past 3 months
a

n
Mean
number
t n
Median
number
(Q1, Q3)
b

z

n
Mean
score
t

n
Median
Score
(Q1, Q3)
b

z
All 108 6.78 --- 108 1 (1, 2) --- 108 3.33 --- 102 2 (1, 3) ---

Gender
Male 53 8.06 -2.87
**
53 1 (1, 2) 2.54
*
53 3.34 -0.07 50 2 (1, 3) 0.89
Female 55 5.55 55 1 (1, 1) 55 3.33 52 2 (1, 3)

Ethnicity
Latino 60 5.49 3.33
**
60 1 (1, 1.1) 1.40 60 3.25 1.08 57 2 (1, 3) -0.10
Non-Latino 48 8.38 48 1 (1, 2) 48 3.44 45 2 (1, 3)

Age at Sexual
Onset
c

≤15 years 62 8.44 -4.91
**
62 1 (1, 2) -2.77
**
62 3.24 1.08 59 2 (1, 3) 0.28
16+ years 44 4.48 44 1 (1, 1) 44 3.43 42 2 (1, 4)

Age Differential
with First Sexual
Partner
c

3+ years 17 9.53 -2.04
†
17 1 (1, 2) 1.71 17 3.06 1.38 17 2 (1, 4) -0.15
<3 years 91 6.26 91 1 (1, 2) 91 3.38 85 2 (1, 3)
†
p < 0.10,
*
p < 0.05,
**
p < 0.01
Note: some frequencies do not sum to n due to missing values
a
Range: 1 (never) to 5 (always);
b
Q1, Q3 refers to the first and third quintiles
c
Dichotomized at median

48

Table 8. Associations between depressive symptomatology and sexual behaviors

Lifetime number of
sexual partners

Number of sexual partners,
past 3 months
Frequency of condom
use with vaginal/anal
intercourse, lifetime
a

Frequency of condom use with
vaginal/anal intercourse,
past 3 months
a

n
Mean
number
t n
Median
number
(Q1, Q3)
b

z

n
Mean
score
t

n
Median
Score
(Q1, Q3)
b

z
Baseline only
c

Yes 41 5.55 2.10
**
41 1 (1, 2) -0.98 41 3.34 0.23 38 1 (1, 4) -0.79
No 60 7.47 60 1 (1, 2) 60 3.38 57 2 (1, 3)

Follow-up only
Yes 33 8.58 -2.72
**
33 1 (1, 2) 1.86
†
33 3.39 -0.46 31 3 (1, 4) 1.29
No 75 5.99 75 1 (1, 1.1) 75 3.31 71 2 (1, 3)

Either time point
Yes 58 6.70 -0.01 58 1 (1, 2) 0.10 58 3.34 0.28 53 2 (1, 4) -0.14
No 43 6.69 43 1 (1, 2) 43 3.40 42 2 (1, 3)

Both time points
Yes 15 7.48 -0.71 15 1 (1, 2) -0.16 15 3.40 -0.16 15 2 (1, 4) 0.23
No 86 6.56 86 1 (1, 2) 86 3.36 80 2 (1, 3)
†
p < 0.10,
*
p < 0.05,
**
p < 0.01
Note: some frequencies do not sum to n due to missing values
a
Range: 1 (never) to 5 (always)
b
Q1, Q3 refers to the first and third quintiles
c
One-sided

49
median number of partners. Frequency of lifetime and recent condom use during anal and/or
vaginal intercourse was not associated with depressive symptomatology.
In bivariate analyses, gender was significantly associated with both depressive
symptomatology and number of sexual partners. The associations between depressive
symptomatology and number of sexual partners stratified by gender were tested. As evident
in Table 9, distinct gender differences exist. Males who exhibited depressive symptoms at
follow-up (M = 11.9 v. 6.93; t = -3.67, p < 0.01) or at either time point (M = 9.83 v. 6.90; t =
-2.27, p < 0.05) reported a significantly higher number of lifetime sexual partners as non-
depressed males. However, females who exhibited depressive symptoms at follow-up
(Median = 1 [1, 2] v. 1 [1, 1]; z = 2.24, p < 0.05) or at either time point (Median = 1 [1, 2] v.
1 [1, 1]; z = 1.80, p < 0.10) reported a higher number of sexual partners in the past 3 months
as non-depressed females.
Discussion
Previous research conducted with mostly females and among adolescent populations
in diverse settings has resulted in discrepant findings on the relationships between depressive
symptomatology and the number of sexual partners and frequency of condom use during
vaginal and/or anal intercourse. Research on the strength of these relationships among
Latinos and youth in continuation high schools has been limited.
Depressive symptomatology has been significantly associated with an earlier age at
sexual onset (Bachanas et al., 2002; Ethier et al., 2006; Kowaleski-Jones & Mott, 1998;
Smith, 1997; Rink, Tricker, & Harvey, 2007). Consistent with previous literature and
extending to the literature on CHS youth, persons who had initiated sex by 15 years of age
exhibited greater depressive symptoms at both time periods and at four-year follow-up. The
depressive symptomatology scale assesses depressive symptoms in the past seven days and
is thus a measure of acute depressive symptoms. Exhibiting depressive symptoms at two

50
Table 9. Associations between depressive symptomatology and number of sexual partners by gender
Lifetime number of
sexual partners
Number of sexual partners,
past 3 months

Females Males Females Males

n
Mean
number
t n
Mean
number
t

n
Median
number
(Q1, Q3)
a

z

n
Median
Score
(Q1, Q3)
a

z
Baseline only
b

Yes 32 4.96 1.00 9 7.67 0.27 32 1 (1, 2) 0.98 9 2 (1, 2) 0.92
No 20 6.17 40 8.12 20 1 (1, 1) 40 2 (1, 3)

Follow-up only
Yes 21 6.67 -1.48 12 11.90 -3.67
**
21 1 (1, 2) 2.24
*
12 2 (1, 3) 1.51
No 34 4.85 41 6.93 34 1 (1, 1) 41 1 (1, 2)

Either time point
Yes 39 5.17 0.74 19 9.83 -2.27
*
58 1 (1, 2) 0.10 53 2 (1, 4) -0.14
No 13 6.19 30 6.90 43 1 (1, 2) 42 2 (1, 3)

Both time points
Yes 14 6.94 -1.57 1 15 -1.56 14 1 (1, 2) 1.80
†
1 2 (2, 2) 0.69
No 38 4.86 48 7.90 38 1 (1, 1) 48 1 (1, 2)
†
p < 0.10,
*
p < 0.05,
**
p < 0.01
a
Q1, Q3 refers to the first and third quintiles
b
One-sided

51
different time points may be indicative of maladjustment beyond an acute phase that may be
partially manifesting through some sexual risk-taking.
The bivariate relationship between depressive symptomatology and an early age at
sexual onset should be interpreted with caution. One may argue that recalling age at sexual
initiation may influence current depressive symptoms. However, depressive
symptomatology was assessed prior to sexual behaviors. The relationship may solely reflect
the distal factor (i.e., early age at sexual initiation) and the current outcome (i.e., depressive
symptomatology) without accounting for other proximal and distal factors between the two
variables.
Depressive symptomatology at baseline was not associated with age at sexual onset.
Given the literature on the relationship between the presence of depressive symptoms and an
early sexual initiation, a significant association between the two items was expected. If an
earlier age at sexual onset occurred, the behavior may have been either explicitly or
implicitly socially accepted by one’s peers. The majority of the sample reported having had
sex by 15 years of age. In an environment with a high proportion of risk behaviors and a
high proportion of sexually active peers as well as a time period where peer influence is key,
initiating sex at an earlier age may have been condoned and not manifested through
depressive symptoms.
It was hypothesized that the presence of depressive symptomatology at either time
period would be associated with high-risk sexual behaviors. In the literature, depressive
symptoms have been shown to be predictive of subsequent multiple sexual partners among
ethnic minority females (Ethier et al., 2006) whereas another study have found no predictive
effects (Hallfors et al., 2005). Contrary to both studies, the presence of depressive
symptomatology at baseline among CHS youth was associated with fewer lifetime sexual
partners reported at four-year follow-up. A different relationship emerged when assessing

52
the cross-sectional relationship between depressive symptoms and number of sexual
partners. Depressive individuals at follow-up reported higher numbers of lifetime and recent
sexual partners.
The predictive nature of depressive symptomatology on the lifetime number of
sexual partners appears to be more reflective of gender differences rather than a true
protective effect. Once gender is controlled, the significant relationship between baseline
depressive symptoms and lifetime number of sexual partners disappears. However, other
gender differences emerged in stratified analyses. Males who were depressive at either time
point or at follow-up only reported significantly higher numbers of lifetime sexual partners.
Females who were depressive at both time points or at follow-up only reported significantly
higher numbers of recent sexual partners.
The majority of extant research on depressive symptomatology has focused mostly
on females. One possible explanation is that females tend to internalize symptom expression
(e.g., depression) whereas males tend to externalize symptom expression (e.g., substance
use; Galaif et al., 1998). The current study provides evidence that males who were in
continuation high schools may also internalize symptoms. The majority of persons
exhibiting depressive symptomatology were female. In fact regardless of time frame,
females not only had fewer sexual partners than males, but they also exhibited greater
depressive symptomatology. However, close to 40% of males exhibited depressive
symptoms at either time point with close to one-quarter exhibiting symptoms at follow-up.
Findings from the current study provide support for the role that depressive symptomatology
may have not only on CHS youth but also on males, an understudied population within this
area of research.
In support of Bachanas et al. (2002) and Ethier el a. (2006), depressive
symptomatology was not associated with frequency of condom use. Inconsistent condom

53
usage during anal and/or vaginal intercourse was common regardless of depressive
symptomatology and demographic and sexual debut characteristics. Although recent condom
use was more sporadic than lifetime condom use, the mean number of recent sexual partners
may be indicative of serial monogamy. For some, practicing monogamy was the safer sex
alternative to consistent condom use.
Despite the study findings, several limitations need to be considered. The ability to
find true significant predictive relationships and associations is limited by the sample size.
The small sample size does not provide sufficient power to detect significant findings.
Attrition was high with only 15% of the baseline having completed a four-year follow-up
survey. Although efforts were taken to contact a larger sample size, tracking participants
proved to be difficult, many of whom were lost to follow-up. However, the final sample size
is representative of a CHS sample that is about 50% Latino and 50% female. Given the high
attrition, baseline participants who were not interviewed in the most recent follow-up (85%
of the sample) may have sexual behaviors that differ from the current sample thereby
limiting the generalizability of the findings.
Validity of the data could also be affected by the self-report nature of the study,
especially with sensitive questions on sexual behaviors. Participants were assured of the
confidentiality of all data, and measures were taken prior to commencing and during the
survey to display this confidentiality to all participants. Furthermore, measures that were
used had been validated previously in adolescent samples. Therefore, there is no reason to
believe that participants were differentially dishonest in their answers.
Despite these limitations, results from this study not only support the important role
that depressive symptomatology plays on the number of sexual partners but also indicates
the potential role that other emotions or states can play on sexual risk-taking. Future research
is needed to understand the roles that emotions or states of being play on sexual risk-taking

54
among youth who are in continuation high schools, especially those that may trigger an
individual to engage in sexual behaviors that place one at risk for HIV or other sexually
transmitted infections. Identifying emotions or states of being that place a CHS youth at
sexual risk bear potential implications for the prevention of sexual risk-taking. Interventions
aimed at youth in continuation high schools could benefit by incorporating comprehensive
sexual education, exercises on identifying sexual triggers, and development of key skills
such as healthier coping strategies and problem-solving into prevention curriculum.
Although it seems intuitive that removal of the trigger to sexual risk-taking would reduce
sexual risk-taking, future research is needed to determine whether these types of
interventions can actually reduce sexual risk-taking.

55
CHAPTER FOUR: SOCIOCOGNITIVE CORRELATES OF CONDOM USE
AMONG YOUNG ADULTS FROM CONTINUATION HIGH SCHOOLS

In 2006 one-third of the new HIV infections in the United States were among youth
and adults between the ages of 13 to 29 years, the age group with the largest number of new
infections (CDC, 2007). Additionally, the age group of 20 to 29 years was one of the age
groups with an increase in the number of new cases with an AIDS diagnosis (CDC, 2007).
Given the median time of ten years between HIV infection and AIDS diagnosis, many young
adults and adolescents were most likely infected during their adolescent years (Bacchetti &
Moss, 1989). These data provide support that adolescents and young adults continue to be an
at-risk age group for HIV.
Youth in continuation high schools (CHS) present as a particularly at-risk group for
HIV. According to national epidemiologic data, the majority of CHS youth (87.8%) had
already had sexual intercourse with more than two-thirds (68.5%) currently sexually active
(CDC, 1999). In comparison, half of youth in regular high schools around the same time
period had sex, and about one-third (36.3%) were currently sexually active (CDC, 2000).
Less than half (45.9%) of CHS youth reported condom use at last sexual intercourse, and
close to one-third (29.5%) had either been pregnant or gotten someone pregnant, a marker of
unprotected sex. Contrastingly, the majority (58%) of youth in regular high schools had used
a condom at last sexual intercourse and only 6.3% had either been pregnant or gotten
someone pregnant.
CHS youth may not only be in an environment with high sexual risk-taking
behaviors but may also perceive a greater support for those behaviors. A primary HIV
prevention strategy among sexually active individuals is to use condoms correctly and
consistently during sexual intercourse. In extant literature, several sociocognitive correlates

56
of condom use among adolescents have been identified and include communication on
sexual safety with a parent, perceived susceptibility to HIV, peer norms for safer sex, and
condom use self-efficacy skills.
Parent-adolescent communication
Parents exert a positive influence on adolescent safer sexual behaviors by
encouraging less favorable attitudes towards sexual risk-taking and less socializing with
risky peer groups (Brown, Mounts, Lamborn, & Steinberg, 1993; Patterson, DeBaryshe, &
Ramsey, 1989). Parent-adolescent communication on sexual behaviors has been effective in
delaying age at sexual initiation among adolescents (Fasula and Miller, 2006; Karofsky,
Zeng, & Kosorok, 2000; Resnick, Bearman, Blum, Bauman, Harris, Jones, et al., 1997).
Although DiClemente et al. (1996) found no correlation, other studies have
demonstrated that parent-adolescent communication on sexuality has a protective effect on
adolescent sexual risk-taking, but findings have varied. In a cross-sectional study of African-
American adolescents, parent-adolescent communication was correlated with condom use
regardless of the parent’s or adolescent’s gender (Romer, Stanton, Galbraith, Feigelman,
Black, and Li, 1999). Adolescents who reported more frequent communication on sexual
behaviors with either or both parents used condoms more consistently.
Mother-daughter communication is also strongly correlated with frequency of
condom use. Higher levels of mother-daughter sexual risk communication on sex, birth
control, HIV/sexually transmitted diseases (STDs), and condoms were associated with fewer
episodes of recent unprotected sex by African-American and Latina female adolescents and
young adults (Hutchinson et al., 2003). This relationship between gender-matched dyads
was also partially mediated by condom use self-efficacy where greater communication was
associated with a young female’s increased confidence in her ability to use condoms.

57
Using national adolescent health data, Henrich, Brookmeyer, Shrier, & Shahar
(2006) found a predictive effect between communication and subsequent sexual risk-taking
with communication focused on the mother-adolescent dyad. Greater communication on sex,
birth control, the negative effects of pregnancy, and sexually transmitted diseases was
predictive of lower sexual risk-taking by female youth one year later. Communication had
no impact on male sexual behaviors. Sexual risk-taking was a composite variable of five
sexual risk behaviors that included unprotected sex. The impact on the frequency of condom
use alone is unknown.
Perceived susceptibility to HIV
HIV awareness and knowledge are high among adolescents (Facente, 2001; Harwell
et al., 1999; Liverpool, McGhee, Lollis, Beckford, & Levine, 2002; Rew, Fouladi, &
Yockey, 2002; Strunin, 1991) yet national and local trends of increasing HIV incidence and
prevalence provide evidence of a discrepancy between HIV knowledge and perceived
susceptibility to HIV. As a component of the Health Belief Model, perceived susceptibility
of HIV measures an individual’s subjective perception that s/he is at risk of contracting HIV
(Janz & Becker, 1984).
Perceiving that one is susceptible to HIV can influence a change in sexual habits
such as opting for abstinence and reducing the number of sexual partners (Goodman &
Cohall, 1989; Hingson, Strunin, Berlin, & Heerin, 1990; Langer & Tubman, 1997). Many
adolescents engage in sexual HIV-risk behaviors such as having multiple sexual partners and
using substances prior to sexual intercourse yet they do not perceive themselves at risk for
HIV, highlighting a disconnect between the two components (Ellen, Boyer, Tschann, &
Shafer, 1996; Facente, 2001; Harwell et al., 1999; Morrison-Beedy, Carey, & Aronowitz,
2002).

58
One would expect that a greater perceived susceptibility to HIV would be associated
with more consistent condom use. In cross-sectional studies with ethnically diverse samples,
perception of HIV risk was not significantly associated with condom use despite large
proportions of HIV risk behaviors. In Ellen et al. (1996), two-thirds of sexually experienced
teenagers from urban high schools reported a multiple number of lifetime sexual partners.
Less than half (43%) reported consistent condom usage all the time, and only one-third
perceived themselves at risk. In Morrison-Beedy, Carey, & Aronowitz (2002), low
perception of HIV vulnerability and inconsistent condom use were evident among female
adolescent and young adults attending a family planning clinic despite reported alcohol
(42%) and/or substance use (23%) prior to sex or having multiple sexual partners (17%).
In some studies, perception of greater HIV risk was actually associated with higher
proportions of unprotected sexual activity in both females and males (Millstein & Moscicki,
1995; Murphy, Rotheram-Borus, and Reid, 1998). Due to the temporal ambiguity nature of
cross-sectional students, a different interpretation may be that those engaging in sexual risk-
taking recognized their risk and reported themselves as being at greater risk for HIV.
Using a prospective study design, perceived susceptibility to HIV has been shown to
have no predictive effect on subsequent condom use. No relationship between perception of
risk and consistent condom use existed among African-American female adolescents and
young adults living in publicly funded housing developments, a group that has been shown
to be at increased risk for HIV (DiClemente et al., 1996; Stanton et al., 1996).
Peer norms on safer sex
Peer norms are important contributors to shaping behaviors in social environments
and have been associated with an earlier age at sexual onset (Bachanas et al., 2002; Brown,
Ladin L’Engle, Pardun, Guo, Kenneavy, & Jackson, 2006; Carvajal, Parcel, Basen-Engquist,
Banspach, Coyle, Kirby, et al., 1999; DiIorio, Dudley, Kelly, Soez, Mbwara, & Sharpe

59
Potter, 2001; Kinsman et al., 1998; Marín, Coyle, Gómez, Carvajal, & Kirby, 2000;
O’Donnell, Myint-U, O’Donnell, & Stueve, 2003; Romer, Black, Ricardo, Feigelman,
Kalijee, Galbraith, et al., 1994; Santelli et al., 2004; Stanton, Li, Black, Ricardo, Galbraith,
Feigelman, et al., 1996; Walter et al., 1995). For males especially, youth are more likely to
initiate sexual intercourse at an earlier age if they perceive high peer sexual activity and high
peer support for sexuality. Similarly youth are more likely to delay sexual initiation and
remain abstinent if there is a perception of high peer disapproval towards sexual intercourse,
a relationship that has been commonly seen among females.
Peer norms for condom use is an integral component of the Theory of Reasoned
Action/Theory of Planned Behavior framework and measures whether or not important
referent peers approve of condom use either explicitly or implicitly (Fishbein, 1990).
Research provides support for a significant relationship between the two variables where
persons who perceived peer norms as supporting condom use were more likely to report
consistent condom usage.
Discrepant findings by gender exist. Among a sample of predominantly African-
American and Latino adolescents and young adults from three urban cities, peer social
norms and protected sexual activity were not only higher among males than females but also
positively correlated with one another (Rotheram-Borus, Marelich, & Srinivasan (1999).
When additional sexual behaviors (e.g., number of sexual partners) were combined with
frequency of condom use, the relationship between peer norms and sexual risk-taking was
evident among females only (Murphy, Rotheram-Borus, and Reid, 1998). Females reported
significantly more unprotected sex acts than males, but sexual risk-taking was significantly
lower among females who perceived peer norms as supporting safer sex.
In studies that reported results on females only, perception of peer norms that
support condom use and protected sexual behaviors have been predictive of condom use.

60
Stanton et al. (1996) and DiClemente et al. (2006) conducted separate prospective studies on
African-American adolescents and young adults living in public housing. Persons who
perceived peer norms as supporting condom use at baseline were more likely to report
consistent condom use at six-month follow-up. The protective effect of peer norms on
protected sexual activity was evident among youth between the ages of 9 and 15 years
(Stanton et al., 1996) and adolescents and young women between the ages of 12 and 21
years (DiClemente et al., 1996). A similar relationship existed in a sample of mostly Latina
adolescents, ages 15 to 18 years old, in a school-based young parents program (Koniak-
Griffin and Stein, 2006). Unlike Stanton et al. (1996) and DiClemente et al. (2006), Sikkema
et al. (2004) found that peer social norms were not significantly correlated with sexual risk
behaviors among African-American and Asian adolescents (ages 12 to 17 years) living in
low-income housing developments.
Condom use self-efficacy
As a concept of Social Learning Theory, condom use self-efficacy measures an
individual’s perception and confidence in one’s ability to perform safer sex by using
condoms (Bandura, 1994). Condom use self-efficacy can measure: an individual’s ability of
putting a condom on self or other (“mechanics”); partner’s approval of condom use; one’s
ability to persuade a partner to use a condom (“negotiation”); and one’s ability to use
condoms while under the influence (Brafford & Beck, 1991).
Greater condom use self-efficacy is expected to ultimately be associated with greater
consistency in condom use and decreases in the number of unprotected sexual activities.
However, results on these relationships vary. DiClemente et al. (1996), Wingood and
DiClemente (1998), and Koniak-Griffin and Stein (2006) found that female adolescents and
young adults who were confident in their ability to negotiate and demand condom use were
more likely to report consistent condom use.

61
Findings vary when self-efficacy is defined as condom mechanics or confidence in
one’s ability to put a condom on self or other correctly. In a cross-sectional study of
predominantly young African-American adolescents, persons with greater confidence in
their ability to put on a condom correctly were twice as likely report consistent condom use
(DiIorio et al., 2001). Among older White adolescents, a higher proportion of female
participants reported recent unprotected sexual activity despite greater condom mechanics
self-efficacy and high-risk sexual behaviors (e.g., multiple sexual partners; Morrison-Beedy,
Carey, & Aronowitz, 2002). In Rotheram-Borus, Marelich, & Srinivasan (1999), confidence
in one’s ability to put on a condom was positively correlated with protected sexual activity
among males only.
Condom use self-efficacy has also been shown to be an insignificant correlate and
factor of condom use consistency. Although females had greater self-efficacy than males,
there was no relationship between self-efficacy and sexual risk-taking in a sample of
adolescents from metropolitan areas (Murphy, Rotheram-Borus, and Reid, 1998). Condom
use self-efficacy was also not a predictive factor of subsequent protected sexual activity
among youth between the ages of 9 and 15 years (Stanton et al., 1996) and adolescents
between the ages of 12 and 17 years (Sikkema et al., 2004) from public housing.
Current study
A tremendous amount of research on these sociocognitive correlates has been
conducted among a variety of adolescent samples (e.g., clinic-based) and utilizing different
methodologies. However, discrepant findings by gender and ethnicity exist. Furthermore, a
review of these correlates on the frequency of condom use among CHS youth is limited.
Understanding the correlates that influence consistent condom use by adolescents is
important for developing effective behavioral interventions aimed at this group. In an effort
to further elucidate the literature among youth in continuation high schools, a study to

62
identify sexual behaviors among young adults who were in continuation high schools was
conducted. The roles of these sociocognitive correlates on the frequency of condom use were
of special interest. It is expected that parent-adolescent communication on sexual safety,
perceived susceptibility to HIV, peer norms on safer sex, and condom use self-efficacy will
be positively associated with frequencies of lifetime and recent condom use among CHS
young adults. It is further expected that each relationship will be moderated by gender and
race/ethnicity whereby positive associations will be evident for males and non-Latinos.
Methods
Sample
The present study uses baseline and four-year follow-up data from a sample of 111
young adults originally recruited from fourteen continuation high schools in southern
California. Young adults were in ninth through twelfth grades, participating in a larger
intervention study (Project Towards No Drug Abuse [TND]) of the roles of associative
memory and implicit cognition on substance abuse among at-risk youth. Surveys assessed
demographic factors and included psychosocial and behavioral measures such as substance
use, sensation-seeking, and sexual behaviors.
School selection
Continuation high schools were selected from 29 southern California school districts
that contain continuation high schools. Schools were randomly sampled from the sample of
continuation high schools not receiving Project TND through Project TND3. Within the
schools, half of the classes were randomly selected. Participants were then randomly
assigned to the different assessment conditions.

63
Student recruitment
Baseline
After receiving approval from school administrators, invited classrooms were
selected by each school. All students listed in the classroom’s roster were invited to
participate in the study. Both parental consent forms and student assent forms were collected
for students under the age of 18 years. Students and parents were informed that the study
related to health behaviors, some of which may be considered sensitive and/or illegal.
Consenting students were randomized on the day of data collection. All materials and
procedures received approval from the Institutional Review Board (IRB) at the University of
Southern California’s Health Sciences Campus (USC HSC).
Four-year follow-up
All materials and procedures from the current study received IRB approval from the
USC HSC in May 2007. A letter signed by the TPRC Principal Investigator was sent to all
eligible subjects notifying them of the telephone-based follow-up study. Subjects were given
the ability to opt-out of being contacted by study staff by returning a postage-paid postcard
by a designated date indicating their decision to not participate. Those subjects who did not
actively refuse to be contacted were eligible for the study.
Student tracking
Baseline
TPRC participants provided contact information and received reminder slips with
the scheduling of future TPRC assessments. The participant’s contact information includes:
(1) participant’s name, (2) current address, (3) phone number, (4) contact person’s name and
phone number (friend or relative who knows how to reach the participant), (5) participant’s
parent(s) name, address, and phone number, (6) participant’s social security number, (7)
participant’s driver’s license number, (8) e-mail address, (9) dates and status of last

64
attempted contact (i.e., contacted, measured, refused), and (10) an extensive notes section
where all attempts to reach participants are recorded.
Four-year follow-up
For the current study, sampling was limited to students who were at least 18 years
old at the time of sampling and self-identified as Latino/a, White, or African-American
(n=747). The sample size was further limited to those individuals who had some reliable
contact information in the TPRC participant database (i.e., phone number or address) and the
name of at least one parent or guardian with whom an online search could be made for
additional information. Given the additional criteria, the final base sample was 391 CHS
youth.
The online search databases of White Pages, Net Detective, and PeopleFinders were
used to update addresses and phone numbers. Searches on each site were made using the
names of the parent(s) or guardian(s) and the student. Reverse searches by address and
phone numbers listed in the TPRC participant database were also conducted in the White
Pages database. If available, phone calls to contacts listed in the TPRC participant database
were made. A final search of the White Pages database was conducted in September 2008 to
update any information.
TPRC participants were considered “lost to follow-up” if the address and phone
number listed in the TPRC participant database were no longer valid, and updated
information was not available for both the parent or guardian and TPRC participant. TPRC
participants were considered “unable to reach” if phone calls were not returned; there was
never any answer; or no phone number was listed publicly (e.g., address listed but phone
number unlisted). A minimum of ten phone call attempts were made to each participant, but
messages were left only at every third call or after a few weeks of leaving a last message.

65
Procedure
Baseline
Self-administered paper-and-pencil surveys were completed in the classrooms at the
respective CHS. Since classes were taught in English, it was assumed that the student had
the ability to read and complete a survey in English. The survey took approximately 30
minutes to complete. Baseline data were collected between February and November 2004.
Four-year follow-up
Between February and April 2008, letters were sent to all individuals in the final
base sample. Random sampling by gender and race (Latino and White only) was conducted
with ultimate sampling of all TPRC participants. The data collection team consisted of the PI
and five undergraduate students doing directed research under the aegis of the Chair of the
PI’s Doctoral Committee. Each undergraduate research assistant participated in 5 hours of
training prior to contacting TPRC participants. The training was conducted by the PI and
consisted of a 2-hour lecture on the study proposal; 1-hour training on the study materials
and protocol; and 1-hour training of conducting mock interviews with one another. Once
phone calls to TPRC participants started, the PI worked alongside each research assistant for
the first few days for quality assurance purposes.
Once the participant was reached by telephone, an initial screener with verbal
consent was read giving basic information on the purpose of the assessment. The survey was
administered by a trained interviewer once the participant consented verbally to participate.
The survey took approximately 15 minutes to complete. Subjects were given the same
unique identification number so that the first and second waves of data could be linked. To
minimize response biases and prior to asking the sexual behaviors questions, participants
were reminded that their name is not written on the survey, and they can refuse to answer
any question. A $15 gift card was mailed to each participant for any out-of-pocket expenses

66
related to completing a survey. Follow-up data were collected between March and
September 2008. Raw data were entered by the PI into an Access database which was then
imported into SAS.
Measures
Dependent variables. Frequencies of lifetime and recent condom use during anal
and/or vaginal intercourse were assessed as separate outcome variables. Two single-item
indicator variables measured the frequency of condom usage during anal and/or vaginal sex
over one’s lifetime and within the past 3 months. These questions were adapted from the
National Institute of Mental Health Multisite HIV Prevention Trial (Fishbein & Coutinho,
1997). The response format was a 5-point Likert-type scale ranging from “All of the time”
(1) to “Never” (5). The item was reverse-coded so that a higher rating indicated greater
frequency in condom use.
Independent variables. Exploratory factor analyses were performed to select reliable
items and create scales for five sociocognitive constructs. Items with a minimum factor
loading of 0.40 on at least one factor were retained in the final scale. Each final scale
consisted of the weighted sum of the values of the variables based on factor loadings.
Parent-adolescent communication. The Parent-Adolescent Communication Scale
(PACS) is a five-item scale assessing an adolescent’s frequency of sexual communication
with one’s parents in the previous six months (McDermott Sales, Milhausen, Wingood,
DiClemente, Salazar, & Crosby, 2008). The sexually related topics are sex, how to use
condoms, protecting self from sexually transmitted diseases, protecting self from HIV, and
preventing pregnancy. It has been validated among a sample of African-American
adolescent females, ages 14 to 18 years, and demonstrated satisfactory internal consistency
(Cronbach α = 0.88).

67
The PACS scale was used to assess sexual communication within the six months
prior to the four-year follow-up survey (Cronbach α = 0.91). Although the scale was
previously validated in an African-American sample of females, it showed high internal
consistency among non-Latinos (Cronbach α = 0.92), Latinos (Cronbach α = 0.91), females
(Cronbach α = 0.91), and males (Cronbach α = 0.91). The scale was further modified to
retrospectively assess communication during the participant’s adolescent years (Cronbach α
= 0.91) and was a reliable measure regardless of ethnicity and gender (non-Latinos,
Cronbach α = 0.94; Latinos, Cronbach α = 0.88; females, Cronbach α = 0.90; males,
Cronbach α = 0.92). Each statement was on a 4-point Likert-type scale ranging from
“Never” (1) to “Often” (4), assessing the frequency of communication per topic. A final
scale by developmental period was the sum of the products of each item and its factor score
and was used in regression models.
Perceived HIV susceptibility. The Health Belief Model Perceived Susceptibility
(HMBP) scale (Lux & Petosa, 1994) is a six-item scale assessing perceived susceptibility to
HIV. The HMBP has been previously validated in a sample of African-American and White
incarcerated youth between the ages 13 to 18 years (Cronbach α = 0.72). The statements
pertain to perceptions that people similar to self in age or other ways do not get HIV; peers
and self are too young to get HIV; one’s good health can fight HIV infection; and small
personal concern over contracting HIV. The statements are on a 4-point Likert-type scale
ranging from “Agree” (1) to “Disagree” (4), assessing level of agreement to each statement.
A higher score indicates greater perceived HIV susceptibility. The final perceived HIV risk
scale (Cronbach α = 0.84) was the sum of the products of each item and its factor score.
Peer norms for condom use. The Sexual Risks Scale – Norms (SRSN) scale is a
seven-item scale that assesses peer norms for condom use (DeHart & Birkimer, 1997). The
SRSN was previously validated in a sample of college students (Cronbach α = 0.84). The

68
statements pertain to peer anger or disappointment in self for not using a condom; frequent
communication on safer sex; peer encouragement to practice safer sex; peer concern over
condom use; mutual communication on carrying a condom (2 items); and communication on
whether or not a condom was used during sexual activity. The item on peer concern over
condom use was reverse-coded to be positively oriented. The statements are on a 5-point
Likert-type scale ranging from “Strongly Disagree” (1) to “Strongly Agree” (5), assessing
level of agreement to each statement. A higher score indicates safer sex peer norms. The
final peer norms for condom use scale (Cronbach α = 0.90) was the sum of the products of
each item and its factor score.
Condom use self-efficacy. The Condom Use Self-Efficacy Scale (CUSES) scale is a
seven-item scale that assesses the mechanics of putting a condom on self or other (4 items)
and one’s assertiveness or ability to persuade a partner to use a condom (3 items; Brafford &
Beck, 1991). The CUSES was validated among a diverse sample of college students (entire
scale, Cronbach α = 0.91; mechanics, Cronbach α = 0.78; assertiveness, Cronbach α = 0.80).
In the current study, all items were reverse-coded to be positively oriented. Statements are
on a 5-point Likert-type scale with a possible range from “Strongly Disagree” (1) to
“Strongly Agree” (5). Both the mechanics (Cronbach α = 0.90) and the assertiveness
(Cronbach α = 0.93) subscales had high internal consistency. All items loaded onto one
factor, and the final condom use self-efficacy scale was the sum of the products of each item
and its factor score (Cronbach α = 0.94).
Covariates. Several covariates were assessed. Gender and race/ethnicity were
assessed as moderators. If not statistically significant, then the variables were treated as
covariates.
Age at sexual onset. Age at sexual onset was dichotomized at the mean age of 15
years where sexual initiation by age 15 years was coded as ‘1.’ Age at first intercourse was

69
assessed as a moderator on the relationships between the independent and dependent
variables.
Age differential of first sexual partner. Questions on if the first sexual partner was
the same age, older, or younger and by how many years were asked in order to determine the
age differential at sexual onset. A dichotomous composite variable (Cronbach α = 0.78) was
created where an older partner (“Older, 3 or more years”) was coded as ‘1’ and a partner that
is younger, same age, or no more than 2 years older was treated as the referent category
(coded as ‘0’).
Sexual maturation. The Adolescence Scale-Imperial College School of Medicine
(AS-ICSM), a brief retrospective assessment of pubertal milestones, was used to assess
sexual maturation (Kaiser & Gruzelier, 1999). The AS-ICSM consists of one common item
(growth spurt); one female-specific item (menarche); and three male-specific items (voice
break, regular shaving, and first nocturnal emission). The AS-ICSM was normed among
young adults (mean age = 20.7 years) and shown to be a reliable retrospective assessment for
both females (Cronbach α = 0.87) and males (Cronbach α = 0.83; Kaiser & Gruzelier, 1999).
In the current study, the mean age at sexual maturation for females was computed
using both growth spurt and menarche (Cronbach α = 0.31). Almost one-fourth of males
(23%) either did not remember or had not experienced a nocturnal emission, and thus, this
item was omitted. For males, the mean age at sexual maturation was computed using ages at
growth spurt, voice break, and regular shaving (Cronbach α = 0.61).
Gender and race/ethnicity. Information on self-reported gender and race/ethnicity
were collected during the TPRC baseline assessment. Gender and race/ethnicity were
dichotomously coded. Females were coded as ‘0,’ and males were coded as ‘1.’ Latinos,
coded with a value of ‘1,’ were compared to non-Latinos (coded as ‘0’).

70
Age. Using birth date and follow-up interview date, age in years at the time of the
four-year follow-up assessment was computed.
Behavioral sexual orientation. Behavioral sexual orientation was defined using four
variables: self-identified sexual orientation; subject’s gender; and reported gender(s) of
lifetime and recent sexual partners. The three behavioral sexual orientation categories were
men who had sex with both men and women (‘1’), women who had sex with men and/or
women (‘2’), and men who had sex with women (‘3’). If a male self-identified as
“heterosexual” but engaged in sex with both males and females, then he was coded as
behaviorally having had sex with both men and women (‘1’).
Acculturation. Acculturation is the process by which the values, norms, attitudes,
and behaviors of an individual of a minority culture are fused with those from the majority
culture and incorporated into one’s existence (Marín, Sabogal, Marín, Otero-Sabogal, &
Perez, Stable, 1987). Marín et al. (1987) created an acculturation scale that focused on the
extent of which languages were spoken (i.e., English, Spanish) in particular settings.
Although the Marín et al. (1987) linguistic acculturation scale was normed with Latino
adults (Cronbach α = 0.90), its subsequent use among Latino youth has yielded high
reliability (e.g., Cronbach α = 0.90 in Ford & Norris, 1993).
Both TPRC and the current study used the Marín et al. (1987) short acculturation
scale (Cronbach α = 0.88, both periods). The four items assessing linguistic acculturation
were language(s) most often read and spoken; language(s) spoken at home; language(s)
spoken with friends; and language(s) to which movies, television, and radio shows are
watched and listened. The response categories were: (1) Only English; (2) English more than
another language; (3) English and another language equally; (4) Another language more than
English; and (5) Only another language (not English). The final acculturation item was the
mean value of the four items.

71
Data analysis
Since baseline data were nested within schools, intraclass correlation (ICC)
coefficients were computed on each dependent variable to address the potential cluster effect
of schools. ICC coefficients provide an estimate of the degree of commonality in
observations within a given unit (e.g., school). School-level ICCs that are less than 0.03
indicate low degree of dependence in observations (Murray and Blitstein, 2003). ICCs were
calculated for variables in the model, and all were found to be less than 0.03. Also four-year
follow-up data were collected when the participant was no longer in continuation high
school, providing further support that controlling for school may not be pivotal in the present
analyses. Therefore, linear regression models were performed without accounting for school
in the modeling.
Descriptive statistics were computed for all demographic variables. Student’s t-tests
were used to test associations between demographics and sexual onset characteristics and
sexual maturation; substance use; parent-adolescent communication on sexual safety;
perceived susceptibility to HIV; peer norms on safer sex; condom use self-efficacy;
frequency of lifetime condom use; and the number of lifetime sexual partners.
Nonparametric Wilcoxon signed rank tests were used to test associations between
demographics and sexual onset characteristics and frequency of condom use in the past 3
months during vaginal and/or anal intercourse and the number of sexual partners in the past
3 months.
Separate models were developed to assess whether or not each sociocognitive factor
was associated with the frequency of lifetime and recent condom use during anal and/or
vaginal sex. Regression diagnostics showed that the variable for frequency of lifetime
condom use was normally distributed whereas the variable for frequency of recent condom
use was positively skewed. Due to a high level of skewness in the variable for recent

72
condom use, this variable was log-transformed for use in the linear regression models.
Regression diagnostics showed that the log-transformation was appropriate. The log-
transformed outcome variable for recent condom use was used in all regressions analyses.
Results on this variable are reported on the log scale. Interaction terms were included
separately for each independent variable and covariate (i.e., gender, race/ethnicity) in order
to evaluate if there is a difference in sexual risk-taking by the covariate. If significant, the
model was refit by the levels of the covariate. The criterion for statistical significance was
set to a 0.05 level. All analyses were conducted using SAS version 9.1.
Results
Demographic characteristics of the study sample are displayed in Table 10. The
sample was mostly Latino (56.8%), female (51%), and heterosexual (96.4%). The mean age
at four-year follow-up was 20.6 years. The majority of the sample had already had
consensual sex (98.2%) with close to 60% initiating sex by 15 years of age. Larger
proportions of females and non-Latinos had an early sexual onset. At first sexual experience,
15.6% of the sample had an older (3+ years) sexual partner. Overall, participants reported
using condoms most of the time in their lifetime (Median score = 4) and rarely in the past 3
months (Median score = 2).
Regardless of demographic and sexual debut characteristics, participants rarely
spoke with a parent or parental figure on sexual safety in the previous six months (Mean
score = 1.8). Sexual safety communication was higher during adolescent years (Mean score
= 2.2) with non-Latinos more likely to speak with a parent on sexual safety during their
adolescent years than Latinos (Mean score = 2.5 v. 2.1; t = 2.2, p = 0.03).
Gender and sexual debut characteristics were associated with other sociocognitive
items. Females were more likely to perceive themselves as susceptible to HIV as males
(Mean score = 3.7 v. 3.5; t = 2.1, p = 0.04). Although their condom use self-efficacy was

73
Table 10. Characteristics of the sample
Variable Overall
N=111
Female
N=56 (50.5%)
Male
N=55 (49.5%)
Latino
N=63 (56.8%)
Non-Latino
N=48 (43.2%)

N (%) N (%) N (%) N (%) N (%)
Age at sexual onset, ≤15 years
(n=107)
63 (58.9) 34 (64.2) 29 (53.7) 28 (47.5) 35 (72.9)
Older (3+ years) sexual partner at
sexual onset (n=109)
17 (15.6) 14 (25.5) 3 (5.6) 9 (14.8) 8 (16.7)
Sexual orientation
Men who have sex with men 4 (3.6) --- --- 4 (7.3) 3 (4.8) 1 (2.1)
Women who have sex with men 56 (50.5) 56 (100.0) --- --- 33 (52.4) 23 (47.9)
Men who have sex with women 51 (45.9) --- --- 51 (92.7) 27 (42.9) 24 (50.0)
Ever had sex
Lifetime 109 (98.2) 55 (98.2) 54 (98.2) 61 (96.8) 48 (100.0)
Past 3 months 103 (92.8) 52 (92.9) 51 (92.7) 51 (81.0) 43 (89.6)

Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Range
Age (years)

20.6 (1.2) 20.3 (1.3) 20.9 (1.1) 20.6 (1.2) 20.5 (1.3) 18-23
Acculturation (score) 1.8 (0.9) 1.9 (0.9) 1.8 (0.8) 2.4 (0.7) 1.0 (0.1) 1-5
Sexual maturation (years) 13.5 (1.1) 13.1 (0.9) 14.0 (1.1) 13.5 (1.1) 13.5 (1.1) 11-18
Parent-adolescent sex and sexual
safety communication, past 6
months
1.8 (0.9) 1.9 (0.9) 1.7 (0.8) 1.9 (0.9) 1.7 (0.9) 1-4
Parent-adolescent sex and sexual
safety communication, adolescent
years
2.2 (0.9) 2.2 (0.9) 2.3 (0.9) 2.1 (0.9) 2.5 (0.9)
*
1-4
Perceived susceptibility to HIV 3.6 (0.6) 3.7 (0.5) 3.5 (0.6)
*
3.6 (0.6) 3.7 (0.5) 1-4
Peer norms on safer sex 3.6 (0.9) 3.6 (0.8) 3.5 (0.9) 3.6 (0.9) 3.6 (0.9) 1-5
Condom use self-efficacy 4.2 (0.7) 4.0 (0.7) 4.4 (0.5)
**
4.2 (0.6) 4.2 (0.8) 1-5

Median Q1,Q3 Median Q1,Q3 Median Q1,Q3 Median Q1,Q3 Median Q1,Q3 Range
Frequency of condom use
Lifetime (n=109) 4 (3, 4) 3 (3, 4) 4 (3, 4) 3 (3, 4) 4 (3, 4) 1-5
Past 3 months (n=103) 2 (1, 3) 2 (1, 3) 2 (1, 4) 2 (1, 3) 2 (1, 3) 1-5
*
p < 0.05,
**
p < 0.01

74
significantly lower than males, females reported a strong confidence in their ability to
practice safer sex (Mean score = 4.0 v. 4.4; t = 2.2, p = 0.03). Peer norms on safer sex were
significantly lower among early sexual initiators than among those who delayed sexual onset
(Mean score = 3.4 v. 3.9; t = 3.6, p = 0.0005). A trend towards greater self-efficacy with
condom use was exhibited by both late sexual initiators (Mean score = 4.4 v. 4.1; t = 1.8, p =
0.07) and persons with a peer-aged partner at sexual onset (Mean score = 4.3 v. 3.8; t = 1.9,
p = 0.07).
Regardless of gender and ethnicity, the parent-adolescent communication on sexual
safety scales at both time periods and the peer norms on safer sex and the condom use self-
efficacy scales were significantly correlated (Tables 11 and 12, respectively). Safer sex peer
norms were also positively correlated with perceived susceptibility to HIV for both Latinos
and non-Latinos.
Gender and ethnic differences in correlations exist. For males, perceived HIV risk
was positively correlated with both parent-adolescent communication on sexual safety
during adolescence and safer sex peer norms. For non-Latinos, parent-adolescent
communication on sexual safety during either period was significantly positively correlated
with safer sex peer norms, and parent-adolescent communication on sexual safety during
adolescence was significantly positively correlated with condom use self-efficacy.
In multiple linear regressions, greater lifetime condom usage during anal and/or
vaginal sex was significantly associated with parent-adolescent communication on sexual
safety during adolescence (Std. β = 0.05, p = 0.01; Table 13) and within the past 6 months
(Std. β = 0.06, p = 0.008) after controlling for covariates. Condom use within the past 3
months was also significantly associated with parent-adolescent communication during
adolescence (Std. β = 0.03, p = 0.02) and within the past 6 months (Std. β = 0.04, p = 0.009).

75
Table 11. Correlations among sociocognitive variables by gender
†

PACS,
past 6 months
PACS,
adolescent years
Perceived
HIV risk
Safer sex
peer norms
Condom use
self-efficacy
PACS, past 6 months
‡
-- 0.426
**
0.007
*
0.136
*
0.018
*

PACS, adolescent years 0.631
**
-- 0.268
**
0.252
*
0.180
*

Perceived HIV susceptibility -0.150** 0.024* -- 0.551
**
0.192
*

Safer sex peer norms 0.234
**
0.139* 0.092* -- 0.455
**

Condom use self-efficacy 0.242* 0.255* 0.039* 0.565
**
--
†
Values for males are above the diagonal.
‡
PACS: Parent-adolescent communication on sex
*
p < 0.05,
**
p < 0.0001

Table 12. Correlations among sociocognitive variables by ethnicity
†

PACS,
past 6 months
PACS,
adolescent years
Perceived
HIV risk
Safer sex
peer norms
Condom use
self-efficacy
PACS, past 6 months
‡
-- 0.572
***
-0.095
*
0.007
*
0.016
*

PACS, adolescent years 0.533
**
-- 0.108
**
-0.012
*
0.101
*

Perceived HIV susceptibility 0.073 * 0.171* -- 0.337
*
0.040
*

Safer sex peer norms 0.439
**
0.439
**
0.440
**
-- 0.414
*

Condom use self-efficacy 0.209 * 0.372
**
0.057* 0.508
**
--
†
Values for Latinos are above the diagonal.
‡
PACS: Parent-adolescent communication on sex
*
p < 0.05,
**
p < 0.0001

76
Table 13. Regression analyses
Psychosexual Construct

PACS,
adolescence
PACS, past
6 months
Perceived
HIV
susceptibility
Peer norms
on safer sex
Condom
use self-
efficacy
Std. β Std. β Std. β Std. β Std. β
Condom use,
lifetime
a

Overall 0.05
*
0.06
**
0.04 0.04
*
0.03
IV-by-gender 0.01 0.06
*
0.18
*
0.05 0.09
†

Gender-stratified
Males 0.06
†
0.06 0.08
†
0.01 0.03
Females 0.04 0.06
†
0.08 0.07
*
0.07
*

IV-by-ethnicity -0.06 -0.03 0.05 0.07
†
0.07
†

Ethnicity-
stratified

Latinos 0.02 0.05 0.02 0.01 0.01
Non-Latinos 0.08
**
0.06
*
0.15
**
0.07
**
0.05
*

Condom use,
recent
b

Overall 0.03
*
0.04
**
-0.01 0.01 0.01
IV-by-gender 0.03 0.01 0.13
*
-0.01 -0.01
Gender-stratified
Males 0.04
†
0.03 0.01 0.01 0.01
Females 0.01 0.03 -0.09
*
0.01 0.01
IV-by-ethnicity 0.04 0.02 -0.08 -0.03 -0.04
Ethnicity-
stratified

Latinos 0.01 0.04
†
-0.04 -0.01 -0.02
Non-Latinos 0.05
*
0.03 0.11
*
0.03 0.02
†
p < 0.10,
*
p < 0.05,
**
p < 0.01
Note. All parameter estimates (standardized betas) are adjusted for age, gender, ethnicity, age at
sexual debut, age differential with first sexual partner, pubertal age, and acculturation.

a
Further adjusted for lifetime number of sexual partners
b
Further adjusted for number of sexual partners in past 3 months

Although most communication-by-gender and communication-by-ethnicity
interactions were not statistically significant, models were refit by levels of the covariate. In
stratified analyses, a higher frequency of sexual safety communication during adolescence
was a significant correlate of greater lifetime (Std. β = 0.08, p = 0.002) and recent condom
use (Std. β = 0.05, p = 0.03) among non-Latinos only. It was also marginally associated with
higher lifetime (Std. β = 0.06, p = 0.06) and recent condom use (Std. β = 0.04, p = 0.06)
among males. A significant relationship between greater recent sexual communication and

77
more consistent lifetime condom usage remained among non-Latinos (Std. β = 0.06, p =
0.04).
Sexual safety communication during adolescence was not a significant correlate of
condom use among females and Latinos; however, recent sexual safety communication was
correlated with condom use. A trend-level effect between a greater frequency of recent
sexual communication and higher lifetime condom usage was evident among females (Std. β
= 0.06, p = 0.05). A trend-level effect between a greater frequency of recent sexual
communication and higher condom usage within the past 3 months was evident among
Latinos (Std. β = 0.04, p = 0.05).
Greater peer norms on safer sex were significantly associated with greater frequency
of lifetime condom use (Std. β = 0.04, p = 0.01). The effect of peer norms on lifetime
condom use remained for females (Std. β = 0.07, p = 0.02) and non-Latinos (Std. β = 0.07, p
= 0.002) only. There was no relationship between peer norms and recent condom use.
Overall, frequency of condom use in either time frame was not associated with
either condom use self-efficacy or perceived susceptibility to HIV. When stratified by
gender and ethnicity, greater belief in one’s ability to use a condom was significantly
associated with greater lifetime condom use among females (Std. β = 0.07, p = 0.02) and
non-Latinos (Std. β = 0.05, p = 0.04). Condom use self-efficacy continued to have no
relationship with recent condom use. In ethnicity-stratified analyses, greater perceived
susceptibility to HIV was associated with greater lifetime (Std. β = 0.15, p = 0.005) and
recent (Std. β = 0.11, p = 0.03) condom use among non-Latinos. It was also marginally
associated with more consistent lifetime condom usage among males (Std. β = 0.08, p =
0.09).
Despite having a greater perception of susceptibility to HIV, the frequency of recent
condom use was significantly lower among females (Std. β = -0.09, p = 0.04). The

78
relationship between perceived susceptibility to HIV and recent condom use was explored
further by stratifying by the number of sexual partners in the past 3 months (i.e., one partner
v. two or more partners) within the gender strata. For persons who reported one sexual
partner, the significant negative relationship between inconsistent condom usage and high
perceived susceptibility to HIV maintained for females (Std. β = -0.10, p = 0.04). A different
relationship between recent condom usage and perceived susceptibility to HIV emerged for
persons who reported multiple sexual partners in the past 3 months. Greater perceived
susceptibility to HIV was associated with greater recent condom use among males (Std. β =
0.05, p = 0.04).
Discussion
An exploration of the relationships between several psychosocial constructs and
frequency of condom usage revealed not only that some constructs remained consistently
significant factors but that these relationships also varied by gender and ethnicity.
Participants rarely discussed sexual safety with a parental figure during adolescence, and
communication was even rarer within the previous six months. However infrequent,
communication on sex and condoms between a parent and a child was associated with a
greater likelihood of consistent lifetime and recent condom use regardless of when the
communication took place. Like Stanton et al. (1999), communication was associated with
condom use regardless of the recipient’s gender, but unlike their sample, these findings can
be extended to Latino and non-Latino young adults.
Parent-adolescent communication also had a predictive effect. Young adults who
reported greater frequency in communication with a parent during their adolescence reported
more consistent lifetime and recent condom use at follow-up. Similarly, this effect was seen
for non-Latinos. Although not significant, some interesting patterns by gender and ethnicity
emerged. Communication on sexual safety during adolescence was not significantly different

79
between males and females and between Latinos and non-Latinos. Henrich et al. (2006)
found communication to be predictive of sexual risk-taking among females only. Among our
sample of youth who were in continuation high schools, the positive predictive effect of
communication during adolescence on subsequent condom use appeared to be stronger for
males. For females and Latinos, the impact of parental influence on their condom use was
evident with recent communication.
Peers also played a key role in condom use in that individuals who perceived peer
norms as encouraging safer sex were more likely to practice safer sex in their lifetime. This
relationship between peer norms and lifetime condom use was further moderated by gender
and ethnicity. Although peer norms were associated with a greater frequency of lifetime
condom use regardless of gender and ethnicity, this relationship was statistically significant
for females and non-Latinos only. Individuals had peer networks that discussed and
encouraged safer sex, but peer norms were a more significant factor in influencing lifetime
condom use for females and non-Latinos.
These gender and ethnic differences were also reflected in the relationship between
lifetime condom use and condom use self-efficacy. For females and non-Latinos, greater
confidence in one’s ability to use condoms was significantly associated with consistent
condom use. Having peers that encourage safer sex as well as having the ability to practice
safer sex, which may be a skill learned and honed by talking with peers, contributed to being
more likely to use condoms consistently.
Although they had a significant effect on lifetime condom usage, peer norms on
safer sex and condom use self-efficacy did not have a significant impact on recent condom
use. Unlike Rotheram-Borus, Marelich, and Srinivasan (1999), there was no correlation
between positive peer norms and greater condom use among males. Instead, peer norms that
support safer sex were positively correlated with greater lifetime condom use among females

80
as well as non-Latinos. Condom use self-efficacy was also positively correlated with greater
lifetime condom use among females and non-Latinos. Females and non-Latinos who were
more confident with their condom mechanics and negotiation skills were more likely to use
condoms consistently over their lifetime. Our findings support previous findings on the
relationship among females (DiClemente et al., 1996; Wingood and DiClemente, 1998).
Ellen et al. (1996) and Morrison-Beedy, Carey, and Aronowitz (2002) did not find a
significant correlation between perceived susceptibility to HIV and condom use. Although
there was no overall significant effect, perception of one’s susceptibility to HIV played a
significant role for males, females, and non-Latinos. For males, greater perception of one’s
susceptibility to HIV was associated with more consistent lifetime condom use. A seemingly
dissonant relationship between perceived susceptibility to HIV and recent condom use was
present among females. Supporting previous findings from Millstein and Moscicki (1995)
and Murphy, Rotheram-Borus, and Reid (1998), females perceived themselves to be highly
susceptible to HIV yet their reported condom use in the past 3 months was infrequent. Upon
further exploration, this negative relationship maintained for females with one sexual partner
and lower HIV risk factors (i.e., self-reported monogamy).
There are many possible reasons for the apparent discrepancy between perception of
HIV risk and condom use. Lower risk individuals may acknowledge the potential for HIV
risk regardless of one’s safer sex behaviors. It can be argued that they may not think that
they are at risk for HIV. Rather, they are not ruling out the possibility that they are at risk of
HIV because they are young. Since they are more aware of the potential risk, safer sexual
behaviors are selected (e.g., monogamy). Furthermore, the perception of HIV risk is known
among the sample, but the likelihood of HIV infection, a potential moderator of the
relationship, is unknown.

81
Individuals could have also underreported sexual risk behaviors compromising the
validity of the data. However, participants were assured of the confidentiality of all data, and
measures were taken to display this confidentiality to all participants. Furthermore, measures
that were used had been validated previously in adolescent samples. Therefore, there is no
reason to believe that participants were differentially dishonest in their answers.
Individuals may also suspect or know that their sexual partner has other sexual
partners, information that was not collected. Although choosing monogamy, a partner’s
sexual practices pose an HIV risk to the study participant. Lack of consistent condom use in
light of these thoughts further poses an HIV risk.
Findings should be considered in light of additional study limitations. Causality
cannot be deduced from these analyses due to their cross-sectional nature. Most of the
sociocognitive items assessed current sociocognitions. It is quite plausible that some
behaviors may lead to social cognitions. For example lack of consistent recent condom use
may have increased one’s perception of HIV susceptibility. Furthermore, social-cognitions
are dynamic and change over time in response to new environmental stimuli (e.g.,
knowledge, peers). However with the recent items, it makes theoretical sense to look at the
association as social-cognitions leading to current sexual behaviors.
Although efforts were taken to contact a larger sample size, tracking participants
was difficult, many of whom were lost to follow-up. The final sample size is representative
of a CHS sample that is about 50% Latino and 50% female. However, baseline participants
who were not interviewed in the most recent follow-up (over 80% of the sample) may have
sexual behaviors that differ from the current sample thereby limiting the generalizability of
the findings.
Results from this study support the association of key sociocognitive constructs on
consistency of using condoms. These conclusions bear potential implications for the

82
prevention of sexual risk-taking that place an individual at risk of HIV and other sexually
transmitted diseases (STD). Interventions aimed at youth in continuation high schools could
benefit by incorporating comprehensive sexual education into curriculum. Sexual education
should not only focus on delaying sexual onset but also on using condoms correctly and
consistently as a way of reducing one’s risk of HIV and other STD. Sexual education
curriculum should also include information to enhance HIV knowledge and perception of
risk (e.g., HIV prevalence among youth) while normalizing sexuality and its discussion
among peers. Furthermore, interventions should include parental involvement. Parents and
parental figures should be provided with the knowledge and tools on how to effectively
communicate with children not only about HIV risk but also about sexuality in general.
Future research is needed to determine whether this type of intervention can reduce sexual
risk-taking among this at-risk population.

83
CHAPTER FIVE: CONCLUSION

In reviewing the literature on select behaviors, feelings, and sociocognitive factors
and correlates of sexual behaviors among youth, research focusing on youth in continuation
high schools is lacking. According to national epidemiologic data, youth from continuation
high schools engage in riskier HIV behaviors such as a higher prevalence of substance use
and sexual risk behaviors than youth in regular high schools. Based on this information, each
of the three preceding studies addressed an issue meriting further investigation in order to
elucidate the literature on CHS youth. The aims of the study were to assess the long-term
effects of substance use and sensation-seeking during adolescence on sexual behaviors
through young adulthood and the role of age at sexual initiation within these pathways; the
role of depressive symptomatology on sexual risk-taking behaviors; and the correlations
between select HIV sociocognitive factors and the frequency of condom use overall and by
gender and ethnicity.
In the assessment of adolescent substance use and sensation-seeking on number of
sexual partners, we found that substance use had an indirect effect on the number of lifetime
and recent sexual partners reported at four-year follow-up. It was hypothesized that
substance use and sensation-seeking would not only have direct effects on number of sexual
partners but the pathways stemming from each would be at least partially mediated by age at
sexual initiation. Unlike research among other adolescent samples, substance use had no
direct effect on number of sexual partners reported by young adults who were in
continuation high schools. The relationship between substance use and the number of sexual
partners was fully mediated by age at sexual initiation. Greater substance use was associated
with an earlier age at sexual onset. An earlier age at sexual onset in turn was related to
greater numbers of lifetime and recent sexual partners over a four-year period.

84
Sensation-seeking had neither direct nor indirect effects on number of sexual
partners. The literature on the relationship between sensation-seeking and sexual risk-taking
among adolescents has been discrepant. Findings from the current study illustrate that
sensation-seeking is not related to number of sexual partners either directly or indirectly
among CHS youth. If a relationship between sensation-seeking and number of partners exists
in this sample, it could be mediated by other variables. In Crockett, Raffaelli, and Shen
(2006), the relationship between sensation-seeking and sexual risk-taking was completely
mediated by substance use. In a sample with a high prevalence of substance use such as the
current study, substance use may mediate the relationship between sensation-seeking and
number of sexual partners among CHS youth. Lack of a significant relationship could also
be related to the length of time between the collection of the two variables, perhaps it was
too long of a period to detect any significant relationship. Sensation-seeking may not have a
long-term effect (i.e., four years in the current study); rather, the effect of sensation-seeking
on sexual risk-taking may be more salient within a shorter time period.
In testing the associations between depressive symptomatology and sexual
behaviors, it was expected that depressive symptomatology at either time period, and
especially if exhibited at both time periods, would be correlated with a higher number of
sexual partners and lower frequency of condom use. Depressive symptomatology at both
time points was not correlated with the number of sexual partners or the frequency of
condom use.
However, some interesting outcomes were observed. Individuals who exhibited
depressive symptomatology only at baseline reported a significantly lower number of
lifetime sexual partners. Our findings are contrary to Ethier et al. (2006) who found
depressive symptomatology to be predictive of a greater number of sexual partners. Like
Mazzaferro et al. (2006), individuals who exhibited depressive symptomatology at follow-up

85
reported a significantly higher number of lifetime sexual partners, but unlike Mazzaferro et
al. (2006) and other previous findings (Kowaleski-Jones & Mott, 1998; Shrier et al., 2001),
depressive symptomatology had no impact on condom use.
Intrapersonal and interpersonal health behavior theories have supported several HIV
sociocognitive concepts as significant correlates of sexual behaviors. Results from the
current study support significant main effects of select sociocognitive correlates on condom
use whereas the relationships for others vary by gender and/or ethnicity. Regardless of
gender, ethnicity, and time period, parent-adolescent communication on sex, pregnancy,
HIV/STDs, and condoms was positively correlated with lifetime and recent condom use.
Additionally, peer norms that support safer sex were positively correlated with lifetime
condom use, and this relationship was more significant for females and non-Latinos. Parents
and peers played significant roles in lifetime condom use among youth who were in
continuation high schools.
Perceived susceptibility to HIV and condom use self-efficacy had no main effects
relationships with condom use. When stratified by ethnicity and gender, different
relationships emerged. Non-Latinos who had a greater perception of HIV risk were more
likely to consistently use condoms over their lifetime and within the past 3 months. For
females, greater perception of HIV susceptibility was negatively correlated with recent
condom use. Although females perceived themselves at risk for HIV, this perception did not
translate into more consistent condom use. When further explored, this relationship
maintained for females with one recent sexual partner but was no longer significant for
females with multiple recent sexual partners. Females and non-Latinos who had a greater
confidence in their abilities to practice safer sex reported more consistent lifetime condom
usage.

86
Parent-adolescent communication on sexual safety, perceived susceptibility to HIV,
peer norms on safer sex, and condom use self-efficacy were positively correlated with
lifetime condom use for non-Latinos. Additionally, parent-adolescent communication during
adolescence and perceived susceptibility to HIV were positively correlated with recent
condom use for non-Latinos. These sociocognitive correlates significantly impacted condom
use among a sample of high-risk youth. These sociocognitive correlates were not significant
for Latino youth. Current communication with a parent on sexual safety had the tendency to
positively influence recent condom use, but no other constructs were significant correlates
among Latino youth.
Limitations
Participants were recruited into the parent study in 2004. There was a four-year gap
between baseline and the most recent follow-up assessment. During this time, participants
proved to be quite mobile and difficult to reach. Efforts were taken to contact a larger sample
size, but tracking participants proved to be challenging. The final base sample at follow-up
was 15% of the baseline sample. However, the final sample size was representative of a CHS
sample that is about 50% Latino and 50% female. Considering that over 80% of the sample
did not complete a follow-up survey, generalizability of the findings is limited. Baseline
participants who were lost-to-follow-up may have sexual behaviors that differ from the
follow-up sample. Additionally, the small sample size does not provide sufficient power to
detect significant findings.
Issues with some of the measurement items exist. Information on substance use and
sensation-seeking were collected at baseline only. Current substance use and sensation-
seeking as well as the chronicity of each are unknown. Participants may now have different
substance-using behaviors or sensation-seeking scores which may positively or negatively
impact sexual behaviors. However, the aim for this study component was to determine if

87
adolescent substance use and sensation-seeking had long-term effects on sexual risk-taking
and if they were predictive of the number of sexual partners reported in young adulthood.
Our findings satisfy this study aim.
Implications
A large portion of studies focusing on HIV protective and risk factors and behaviors
among adolescents have been conducted with a variety of adolescent samples of varying risk
(e.g., regular school-based, homeless). These studies have made significant contributions to
the literature by reporting on the prevalence of behaviors that place adolescents at risk for
HIV; identifying modifiable risk and protective factors for these behaviors; and identifying
points along pathway where one can intervene to reduce the risk of HIV infection.
Exploration of these factors and behaviors among CHS youth is warranted given that
they are at a higher risk of dropping out of school, making them available to participate in
deleterious behaviors; they have a higher prevalence of HIV risk behaviors; and they are
exposed to a social environment where HIV risk behaviors are high. To the best of our
knowledge, two studies have addressed sexual behaviors among youth in continuation high
schools: the national Alternative High School Youth Risk Behavior Survey (ALT-YRBS;
CDC, 1999) and O’Hara et al. (1998). Both studies have been instrumental in providing
information on the prevalence of risk and protective behaviors among ethnically diverse
CHS youth. However, these studies do not provide further insight into the factors that
encourage or discourage sexual risk-taking. Additionally, these studies are dated. ALT-
YRBS data were collected in 1998 whereas data from O’Hara et al. (1998) were collected in
the early 1990’s.
This study was undertaken in order to extend previous findings from other
adolescent samples to CHS youth and explore the risk and protective factors and correlates
of two HIV sexual risk behaviors: number of sexual partners and frequency of condom use.

88
Studies one and two provide support that negative behaviors or states have deleterious
effects on sexual behaviors whereas study three highlights protective correlates of sexual
behaviors. Study one supports the fact that greater substance use has an impact on a higher
number of sexual partners. Unlike other studies, it did not have a direct effect on number of
sexual partners. Age at sexual initiation proved to be a critical component and mediator of
the relationship between substance use and number of sexual partners. This relationship
provides two time points and at least two variables on which to intervene in order to produce
a more salutary outcome. Targeting substance use may not only impact health outcomes
associated with it (e.g., addiction) but may also impact sexual risk-taking like an earlier age
at sexual debut, an independent predictor of sexual behaviors (Harwell et al., 1999;
Hutchinson et al., 2003; Krantz, Lynch, & Russell, 2002; Miller, Clark, & Moore, 1997;
National Research Council, 1987; O’Donnell, O’Donnell, & Stueve, 2001; O’Donnell et al.,
1999; O’Hara et al., 2004; Stock et al., 1997).
Study two provides insight into the varying effects that depressive symptomatology
may take. The presence of depressive symptoms can influence the number of sexual partners
where some individuals will have a higher number of multiple sexual partners but others will
have fewer numbers. Therefore it seems that depressive symptomatology may be a risk
factor for sexual risk-taking only for some people. Depressive symptomatology may still
pose a health risk even if it does not necessarily lead to sexual risk-taking directly. The
presence of depressive symptomatology may lead to substance use as a way of self-
medicating the symptoms. Depressive symptomatology may be another distal factor of
number of sexual partners along with substance use and age at sexual initiation.
Study three highlights the importance of several intrapersonal and interpersonal
correlates of sexual behaviors. Many health behaviors originate in childhood, solidify in
adolescence, and carry through to adulthood. During adolescence, behaviors are shaped and

89
formed and have lasting effects, and peers begin to take on a larger role. In addition, CHS
youth are in an environment where substance use and sexual risk behaviors are high. It is
during this time that parents continue to play a vital role. In the current study, parent-
adolescent communication on sexual safety proved to be the constant significant correlate of
lifetime and recent condom use. These findings may be extrapolated further to suggest that
for youth exposed to a high-risk social environment, parental communication on sexual
safety may act as a buffer between the influence of risky peers and sexual risk behaviors.
Our findings also suggest that by young adulthood, both peers and parents exert influence.
Having positive extrinsic influences may influence the development and honing intrinsic
knowledge (e.g., perception of risk) and skills (e.g., condom use self-efficacy).
The significance of these studies lies in the fact that these findings offer points of
intervention and future research on sexual behaviors among CHS youth. First, teaching
effective refusal skills to reduce substance use and teaching coping strategies to improve
negative mood and to reduce sensation-seeking may be useful components in prevention of
sexual risk behaviors such as an earlier age at sexual onset, multiple sexual partners, and
inconsistent condom usage. HIV knowledge and education in various media need to be
continued to be disseminated in order to affect perception of susceptibility to HIV. Teaching
skills on condom mechanics, safer sex negotiation, and communication on sexuality to both
adolescents and parents are also useful prevention components. Additional research is
needed to identify the sociocognitive correlates and factors that affect sexual risk-taking and
preventative behaviors among Latino youth.
Individual and family-level interventions are important to reduce sexual risk-taking
among CHS youth, but these need to occur within the context of societal and cultural
influences on sexual behaviors. One can effect change among individuals and families to a
certain extent, but larger influences can challenge the initiation and maintenance of safer sex

90
behaviors. Individual- and family-level interventions may need to be developed along with
community-level interventions that normalize sexuality, invite open communication on it,
and move beyond abstinence only. A cultural shift needs to occur not only on the topic of
sexuality but also on gender and ethnic power differentials that have negatively impacted
sexual behaviors. Continued investigation of risk and protective factors from various levels
within this population would be helpful in improving the health of our youth and into
adulthood.

91
BIBLIOGRAPHY

Andresen, E.M., Malmgren, J.A., Carter, W.B., & Patrick, D.L. (1994). Screening for
depression in well older adults: evaluation of a short form of the CES-D (Center for
Epidemiologic Studies Depression Scale). American Journal of Preventive
Medicine, 10, 77-84.

Bacchetti, P, & Moss, A.R. (1989). Incubation period of AIDS in San Francisco. Nature,
338(6212), 251-253.

Bachanas, P.J., Morris, M.K., Lewis-Gess, J.K., Sarett-Cuasay, E.J., Flores, A.L., Sirl, K.S.,
et al. (2002). Psychological adjustment, substance use, HIV knowledge, and risky
sexual behavior in at-risk minority females: developmental differences during
adolescence. Journal of Pediatric Psychology, 27(4), 373-384.

Bailey, S.L., Gao, W., & Clark, D.B. (2006). Diary study of substance use and unsafe sex
among adolescents with substance use disorders. Journal of Adolescent Health, 38,
13-20.

Bandura, A. (1994). Social cognitive theory and exercise of control over HIV infection. In
Preventing AIDS: Theories and methods of behavioral interventions. AIDS
prevention and mental health, DiClemente, RJ, and Peterson, JL (Eds). New York:
Plenum Press, pp 25-29.

Baranowski, T., Perry, C.L., & Parcel, G.S. (1997). How individuals, environments, and
health behavior interact. Social Cognitive Theory. In Health Behavior and Health
Education, Glanz, K., Lewis, F.M., and Rimer, B.K. (Eds). San Francisco, CA:
Jossey-Bass Inc., pp 153-178.

Bentler, P. (1990). Comparative fit indexes in structural models. Psychological Bulletin.,
107, 238-246.

Brafford, L.J., & Beck, K.H. (1991). Development and validation of a condom self-efficacy
scale for college students. Journal of American College Health, 39, 219-225.

Brewster K.L., Cooksey, E.C., Guilkey, D.K., & Rindfuss, R.R. (1998). The changing
impact of religion on the sexual and contraceptive behavior of adolescent women in
the United States. Journal of Marriage and the Family, 60(2), 493-504.

Brown, B.B., Mounts, N., Lamborn, S.D., & Steinberg, L. (1993). Parenting practices and
peer group affiliation in adolescence. Child Development, 64, 467-482.

Brown, J.D., Ladin L’Engle, K., Pardun, C.J., Guo, G., Kenneavy, K., & Jackson, C. (2006).
Sexy media matter: exposure to sexual content in music, movies, television, and
magazines predicts Black and White adolescents’ sexual behavior. Pediatrics,
117(4), 1018-1027.

92
Brown, L.K., DiClemente, R.J., & Beausoleil, N.I. (1992). Comparison of human
immunodeficiency virus related knowledge, attitudes, intentions and behaviors
among sexually active and abstinent young adolescents. Journal of Adolescent
Health, 13, 651-657.

Browne, M., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. Bollen &
J. Long (Eds.), Testing structural equation models (pp. 136-162). Newbury Park:
Sage Publications.

Buzi, R.S., Tortolero, S.R., Roberts, R.E., Ross, M.W., Addy, R.C., & Markham, C.M.
(2003). The impact of a history of sexual abuse on high-risk sexual behaviors among
females attending alternative schools. Adolescence, 38(152), 595-605.

California State Office of AIDS, HIV/AIDS Case Registry Section (2005a). California AIDS
case surveillance report. Sacramento: California Department of Health Services.

California State Office of AIDS, HIV/AIDS Case Registry Section (2005b). California HIV
case surveillance report. Sacramento: California Department of Health Services.

California State Office of AIDS, HIV/AIDS Epidemiology Branch (2005c). FAQs
(Frequently Asked Questions) about HIV reporting. Sacramento: California
Department of Health Services.

Carvajal, S.C., Parcel, G.S., Basen-Engquist, K., Banspach, S.W., Coyle, K.K., Kirby, D., et
al. (1999). Psychosocial predictors of delay of first sexual intercourse by
adolescents. Health Psychology, 18(5), 443-452.

Centers for Disease Control and Prevention (1992). Sexual behavior among high school
students – United States, 1990. MMWR, 40, 885-888..

Centers for Disease Control and Prevention (1994). Health risk behaviors among persons
aged 12-21 years – United States, 1992. MMWR, 43, 231-235.

Centers for Disease Control and Prevention (1994a). Health risk behaviors among
adolescents who do and do not attend school – United States, 1992. MMWR, 43(8),
1-3.

Centers for Disease Control and Prevention (1994b). Sexual behaviors and drug among
youth in dropout-prevention program – Miami, 1994. MMWR, 43(47), 873-876.

Centers for Disease Control and Prevention (1999). Youth Risk Behavior Surveillance –
National Alternative High School Youth Risk Behavior Survey, United States, 1998.
MMWR, 48(SS07), 1-44.

Centers for Disease Control and Prevention (2000). Youth Risk Behavior Surveillance –
United States, 1999. MMWR, 49(SS05), 1-104.

93
Centers for Disease Control and Prevention (2001). HIV incidence among young men who
have sex with men – Seven U.S. cities, 1994-2000. Morbidity and Mortality Weekly
Report, 50(21), 440-444.

Centers for Disease Control and Prevention (2004). HIV/AIDS Surveillance Report, 2003,
Vol. 15. Atlanta: US Department of Health and Human Services, Centers for
Disease Control and Prevention.

Centers for Disease Control and Prevention (2005). HIV/AIDS Surveillance Report, 2004,
Vol. 16. Atlanta: US Department of Health and Human Services, Centers for
Disease Control and Prevention.

Centers for Disease Control and Prevention (2006a). Youth Risk Behavior Surveillance –
United States, 2005. MMWR, 55(SS-5).

Centers for Disease Control and Prevention (2006b). YRBSS National Youth Risk Behavior
Survey: 1991-2005, Trends in the prevalence of sexual behaviors. Available at
http://www.cdc.gov/HealthyYouth/yrbs/pdf/trends/2005_YRBS_Sexual_Behaviors.
pdf. Accessed July 11, 2006.

Centers for Disease Control and Prevention (2009). HIV/AIDS Surveillance Report, 2007,
Vol. 19. Atlanta: US Department of Health and Human Services, Centers for
Disease Control and Prevention.

Crockett, L. J., Raffaelli, M., & Shen, Y. L. (2006). Linking self-regulation and risk
proneness to risky sexual behavior: pathways through peer pressure and early
substance use. Journal of Research on Adolescence, 16, 503-525.

Darroch, J.E., Landry, D.J., & Oslak, S. (1999). Age differences between sexual partners in
the United States. Family Planning Perspectives, 31(4):160-167.

Deas-Nesmith, D, Brady, KT, White, R, & Campbell, S. (1998). HIV-risk behaviors in
adolescent substance abusers. Journal of Substance Abuse Treatment, 16(2), 169-
172.

DeHart, D.D., & Birkimer, J.C. (1997). Trying to practice safer sex: development of the
sexual risks scale. The Journal of Sex Research, 34, 11-25.

Dent, CW, Sussman, S., & Stacy, AW. (2001). Project Towards No Drug Abuse:
Generalizability to a general high school sample. Preventive Medicine, 32, 514-520.

Devieux, J, Malow, R, Stein, JA, Jennings, TE, Lucenko, BA, et al. (2002). Impulsivity and
HIV risk among adjudicated alcohol- and other drug-abusing adolescent offenders.
AIDS Education and Prevention, 14, Supplement B, 24-35.

DiClemente, R.J. (1991). Predictors of HIV-preventive sexual behavior in a high-risk
adolescent population: the influence of perceived peer norms and sexual
communication on incarcerated adolescents’ consistent use of condoms. Journal of
Adolescent Health, 12(5), 385-390.

94

DiClemente, R.J., Lodico, M., Grinstead, O.A., Harper, G., Rickman, R.L., Evans, P.E., et
al. (1996). African-American adolescents in high-risk urban environments do use
condoms: Correlates and predictors of condom use among adolescents in public
housing developments. Pediatrics, 98(2), 269-278.

DiIorio, C., Dudley, W.N., Kelly, M., Soez, J.E., Mbwara, J., & Sharpe Potter, J. (2001).
Social cognitive correlates of sexual experience and condom use among 13- though
15-year-old adolescents. Journal of Adolescent Health, 29, 208-216.

DuRant, R.H., Ashworth, C.S., Newman, C., & Gaillard, G. (1992). High school students’
knowledge of HIV/AIDS and perceived risk of currently having AIDS. Journal of
School Health, 62(2), 59-63.

Ellen, J.M., Boyer, C.B., Tschann, J.M., & Shafer, M.A. (1996). Adolescents’ perceived risk
for STDs and HIV infection. Journal of Adolescent Health, 18, 177-181.

Ellen, J.M., Gurvey, J.E., Pasch, L., Tschann, J., Nanda, J.P., & Catania, J. (2002). A
randomized comparison of A-CASI and phone interviews to assess STD/HIV-
related risk behaviors in teens. Journal of Adolescent Health, 31, 26-30.

Ellickson, P.L., Collins, R.L., Bogart, L.M., Klein, D.J., & Taylor, S.L. (2005). Scope of
HIV risk and co-occurring psychosocial health problems among young adults:
violence, victimization, and substance use. Journal of Adolescent Health, 36, 401-
409.

Elze, D.E., Auslander, W., McMillen, C., Edmond, T., & Thompson, R. (2001). Untangling
the impact of sexual abuse on HIV risk behaviors among youths in foster care. AIDS
Education and Prevention, 13(4), 377-389.

Ethier, K.A., Kershaw, T.S., Lewis, J.B., Milan, S., Niccolai, L.M., & Ickovics, J.R. (2006).
Self-esteem, emotional distress and sexual behavior among adolescent females:
inter-relationships and temporal effects. Journal of Adolescent Health, 38, 268-274.

Facente, A.C. (2001). Adolescents and HIV: knowledge, behaviors, influences, and risk
perceptions. Journal of School Nursing, 17(4), 198-203.

Fasula, A.M., & Miller, K.S. (2006). African-American and Hispanic adolescents’ intentions
to delay first intercourse: parental communication as a buffer for sexually active
peers. Journal of Adolescent Health, 38, 193-200.

Fishbein, M. (1990). AIDS and behavior change: An analysis based on the theory of
reasoned action. Interamerican Journal of Psychology, 24, 37-56.

Fishbein, M., & Coutinho, R. (1997). NIMH Multisite HIV Prevention Trial. AIDS, 11, S2.

Ford, K., & Norris, A.E. (1993). Urban Hispanic adolescents and young adults: relationship
of acculturation to sexual behavior. The Journal of Sex Research, 30(4), 316-323.

95
Galaif, E.R., Sussman, S., Chou, C., & Wills, T.A. (2003). Longitudinal relations among
depression, stress, and coping in high risk youth. Journal of Youth and Adolescence,
32(4), 243-258.

Galaif, E.R., Chou, C., Sussman, S., & Dent, C.W. (1998). Depression, suicidal ideation, and
substance use among continuation high school students. Journal of Youth and
Adolescence, 27(3), 275-299.

Garofalo, R., Wolf, R.C., Kessel, S., Palfrey, J., & DuRant, R.H. (1998). The association
between health risk behaviors and sexual orientation among a school-based sample
of adolescents. Pediatrics, 101(5), 895-902.

Goodman, E., & Cohall, A.T. (1989). Acquired immunodeficiency syndrome and
adolescents: knowledge, attitudes, beliefs, and behaviors in a New York City
adolescent minority population. Pediatrics, 84(1), 36-42.

Graham, J.W., Flay, B.R., Johnson, C.A., Hansen, W.B., Grossman, L., & Sobel, J.L.
(1984). Reliability of self-report measures of drug use in prevention research:
evaluation of the project SMART questionnaire via the test-retest reliability matrix.
Journal of Drug Education, 14, 175-193.

Greene, K., Krcmar, M., Walters, L.H., Rubin, D.L., Hale, J., & Hale, L. (2000). Targeting
adolescent risk-taking behaviors: the contributions of egocentrism and sensation-
seeking. Journal of Adolescence, 23, 439-461.

Grunbaum, JA, & Basen-Engquist, K. (1993). Comparison of health risk behaviors between
students in a regular high school and students in an alternative high school. Journal
of School Health, 63, 421-425.

Hallfors, D.D., Waller, M.W., Ford, C.A., Halpern, C.T., Brodish, P.H., & Iritani, B. (2004).
Adolescent depression and suicide risk: association with sex and drug behavior.
American Journal of Preventive Medicine, 27(3), 224-230.

Hallfors, D.D., Waller, M.W., Bauer, D., Ford, C.A., & Halpern, C.T. (2005). Which comes
first in adolescence – sex and drugs or depression? American Journal of Preventive
Medicine, 29(3), 163-170.

Harwell, T., Trino, R., Rudy, B., Yorkman, S., & Gollub, E. (1999). Sexual activity,
substance use, and HIV/STD knowledge among detained male adolescents with
multiple versus first admissions. Sexually Transmitted Diseases, 26(5), 265-271.

Henrich, C.C., Brookmeyer, K.A., Shrier, L.A., & Shahar, G. (2006). Supportive
relationships and sexual risk behavior in adolescence: an ecological-transactional
approach. Journal of Pediatric Psychology, 31(3), 286-297.

Hingson, R.W., Strunin, L., Berlin, B.M, & Heeren, T. (1990). Beliefs about AIDS, use of
alcohol and drugs, and unprotected sex among Massachusetts adolescents. American
Journal of Public Health, 80(3), 295-299.

96
Houck, C.D., Lescano, C.M., Brown, L.K., Tolou-Shams, M., Thompson, J., et al. (2006).
“Islands of risk:” subgroups of adolescents at risk for HIV. Journal of Pediatric
Psychology, 31(6), 619-629.

Hutchinson, MK, Jemmott, JB, Sweet Jemmott, L, Braverman, P, & Fong, GT. (2003). The
role of mother-daughter sexual risk communication in reducing sexual risk
behaviors among urban adolescent females: A prospective study. Society for
Adolescent Medicine, 33, 98-107.

Janz, NK, and Becker, MH. (1984). The Health Belief Model: A decade later. Health
Education Quarterly, 11, 1-47.

Jemmott, JB, Jemmott, LS, & Fong, GT. (1998). Abstinence and safer sex HIV risk-
reduction interventions for African American adolescents, JAMA, 279, 1529-1536.

Jessor, R. (1984). Adolescent development and behavioral health. In Behavioral Health:
Handbook of Health Enhancement and Disease Prevention, Matarazzo, JD, Weiss,
S, Herd, JA, et al (Eds). New York, NY: John Wiley, pp 69-90.

Kaiser, J., & Gruzelier, J.H. (1999). The Adolescence Scale (AS-ICSM): A tool for the
retrospective assessment of puberty milestones. Acta Pædiatrica, Supplement 429,
64-68.

Kalichman, S.C., Heckman, T., & Kelly, J.A. (1996). Sensation seeking as an explanation
for the association between substance use and HIV-related risky sexual behaviors.
Archives of Sexual Behavior, 25(2), 141-154.

Karofsky, P.S., Zeng, L., & Kosorok, M.R. (2000). Relationship between adolescent-
parental communication and initiation of first intercourse by adolescents. Journal of
Adolescent Health, 28(1), 41-45.

Kinsman, S.B., Romer, D., Furstenberg, F.F., & Schwarz, D.F. (1998). Early sexual
initiation: the role of peer norms. Pediatrics, 102(5), 1185-1192.

Kline, R. (1998). Principles and Practices of Structural Equation Modeling. New York, NY:
Guilford Press.

Koniak-Griffin, D., & Stein, J.A. (2006). Predictors of sexual risk behaviors among
adolescent mothers in a human immunodeficiency virus prevention program.
Journal of Adolescent Health, 38, e1-e11.

Kowaleski-Jones & Mott (1998). Sex, contraception, and childbearing among high-risk
youth: do different factors influence males and females? Family Planning
Perspectives, 30(4), 163-169.

Krantz, SR, Lynch, DA, & Russell, JM. (2002). Gender-specific profiles of self-reported
adolescent HIV risk behaviors. Journal of the Association of Nurses in AIDS Care,
13(6), 25-33.

97
Langer, L.M., & Tubman, J.G. (1997). Risky sexual behavior among substance-abusing
adolescents: psychosocial and contextual factors. American Journal of
Orthopsychiatry, 67(2), 315-322.

Leigh, B, & Stall, R. (1993). Substance use and risky sexual behavior for exposure to HIV:
issues in methodology, interpretation, and prevention. American Psychologist, 48,
1035-1043.

Li, X., Stanton, B., Cottrell, L., Burns, J., Pack, R., & Kaljee, L. (2000). Patterns of initiation
of sex and drug-related activities among urban low-income African-American
adolescents. Journal of Adolescent Health, 28, 46-54.

Liverpool, J., McGhee, M., Lollis, C., Beckford, M., & Levine, D. (2002). Knowledge,
attitudes, and behavior of homeless African-American adolescents: implications for
HIV/AIDS prevention. Journal of the National Medical Association, 94(4), 257-263.

Luster, T., & Small, S.A. (1997). Sexual abuse history and number of sex partners among
female adolescents. Family Planning Perspectives, 29(5), 204-211.

Lux, K.M., & Petosa, R. (1994). Preventing HIV infection among juvenile delinquents:
educational diagnosis using the health belief model. International Quarterly of
Community Health Education, 15, 145-163.

Manlove J. (1998) The influence of high school dropout and school disengagement on the
risk of school-age pregnancy. Journal of Research on Adolescence, 8:187-220.

Marín, B.V., Coyle, K.K., Gómez, C.A., Carvajal, S.C., & Kirby, D.B. (2000). Older
boyfriends and girlfriends increase risk of sexual initiation in young adolescents.
Journal of Adolescent Health, 27, 409-418.

Marín, G., Sabogal, F., Marín, B., Otero-Sabogal, R., & Perez, Stable, E.J. (1987).
Development of a short acculturation scale for Hispanics. Hispanic Journal of
Behavioral Sciences, 9, 183-205.

Mark, H.D., Sifakis, F., Hylton, J.B., Celentano, D.D., MacKellar, D.A., et al. (2005). Sex
with women as a risk factor for herpes simples virus type 2 among young men who
have sex with men in Baltimore. Sexually Transmitted Diseases, 32(11), 691-695.

Mauldon J, & Luker K. (1996) The effects of contraceptive education on method use at first
intercourse. Family Planning Perspectives, 28:19-24.

Mazzaferro, K.E., Murray, P.J., Ness, R.B., Bass, D.C., Tyus, N., & Cook, R.L. (2006).
Depression, stress, and social support as predictors of high-risk sexual behaviors and
STIs in young women. Journal of Adolescent Health, 39, 601-603.

McDermott Sales, J., Milhausen, R.R., Wingood, G.M., DiClemente, R., Salazar, L.F., &
Crosby, R.A. (2008). Validation of a parent-adolescent communication scale for use
in STD/HIV prevention interventions. Health Education & Behavior, 35(3), 332-
345; Epub 15 December 2006.

98

Millstein, S.G., & Moscicki, A.B. (1995). Sexually-transmitted disease in female
adolescents: effects of psychosexual factors and high risk behaviors. Journal of
Adolescent Health, 17, 83-90.

Miller, KS, Clark, LF, & Moore, JS. (1997). Sexual initiation with older male partners and
subsequent HIV risk behavior among female adolescents. Family Planning
Perspectives, 29(5), 212-214.

Morrison-Beedy, D., Carey, M.P., & Aronowitz, T. (2002). Psychosocial correlates of HIV
risk behavior in adolescent girls. Journal of Obstetric, Gynecologic, and Neonatal
Nursing, 32(1), 94-101.

Mott, F.L., & Haurin, R.J. (1988). Linkages between sexual activity and alcohol and drug
use among American adolescents. Family Planning Perspectives, 20, 128-136.

Mott, F.L., Fondell, M.M., Hu, P.N., Kowaleski-Jones, L., & Menaghan, E.G. (1996). The
determinants of first sex by age 14 in a high-risk adolescent population. Family
Planning Perspectives, 28(1), 13-18.

Murphy, D.A., Rotheram-Borus, M.J., & Reid, H.M. (1998). Adolescent gender differences
in HIV-related sexual risk acts, social-cognitive factors and behavioral skills.
Journal of Adolescence, 21, 197-208.

Murray, D.M., & Blitstein, J.L. (2003). Methods to reduce the impact of intraclass
correlation in group-randomized trials. Evaluation Review, 27(1), 79-103.

National Research Council (1987). Panel on adolescent pregnancy and child bearing. In
Risking the Future: adolescent sexuality, pregnancy, and child bearing, Hayes, CD
(Ed). Washington, DC: National Academy Press.

Newcomb, MD, Maddahian, E, Skager, R, & Bentler, PM. (1987). Substance abuse and
psychosocial risk factors among teenagers: Associations with sex, age, ethnicity, and
type of school. American Journal of Drug and Alcohol Abuse, 13, 413-433.

O’Donnell, L., Stueve, A., San Doval, A., Duran, R., Haber, D., et al. (1999). The
effectiveness of the Reach for Health Community Youth Service learning program
in reducing early and unprotected sex among urban middle school students.
American Journal of Public Health, 89(2), 176-181.

O’Donnell, L., O’Donnell, C.R., & Stueve, A. (2001). Early sexual initiation and subsequent
sex-related risks among urban minority youth: the Reach for Health study. Family
Planning Perspectives, 33(6), 268-275.

O’Donnell, L., Myint-U, A., O’Donnell, C.R., & Stueve, A. (2003). Long-term influence of
sexual norms and attitudes on timing of sexual initiation among urban minority
youth. Journal of School Health, 73(2), 68-75.

99
O’Hara, P., Messick, B.J., Fichtner, R.R., & Parris, D. (1996). A peer-led AIDS prevention
program for students in an alternative school. Journal of School Health, 66(5), 176-
182.

O’Hara, P., Parris, D., Fichtner, R.R., & Oster, R. (1998). Influence of alcohol and drug use
on AIDS risk behavior among youth in dropout prevention. Journal of Drug
Education, 28(2), 159-168.

Otto-Salaj, LL, Gore-Felton, C, McGarvey, E, & Canterbury, RJ (2002). Psychiatric
functioning and substance use: factors associated with HIV risk among incarcerated
adolescents. Child Psychiatry and Human Development, 33(2), 91-106.

Padian, N.S., Shiboski, S.C., & Jewel, N.P. (1991). Female-to-male transmission of human
immunodeficiency virus. Journal of the American Medical Association, 266, 1664-
1667.

Parson, J.T., Halkitis, P.N., Bimbi, D., & Borkowski, T. (2000). Perceptions of the benefits
and costs associated with condom use and unprotected sex among late adolescent
college students. Journal of Adolescence, 23, 377-391.

Patterson, G.R., DeBaryshe, B.D., & Ramsey, E. (1989). A developmental perspective on
antisocial behavior. American Psychologist, 44, 329-335.

Reitman, D., St. Lawrence, J.S., Jefferson, K.W., Alleyne, E., Brasfield, T.L., & Shirley, A.
(1996). Predictors of African American adolescents’ condom use and HIV risk
behavior. AIDS Education & Prevention, 8(6), 499-515.

Resnick, M., Bearman, P., Blum, R.W., Bauman, K.E., Harris, K.M., Jones, J., et al. (1997).
Protecting adolescents from harm: findings from the national longitudinal study of
adolescent health. Journal of the American Medical Association, 278(10), 823-832.

Rew, L., Fouladi, R.T., & Yockey, R.D. (2002). Sexual health practices of homeless youth.
Journal of Nursing Scholarship, 34(2), 139-145.

Rink, E., Tricker, R., & Harvey, S.M. (2007). Onset of sexual intercourse among female
adolescents: the influence of perceptions, depression, and ecological factors. Journal
of Adolescent Health, 41, 398-406.

Robbins, R.N., & Bryan, A. (2004). Relationships between future orientation, impulsive
sensation seeking and risk behaviors among adjudicated adolescents. Journal of
Adolescent Research, 19(4), 428-445.

Romer, D., Black, M., Ricardo, I., Feigelman, S., Kalijee, L., Galbraith, J. et al. (1994).
Social influences on the sexual behavior of youth at risk for HIV exposure.
American Journal of Public Health, 84(6), 977-985.

Romer, D., Stanton, B., Galbraith, J., Feigelman, S., Black, M., & Li, X. (1999). Parental
influence on adolescent sexual behavior in high-poverty settings. Archives of
Pediatrics & Adolescent Medicine, 153(10), 1055-1062.

100

Rosenbaum, E., & Kandel, D.B. (1990). Early onset of adolescent sexual behavior and drug
involvement. Journal of Marriage and the Family, 52, 783-798.

Rotheram-Borus, M.J., Marelich, W.D., & Srinivasan, S. (1999). HIV risk among
homosexual, bisexual, and heterosexual male and female youths. Archives of Sexual
Behavior, 28(2), 159-177.

Salazar, L.F., DiClemente, R.J., Wingood, G.M., Crosby, R.A., Harrington, K., Davies, S., et
al. (2004). Self-concept and adolescents’ refusal of unprotected sex: a test of
mediating mechanisms among African-American girls. Prevention Science, 5(3),
137-149.

Santelli, J.S., Kaiser, J., Hirsch, L., Radosh, A., Simkin, L., & Middlestadt, S. (2004).
Initiation of sexual intercourse among middle school adolescents: the influence of
psychosocial factors. Journal of Adolescent Health, 34(3), 200-208.

Shrier, L.A., Harris, S.K., Sternberg, M., & Beardslee, W.R. (2001). Associations of
depression, self-esteem, and substance use with sexual risk among adolescents.
Preventive Medicine, 33, 179-189.

Shrier, L.A., Harris, S.K., & Beardslee, W.R. (2002). Temporal associations between
depressive symptoms and self-reported sexually transmitted diseases among
adolescents. Archives of Pediatric and Adolescent Medicine, 156, 599-606.

Shrout, P.E., & Yager, T.J. (1989). Reliability and validity of screening scales: effect of
reducing scale length. Journal of Clinical Epidemiology, 42(1), 69-78.

Smith, C.A. (1997). Factors associated with early sexual activity among urban adolescents.
Social Work, 42(4), 334-346.

Substance Abuse and Mental Health Services Administration (2004). 2004 National Survey
on Drug Use and Health. Available at http://www.oas.samhsa.gov/nhsda.htm.
Accessed March 15, 2006.

Sikkema, KJ, Brondino, MJ, Anderson, ES, Gore-Felton, C, Kelly, JA, et al. (2004). HIV
risk behavior among ethnically diverse adolescents in low-income housing
developments. Journal of Adolescent Health, 35, 141-150.

Sneed, C.D., Morisky, D.E., Rotheram-Borus, M.J., Ebin, V., Malotte, C.K., Lyde, M.
(2001). “Don’t Know” and “didn’t think of it:” condom use at first intercourse by
Latino adolescents. AIDS Care, 13(3), 303-308.

Stacy, A. W., Ames, S. L., Sussman, S., & Dent, C. W. (1996). Implicit cognition in
adolescent drug use. Psychology of Addictive Behaviors,10(3), 190–203.

101
Stanton, B., Li, X., Black, M., Ricardo, I., Galbraith, J., Feigelman, S., et al. (1996).
Longitudinal stability and predictability of sexual perceptions, intentions, and
behaviors among early adolescent African-Americans. Journal of Adolescent
Health, 18, 10-19.

Staton, M., Leukefeld, C., Logan, T.K., Zimmerman, R., Lynam, D., Milich, R., et al.
(1999). Gender differences in substance use and initiation of sexual activity.
Population Research and Policy Review, 18, 89-100.

Stock, J.L., Bell, M.A., Boyer, D.K., & Connell, F.A. (1997). Adolescent pregnancy and
sexual risk-taking among sexually abused girls. Family Planning Perspectives, 29,
200-203, 227.

Strunin, L. Adolescents’ perception of risk for HIV infection: implications for future
research. Social Science & Medicine, 32(2), 221-228.

Stueve, A., & O’Donnell, L. (2005). Early alcohol initiation and subsequent sexual and
alcohol risk behaviors among urban youths. American Journal of Public Health,
95(5), 887-893.

Sussman, S. (2005). The relations of cigarette smoking with risky sexual behavior among
teens. Sexual Addiction and Compulsivity, 12, 181-199.

Sussman, S., Dent, CW, Simon, TR, Stacy, AW, Galaif, ER, et al. (1995a). Immediate
impact of social influence-oriented substance abuse prevention curricula in
traditional and continuation high schools. Drugs and Society, 8, 65-81.

Sussman, S., Stacy, AW, Dent, CW, Simon, TR, Galaif, ER, et al. (1995b). Continuation
high schools: Youth at risk for drug abuse. Journal of Drug Education , 25(3), 191-
209.Sussman, S. (1996). Development of a drug abuse prevention curriculum for
high risk youth. Journal of Psychoactive Drugs, 28, 169-182.

Sussman, S. (1996). Development of a drug abuse prevention curriculum for high risk youth.
Journal of Psychoactive Drugs, 28, 169-182.

Sussman, S., Dent, CW, Simon, TR, Stacy, A.W., & Craig, S. (1998). One-year outcomes of
Project Towards No Drug Abuse. Preventive Medicine, 27, 632-642.

Sussman, S., Simon, TR, Dent, CW, Steinberg, J.M., & Stacy, A.W. (1999). One-year
prediction of violence perpetration among high-risk youth. American Journal of
Health Behavior, 23(5), 332-344.

Sussman, S., Dent, CW, & Stacy, AW. (2002). Project Towards No Drug Abuse: A review
of the findings and future directions. American Journal of Health Behavior, 26(5),
354-365.

Thiede, H., Valleroy, L.A., MacKellar, D.A., Celentano, D.D., Ford, W.L., et al. (2003)
Regional patterns and correlates of substance use among young men who have sex
with men in 7 US urban areas. American Journal of Public Health, 93, 1915-1921.

102
Thompson, N.J., Sharpe Potter, J., Sanderson, C.A., Maibach, E.W. (1997). The relationship
of sexual abuse and HIV risk behaviors among heterosexual adult female STD
patients. Child Abuse & Neglect, 21(2), 149-156.

Turner, C.F., Ku, L., Rogers, S.M., Lindberg, L.D., Pleck, J.H., & Sonenstein, F.L. (1998).
Adolescent sexual behavior, drug use, and violence: increased reporting with
computer survey technology. Science, 280, 867-873.

Valdiserri, R.O., Lyter, D., Leviton, L.C., Callahan, C.M., Kingsley, L.A., Rinaldo, C.R., et
al. (1988). Variable influencing condom use in a cohort of gay and bisexual men.
American Journal of Public Health, 78(7), 801-805.

Valleroy, L.A., MacKellar, D.A., Karon, J.M, Janssen, R.S., & Hayman, C.R. (1998). HIV
infection in disadvantaged out-of-school youth: prevalence for U.S. Job Corps
entrants, 1990 through 1996. Journal of Acquired Immune Deficiency Syndromes &
Human Retrovirology, 19(1), 67-73.

Valleroy, L.A., MacKellar, D.A., Karon, J.M, Rosen, D.H., McFarland, W., et al. (2000).
HIV prevalence and associated risks in young men who have sex with men. Journal
of the American Medical Association, 284(2), 198-204.

Walter, H.J., Vaughan, R.D., Armstrong, B., Krakoff, R.Y., Maldonado, L.M., Tiezzi, L., et
al. (1995). Sexual, assaultive, and suicidal behaviors among urban minority junior
high school students. Journal of the American Academy of Child and Adolescent
Psychiatry, 34(1), 73-80.

Whaley, AR. (1999). Preventing the high-risk sexual behavior of adolescents: Focus on
HIV/AIDS transmission, unintended pregnancy, or both? Journal of Adolescent
Health, 24, 376-382.

Wingood, G.M., & DiClemente, R.J. (1998). Partner influences and gender-related factors
associated with noncondom use among young adult African-American women.
American Journal of Community Psychology, 26(1), 29-51.

Zuckerman, M., Kuhlman, D.M., Thornquist, M., & Kiers, H. (1991). Five (or three) robust
questionnaire scale factors of personality without culture. Personality and Individual
Differences, 12, 929-941.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Sociocultural stress, coping and substance use among Hispanic/Latino adolescents

PDF

A closer look at intention to use condoms and risky sexual behaviors among homeless people: application of the theories of reasoned action and planned behavior

PDF

Cultural risk and protective factors for tobacco use behaviors and depressive symptoms among American Indian adolescents in California

PDF

The link between maternal depression and adolescent daughters' risk behavior: the mediating and moderating role of family

PDF

Exploring the role of peer influence, linguistic acculturation, and social networks in substance use

PDF

Energy drink consumption, substance use and attention-deficit/hyperactivity disorder among adolescents

PDF

Stage of readiness and learning styles -- a tailored HIV/AIDS prevention intervention for youth detained in the Los Angeles County Juvenile Justice system

PDF

The dynamic relationship of emerging adulthood and substance use

PDF

Risky health behaviors and neurocognitive function among people living with HIV

PDF

Role transitions, past life events, and their associations with multiple categories of substance use among emerging adults

PDF

Childhood cancer survivorship: parental factors associated with survivor's follow-up care behavior and mental health

PDF

The Internet activities, gratifications, and health consequences related to compulsive Internet use

PDF

Motivation and the meanings of health behavior as factors associated with eating behavior in Latino youth

PDF

Anxiety symptoms and nicotine use among adolescents and young adults

PDF

Sex talks: an examination of young Black women's social networks, sexual health communication, and HIV prevention behaviors

PDF

An examination of the association between spousal support and type 2 diabetes self-management

PDF

The influence of contextual factors on the processes of adoption and implementation of evidence-based substance use prevention and tobacco cessation programs in schools

PDF

A penta-dimensional longitudinal analysis of the predictors of compulsive internet use among adolescents using linear mixed model (LMM)

PDF

Braver together: exploring the communicative and psychological experiences of LGBTQ youth and families

PDF

The role of dyadic and triadic factors on psychosocial wellbeing and healthcare interactions among childhood cancer survivors, parents, and medical providers

Asset Metadata

Creator Espinoza, Lilia (author)

Core Title Factors and correlates of sexual behaviors among young adults from continuation high schools

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Preventive Medicine (Health Behavior)

Publication Date 08/03/2009

Defense Date 06/17/2009

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag age at sexual onset,condom use,condom use self-efficacy,continuation high schools,depressive symptomatology,HIV risk,Latinos,OAI-PMH Harvest,parent-adolescent communication,perceived HIV risk,safer sex peer norms,sensation-seeking,sexual partners,substance use,youth

Place Name California (states)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Richardson, Jean R. (committee chair), Baezconde-Garbanati, Lourdes (committee member), Chou, Chih-Ping (committee member), Ferguson, Kristin M. (committee member), Stacy, W. Alan (committee member)

Creator Email lespinoz@usc.edu,lilespi@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2455

Unique identifier UC176709

Identifier etd-Espinoza-3075 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-172388 (legacy record id),usctheses-m2455 (legacy record id)

Legacy Identifier etd-Espinoza-3075.pdf

Dmrecord 172388

Document Type Dissertation

Rights Espinoza, Lilia

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu