Citation
Shared sexual health conversations

Material Information

Title:
Shared sexual health conversations intentions, behaviors, and strategies for change
Creator:
Kalinka, Christina Jayne ( author )
Language:
English
Physical Description:
1 electronic file (104 pages). : ;

Subjects

Subjects / Keywords:
Sexual health ( lcsh )
Sexually transmitted diseases ( lcsh )
Communication in sex ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Over 19 million new STDs infections are diagnosed each year, which costs the U.S. healthcare system $16.4 billion annually (CDC, 2010). This epidemic may be a result of, in part, the lack of sexual health communication between partners, as STD exposure history and current risky behaviors are not always disclosed before engaging in sexual activity. This study aims to use the Theory of Planned Behavior (TPB) to predict individuals intentions to engage in a Shared Sexual health Conversation (SSHC), reported SSHC behaviors with their last partner (i.e., depth of sexual health topics covered and ease with which participants engaged in the conversation), and the relation of SSHC behaviors to safer sex and healthy relationship outcomes. Data were collected from 196 sexually active undergraduate students. Structural equation modeling revealed that TPB predictors (attitudes, subjective norms, and perceived behavioral control) accounted for 53% of the variance of intentions to have a SSHC with any new sexual partner, and 67% of the variance of intention to have a SSHC with one s last sexual partner. Both types of intentions predicted the depth with which they talked about sexual health topics with their last partner; only intentions specific to their last sexual partner was able to predict the ease of having a SSHC with that partner (general intentions did not). The relation between having a SSHC before sex and using a condom was mixed. Whereas under some conditions the association was not significant, for individuals that reported having vaginal or anal sex at some point with their last partner results suggested that having a SSHC before their first sexual encounter was associated with a greater likelihood of condom use. Engaging in SSHC before sex was also associated with better relationship health, increased safety, and more positive feelings toward each individual before and after sex.
Thesis:
Thesis (Ph.D.)--University of Colorado Denver.
Bibliography:
Includes bibliographic references.
System Details:
System requirements: Adobe Reader.
General Note:
Department of Psychology
Statement of Responsibility:
by Christina Jayne Kalinka.

Record Information

Source Institution:
University of Colorado Denver
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
898038986 ( OCLC )
ocn898038986

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

SHARED SEXUAL HEALTH CONVERSATIONS: INTENTIONS, BEHAVIORS, AND STRATEGIES FOR CHANGE by CHRISTINA JAYNE KALINKA B.A., University of Michigan (Ann Arbor), 2004 M.A., Teachers College Columbia University, 2006 Ed.M., Teachers College Columbia University, 2 006 M.A., University of Colorado Denver, 2012 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Clinical Health Psychology 2014

PAGE 2

ii 2014 CHRISTINA JAYNE KALINKA ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Doctor of Philosophy degree by Christina Jayne Kalinka has been approved for the Clinical Health Psychology Program by Kristin Kilbourn, Chair Elizabeth Allen, Advis or Evelinn Borrayo Kevin Masters April 22 2014

PAGE 4

iv Kalinka, Christina Jayne (Ph.D., Clinical Health Psychology) Shared Sexual Health Conversations: Intentions, Behaviors, and Strategies for Change Dissertation directed by Assistant Professor Kristin Kilbour n. ABSTRACT Over 19 million new STDs infections are diagnosed each year which costs the U.S. healthcare system $16.4 billion annually (CDC, 2010) This epidemic may be a result of, in part, the lack of sexual health communication between partners as S TD exposure history and current risky behaviors are not always disclosed before engaging in sexual activity. Thi s study aims to use the Theory of Planned Behavior (TPB) to predict individuals' intentions to engage in a Shared Sexual Health Conversation (S SHC), reported SSHC behaviors with their last partner (i.e., depth of sexual health topics covered and ease with which participants engaged in the conversation), and the relation of SSHC behaviors to safer sex and healthy relationship outcomes Data were collected from 196 sexually active undergraduate students. Structural equation modeling revealed that TPB predictors ( attitudes, subjective norms, and perceived behavioral control ) accounted for 53% of the variance of intention s to have a SSHC with any ne w sexual partner and 67% of the variance of intention to have a SSHC with one's last sexual partner Both types of intentions predicted the depth with which they talked about sexual health topics with their last sexual partner; only i ntentions specific t o their last sexual partner was able to predict the ease of having a SSHC with that partner (general intentions did not) The relation between having a SSHC before sex and using a condom was mixed W hereas under some conditions the association was not si gnificant, for individuals that reported having vaginal or anal sex at some point with their last partner,

PAGE 5

v results suggested that having a SSHC before their first sexual encounter was associated with a greater likelihood of condom use Engaging in a SSHC before sex was also associated with better relationship health, increased safety, and more positive feelings toward each individual before and after sex. The form and content of this abstract are approved. I recommend its publication. Approved: Elizabet h Allen

PAGE 6

vi ACKNOWLEDGEMENTS It has been an honor to work with the University of Colorado's psychology department faculty and students to complete this dissertation I couldn't have done it without you I would also like to express sincere gratitude to my advisor Elizabeth Allen, who has been an incredible mentor during my graduate school journey ; m y stats guru, Krista Ranby whose delightful patience and wittiness made learning Mplus fun ; my undergraduate research assistant, Amanda Dunn, who spent coun tless hours helping me figure out complicated skip patterns and color coding into Qualtrics ; my fiancÂŽ Jon Bathgate, who has unconditionally supported my dreams and is simply the greatest man alive; our dog, Coco, the most loveable pup who will never pass up an opportunity to see what I am working on ; and my loving pare nts, John and Dac Kalinka, who taught me that confidence does not come from having all of the answers, but from being open to all of the questions.

PAGE 7

vii CONTENTS CHAPTER I. B ACKGROUND AND SIGNIF ICANCE ................................ ................................ ...... 1 Sexually Transmitted Disease Risk and C ommunication ................................ ............... 1 Sexually Transmitted D isease Self Disclosure Factors ................................ .................. 3 Theory o f Planned Behavior ................................ ................................ ........................... 7 Specific Aims ................................ ................................ ................................ ................ 12 II. METHOD ................................ ................................ ................................ .................... 15 Participants ................................ ................................ ................................ .................... 15 Design a nd Procedure ................................ ................................ ................................ ... 16 Measures ................................ ................................ ................................ ....................... 17 Data Analysis ................................ ................................ ................................ ................ 34 III. RESULTS ................................ ................................ ................................ .................. 38 Descriptive Data ................................ ................................ ................................ ............ 38 Correlates of Shared Sexual Health Conversations ................................ ...................... 41 Predictors of SSHC De pth and Ease: Structural Equation Modeling ........................... 44 SSHC Engagement and Condom Use ................................ ................................ ........... 50 SSHC Engagement and Healthy Relationship Qualities ................................ .............. 54 SSHC Engagement and Interpersona l Connectedness ................................ .................. 55 IV. DISCUSSION ................................ ................................ ................................ ............ 57 Overall Predictive Power of the TPB Models ................................ ............................... 58 Predictive Power within the General TPB Models ................................ ....................... 59 Pr edictive Power within the Last Partner S pecific TPB Models ................................ .. 60 Associations B etween SSHC Engagement and Condom Use ................................ ...... 62

PAGE 8

viii SSHCs, Healthy Relationship Qualities, and Interpersonal Connectedness ................. 63 Study Limitations ................................ ................................ ................................ .......... 66 Future Directions ................................ ................................ ................................ .......... 72 REFERENCES ................................ ................................ ................................ ................. 75 APPENDIX A. Participant Demographics ................................ ................................ ....................... 81 B. General Th eory of Planned Behavior (TPB) Questions ................................ .......... 82 General intentions ................................ ................................ ................................ ... 82 General attitudes ................................ ................................ ................................ ..... 82 General subjective norms ................................ ................................ ....................... 83 General perceived beh avioral control (PBC) ................................ ......................... 83 C. Partner Demographics ................................ ................................ ............................. 84 D. Shared Sexual Health Conversation (SSHC) Questions ................................ ......... 86 E. Last Partner Specific Theory of Planned Behavior (T PB) Questions ..................... 88 Last partner specific intentions ................................ ................................ .............. 88 Last partner specific attitudes ................................ ................................ ................ 88 Last partner specific subjective norms ................................ ................................ ... 88 Last p artner specific perceived behavioral control (PBC) ................................ ..... 89 F. Healthy Relationship Qualities (Before Sex) ................................ .......................... 90 G. Interpersonal Connectedness (Before Sex) ................................ ............................. 91 H. Healthy R elationship Qualities (After Sex) ................................ ............................ 92 I. Interpersonal Connectedness (After Sex) ................................ ................................ 93

PAGE 9

ix LIST OF TABLES TABLE 1. Items and R eliability C oefficients of Affective and Instrume ntal Attitudes to Engage in a SSHC ................................ ................................ ................................ 21 2. Items and Reliability Coefficients of Subjective Norms to Engage in a SSHC .. 23 3. Items and Reliability Coefficients of Perceived Behavioral Control to Engage i n a SSHC ................................ ................................ ................................ .................... 25 4. Items and Reliability Coefficients of Intentions to Engage in a SSHC ............... 26 5. Comparison Between HSPC Items and SSHC Items by Topic Area. .................. 29 6. Healthy Relations hip Quality Constructs, Ordered by Size of Factor Loading. ... 32 7. Items and Reliability Coefficients of Interpersonal Connectedness Measures. .... 34 8. Demographics and Characteristics of the Final Sample ( N = 196). ...................... 39 9. Means and Standard Deviations of the TPB Variables. ................................ ........ 41 10. Correlations Between TPB Variables. ................................ ................................ .. 41 11. Summary of Last Partner Specific TPB Analyses. ................................ ............... 50 12. SSHC Engagement and Condom Use: Penetrat ive Sex During First Encounter. 51 13. SSHC Engagement a nd Condom Use: Penetrative Sex at Some Point with Partner. ................................ ................................ ................................ ............................... 52 14. Means and Standardard Deviations o f Healthy Relationship Quality Variables. 54 15. Means and Standardard Deviations o f Interpersonal Connectedness Variables. .. 56

PAGE 10

x LIST OF FIGURES FIGURE 1. Azjen's Theory of Planned Behavior (TPB). ................................ .......................... 8 2. General TPB M odel for SSHC D epth. ................................ ................................ .. 45 3. General TPB M odel for SSHC E ase. ................................ ................................ .... 46 4. Hypothesized L ast P artner S pecific TPB M odel for SSHC D epth. ..................... 47 5. Modified L ast P artner S pecific TPB M odel for SSHC D epth. ............................. 48 6. Hypothesized L ast P artner S pecific TPB M odel for SSHC E ase. ........................ 49 7. Modified L ast P artner S peci fic TPB M odel f or SSHC E ase. ............................... 50 8. Fixed S hared S exual H ealth C onversation D efinition T extbox from O nline S urvey. ................................ ................................ ................................ .................. 71

PAGE 11

xi LIST OF ABBREVIATIONS AIDS Acquired Immunodeficiency Syndrome CDC Centers for Disease Control and Prevention HIV Human Immunodeficiency Virus HPSC Health Protective Sexual Communication HPV Human Papillomavirus PBC Perceived Behavioral Control SSHC Shared Sexual Health Conversation STD Sexually Transmitted Disease STI Se xually Transmitted Infection TPB Theory of Planned Behavior UCD University of Colorado Denver

PAGE 12

1 CHAPTER I BACKGROUND AND SIGNIFICANCE Sexually Transmitted Disease Risk and Communication In 2010, the Centers for Disease Control and Prevention reported th at sexually transmitted disease (STD) transmission and infection are major public health issues, as there are approximately 19 million new STD infection diagnoses each year which subsequently cost the U.S. healthcare system $16.4 billion annually ( Centers for Disease Control and Prevention, 2010 ) This epidemic may be a result of, in part, the lack of sexual health communication between partners about S TDs, including risk of infection, testing options, prevention methods, history of exposure, and current disease s tatus. It has long been known that STDs are often transmitted within close relationships (e.g., Seal, 1997) with data suggesting that STDs often go undetected and untreated eve n in close relationships, often because one partner is not fully aware of their partner's risky behavior and STD history ( Hernandez et al., 2008 ; Witte, El Bassel, Gilbert, Wu, & Chang, 2010 ) For example, one study of couples in a close relationship found a number of individuals were unaware that their partner recently committed infidelity in the relationship, injected drugs, received a recent STD diagnosis, or was diagnosed as HIV positive (Witte, et al., 2010) As another example, Desiderato and Crawford (199 5) s ampled 231 sexually active students and found that almost half who had previous sexual partners did not disclose this to a current partner and more than half failed to use condoms with this previous partner and did not disclose this to their current partne r O f the 22% of individuals in th is sample who w ere diagnosed with a STD in the past over one third did not disclose this to their current sexual partner. Unfortunately, individuals often use familiarity and a sense that they "just know" their partner as a basis of trust

PAGE 13

2 when considering their sexual partner's safety ( Masaro, Dahinten, Johnson, Ogilvie, & Patrick, 2008 ) However, individuals cannot assume sexu al health safety if they are only aware of their own actions; rather, the process of establishing sexual safety must include an explicit awareness of partners' current and past sexual risk factors Most guidelines ( e.g., Noar & Edgar, 2008 ; SIECUS National Guidelines Tas kforce, 2004) indicate that such awareness should be developed through open shared conversation s of sexual health between partners, as well as testing for STDs. These conversations should not only be candid regarding sexual risk factors, but should also ad dress the emotional climate of the relationship, including a sense of comfort, readiness, and lack of coercion toward engaging in sexual behaviors. Unfortunately, this type of conversation appears to be easier to discuss in theory than it i s to engage in on a day to day basis In 2002, a group of young college women was interview ed about the health protective sexual communication that did ( or did not ) happen with their most recent sexual partner prior to first intercourse ; the women shared that they did n ot usually talk about sexual health topics with their partners because it was uncomfortable divulging personal details to a partner they did not know very well ( Cleary, Barhman, MacCormack, & Herold, 2002 ) However, they said they would be more likely to talk about sexual health if they felt their partner would be open and willing to engage i n a conversation, but very few reported having a sexual partner that made them feel that way ( Cleary, et al., 2002 ) It seems that comfort level with a partner may be a key factor that may inhibit or facilitate individual willingness to initiate open communication around sexual health before engaging in sexual activity with another person

PAGE 14

3 The focus of t he current stu dy is to examine a number of factors related to engagement in shared sexual health conversations ( SSHCs) before engaging in sexual intimacy for the first time with a partner Catania, Binson, & Stone ( 1996 ) used a similar term, Health Protective Sexual Communication (HPSC), which refers to communication content that includes discussing sexually transmitted infections (STIs), human immunodeficiency virus (HIV), and birth control However, HPSC as defined in this way does not reference interpersonal aspects of feeling ready, willing, and comfortable with engaging in sexual behaviors. Adding to Catania et al.'s ( 1996 ) definition I c onceptualize SSHCs as communication between partners about previous and current drug use, sexual experiences, and risky behaviors, exposure to STDs, a sense of interpersonal safety regarding sex, and the use of preventative safe sex behaviors. My focus is the degree to which this type of conversation occurs before the first sexual encounter. Sexually Transmitted Disease Self Disclosure Factors To better understand how we can make SSHCs feasible and successful, it is important to consider what factors hav e currently been linked to increased STD self disclosure, which is a major component of SSHCs. A recent meta analysis on self disclosure discovered four themes related to whether or not individual s will disclose their drug and sexual history to a new part ner: 1) personal characteristics, 2) relationship type, 3) motivation, and 4) knowledge of the disease (Montgomery, Gonzales, & Montgomery, 2008).

PAGE 15

4 Personal Characteristics. The first theme emerging from the literature is that certain personal character istics may influence STD disclosure decisions. Research with HIV positive individuals found that individuals who disclose their HIV status tend to have high ethical and moral standards, be younger in age, more spiritual, or have participated in disclosure intervention groups ( Best, 2002 ; DeRosa & Marks, 1998 ; Knight et al., 2005 ; Simoni, Demas, Mason, Drossman, & Davis, 2005 ) Gender and gender role expectations have been hypothesized to have an influence on general disclosure behavior; for instance, men being le ss likely to self disclose because they are expected to appear tough, objective and emotionally unexpressive, and women being more likely to self disclose due to their more nurturing, comforting, and emotionally driven nature ( Jourard, 1971 ) However, studies using gender and gender role expectations as predictors of STD disclosure failed to sho w consistent evidence to support these hypotheses (e.g., Desiderato & Crawford, 1995; Green et al., 2003) Relationship Type. Self disclosure behaviors also appear to be highly dependent on the type of relationsh ip an individual has with his or her partner ; greater rates of disclosure were seen in more intimate relationships ( Fife & Weeks, 2010 ; Keller, von Sadovszky, Pankratz, & Hermsen, 2000 ) and serodiscordant relationships ( Wolitski, Bailey, O'Leary, Gomez, & Parsons, 2003 ) In a sample of individuals diagnosed with HPV, Keller et al. ( 2000 ) found that self disclosure was more likely to occur with steady partners (95% disclosure rate) than among those with casual partners (31% disclosure rate). Similarly, Perrin et al. ( 2006 ) found that after being diagnosed with HPV, 65% of people in serious relationships chose to disclose this to their current partner. Moreover, in a sample of HIV positive men, Bairan et al. ( 2007 ) found that the men tended to

PAGE 16

5 disclose their serostatus in long term, non casual, committed relationships but would often lie about their status to casual or short term sex partners. Among these men, there w as a consensus that disclosure should occur before sex at the beginning of a relationship, but many tended to wait until a few weeks into the relationship to disclose. This suggests a discrepancy between their attitude toward disclosure and their actual d isclosure behaviors, and that even though disclosure is more likely in a committed, long term relationship, the disclosure may take place after sexual activity has already begun. Motivation. In the context of disclosure, motivation is often categorized as either self or other focused. Self focused motivations are generally thought of as having a focus on the benefit to the self, and include disclosure in order to gain social support or letting out a "secret" to unburden one's self ( Duncan, Hart, Scoular, & Bigrigg, 2001 ; Frayley, 2002 ; Holt et al., 1998 ; Keller, et al., 2000 ; McCaffery, Waller, Nazroo, & Wardle, 2006 ) Other focused motivations are conceptualized as focused on the benefit (or punishment) of the partner or the relationship, and include disclosing for the purpose of educating others and fulfilling a self imposed duty to inform partners ( Green, et al., 20 03 ; Kahn et al. 2005 ; Keller, et al., 2000 ) desiring honesty and truth in their relationships ( Green, et al., 2003 ) feeling morally obligated to do so ( Keller, et al., 2000 ) or assigning blame to their partner for potentially giving them an STD ( Kahn, 2005 ) There are a lso motivations to not disclose, such as stigma, shame, fear of rejection, or to protect the other individual from the psychological stress that tends to accompany an STD diagnosis ( Waller, Marlow, & Wardle, 2007 ) Knowledge of the Disease Lastly, knowledge of the disease appears to impact disclosure behaviors. Numer ous studies have found that people often report not having

PAGE 17

6 adequate knowledge about STDs, so they are hesitant to disclose their disease status to partners. Additionally, if individual s ha ve an STD that presents asymptomatically or on an intermittent basi s, they tend to believe that they are not contagious all the time, which leads them not to disclose ( Green, et al., 2003 ; Keller, et al., 2000 ; McCaffery, et al., 2006 ) On the other hand, if individual s h ave an STD that presents more visibly and gets increasingly worse, they tend to feel forced to disclose be cause their symptoms are more salient ( Holt, et al., 1998 ; Sullivan, 2005 ) Related to th is, some individuals who are HIV positive have reported wanting to hide their positive serostatus for as long as possible, and only disclosing when the ir illness has forced them into hospitalization or when death is imminent ( Derlega, Greene, Petronio, & Gust, 2003 )

PAGE 18

7 Theory of Planned Behavior Although we know that there are various factors, such as personal characteristics, relationship type, motivations, and knowledge about the disease that relate to disclosure behavior (Montgomery et al., 2008), there is little known about how specific attitudes, norms, and perceived behavioral control relate to whether or not an individual will engage in an SSHC with a partner before sexual ac tivity The premise of the current study is that if individuals believe that talking about sexual health with their partner is a good thing (attitudes), see that people similar to them and important to them also believe that talking about sexual health is a good thing (subjective norms), and think that they will be able to engage in a sexual health conversation with their partner with relative ease because it is within their control (perceived behavioral control), then they will be higher in intentions to engage in a SSHC with their partner (behavioral intention) and ultimately be more likely to actually engage in an SSHC (behavior). These factors represent the general Theory of Planned Behavior (TPB) model (Azjen, 1985, 1991; Azjen & Madden, 1986; see Figu re 1 and described further below), which posits that if an individual has favorable attitudes about the behavior, believes in greater norms toward a behavior, and has greater perceived behavioral control toward engaging in the behavior, then the intention to perform that given behavior will be stronger, and this intention will predict greater probabil ity of engaging in the behavior.

PAGE 19

8 F igure 1 Azjen' s Theory of Planned Behavior (TPB). The TPB model has been used with various populations in health psychology and behavioral medicine and is widely accepted as an efficacious model of behavioral intentions and actual behavioral engagement In a meta analysis of 185 studies using TPB since 1997, TPB accounted for 27% o f the variance in predicting behavior and 39% of the variance in predicting intentions with a medium to large effect size ( Armitage & Conner, 2001 ) TPB has also been widely used to predict safe sex n egotiation intentions and behaviors ( Albarracin, Johnson, Fishbein, & Muellerleile, 2001 ) With regard to condom use, Albarracin et al. ( 2001 ) found that wh ereas attitudes, subjective norm, and perceived behavioral control all related to individuals intentions to use condom s attitudes had the strongest impact on predicting intentions. Additionally, after placing perceived behavioral control in a larger path model controlling for intention, they found that perceived behavioral control did not exert unique variance on condom use behavior Overall, these results su ggest that, for condom use, interventions that focus on changing an individual's attitudes surrounding condom use may be most effective. The goal of the current study seeks to address similar questions regarding the associations of TPB factors as well as the relative strengths of these associations to predict the construct

PAGE 20

9 of SSHCs. Below is a break down of the main pieces of the TPB model and how they currently align with SSHCs. Attitudes. Attitudes towards a behavior reflect the global positive or ne gative evaluations an individual has about performing a particular behavior; in general, the more favorable the attitude toward a behavior, the stronger intention an individual will have to perform it ( Armitage & Conner, 2001 ; Azjen, 1985 1991 ) As a logical extension, individuals who have negative attitudes towar d talking about issues such as STDs would theoretically be less likely to engage in conversations about STDs. A range of factors may underlie a negative attitude about having such a conversation For example, many people worry about disclosing their sexu al health history to a close partner, especially when they have incurable (although manageable) STDs, because there is a fear and stigma associated with STDs, and they are unsure about what kind of reaction they will receive from their partner and the pote ntial negative impact it could have on their sexual relationship ( McCaffery, et al., 2006 ; Waller, McCaffery, Nazroo, & Wardle, 2005 ) I nvestigating these findings from a qualit ative perspective, Cleary et al. (2002) interviewed a sample of young women who were in a relationship but had not yet had intercourse with their partner M any of the women reported t hat they rarely perceived a need to talk about sexual health issues if they knew they were already protecting themselves in another way, such as using birth control pills or condoms ( Cleary et al., 2002 ) They also found that these young women tended to believe everything that a potential partner told them, and based their decisions about safe sex on this information Unfortunately, people can be deceitful when it comes to STD disclosure ( Desiderato & Crawford, 1995 ) and may lie about infidelity behavi ors, even to those with whom they

PAGE 21

10 are in a close relationship ( Williams, 2001 ) Although this research suggests that some people may lie and be deceitful even if they are ostensibly having an SSHC engaging in a SSHC may serve to increase the probability that STD disclosure occurs for those who do not lie or may allow for other relevant sexual health infor mation and topics to be discussed thus people may have the chance to make a more informed sexual decision Following Carmack & Lewis Moss ( 2009 ) the current study measured both a ffective attitudes (whether an individual likes or dislikes the behavior) and instrumental attitudes (whether an individ ual believes the behavior is beneficial), as both have been shown to uniquely contribute to attitudes as a latent factor. Subjective Norms. Subjective norms reflect an individual's perception of the expectations of those similar and/or important to him or her to perform (or not to perform) a given behavior; in general, perceiving that significant people in one's life believe that one should engage in a particular behavior will make someone more likely to engage in that behavior ( Armitage & Conner, 2001 ; Azjen, 1985 1991 ) Therefore, we must take into considerati on the role of peers and other important influencing people in an individual's life to see if SSHCs are a priority among those circles. According to Cleary et al.'s ( 2002 ) qualitative interviews with young women, peers may play a big role in the reasons why women might not talk with their partners about sexual health. Young women in this study reported that they ofte n do not know how to talk about sexual health with a partner because they do not even talk about these personal issues with friends; wh ereas many aspects of sexuality are discussed with peers, health related behaviors and risks topics are not discussed wit h peers because those are "scary" topics. Moreover, the interviewed women said that they were not even sure if their peers discussed sexual

PAGE 22

11 health with their partners ( Cleary, et al., 2002 ) This suggests that the current perception surrounding sexual health is that it is somewhat taboo to b ring up to peers and partners. In this study, subjective norms were measured as an aggregate of i njunctive norms ( whether the behavior is approved by important other s) and descriptive norms ( whether the behavior is perf ormed by important others) Perceived Behavioral Control Perceived behavioral control (PBC) is the extent to which people feel they are able to enact a behavior and is an aggregate of self efficacy (how confident a person feels about being able to perform a behavior) and controllability (how much a person has control over a behavior). PBC has both direct effects on actual behavior and in direct effects through intentions. Depending on how strongly an intention is held, moving one's intention to actual behavior is at least partially determined by environmental and personal barriers ( Armitage & Conner, 2001 ) In general, individuals are more likely to form a behavioral intention to do something if they think that they are able to achieve it ( Bandura, 1977 ) The refore, it is important to gauge whether or not individuals feel they have adequate resources and opportunities to engage in SSHCs, and the anticipated impediments or obstacles that may prevent an SSHC from happening. That is we must evaluate the role th at self efficacy and controllability contribute to this process. Self efficacy. Cleary et al. ( 2002 ) found that m ost young women in their study did not know how to talk about sexual health (thus low self efficacy in terms of skills and/or ability) because they have not been exposed to adequate formal and informal sexual education experiences. They reported low conf idence in their communication skills, and as a result, felt it was a barrier to initiating health protective sexual

PAGE 23

12 conversations with their partner before having sex with them for the first time Recent evidence affirms that communicating about sexual he alth is still difficult for young women, such that only 26% of women going in for STD testing felt able to talk with their partner about the test ahead of time ( Friedman & Bloodgood, 2010 ) Controllability. Previous research suggests that young women feel they have limited control over bringing up sexual health conversations Cleary et al. ( 2002 ) found that very few women reported having a partner that was willing and comfortable to discuss sexual health issues with them, even though it was something that they wanted to talk with their partner about. This suggests potential obstacles within dyadic relationships that limit one's confidence and ability to engage in a SSHC. More research is needed to shed light on how influ ential PBC is in the process of deciding to engage in and actually following through with a SSHC. Intentions. Intentions are assumed to capture the various motivational factors influencing a particular behavior ( e.g., perceived attitudes, subjective norm s, and perceived behavioral control) and can indicate how much effort or willingness an individual will put forth to perform that behavior ( Azjen, 1991 ) One g oal of the current study is to gain a better understanding of predictors of intention s to engage in a SSHC and the relationship of intention s to SSHC behavior s Specific Aims The goal of this study is to improve scientific knowledge surrounding sexual h ealth communication and disclosure by illuminating how direct attitudes, subjective norms, and perceived behavioral control are related to intentions towards engaging in a

PAGE 24

13 SSHC, how these variables relate to SSHC behavior, and how SSHC behavior relates to condom usage, healthy relationship qualities, and interpersonal connectedness within the relationship. The following specific aims and hypotheses helped guide this study: Aim 1. To e xamine the relationship between the components of Azjen's Theory of Plan ned Behavior Model (e.g., direct attitudes, subjective norms, and perceived behavioral control) and determine if this model can adequately "predict" individuals' intentions to engage in a SSHC before sex and reported SSHC behaviors before sex, such as perc eived depth and ease of SSHC conversation topics. Sex in this project is defined as oral, vaginal, and/ or anal sex. Hypothesis 1 : More positive ratings of attitudes, subjective norms, and perceived behavioral control toward engaging in a SSHC before sex will be significantly related to (1) increased intentions to have a SSHC with new partners before sex and (2) recalled increased intentions of having a SSHC with his/her last partner before sex. Hypothesis 2 : Higher levels of intentions and perceived be havioral control toward engaging in a SSHC before sex will be significantly related to ( 1) g reater reported depth of SSHC topics discu ssed with his/her last partner before sex and ( 2 ) g reater reported ease of engaging in SSHC topics with his/her last partn er before sex Aim 2. Determine the relationship between when /whether an individual engages in an SSHC and self reports of condom ( or other barrier method ) usage healthy relationship qualities, and interpersonal connectedness within the relationship. H ypothesis 3 : Individuals who reported engaging in a SSHC before their first sex encounter with their last partner will report: (1) more frequent condom use, (2) increased healthy relationship qualities (before and after sex) and (3) increased interperson al

PAGE 25

14 connectedness within the relationship (before and after sex) compared to individuals who did not have a SSHC before their first sex enc ounter with their last partner.

PAGE 26

15 CHAPTER II METHOD Participants This study recruited 247 undergraduate men and women of any racial/ethnic background, who were enrolled in a psychology course at the University of Colorado Denver, and were at least 18 years old Participants were told that they needed to have access to a computer where they could privately answer survey questions. In exchange for their participation, participants were eligible to receive extra credit in their current psychology course, as determined by their instructor. Undergraduate students were a sample of interest and an important target population given their 1) potential ethnic/racial diversity and 2) their average age, which has been shown to correspond with high rates of sexual activity with new partners (Chandra, Mosher, Copen, & Sionean, 2011). The racial/ethnic breakdown of University of Colo rado Denver's (UCD) undergraduate students is comparable to Denver metropolitan and National estimates, with the exception of having a higher reported Asian population (UCD: 10%, Denver metropolitan: 3.2%, National: 4.2%) and lower reported African America n/Black population (UCD: 5%, Denver metropolitan: 12%, National: 12.9%). These numbers are based on the projected racial/ethnic breakdown from the U.S. Census (2000) and current racial/ethnic breakdown reported by the College of Liberal Arts and Sciences (CLAS) at UCD. However, all conclusions will be limited to a college population. As noted, the age of this population is also relevant The average age of a UCD undergraduate CLAS student is 23.6 yea rs old, which correspond with an increased likelihood of having engaged in sexual activity Chandra, Mosher, Copen and Sionean ( 2011 ) estimated th at 70% of females (75.8% of males) between the

PAGE 27

16 ages of 18 and 19 years old have engaged in sexual activity, and 87.7% of females (85.7% of males) between the ages of 20 and 24 years old have had some form of opposite sex sexual contact. Design and P roced ure All participants included in this study gave their informed consent and research was conducted in compliance with the university's Institutional Review Board Online self report surveys were used with participants to assess the constructs of interest The surveys were given at one time point, thus all data are cross sectional and will rely on "postdiction" modeling (as opposed to prediction). L imitations inherent to this design will be described in the discussion section Online data collection was chosen as there have been strong indications this method is suitable for sensitive research topics, such as sexual behaviors and disease transmission, because it is less prone to social desirability biases than other methods of data collection (see van Gel der, Bretveld, & Roeleveld, 2010) Collecting data online also helped reduce the possibility of misplacing paper and pencil survey responses and decreased opportunities for human data entry errors. Participants completed the online survey anonymously (wh ich has also been shown to be important in eliciting honest responses to sensitive questions) and were provided with a "proof of completion" page upon reaching the end of the survey to bring to their instructor for extra credit Survey response s were down loaded from the online survey provider, Qualtrics, and placed into a secure, password protected database. This database was then imported into SPSS version 21 ( IBM Corp, 2012) and MPlus version 7.0 ( MuthÂŽn & MuthÂŽn, 1998 2012 ) Before

PAGE 28

17 conducting statistical analyses raw data were checked to ensure accuracy of the download. Measures The measures were given in a fixed order and followed an appropriate skip logic pattern based on the respon ses prov ided by the participant The online questionnaire contained four sections The first section contained demographic questions about the participant and measured general TPB variables using any new sexual partner as the referent. Participants received t he following prompt, which was paired with specific instructions before each TPB construct: "The next questions relate to how you generally feel in relationships with sexual/romantic partners (e.g., girlfriends, boyfriends, spouses, etc.) There are no ri ght or wrong answers." Participants were also introduced to the concept of SSHCs A fixed text box with the definition of SSHC was placed on the right hand side of the online survey for participants to reference at any point during the survey Particip ants received the following prompt: "We'd also like to introduce you to a new term we'll be using throughout the survey called a Shared Sexual Health Conversation, or "SSHC" for short. SSHCs refer to the conversations we have with others about condoms, b irth control, STDs, drug use, history of sexual partners and experiences, and a sense of readiness to engage in sexual intimacy Throughout the survey in places where we use the term "SSHC" we will place a box on the right hand side of your screen, just l ike you see on this page, which will contain this definition for you in case you forget. The second section contained demographic questions about the participant's last partner where oral, vagi nal, or anal sex was considered or approached, "followed by a series of

PAGE 29

18 questions focused on Shared Sexual Health Conversation s (SSHC s ) Participants received the following prompt to identify this referent person of interest : "For the rest of the survey, we are interested in learning more about potential experience s people might have with another individual where oral, vaginal, or anal sex either occurred or was a real possibility This individual could have been a casual partner, like a hook up at a party, to someone more serious, such as someone you've recently d ated (or are still dating) Essentially, we want you to think of anyone that you felt that there: 1) was a real chance that you would have oral, vaginal, or anal sex with, OR 2) someone that you actually did have oral, vaginal, or anal sex with." The thir d and fourth sections continued to use the participant's last partner where oral, vaginal, or anal sex was considered or approached as the referent person of interest (see prompt above). S ection three contained questions measuring TPB variables regardin g this person before the first potential or actual sexual encounter using the following prompt: "We're curious about the SSHC experience you had with your last partner Let's start off with the some questions that focus on the period of time when you were considering or approaching having oral, vaginal, or anal sex with this person. S ection four contained questions focused on condom usage healthy relationship qualities, and interpersonal connectedness within the relationship before and after sexual activ ity Demographics. Basic participant characteristics were collected, such as age, sex, race family economic status, and sexual history (e.g., whether or not they have ever engaged in oral, vaginal, and/or anal sex, number of past partners, and first sex ual encounter with his or her last partner) For the purposes of this study, only participants that reported being sexually active (i.e., having engaged in oral, vaginal, and/ or anal sex with another partner) were included in the analyses. Participant's sexual orientation was

PAGE 30

19 assessed using the 7 item Continuum of Sexual Orientation Scale, a widely used scale assessing a person's same versus other sex orientation ( Kinsey, Pomeroy, & Martin, 1948 ; Kinsey, Po meroy, Martin, & Gebhard, 1953 ) Participants were asked to provide information about their relationship status before their first sexual encounter with their last partner using the following check all that apply" choices: "o ne night stand ," "i nfatua ted with each other ( lust and physical attraction were definite driving fo rces), "relaxed" ( in the proces s of getting to know each other a nd not rushing into anything), "c asual friends ( friend s but never hooked up before), "f riends with benefits ( casual ly hooked up from time to time), "d ating ( had already been on a few dates together), "exclusive" (only dating each other), "l ove birds ( considered ourselves actively falling in love), and/or "c ommitted ( in a s erious, committed relationship). To create a more meaningful variable that captured general relationship status before first sexual encounter I hand coded the above choices into 4 main categories: one night stand (if they selected "one night stand"), casual, non exclusive dating (if they selected a ny combination of "infatuated with each other," "relaxed," "casual friends," or "friend s with benefits"), exclusively dating (if they selected "dating" or "exclusive" and any combination of other categories), and serious and committed relationship (if they selected "committed" and any combination of other categories). There were no participants that only selected "love birds," so I considered th eir other responses to place them into an appropriately fitting category. Participants also provided a "YES" or NO" response to whether they were sexual ly active with other people at the time of their first sexual encounter with their last

PAGE 31

20 partner Those that responded, "YES" received a follow up question about whether their last partner was aware of this using the following prompt: D id you r partner know that you were sexually active with others at that time [ referencing before oral sex/vaginal sex/anal sex ]? Participants could then select either "Yes my partner knew I had concurrent sexual partners at that time ," or "No my partner did not know I had other concurrent sexual partners at that time". Theory of Planned Behavior Measures. The Theory of Planned Behavior (TPB) measures were created for this study in accordance with the guidelines from Azjen (1991) and Francis et al. (2004). Attitudes measure. Seven items were developed to measure a person's general attitudes toward engaging in a SSHC with any new sexual partner (see Table 1) An additional parallel seven items were developed to measure a person's r ecollections of last partner specific attitudes toward engaging in a SSHC, using his or her last partner where sex was considered or approached as the referent. P articipants were asked to rate 3 pairs of affective attitude words (i.e., the degree to which a behavior is thought to be comfortable ) and 4 pairs of instrumental attitude words (i.e., the degree to which a behavior is thought to be beneficial ), on a 7 point scale Consideration was given to whether attitudes should be used as a stand alone indica tor variable (using the mean across all affective and instrumental attitude items) or as a latent variable containing two separate factors (affective attitudes and instrumental attitudes) Confirmatory factor analyses were conducted in MPlus to compare th e 1 factor model (all affective and instrumental attitude items combined as a measured factor ) to the 2 factor model (attitudes as a latent variable containing 2 measured factors

PAGE 32

21 affective attitudes and instrumental attitudes) Since these models are nes ted, a chi squared difference test was conducted between the 1 factor ( general 2 = 258.58 df = 14; last partner specific 2 = 125.98, df = 14 ) and 2 factor models (general 2 = 35.75, df = 13; last partner specific 2 = 29.86, df = 13) The results of t hese tests suggested that we reject the simpler 1 factor model and utilize the more specific 2 factor model Therefore, I concluded that separate composite scores for affective attitude items and instrumental attitude items would be the most appropriate wa y to represent direct attitudes. Four c omposite attitude scores which had adequate to excellent internal consistency were created : general affective ( = .83) general instrumental ( = .89) last partner specific affective ( = .74) last partner specific instrumental ( = .92) ; s ee Table 1 below Higher affective composite scores indicate more positive affective attitudes toward engaging in the SSHC before sex; positive affective attitudes, in this case, generally refer to more comfort regarding the SSHC. H igher instrumental composite scores indicate more perceived benefits of engaging in the SSHC before sex. Tab le 1 Items and R eliab ili ty C oefficie nts of A ffective and Instrumental A ttitudes to E ngage in a SSHC. Variable Item General affective a ttitudes For me, having an SSHC with any new sexual partner before oral, vaginal, or anal sex is generally .83 anxiety provoking comfortable unnatural natural something that ruins the mood something that sets the mood General instrumental a ttitudes For me, having an SSHC with any new sexual partner before oral, vaginal, or anal sex is generally .89 b eneficial /harmful + worthless/useful necessary/unnecessary + unimportant/important

PAGE 33

22 Table 1 (cont.) Varia ble Item Last p artner specific affective a ttitudes For me, I thought that having an SSHC with this person would be .74 anxiety provoking comfortable unnatural natural something that ruins the mood something that sets the mood Last p artner spe cific instrumental a ttitudes For me, I tho ught that having an SSHC with this person would be .92 b eneficial /harmful + worthless/useful necessary/unnecessary + unimportant/important Note: Attitude items were rated on a 1 7 scale with the le ft side word of the pair receiving a score of 1 and the right side word of the pair receiving a score of 7. In reference to the last partner he or she considered/approached having oral, vaginal, or anal sex with. + Reverse scored item. Subjective norms measure Three items were developed to measure a person's social group's perceived general beliefs toward engaging in a SSHC with a ny new sexual partner An additional parallel three items were developed to measure person's social group's perceived last partner specific beliefs toward engaging in a SSHC using his or her last partner where sex was considered or approached as the referent P articipants rated two injunctive norms (i.e., whether the behavior is app roved by important others ; that is, others think respondent should engage in the behavior ) and one descriptive norm (i.e., whether the behavior is performed by peers ), on a scale of 1 to 7 (1= strongly disagree, 7 = strongly agree). Two composite subjective norms scores each with good internal consistency were created : general subjective norms ( = .81) and last partner specific subjective norms ( = .82); see Table 2 below Higher composite scores indicate greater perceived social pressure to engage in an SSHC be fore sex

PAGE 34

23 Table 2 Items and Reliability C oefficients of Subjective N orms to E ngag e in a SSHC. Variable Item General subjective norms General i njunctive + descriptive norms (combined) .81 Injunctive norms Most people who are important to me think that people should have a SSHC with any new sexual partner before oral, vaginal, or ana l sex. It is expected of me that I should have an SSHC with any new sexual partner before oral, vaginal, or anal sex Descriptive norms Most people who are like me have an SSHC with any new sexual partner before oral, vaginal, or anal sex. Last pa rtner specific subjective norms Last p artner specific injunctive + descriptive norms (combined) .82 Injunctive norms Most people who are important to me probably think that I should have had an SSHC with this person before oral, vaginal, or anal sex with him or her. In this situation it would have been expected of me to have an SSHC before oral, vaginal, or anal sex with this person Descriptive norms Most people who are like me to me would have had an SSHC with this person befor e oral, vaginal or anal sex. Note: Subjective norm items rated on a 1 7 scale (1 = strongly disagree 7 = strongly agree ) In reference to the last partner he or she considered/approached having oral, vaginal, or anal sex with. Perceived behavioral c ontrol (PBC) meas ure Seven items were developed to measure general PBC toward engaging in a SSHC with any new sexual partner An additional parallel seven items were developed to measure recollections of last partner specific PBC toward engaging in a SSHC using the re spondent's last partner where sex was considered/ approached as the referent P articipants were asked to rate how much self efficacy (i.e., an individual's sit uation specific self confidence ) and controllability (i.e., the extent to which an individual has control ) they experience toward engaging in an SSHC on a scale of 1 to 7 (1= strongly disagree, 7 = strongly agree). Consideration was given to whether perceived behavio ral control should be used as a stand alone indicator variable (using the mean acros s all self efficacy and controllability items) or as a latent variable containing two separate factors (self efficacy and controllability). Reliability analyses indicated that the general PBC scale as a whole

PAGE 35

24 had adequate internal consistency ( = .75) although the last partner specific PBC was just short of the cutoff for adequacy at = .68. Whereas deleting one item from the last partner specific scale did raise the level to just above the cutoff, I decided to retain the whole set of items g iven how close the overall alpha was to .70 and to retain the parallel structure between the general and the last partner specific set. Exploratory factor analyses using 2 fixed factors indicated that items for self efficacy and controllability were cross loaded and the scree plot suggested that there was only 1 underlying factor. C onfirmatory factor analyses were conducted in MPlus to compare the 1 factor model ( all self efficacy and controllability items combined ) to the 2 factor model ( perceived behavi oral control as a latent variable containing 2 factors self efficacy and controllability ). Since these models are nested, a chi squared difference test was conducted between the 1 factor ( general 2 = 50.71 df = 14 ; last partner specific 2 = 114.63, df = 14 ) and 2 factor models ( general 2 = 49.07, df = 13; last partner specific 2 = 112.00 df = 13 ) The results suggested that we reject the more specific 2 factor model and utilize the simpler 1 factor model Therefore, I conclude d that perceived beha vioral control would be best represented by a unitary composite containing both self efficacy and controllability items. Two composite PBC scores were created : general PBC ( = .75) and last partner specific PBC ( = .68) ; s ee Table 3 below Higher comp osite scores indicate higher perceived control over engaging in an SSHC before sex

PAGE 36

25 Table 3 Items and Reliability C oefficients of Perceived Behavioral C ontrol to E ngage in a SSHC Variable Item General PBC General self efficacy + controllability (c ombined) .75 Self efficacy In general, having an SSHC with any new sexual partner before oral, vaginal, or anal se x is ( 1 = easy, 7 = difficult ) + I am generally confident that I can have a SSHC with any new sexual partner before oral, vaginal, or anal sex. I do not know the best way to have a SSHC with a new sexual partner before oral, vaginal, or anal sex + Controllability The decision to have a SSHC with any new sexual partner before oral, vaginal, or anal sex is beyond my control + Whether o r not I have a SSHC with any new sexual partner before oral, vaginal, or anal sex is entirely up to me. New sexual partners make it hard to have a SSHC before oral, vaginal, or anal sex + New sexual partners won't let me have a SSHC before oral, vagi nal, or anal sex + Last partner specific PBC Last partner specific self efficacy + controllability (combined) 68 Self efficacy Having an SSHC with this person before oral, vaginal, or anal sex was (1 = easy, 7 = difficult) + I was confident that I could have a SSHC with this person before oral, vaginal, or anal sex. I tried to have a SSHC with this person but I didn't know the best way to do it + Controllability The decision to have a SSHC with this person before oral, vaginal, or anal s ex was beyond my control + Whether or not I had a SSHC with this person before oral, vaginal, or anal sex was entirely up to me. I tried to have a SSHC with this person before oral, vaginal, or anal sex, but they made it hard for me to do + I t ried to have a SSHC with this person but they wouldn't let me go there + Note: PBC items rated on a 1 7 scale (1 = strongly disagree, 7 = strongly agree ; exceptions listed). In reference to the last partner he or she considered/approached having ora l, vaginal, or anal sex with. + Reverse scored item Generalized intentions measure Six items were used to measure general intentions toward engaging in a SSHC with any new sexual partner. An additional six items were used to measure recollections of last partner specific intentions toward engaging in a SSHC using the respondent's last partner where sex was considered or

PAGE 37

26 approached as the referent All it ems were rated on a 1 to 7 scale (1= strongly disagree, 7 = strongly agree) Two composite inte ntions scores, each with excellent internal consistency, were created: general intentions ( = .95) and last partner specific intentions ( = .96); see Table 4 below. Higher c omposite scores indicate increased intentions to engage in an SSHC before sex. Table 4 Items and Reliability Coefficients of Intentions to Engage in a SSHC. Variable It em General intentions I always expect to have a SSHC with any new sexual partner before oral, vaginal, or anal sex. .95 I always want to make sure that I have a SSHC with any new sexual partner before oral, vaginal, or anal sex. I always have a SSH C with any new sexual partner before oral, vaginal, or anal sex. I plan to have a SSHC with any new sexual partner before we engage in oral, vaginal, or anal sex. I will have a SSHC with any new sexual partner before we engage in oral, vaginal, or an al sex. I am going to have a SSHC with any new sexual partner before we engage in oral, vaginal, or anal sex. Last p artner specific intentions I expected that I would have a SSHC with this person before oral, vaginal, or anal sex. .96 I was sure th at I would have a SSHC with this person before oral, vaginal or anal sex I fully intended to have a SSHC with this person before oral, vaginal, or anal sex. I planned to have a SSHC with this person before oral, vaginal, or anal sex. I meant to have a SSHC with this person before oral, vaginal, or anal sex. I was going to have a SSHC with this person before oral, vaginal, or anal sex. Note: Intention items r ated on a 1 7 scale (1 = strongly disagree 7 = strongly agree ) In reference to the last partner he or she considered/approached having oral, vaginal, or anal sex with. + Reverse scored item. Shared Sexual Health Conversations Measure (SSHCs) To measure shared sexual health conversations ( SSHCs) that may occur between partners I c reated a 25 item SSHC scale partly based on questions from the Health Protective Sexual Communication

PAGE 38

27 Scale ( Catania, Coates, & Kegeles, 1994 ; Catania, Coates, Kegeles, et al., 1992 ; van der Straten, Catania, & Pollack, 1998 ) The original 10 item HPSC asked people to rate a list of things they t alk about before they have sex with each other for the first time using a 1 to 4 scale (1 = never 4 = always ), and contained questions about condom use, birth control, past sexual experiences, waiting to get to know one another before sex, getting tested for AIDS, drug use, and previous exposure to venereal diseases Previous studies have shown that compared to respondents with high HPSC scores, respondents with low HPSC scores are more likely to engage in to high risk sexual behaviors such as having mu ltiple partners, using alcohol before sex, decreased condom use and condom negotiation skills, and reduced frequency of STD testing ( Catania, et al., 1994 ; Dolcini, Coates, Catania, Kegeles, & Hauck, 1995 ) Those with higher HPSC scores were somewhat more likely to be Black than Hispanic, and were almost three times more likely to be women than men ( Catania, Coates, Stall, et al., 1992 ) While the HPSC provides helpful information about how likely people are to talk about certain sexual health topics, it was unable to provide the level of depth and ease with which they we re abl e to discuss the topics. The scale also had a limited number of items measuring previous sexual health testing, risky behaviors, and interpersonal safety Therefore, to meet the goals of this study, it was necessary to modify the existing HPSC measure by changing the response scale and creating additional items to better capture the desired range of SSHC engagement topics This modified scale, which I refer to as the "SSHC scale", asked people to rate the depth (0 = not at all, 7 = we discussed this topi c extensively) and ease ( after reverse scoring, 0 = did not discuss at all, 1 = difficult to talk about, 2 = somewhat difficult to

PAGE 39

28 talk about, 3 = somewhat easy to talk about, and 4 = easy to talk about ) with which they talked about various sexual health t opics before engaging in oral, vaginal, and/ or anal sex with their last partner. As mentioned in the demographics section, participants indicated whether they sex with their last partner, as well as the first sexual encounter they had with their last partn er. Based on their responses, the online survey would generate a full set of SSHC questions for each type of sexual encounter they considered or approached with their last partner. Participants were prompted with the following to isolate the referent sex ual encounter, as marked by the [bracketed text]: "Here is a list of some things that people might talk about as they are considering/approaching sex with someone new. We call talking about these kinds of things a "Shared Sexual Health Conversation" or S SHC for short. BEFORE [oral sex/vaginal sex/anal sex] with your last partner, we'd like to know" At the end of each set of SSHC questions, participants were asked to indicate whether they thought they engaged in a SSHC with their last partner before that specific sexual encounter. Specifically, they were asked to select "YES" or "NO" after reading the following question: "Given the nature of SSHC topics above, do you think that you and your last partner most likely had some sort of SSHC before consideri ng or approaching [oral sex/vaginal sex/anal sex]?" Using the participant responses to the question above and their report of the first sexual encounter they had with their last partner, I created a binary variable to represent whether or not the participa nt reported engaging in a SSHC before their first sexual encounter with their last partner Two composite scores each with excellent internal consistency

PAGE 40

29 were also created : SSHC depth ( = .94 ) and SSHC ease ( = .96 ) Higher SSHC depth composite scores indicated greater depth coverage of the SSHC topics with their last partner before their first sexual encounter Higher SSHC ease composite scores indicated greater ease with which they were able to talk about the SSHC topics with their last partner before their first sexual encounter See Table 5 below for a comparison of HPSC and SSHC scale questions by topic area. Table 5 Comparison Between HSPC Items and SSHC Items by Topic A rea H ealth Protective Sexual Communication Scale H PSC (10 items ) Shared Sexual Health Conversation Scale SSHC (25 items ) Condoms and Birth Control 1. Asked a new sex partner how (he/she) felt about using condoms before you had intercourse. 2. Told a new part ner that you won't have sex unless a condom is used. 3. Talked to a new sex partner about birth control before having sex for the first time. 1. Shared feelings about condom use. 2. Shared feelings about using birth control, other than condoms. Previous Partner History 4. Asked a new sex partner about the number of past sex partner (he/she) had. 5. Told a new sex partner about the number of sex partners you have had. 6. Talked about whether you or a new sex partner ever had homosexual experiences. 3. Num ber of past sexual partners we each have had. 4. Sexual experiences we have had with the opposite sex. 5. Sexual experiences we have had with the same sex. STD Testing 7. Discussed with a new sex partner the need for both of you to get tested for the AID S virus before having sex. 6. The most recent time that we have been tested for STDs/STIs. 7. The need for us both to get tested for STDs/STIs. 8. The most recent time that we have been tested for HIV/AIDS. 9. The need for us both to get tested for HIV/A IDS. Current & Previous STD Exposure 8. Asked a new sex partner if (he/she) has ever had some type of venereal disease (VD), like herpes, clap, syphilis, or gonorrhea. 10. If we currently have any STDs or STIs. 11. If we have ever had or been intimate w ith someone who, at the time, had a STD or STI. (cont. on next page )

PAGE 41

30 Table 5 (cont.) Health Protective Sexual Communication Scale HPSC (10 items) Shared Sexual Health Conversation Scale SSHC (25 items) Current & Previous STD Exposure (cont.) 12. If we currently have HIV/AIDs. 13. If we have ever had or been intimate with someone who, at the time, had HIV/AIDs. Drug Use 9. Asked a new sex partner if (he/she) ever shot drugs like heroin, cocaine, or speed. 14. Injection drug use/experiences we have had (e.g., heroin, cocaine, speed, methamphetamine). 15. Recreational drug use/experiences we have had (e.g., marijuana, prescription medication misuse, cocaine, salvia, ecstasy). Unprotected Sex (none) 16. Shared feelings about unprotected sex. 17. Unprote cted sex encounters we have had. Emotional Safety 10. Talked with a new sex partner about not having sex until you have known each other longer. 18. The importance of getting to know one another on an emotional level before going any further. 19. Waiting to have sex until we have known each other longer. 20. Things we are nervous about surrounding sex. 21. Things that make us feel physically safe in a relationship. 22. Things that make us feel emotionally safe in a relationship. 23. What our comfort level is engaging in oral/vaginal/anal sex. 24. What our comfort level is engaging in other kinds of sexual intimacy. 25. Each other's expectations of what it would mean for our relationship if we decided to have oral/vaginal/anal sex. Note: STD = sexually tra nsmitted disease. STI = sexually transmitted infection. HIV = human immunodeficiency virus. AIDS: acquired immunodeficiency disease. HPSC = Health Protective Sexual Communication Scale (Catania et al., 1992; Catania et al., 1994) and is measured on a s cale from 1 4 (1 = always, 4 = never). SSHC = Shared Sexual Health Conversation (developed for this study). SSHC Depth measured on a scale from 0 7 (0 = not at all, 7 = we discussed this topic extensively) and SSHC Ease measured on a scale from 0 4 (afte r reverse scoring, 0 = did not discuss at all, 1 = difficult to talk about, 2 = somewhat difficult to talk about, 3 = somewhat easy to talk about, 4 = easy to talk about).

PAGE 42

31 Condom/ Protective Barrier Usage Measure Participants who reported having penetrat ive sex (i.e., vaginal or anal sex) with their last partner were asked to select whether or not they (or their partner) d ecided to use a condom or other barrier method of protection before sex Healthy Relationship Qualities. Twelve items were u sed to assess participant s self report of healthy relationship qualities before their first sexual encounter and after the sexual encounter (if applicable), using the last partner where sex was considered or approached as the referent All items were rated on a 1 to 7 scale (1 = not at all true, 7 = extremely true). Participants that reported having sex received specific "before sex" and "after sex" prompt s Participants that indicated having an SSHC before sex were prompted with the following: Before sex: After I had (or tried to have) the SSHC with my last partner, but before I actually had sex with him/her I felt... After sex: After having oral, vaginal, and/or anal sex with my last partner I felt..." Participants that indicated that they did NOT have a n SSHC before sex were prompted with the following: Before sex: Before I actually had oral, vaginal, and/or anal sex with my last partner, I felt..." After sex: "After having oral, vaginal, and/or anal sex with my last partner I felt..." Exploratory fact or analyses using principal factor extraction indicated that there were 3 extracted factors, which explained approximately 77% of response variance. A d irect oblimin method (oblique rotation) was requested to determine factor loadings for interpretation; t his method was chosen because I expected the factors to correlate and wanted to minimize the covariance of the squared loadings in each factor's distinct column. With a cutoff of .59 for inclusion of a variable in interpretation of a factor, all

PAGE 43

32 variable s loaded onto one of three distinct factors. These factors were conceptualized as relationship health ( 57% of variance), relationship safety ( 10% of variance), and relationship certainty (10% of variance). See Table 6 for an outline of factors and corres ponding variables, which are ordered and grouped by loading size to facilitate interpretation. Table 6 Healthy Relationship Quality Constructs, Ordered by Size of Factor L oading Factor 1: Relationship Health Factor 2: Relationship Safety Factor 3: Relat ionship Certainty The relationship was I felt (with him or her) I felt fun .97 sexually safe .80 conflicted about the relationship + .77 enjoyable .91 physically safe .80 worried about the relationship + .74 exciting. .83 emotionally safe .8 0 intimate. .69 comfortable. .66 not really worth it + .60 healthy .59 Note: Variables rated on a 1 7 scale (1 = not at all true 7 = extremely true ) All 12 original variables were uniquely accounted for across the 3 factors wi th factor loadings > .59. In reference to the last partner he or she considered/approached having oral, vaginal, or anal sex with. + Reverse scored item. Two scales, with good to excellent internal consistency were created for f actor s 1 and 2 : relations hip health ( = .93 ) and relationship safety ( = .86 ) A significant correlation was seen between the two variables in factor 3, so a composite score was created: relationship certainty ( r = .64, p < .001) Higher relationship health composite scores i ndicate increased relationship satisfaction H igher relationship safety composite scores indicate increased physical, s exual, and emotional safeness. Higher relationship

PAGE 44

33 certainty composite scores indicate feeling less worried or conflicted about the rel ationship. Interpersonal Connectedness. Fourteen items were developed for this study to measure a participant's self report of interpersonal connectedness before their first sexual encounter and after the sexual encounter (if applicable), using his or her last partner where oral, vaginal, or anal sex was considered or approached as the referent There were 5 items that measur ed positive feelings toward partner 4 items measured positive feelings from partner and 5 items measured an individual 's readines s and willingness to engage in sexual activity with his or her last partner All items were rated on a 1 to 7 scale (1 = not at all true, 7 = extremely true). Participants that reported having sex received specific "before sex" and "after sex" prompts. Participants that indicated having an SSHC before sex were prompted with the following: Before sex: "After I had (or tried to have) the SSHC with my last partner, but before I actually had sex with him/her I felt..." After sex: "After having oral, vagina l, and/or anal sex with my last partner I felt..." Participants that indicated that they did NOT have an SSHC before sex were prompted with the following: Before sex: "Before I actually had oral, vaginal, and/or anal sex with my last partner, I felt..." A fter sex: "After having oral, vaginal, and/or anal sex with my last partner I felt..." Three composite scales, with adequate to good internal consistency were created: positive feelings felt towards partner (" = .87) positive feelings from partner (" = 77) and readiness and willingness to engage in sexual activity (" = .87); see Table 7 below Higher composite scores indicate increased positive feelings felt toward last partner increased positive feelings from last partner and increased readiness and willingness to

PAGE 45

34 engage in sexual activity with last partner before and after ( if applicable) engaging in oral, vaginal, and/ or anal sex Table 7 Items and Reliability C oefficients of I nterpersonal C onnectedness M easures Variable Item Positive feelings toward partner I felt close to this person .87 I knew this person very well. I was committed to this person I was attracted to this person I was confident about the decisions I made thus far with this person P ositive feelings from partner I felt that this person was committed to me. .77 I was worried that this person would not love me + I felt that this person cared about me. I felt that this person liked me. Readiness & willingness to engage in sexual activit y I was ready to have sex with this person .87 I was willing to have sex with this person I was nervous to have sex with this person *+ I felt pressure from this person to have sex + I felt stressed about having sex with this person *+ Not e: Interpersonal connectedness i tems were rated on a 1 7 scale (1= not at all true 7 = extremely true ) In reference to the last partner he or she considered/approached having oral, vaginal, or anal sex with. + Reverse scored item. Data Analysis Descri ptive analyses (i.e., means, standard deviations, skewness, kurtosis, and percentages) were used to assess the distributional characteristics of the data, including the examination and management of outliers and normal distribution assumptions Additional ly, b ivariate correlational analyses were conducted to examine the direct relationships between all variables of interest in preparation for additional statistical analysis.

PAGE 46

35 To address Aim 1, a serie s of structural equation models were tested with MPlus v ersion 7.11 ( MuthÂŽn & MuthÂŽn, 1998 2012 ) to pred ict depth of the SSHC topics covered and ease of engagement with SSHC topics by participants' attitudes subjective norms, PBC, and intenti ons to engage in a SSHC with any new sexual partner ( general models ) and the last partner where oral, vagi nal, or ana l sex was considered or approached ( last partner specific models ). Simple mean composite sco res ( for each scale or composite, the score is the sum of all items divided by number of items) were constructed for general and last partner specific models; the se included affective attitudes ( 3 items) instrumental attitudes (4 items) subjective norms (3 items) PBC (7 items) and intentions (6 items). MPlus has the capability to test models using a full information (direct) maximum likelihood estimator, which addresses data that display levels of missingness in line with those observed in this study (general model: up to 5 % missing ; last partner specific : up to 5.1% missing ), and is considered the state of the art for dealing with data that are missing at rand om ( Engers & Bandalos, 2001 ; Schafer & Graham, 2002 ) The Chi Square Test of Model Fit statistic w as used as a preliminary measure of fit for each model; a non significant chi square statistic is desirable because it indicates that there is not a significant difference between a model and the associated data. However, the chi square statistic can be i nfluenced by trivial differences, so it is not always a straightforward assessment of fit ( Tabachnick & Fidell, 2007 ) Therefore, three goodness of fit indices were also utilized to examine fit of the models to the data. These included the Comparative Fit Index (CFI), the Root Mean Square Error Approximation (RMSEA), and the Standardized Root Mean Square Residual (SRMR). Evidence of an

PAGE 47

36 appropriate fit is indicated by CFI values of .95 or higher, RMSEA values below .05, and SRMR values below .05 ( Geiser, 2013 ) To address Aim 2, it was fundamental to analyze information about individuals who had a SSHC with their last partner before their first sexual encounter to individuals w ho did not To further understand the relationship between SSHC engagement and condom usage, a series of Pearson's c hi square statistic tests independent t tests and logistic regressions were used. Pearson's chi square statistics were used to test for independence or association of the independent variable (engaging in an SSHC before first sexual encounter or not ) with the dependent variable (condom use or not ) The first chi square test was conducted among participants reporting that their f irst sexu al encounter with their last partner was penetrative sex (i.e., vaginal and/or anal sex), as I thought they would be more likely to report using a condom before their first sexual encounter than those reporting oral sex only The second chi square test wa s conducted on a broader samp le, using participants that reported engaging in vaginal or anal sex at some point with their last partner and their reports of condom use before these encounters. R esults from the se chi square tests help address whether engag ing in a SSHC before the first sexual encounter is associated with condom use before penetrative sex (i.e., vaginal or anal sex). I ndependent t tests were used to compare reported SSHC depth and SSHC ease between individuals that used a condom before penet rative sex with their last partner and those that did not Logistic regression analyses were used to determine if increases in reported SSHC depth and SSHC ease increased the odds of whether or not an individual reported using a condom with their last par tner.

PAGE 48

37 Independent t tests were also used t o further understand the relationship between SSHC engagement, healthy relationship qualities, and interpersonal connectedness These tests were used to compare reported healthy relationship qualities and interpe rsonal connectedness b etween individuals that reported engaging in a SSHC with their last partner before their first sexual encounter and those that did not Effect sizes were computed as Cohen's d values ( Cohen, 1988 )

PAGE 49

38 CHAPTER III RESULTS Descriptive D ata The original dataset collected for this dissertation contained N = 247 participants; howe ver there were 51 participants reporting that they were not sexually active Since the specific aims of this project focus on SSHC engagement with participants last sexual partner, these participants were excluded from analyses, leaving a remaining sampl e of n = 196 sexually active participants. Demographics and general characteristics of the final sample are presented in Table 8. Means and standard deviations of reported TPB variables of interest (affective attitudes, instrumental attitudes, subjective norms, perceived behavioral control, intentions, SSHC depth, and SSHC ease) are presented in Table 9. T he majority of participants were between the ages of 18 27 years old ( M = 23 years, SD = 5.66 years ) and identified as female (7 5.5 %), lower middle in come status (45.9 %) Caucasian (60.2 %) exclusively heterosexual (80.6%) and were not sexually active with other people when their first sexual encounter with their last partner occurred (83.7%) Almost half (45%) reported that they were i n a casual, no n exclusive relationship with the last partner at the time of their first sexual encounter with this person. Participants were also asked additional questions about the last partner where oral, vaginal, or anal sex was considered or approached Using thi s last partner as the referent, the majority of those participants reported that their partner was male (7 4.5%) and within 2 years of their age (6 2.8 %) When examining participants' first sexual encounter with their last partner, the majority reported hav ing either oral and vaginal sex in the same

PAGE 50

39 encounter (36.7%) or "vaginal sex only" (36.1%), followed by oral sex only (2 6 % ), oral and anal sex in the same encounter (0 .6 %), and oral, vaginal, a nd anal sex in the same encounter (0. 6 %). When looking at participant s behaviors with their last partner, most reported engaging in a SSHC before their first sexual encounter (67.3%) and most reported us ing a condom before penetrative sex (70.7% ). Table 8 Demographics and C haracteristics of the F inal S ampl e ( N = 196). Variable M ean SD Age 23.4 5.66 Range = 18 49 years Median = 21.5 years Number of past partners a 9.20 11.48 Range = 1 80 Median = 6 Variable n % Sex Male 48 24.5 Female 148 75.5 Family Economic Status Low income 20 10.2 Lower middle income 90 45.9 Higher middle income 78 39.8 High income 8 4.1 Race White/Caucasian 118 60.2 Black/African American 7 3.6 Asian 18 9.2 Native Hawaiian or Pacific Islander 1 0.5 Hispanic 31 15.8 Other (birac ial, multiracial) 21 10.7 Sexual Orientation Exclusively heterosexual 156 79.6 Predominantly heterosexual with only incidental homosexual 17 8.7 Predominantly heterosexual but more than incidentally homosexual 7 3.6 Equally heterosexual and hom osexual 5 2.6 Predominantly homosexual but more than incidentally heterosexual 3 1.5 Predominantly homosexual with only incidental heterosexual 3 1.5 Exclusively homosexual 5 2.6 Sex of last partner Opposite sex partner 171 95.0 Same sex par tner 9 5.0

PAGE 51

40 Table 8 (cont.) Variable n % Age of last partner Two or more years younger that participant 14 7.1 Within two years of age as participant 123 62.8 Two or more year older than participant 59 30.1 Relationship status with last partne r before first sexual encounter One night stand 20 11.1 Casual dating, non exclusive 81 45.0 Exclusively dating 18 10.0 Serious, committed relationship 61 33.9 SSHC engagement with last partner Yes, had a SSHC before first sexual encounte r 132 67.3 No, did not have a SSHC before first sexual encounter 64 32.7 Had sex with last partner Yes 180 91.8 No 16 8.2 Sexually active with other partners at the time of first sexual encounter with last partner Yes 28 16.3 No 144 83.7 Last partner's knowledge of concurrent sexual partners before first sexual encounter Yes, my last partner knew I had concurrent partners at that time 12 42.9 No, my last partner did not know I had concurrent sexual partners at that time 16 57.1 F irst sexual encounter with last partner Oral sex only 47 26.0 Vaginal sex only 65 36.1 Anal sex only 0 0 Oral and vaginal sex same encounter 66 36.7 Oral and anal sex same encounter 1 0.6 Vaginal and anal sex same encounter 0 0 Oral, vaginal, and anal sex same encounter 1 0.6 Condom usage with last partner before first sexual encounter ( excludes oral sex only ) b Yes, used a condom before first sexual encounter 87 70.7 No, did not use a condom before first sexual encounter 36 29.3 Condom usage with last partner before penetrative sex c Yes, used a condom before first penetrative sex encounter 127 70.9 No, did not use a condom before penetrative sex encounter 52 29.1 Note : SSHC = Shared Sexual Health Conversation; perce ntages reported above reflect "valid percentages," which take into consideration missing variables a Z scores greater than the absolute value of 3 indicated that there were two outliers ( reported having 150 and 269 past sexual partners ); t hese two individ uals were not included when calculating the mean on this variable b Means and percentages only reflect participants that reported engaging in penetrative sex (i.e., vaginal or anal sex) during their first sexual encounter with their last partner; that is participants that reported "oral sex only" as their first sexual encounter with their last partner were not included in this calculation. c Means and percentages reflect participants that reported engaging in penetrative sex (i.e., vaginal or anal sex) a t some point with their last partner.

PAGE 52

41 Table 9 Means and Standard D eviations of the TPB V ariables Variable # of items Range General TPB Last partner s pecific TPB + Mean SD Mean SD Affective attitudes 3 1 7 4.30 1.47 4.33 1.34 Instrumental attitudes 4 1 7 6.32 .90 6.15 1.10 Subjective norms 3 1 7 5.20 1.36 5.09 1.57 Perceived behavioral control 7 1 7 5.15 1.07 5.55 1.03 Intentions 6 1 7 5.40 1.46 4.89 1.84 SSHC depth ^ 25 1 7 3.63 1.61 --SSHC ease ^ 25 1 4 3.20 .81 --Note : TPB = Theory of Planned Behavior ; SSHC = Shared Se xual Health Conversation; SD = s tandard deviation The skewness and kurtosis of the variables above ranged between 2 and +2, indicating no serious violations of normality. Using any new sexual partner as the referent g roup. + Using an individual's last partner where oral, vaginal, or anal sex was considered or approached as the referent group. ^The mean and standard deviation values for SSHC depth and SSHC ease were the same for the general TPB model and last partner spe cific TPB models since participants were providing reports on what happened with their last partner. Correlates of Shared Sexual Health Conversations Bivariate correlations were conducted between all variables in the Theory of Planned Behavior (TPB) mode ls to examine the direct relat ionships between all variables S ee Table 10 for the general TPB model (shaded above the diagonal) and the last partner specific TPB model ( unshaded, below the diagonal) Table 10 Correlations B etween TPB V ariables Variabl e 1 2 3 4 5 6 7 1. Affective attitudes -.3 5 *** .40 *** .6 6 *** .4 8 *** .40 *** 41 *** 2. Instrumental attitudes .39 *** -.46 *** .36 *** 57 *** .2 7 ** .1 9 ** 3. Subjective norms .29 *** .47 *** -.43 *** .6 1 *** .2 9 *** 23 * 4. PBC .42 *** .44 *** .19 ** -. 51 ** 39 *** .4 7 *** 5. Intentions .44 *** .49 *** .78 * .30 *** -. 43 *** .26 ** 6. SSHC depth .42 *** .34 *** .37 *** .20 ** .53 *** -.35 ** 7. SSHC ease .43 *** .28 *** .17 .48 *** .30 *** .35 * -Note: TPB = Theory of Planned Behavior; PBC = Perceived Behavi oral Control. The correlations between the general TPB model variables are provided above the diagonal; correlations between the last partner specific TPB model variables are provided below the diagonal. p < 0.05. ** p < .01. *** p < .001.

PAGE 53

42 General TPB Model C orrelates. The core TPB variables were all statistically significant correlates of intentions to engage in a SSHC with any new sexual partner, with subjective norms being the strongest ( r = .61, p < .001), followed by instrumental attitudes ( r = .5 7, p < .001), PBC( r = .51, p < .001), and affective attitudes ( r = .48, p < .001). Intentions to engage in a SSHC with any new sexual partner were more strongly correlated with the depth of topics covered during a SSHC ( r = .43, p < .001) than the ease w ith which SSHC conversation topics were discussed ( r = .26, p < .05), as revealed by a Fisher's r to z difference test (z diff = 1.97, p < .05). An opposing pattern was seen for PBC, such that PBC appeared more strongly correlated with SSHC ease ( r = .47, p < .001) than SSHC depth ( r = .39, p < .001); however, a Fisher's r to z difference test suggested there was no significant difference (z diff = 1.07, p = .28). S tatistically significant association s were also ob tained between indirect TPB variables (aff ective attitudes, instrume ntal attitudes, and subjective norms) and the outcome variables (SSHC depth and S SHC ease ) For SSHC depth, affective attitudes were the strongest ( r = .40 p < .001) followed by subjective norms ( r = .29 p < .001) and instrumen tal attitudes ( r = .27 p < .01). A similar pattern was seen for SSHC ease, such that affective attitudes were the strongest ( r = .41, p < .001), followed by subjective norms ( r = .23, p < .01) and instrumental attitudes ( r = .19, p < .01). Overall, the se correlations followed the expected patterns Fisher's r to z difference tests showed that the correlation al strength between affective attitudes and the outcome variables ( SSHC depth and SSHC ease) did not differ in magnitude when compared to the corre lations between the intentions, PBC and the outcome variables. Since PBC and intentions are

PAGE 54

43 posited to be direct paths to the outcome variable s, it may be useful to create a direct path between affective attitudes and the outcome variables given the stro ng correlation. Last P artner S pe cific TPB Model C orrelates. For the last partner specific TPB model, t he core TPB variables were all statistically significant correlates of intentions to engage in a SSHC with the last partner where sex was considered or a pproached; subjective norms was the strongest ( r = .78, p < .001), followed by instrumental attitudes ( r = .49, p < .001), affective attitudes ( r = .44, p < .001), and perceived behavioral control ( r = .30, p < .001). When looking at intentions, perceive d behavioral control, SSHC depth, and SSHC ease, the last partner specific model followed the same pattern as the general model I ntentions were more strongly correlated with the depth of topics covered during a SSHC ( r = .53 p < .001) than the ease with which SSHC conversation topics were discussed ( r = .30 p < .001 ) as revealed by a Fisher's r to z difference test (z diff = 2.6 7, p < .01 ) The opposite was true when examining PBC, such that PBC was more strongly correlated with SSHC ease ( r = .48, p < .001) than SSHC depth ( r = .20, p < .01), as revealed by a Fisher's r to z difference test (z diff = 3.08, p < .01). St atistically significant associations were also obtained between indirect TPB variables (affective attitudes, instrumental attitudes, and subjective norms) and the outcome variables (SSHC depth and SSHC ease ) For SSHC depth, affective attitudes were the strongest ( r = .42, p < .001), followed by subjective norms ( r = .37 p < .001), and instrumental attitudes ( r = 34 p < .0 0 1). For SSHC ease, affective attitudes were the strongest ( r = .43, p < .001), followed by instrumental attitudes ( r = .28, p < .001), and subjective norms ( r = .17, p < .05) Overall, these correlations followed the expected

PAGE 55

44 patterns. Similar to what was seen in th e general TPB model, Fisher's r to z difference tests showed that the correlational strength between affective attitudes and the outcome variables (SSHC depth and SSHC ease) did not differ in magnitude when compared to the correlations between the intentio ns, PBC, and the outcome variables. Since PBC and intentions are posited to be direct paths to the outcome variables, it may be useful to create a direct path between affective attitudes and the outcome variables given the strong correlation. Predictors of SSHC Depth and Ease : Structural Equation Modelin g MPlus version 7.11 ( MuthÂŽn & MuthÂŽn, 1998 2012 ) was used to test general and last partner specific TPB models in a sample of 196 sexually active individuals These models are aimed at predicting 1) the depth of the SSHC topics covered and 2) the ease of engagement with SSHC topics Both outcomes are predicted by individuals' affective attitudes, instrumental attitudes, subjective norms, perceived behavioral control, and intentions to engage in a SSHC with any new sexual partner (general model) or wit h the last partner where oral, vaginal, or anal sex was considered or approached (last partner specific model). In the general and last partner specific models, full information maximum likelihood (FIML) estimator was used. E xogenous variables (affective attitudes, instrumental attitudes, subjective norms, perceived behavioral control) were allowed to correlate and standardized regression coefficients are presented to allow for relative comparisons across paths. Residual variances (represented by the sma ll arrows above and pointing towards intentions and the behavior) and R 2 values are also presented for

PAGE 56

45 endogenous variables Residual variances provide an estimate of the proportion of variability not accounted for by predictors in the model whereas R 2 v alues provide an estimate of the proportion of the variability that can be accounted for by predictors in the model General TPB M odels For the general TPB models, no missing data existed on the 5 variables of interest (affective attitudes, instrument al attitudes, subjective norms, perceived behavioral control, intentions); one participant was missing data on the depth of the SSHC topics covered with their last partner and three participants were missing data on the ease with which they discussed SSHC topics with their last partner. General SSHC depth model Figure 2 represents the general TPB model of SSHC depth; it shows depth of the SSHC topics predicted by respondents' affective attitudes, instrumental attitudes, subjective norms, PBC, and intenti ons to engage in a SSHC with any new sexual partner The proposed model fit the data well, 2 ( 3) = 4.75, p = .19, CFI = .99, RMSEA = .05, and SRMR = .02 The model explained 53.4% of the variance in intentions, and 22.3% of the variance in depth of SSHC topics covered. Figure 2 G eneral TPB Model for SSHC D epth. Note: N = 196. Ex ogenous variable correlations not shown. FIML e stimator and yx standardized (STDyx) coefficients. Solid lines indicate signitificant associations; dotted lines indicate non significant ( p > .05) associations. *** p < .001; ** p < .01; p < .05; + p < .10. .31 *** .12 + 18 * Affective Attitudes towards a SSHC wi th any new sexual partner Intentions to engage in SSHC with any new sexual partner SSHC Depth with last sexual partner Instrumental Attitudes towards a SSHC with any new sexual partner Norms for engaging in a SSHC with any new sexual partner PBC f or engaging in a SSHC with any new sexual partner .4 7 78 .32 *** .34 *** .22 **

PAGE 57

46 General SSHC ease model Figure 3 represents the general TPB model of SSHC ease; it shows ease of engagement with SSHC topics by respondents' affective attitudes, instrumental attitudes, subjective norms, PBC, and intentions to engage in a SSHC with any n ew sexual partner The proposed model fit the data well, 2 ( 3) = 4.54, p = .21, CFI = .99, RMSEA = .05, and SRMR = .02 The model explained 53.4% of the variance in intentions, and 22.4% of the variance in ease of engagement with SSHC topics covered. I n contrast with SSHC depth, intentions were not a significant predictor of SSHC ease; the only significant predictor of SSHC ease was PBC. Figure 3 G eneral TPB Model for SSHC E ase. Note: N = 196. Exogenous variable correlations not shown. FIM L estimator and yx standardized coefficients (STDyx). Solid lines indicate signitificant associations; dotted lines indicate non significant ( p > .05) associations. *** p < .001; ** p < .01; p < .05; + p < .10. Last Partner S pecific TPB M odels For the last partner specific TPB models, no missing data existed on the 3 variables of interest (subjective norms, perceived beha vioral control, and intentions). E ight participants were missing data on both affective and instrumental attitudes, one participant w as missing data on the depth of the SSHC topics covered with their last partner, and three participants were missing data on the ease with which they discussed SSHC topics with their last partner. .31 *** .12 + 18 * Affective Attitudes towards a SSHC with any new sexual partner Intentions to engage in SSHC with any new sexual partner SSHC Ease with last sexua l partner Instrumental Attitudes towards a SSHC with any new sexual partner Norms for engaging in a SSHC with any new sexual partner PBC for engaging in a SSHC with any new sexual partner .4 7 78 .01 .34 *** .47 ***

PAGE 58

47 Last partner specific SSHC depth models Figure 4 represe nts the hypothesized last partner specific model of SSHC depth; it shows depth of the SSHC topics covered by respondents' affective attitudes, instrumental attitudes, subjective norms, PBC, and intentions to engage in a SSHC with the last partner where sex was considered or approached The proposed model did not fit the data well, 2 ( 3) = 16.16, p = .001, CFI = .95, RMSEA = .15, and SRMR = .04. When overall model fit is poor, it suggests that predictors are not related pairwise to the outcomes and/or there are important direct paths that are not being accounted for. Previous biv ariate correlation analyses indicate that all of the predictors were related to SSHC depth so we can infer that poor model fit is a result of unaccounted for direct paths. Modification indices suggest the 2 statistic will improve if we account for a dir ect path between affective attitudes and SSHC depth This path is not in opposition to our hypothesized model, but suggests that the strong relationship between affective attitudes and ease is not being fully transmitted through intentions. The modified SSHC depth model will be presented after Figure 4. Figure 4 Hypothesized Last Partner Specific TPB Model for SSHC D epth. Note:N = 196. Exogenous variable correlations not shown. FIML e stimator and yx standardized coefficients (STDyx). Solid lin es indicate signitificant associations; dotted lines indicate non significant ( p > .05) associations. *** p < .001; ** p < .01; p < .05. .07 .20 *** 05 Affective At titudes towards a SSHC with last sexual partner Intentions to engage in SSHC with last partner where oral, vaginal, or anal sex was considered or approached Instrumental Attitudes towards a SSHC with last sexual partner Norms for engaging in a SSHC wi th last sexual partner PBC for engaging in a SSHC with last sexual partner .33 72 .52 *** .68 *** SSHC Depth with last sexual partner .04

PAGE 59

48 Figure 5 represents the modified last partner specific model of SSHC depth, which adds the suggested direct path betw een affective attitudes and SSHC depth. The modified model fits the data well, 2 ( 2) = 2.74, p = .25, CFI = 1.0, RMSEA = .04, and SRMR = .01. The model explained 66.9% of the variance in intentions, and 33% of the variance in depth of SSHC topics covered. Figure 5 Modified Last Partner Specific TPB Model for SSHC D epth. Note: N = 196. Exogenous variable correlations not shown. FIML e stimator and yx standardized coefficients (STDyx). Solid lines indicate signitificant associations; dotted lines indicate non significant ( p > .05) associations. *** p < .001; ** p < .01; p < .05. Last partner specific SSHC ease models Figure 6 represents the hypothesized last partner specific model of SSHC ease; it shows ease of engagement with SSHC topics by respondents' affective attitudes, instrumental attitudes, subjective norms, PBC and intentions to engage in a SSHC with the last partner where sex was considered or approached The proposed model did not fit the data well, 2 ( 3) = 13.15, p = .004, CFI = .96, RMSEA = .13, and SRMR = .04. As previously stated above for the propose d last partner specific SSHC depth model, when model fit is poor, it suggests that predictors are not related pairwise and/or there are important direct paths that are not being accounted for. Previous bivariate correlation analyses indicate that all of t he predictors were related to SSHC ease so we .04 .20 *** 05 Affective Attitudes towards a SSHC with last sexual partner SSHC Depth with last sexua l partner Instrumental Attitudes towards a SSHC with last sexual partner Norms for engaging in a SSHC with last sexual partner PBC for engaging in a SSHC with last sexual partner .33 67 .42 *** .68 *** .21 *** .07 Intentions to engage in SSHC wi th last partner where oral, vaginal, or anal sex was considered or approached

PAGE 60

49 can infer that poor model fit is a result of unaccounted for direct paths. Modification indices suggest the 2 statistic will improve if we account for a direct path between affective attitudes and SSHC ease This path is not in opposition to our hypothesized model but suggests that the strong relationship between affective attitudes and ease is not being full y transmitted through intentions. The modified SSHC ease model will be presented after Figure 6. Figure 6 Hypothesized Last Partner Specific TPB Model for SSHC E ase. Note: N = 196. Exogenous variable correlations not shown. FIML e stimator and yx standardized coefficients (STDyx). Solid lines indicate signitificant associations; dotted lines indicate non significant ( p > .05) associations. *** p < .001; ** p < .01; p < .05 Figure 7 represents the modified last partner specific model of SSHC ease, which adds a direct path between affective attitudes and SSHC ease. The modified model fits the data well, 2 ( 2) = .86, p = .65, CFI = 1.00, RMSEA = .00, and SRMR = .007. The model explained 67.1% of the variance in intentions, and 30.1% of the va riance in ease of engagement of SSHC topics covered. Intentions is no longer a significant predictor of SSHC ease which suggests that intentions does not mediate any of the relation between affective attitudes and SSHC ease ; however, this finding may be a result of having such strong predictors of SSHC ease in the model (i.e., affective attitudes and PBC). .07 .20 *** 05 Affective Attitudes towards a SSHC with last sexual partner Instrumental Attitudes towards a SSHC with last sexual partner Norms for engaging in a SSHC wi th last sexual partner PBC for engaging in a SSHC with last sexual partner .33 75 .17 ** .68 *** .42 *** Intentions to engage in SSHC with last partner where oral, vaginal, or anal sex was considered or approached SSHC Ease with last sexual partn er

PAGE 61

50 Figure 7 Modified Last Partner Specific TPB Model for SSHC E ase. Note: N = 196. Exogenous variable correlations not shown. FIML e stimator and yx standardized coefficients (STDyx). Solid lines indicate signitificant associations; dotted lines indicate non significant ( p > .05) associations. *** p < .001; ** p < .01; p < .05 Table 11 provides a summary comparison of the model fit statisti cs between the hypothesized and modified last partner specific models. Table 11 Summary of Last Partner Specific TPB A nalyses. Model Fit Statistics SSHC depth SSHC ease Hypothesized model Modified model Hypothesized model Modified model 2 ( df ) 16.16 2.74 13.15 .86 p ( 2 ) .001 .25 .004 .65 CFI .95 1.0 .96 .000 RMSEA .15 .04 .13 1.00 SRMR .04 .01 .04 .007 R 2 intentions 66.8% 66.9% 66.8% 67.1% R 2 behavior 28% 33% 25.2% 30.1% Note: N = 196. 2 = chi square statistic (w ant probability of 2 to be non significant); CFI = Comparative Fit Index (want this .95 or higher); RMSEA = Root Mean Square Error Approximation (want this below .05); SRMR = Standardized Root Mean Square Residual (want below .05). SSHC Engagement and C ondom Use Pearson's chi square tests were conducted to test for the independence or association of condom use and SSHC engagement before one's first sexual encounter. The first sample I focused on was participants who reported having penetrative sex (i.e. .34 *** .21 *** 05 Affective Attitudes towards a SSHC with last sexual partner SSHC Ease with last sexual partner Instrumental Attitudes towards a SSHC with last sexual partner Norms for engaging in a SSHC with last sexual partner PBC for eng aging in a SSHC with last sexual partner .33 70 .08 .68 *** .27 *** .07 Intentions to engage in SSHC with last partner where oral, vaginal, or anal sex was considered or approached

PAGE 62

51 vaginal or anal sex) during the first sexual encounter t hey had with their last partner ( n = 123; participants who reported engaging in "oral sex only" during this first sexual encounter were excluded ) A c hi square test was used to compare reported con dom us e among those who had the SSHC before this first sexual encounter and among those who did not The chi square results were not significan t but did show a trend 2 ( 1) = 3.29 n = 123 p = .07 This trend suggesting that among participants who repo rted having penetrative sex during their first sexual encounter with their last partner, having a SSHC before this sexual experience was associated with a greater (but non significant) likelihood of condom use than expected by chance (and not having the SS HC was associated with a lower likelihood of condom use than expected by chance) See Table 12 for cross tabulation results. Table 12 SSHC E ngagement and Condom U s e: Penetrative S ex During First E ncounter Condom used before sex No Yes Total SSHC b efore first sexual encounter No 11 ( 7 ) 14 ( 18 ) 25 Yes 25 ( 28 ) 73 ( 69 ) 98 Total 36 87 123 Note: n = 123. Sample comprised of participants who reported engaging in penetrative sex (i.e., vaginal or anal sex) during their first sexual encounter with thei r last partner; those who reported engaging in "oral sex only" during their first sexual encounter were excluded. Observed counts listed in plain text; expected counts listed inside parentheses in italics and rounded to the closest whole number. The sec ond sample was chosen as a way to expand upon the first sample; rather tha n focusing on participants whose first sexual encounter was penetrative sex, I focused on participants that reported engaging in penetrative sex at some point with their last partner and their reports of condom use before those first penetrative encounters (n = 166) A chi square test was used to compare reported condom use among those who had the SSHC before this first penetrative sexual encounter and among those who did not The c hi square was significant, 2 ( 1) = 5.318, n = 166, p < .05, suggesting that among

PAGE 63

52 participants who reported having penetrative sex at some point with their last partner, having a SSHC before their first sexual encounter was associated with a greater likel ihood of condom use than expected by chance (and not having the SSHC was associated with a lower likelihood of condom use than expected by chance) See Table 13 for cross tabulation results. Table 13 SSHC Engagement and Condom U se : Penetrative S ex at Some P oint with P artner Condom used before sex No Yes Total SSHC before first sexual encounter No 19 ( 13 ) 26 ( 32 ) 45 Yes 29 ( 35 ) 92 ( 86 ) 121 Total 48 118 166 Note: n = 166. Sample comprised of participants who reported engaging in penetrative sex (i.e., vaginal or anal sex) at some point with their last partner Observed counts listed in plain text; expected counts listed inside parentheses in italics and rounded to the closest whole number. Overall, the chi square results suggest that engag ing in a S SHC before the first sexual encounter is associated with a greater likelihood of condom use when penetrative sex occurs. I ndependent t tests were used to examine potential differences on reported SSHC depth and SSHC ease among individuals that us ed a condom before sex with their last partner and those that did not Among the sample of participants who reported engaging in penetrative sex during their first sexual encounter with their last partner, no significant differences were identified among those who used a condom before sex on SSHC depth ( M = 3.90 SD = 1.59 ) or SSHC ease ( M = 3.29 SD = .82 ) compared to those who did not use a condom (SSHC depth: M = 3.49 SD = 1.64; SSHC ease: M = 3.09 SD = .75 ). Similarly, among the sample of participan ts who reported engaging in penetrative sex at some point with their last partner, n o significant difference s were identified among those

PAGE 64

53 who used a condom before sex on SSHC depth ( M = 3.88, SD = 1.56) or SSHC ease ( M = 3.26, SD = .81) compared to those w ho did not use a condom ( SSHC depth: M = 3.43, SD = 1.64; SSHC ease: M = 3.06, SD = .80 ). Of note is that group means follow ed the predicted pattern, such that individuals who used condom s before sex reported increased depth of SSHCs and ease with which t hey had the SSHC with their last partner. Lastly, logistic regression analyses were used to determin e if increases in reported SSHC depth and SSHC ease would increase the odds of whether or not an individual reported using a condom with their last partner Among the sample of participants who reported engaging in penetrative sex during their first sexual encounter with their last partner, o mnibus tests of model coefficients indicated that neither SSHC depth nor SSHC ease were able to significantly increas e the li kelihood of reported condom use The same was true for the sample of participants who reported engaging in penetrative sex at some point with their last partner. However, this may have been due to our high base rate of condom use of approximately 71% Although odds ratios were not significant, results suggested that among the sample of participants who reported engaging in penetrative sex during their first sexual encounter with their last partner, for every one unit increase in SSHC depth, the o dds of using a condom increased by 17.8%, and for every one unit increase in SSHC ease, the odds of using a condom increased by 35.3%. Similar results were seen for the sample of participants who reported engaging in penetrative sex at some point with the ir last partner; for every one unit increase in SSHC depth, the odds of using a condom increased by 19.8% and for every one unit increase in SSHC ease, the odds of using a condom increased by 34.8%.

PAGE 65

54 SSHC Engagement and H ealthy R ela tionship Qualities Ind ependent t tests were used to detect differences on reported healthy relationship qualities (i.e., relationship health, relationship safety, relationship certainty) by comparing individuals who had a SSHC with their last partner before their first sexual e ncounter to individuals who did not Levene's test for equality of variances helped determined which t value and corresponding degrees of freedom would be appropriate. When Levene's test was significant, the assumption of homogeneity of variances was vio lated and results from the "equal variances not assumed "row were used; results from this row have smaller t values and degrees of freedom because they are based on a correction for the lack of homogeneity of variance Compared to individuals who did not have a SSHC before their first sexual encounter, those that did have the SSHC beforehand reported higher levels of relationship health before sex, t (71.88) = 2.08, p < .05, d = .35, and after sex, t (166) = 2.42, p < .05, d = .42; higher levels of perceiv ed safeness in their relationship before sex, t (167) = 2.15, p < .05, d = .37 and after sex, t (59.13) = 2.43, p < .05, d = .51; and higher levels of relat ionship certainty before sex, t (167) = 2.51, p < .05, d = .43 (no significant differences were see n for relationship certainty after sex). Table 1 4 provides a list of means and standard deviations for each of the healthy relationship quality variables. Table 14 Means and S tandard Deviations of H ealth y Relationship Q uality V ariables SSHC with last partner before first sexual encounter No ( n = 46) Yes ( n = 123) Variable M SD M SD Relationship health Before sex 5.78 1.14 6.17 .99 After sex 5.74 1.36 6.24 1.13 (cont. on next page)

PAGE 66

55 Table 14 (cont.) SSHC with last p artner before first sexual encounter No ( n = 46) Yes ( n = 123) Variable M SD M SD Relationship safety Before sex 5.53 1.30 5.99 1.24 After sex 5.66 1.67 6.30 1.06 Relationship C ertainty Before sex 4.84 1.60 5.52 ** 1.5 6 After sex 5.09 1.94 5.45 (ns) 1.96 Note: SSHC = Shared sexual health conversations; ns = non significant difference ** p < .01; p < .05 SSHC Engagement and Interpersonal Connectedness Independent t tests were used to detect differences on reporte d interpersonal connectedness (i.e., positive feelings toward partner, positive feeling from partner, "readiness and willingness" to engage in sexual activity with partner ) by comparing individuals who had a SSHC with their last partner before their first sexual encounter to individuals who did not. Compared to individuals who did not have a SSHC before their first sexual encounter, those who engaged in a SSHC before hand reported more positive f eelings toward their partner before sex t ( 65.30) = 3.5, p < .001 d = .69 and after sex, t ( 62.44) = 3.12, p < .01, d = .63 and more perceived positive feelin gs coming from their partner before sex t ( 167) = 3.38, p < .001 d = .58 and after sex t 165) = 2.58, p < .001, d = .45 There were no before sex or aft er sex differences on "readiness and willingness" to engage in sexual activity with t heir partner but mean differences trended in the hypothesized directed (i.e., individuals who had a SSHC with their last partner before their first sexual encounter repor t higher means than those who did not). Table 1 5 provides a list of means and standard deviations for each of the interpersonal connectedness variables.

PAGE 67

56 Table 1 5 Means and S tandard D eviations of Interpersonal Connectedness V ariables SSHC with last pa rtner before first sexual encounter No ( n = 46) Yes ( n = 123) Variable M SD M SD Positive feelings toward partner Before sex 5.0 2 1.44 5.84 *** 1.09 After sex 5.21 1.64 6.03 ** 1.15 Positive feelings from partner Before sex 5.09 1.26 5.78 *** 1.16 After sex 5.37 1.43 5.95 *** 1.25 "Readiness and willingness" to engage in sexual activity with partner Before sex 5.71 1.32 5.83 (ns) 1.12 After sex 6.15 1.22 6.19 (ns) 1.17 Note: SSHC = Shared sexual health conversations; ns = non significant difference. Bolded numbers indicate which mean value is higher between those who did not and those who did engage in a SSHC with their last partner before their first sexual encounter. *** p < .001; ** p < .01; p < .05

PAGE 68

5 7 CHAPTER IV DISCUSSION The purpose of this study was to examine the applicability of the theory of planned behavior (TPB) in the study of shared sexual health conversations ( SSHCs) among sexually active undergraduate students, and to explore associa tions between SSHC engagement, healthy relationship qualities, and interpersonal connectedness I define S SHC s as the communication between partners about previous and current drug use, sexual experiences, and risky behaviors, exposure to STDs, a sense of interpersonal safety regarding sex, and the use of preventative safe sex behaviors. For this study, I have focused on the degree to which this type of conversation occurs before the first sexual encounter and associated outcomes. Overall, this was a sam ple that was often in a fairly casual relationship with the ir most recent partner (e.g., 56% coded as "one night stand" or "casual, non exclusive" ) but did not have multiple concurrent relationships (e.g., 83.7% noted no concurrent sexual partners at th e time of the first sexual encounter with their last partner ) P articipants in this sample generally endorsed some key protective behaviors ; almost 68% engaged in a SSHC before their first sexual encounter with their last partner, and 71% reported using a condom before the first time they had penetrative sex with their last partner. This appears to be in line with previous reports of condom use in college students residing in the Rocky Mountain region; Murray and Miller (2000) found a high rate of condom use among a sample of 105 college students (n = 60 women) with 88% report ing that they used condoms over 75% of the time.

PAGE 69

58 Overall Predictive Power of the TPB Models The results of this study clearly support the applicability of the theory of planned behavior (TPB) in the study of shared sexual health conversations (SSHCs) among sexually active undergraduate students Over three quarters of the sample engaged in a SSHC before their first sexual encounter with their last partner; most reported that the SSHC topics were "somewhat easy" to "easy" for them to talk about beforehand, but that the depth of their conversation was only moderately extensive The hypothesized general models for SSHC depth and ease, which used the referent group "any new sexual p artner" fit the data well ; however the hypothesized last partner specific models, which used the referent group "last partner where oral, vaginal, or anal sex was considered or approached," did not fit well. Modification indices indicated that a direct p ath between affective attitudes (i.e., the degree to which a behavior is thought to be comfortable) and both SSHC depth and ease should be accounted for and after adding this path the models had adequate fit; I heretofore refer to these models as the "modi fied last partner specific" models. For SSHC depth, the general model explained 53.4% of the variance in intentions to engage in a SSHC with any new sexual partner, and 22.3% of the variance in SSHC depth with one's last sexual partner. The modified last partner specific model explained 66.9% of the variance in intentions to engage in a SSHC with the last sexual partner, and 33% of the variance in SSHC depth with this last sexual partner. The modified last partner specific models were able to account for a larger proportion of the variance in intentions and SSHC depth relative to the general model.

PAGE 70

59 For SSHC ease, the general model explained 53.4% of the variance in intentions to engage in a SSHC with any new sexual partner, and 22.4% of the variance in SS HC ease with one's last sexual partner. The modified last partner specific model explained 67.1% of the variance in intentions to engage in a SSHC with the last sexual partner and 30.1% of the variance in SSHC ease with this last sexual partner. Again, the modified last partner specific models were able to account for a larger proportion of the variance in intentions and SSHC ease relative to the general model. Our data compared favorably with findings of previous meta analyses ( McEachan, Conner, Taylor, & Lawton, 2011 ) which demonstrated that TPB constructs predicted up to 43.3% of the variance i n health related behavior intentions, such as condom use. Predictive Power within the General TPB Models For the general models, 3 out of 4 core TPB variables were significantly associated with intentions to engage in a SSHC with any new sexual partner; subjective norms was the strongest predictor, followed by instrumental attitudes and PBC. Affective attitudes w ere not a significant predictor of intentions, but trended in the appropriate direction. Since bivariate correlational analyses indicated that affective attitudes w ere significantly related to general intentions, the non significant effect of affective attitudes on intentions in the models suggests that affective attitudes does not account for unique variability in intentions above and beyond th at which has already been explained by instrumental attitudes, subjective norms, and PBC. For this general model, having an overall belief that engaging in a SSHC with any new partner would be

PAGE 71

60 beneficial, that others close to them would engage in a SSHC, and perceiving high behavioral control over engaging in a SSHC significantly increased participants' intentions to engage in a SSHC with any new partner. The general TPB models adequately "predicted" participants' reports of SSHC depth and ease Our find ings suggested that for this sample, perceiving high behavioral control to engage in a SSHC with any new sexual partner was strongly associated with increased conversational depth and ease with their last sexual partner. However, intentions to engage in a SSHC with any new sexual partner were an important predictor for depth (but not ease). Having the intention to engage in a SSHC might be a more accurate reflection of the actual act of talking about sexual health topics (related to conversational depth) rather than if one can talk about the sexual health topics with more ease. Predictive Power within the Last Partner S pecific TPB Models For the modified last partner specific models, only 2 out of 4 core TPB variables were significantly associated with i ntentions to engage in a SSHC with one's last partner Subjective norms were the strongest predictor, followed by affective attitudes. Interestingly, instrumental attitudes and PBC were not significant predictors of intentions as they were in the general models Since bivariate correlational analyses indicated that these variables were each associated with intentions, the non significant relationships observed suggest instrumental attitudes and PBC do not exert unique predictive ability above and beyond that which has already been explained by affective attitudes and subjective norms For our sample, having an overall belief that engaging in a SSHC

PAGE 72

61 would be comfortable and that those important to them were also engaging in similar behaviors increased par ticipants' intentions to have a SSHC with their last sexual partner. The modified last partner specific TPB models adequately "predicted" participants' reports of SSHC depth and ease Our findings suggested that for this sample, perceiving that the conve rsation would be comfortable (affective attitudes) was strongly associated with increased conversational depth and ease with their last partner However, a few key differences existed between the SSHC depth and ease models. Having the intention to engage in the conversation with one's last partner was an important predictor for conversational depth (but not ease), and perceiving high behavioral control to engage in the conversation with one's last partner important for conversational ease (but not depth). It is possible that this difference was seen because conversational comfort and ease may be perceived as "in one's own control," but how in depth the conversation may go might not be perceived as fully under one's own control. The findings regarding no rms suggest that supporting an "everybody's doing it" mentality, that is, conveying that everybody is talking about sexual health with their partner before having sex, can be a useful intervention strategy. It is also generally accurate given the rates of SSHC in this sample, where a majority of participants engaged in SSHCs. Additionally, finding ways for people to be more comfortable with the idea of talking about sexual health topics before sex, and allowing practice opportunities to have these conversa tions, could result in increased conversational depth and ease.

PAGE 73

62 Association s between SSHC E ngagement and Condom Use Results also suggest that engaging in a sexual health conversation with your partner before the first sexual encounter is associated with a greater likelihood of condom use when first engaging in penetrative with the partner While one interpretation of this finding may be that our results reflect a more "proactive sexual health individual" who is simply more likely to have a SSHC before s ex and use a condom another pla usible alternative is that engaging in the SSHC makes one more likely to use a condom. The temporal sequencing within the question suggests that t he SSHC should have preceded this particular instance of condom use ("Did you engage in a SSHC conversation before your first sexual encounter with your last partner?"), assuming retrospective accuracy in reporting However, the presence of this sequencing should not be construed as causality, as we cannot infer this from the data T test analyses among participants who had penetrative sex during their first sexual encounter, as well as among participants who had penetrative sex at some point with their last partner, showed that there were no significant conversational depth and ea se differences between condom users and non condom users. However, the direction of the group means followed the predicted pattern; condom users reported increased conversational depth and ease than non condom users It is possible that the fact that alm ost 71% of our sample reported using a condom affects our ability to detect differences between condom users and non condom users Replicating this in l ogistic regression analyses increases in conversational depth and ease did not significantly increase the odds of using a condom or not; however, the odds were in the predicted

PAGE 74

63 direction in terms of the odds of condom use increasing as levels of depth and ease increased. SSHC s Healthy Relationship Qualities and Interpersonal Connectedness Compared to i ndividuals who did not have a sexual health conversation with their last partner before engaging in sexual activity, those that did have a conversation reported better relationship health, increased safety in their relationship, and more positive feelings before and after sex toward their partner and themselves This suggests that the process of engaging in a SSHC could increase positivity and connectedness felt in the relationship before and after sex Alternatively, the positivity and connectedness one feels with a partner could increase likelihood of engaging in an SSHC, as suggested by elements of the TPB model (e.g., direct impact of affective attitudes on conversational depth and ease) Overall, SSHCs are part of a constellation of proactive, healthy relationship and healthy sexual variables In simple univariate analyses, each one of the variables in the TPB model did predict SSHC depth and ease with the last actual sexual partner In the models that accounted for multiple predictors simultaneously SSHC depth and ease were consistently predicted by the subjective norm variables from the TPB. Other predictors, such as affective attitudes, instrumental attitudes, PBC, and intentions, had varying relevance in the model; this varying relevance depend e d on the referent (any new partner vs. last sexual partner) and SSHC outcome (depth or ease). General b eliefs that engaging in a SSHC would be beneficial and useful with any new sexual partner (instrumental attitudes) were significant predictors of SSHC depth

PAGE 75

64 and ease with the last actual sexual partner ; however, instrumental attitudes specifically in relation to one's last sexual partner were not significant predictors of SSHC depth and ease with that last sexual partner Again, this is only true in the overall model, after accounting for the other variables in the model The opposite pattern was true when examining affective attitudes. General b eliefs that engaging in a SSHC would be comfortable with any new partner (affective attitudes) were not sign ificant predictors of SSHC depth and ease with the last actual sexual partner ; however, affective attitudes in relation to one's last sexual partner were significant predictors of SSHC depth and ease with that last sexual partner When considering the se fi ndings about instrumental and affective attitudes, it seems that believing that a SSHC is a beneficial thing to do with any partner (general instrumental attitudes) and how comfortable you feel with that specific partner (partner specific affective atti tudes) are the most important predictors of having an in depth and at ease sexual health conversation Although i ndividuals may feel that talking with a specific partner would be beneficial (partner specific instrumental attitudes) that does not seem to predict conversational depth or ease beyond the degree to which they report being comfortable with that specific partner (partner specific affective attitudes) That is, when it comes to talking with a specific partner about sexual health, anticipating th at the conversation will be comfortable with that person (partner specific affective attitudes) becomes more influential on conversational depth and ease than its perceived benefit with that partner (partner specific instrumental attitudes). That is not t o say that instrumental attitudes are not important ; i t seems that a general understanding of the benefits of having

PAGE 76

65 such a conversation (general instrumental attitudes) does carry over into behavior with a specific partner. Having a sense of general conf idence and ability to have a SSHC ( perceived behavioral control; PBC) and planning to have a SSHC (intentions) with any new partner were both significant predictors of SSHC depth and ease with the last actual sexual partner However, when asking about thi s sense of confidence, ability, and intention to have a SSHC with the last partner specifically, there was variability in what aspect of the SSHC was associated with PBC and intention. Having a sense confidence and ability to have a SSHC with one's last s exual partner was only a significant predictor of SSHC ease with that last sexual partner (and not SSHC depth). The opposite was true for intentions; intending to have a SSHC with one's last sexual partner was only a significant predictor of SSHC depth wi th that last sexual partner (and not SSHC ease). When considering these findings about PBC and intentions, it seems that having the general confidence, ability, and intentions to talk about sexual health with any new partner are important predictors of con versational depth and ease with the last actual sexual partner. However, when asking about the specific partner, a sense of confidence and ability (PBC) does not seem to predict how in depth the conversation will be when controlling for the degree to whic h the individual has plans (intentions) to talk with that specific partner. In contrast, that sense of confidence and ability does seem to predict how at ease the conversation felt, even beyond the other variables in the model Intentions with the last p artner seem to be more important in the actual depth of the conversatio n. Overall, having a clear goal (intentions) may drive the behavior (actual

PAGE 77

66 depth), whereas feeling confident about the goal (PBC) relates to how comfortable it feels to engage in the behavior (actual ease) Study Limitations Results presented here should be interpreted in light of several limitations. The generalizability of this present study is limited to most female, sexuall y active undergraduates Sexually active undergraduates are a population is of interest to study because risky sexual behaviors such as casual sex, having multiple partners, and inconsistent use of condoms have been found to be relatively common on college campus es ( Laska, Pasch, Lust, Story, & Ehlinger, 2009 ; Ravert et al., 2009 ) However, in order to more confidently generalize these results to female and male sexually active unde rgraduates, a larger sample of male respondents would be necessary. D ue to the cross sectional nature of this study, our findings are based on "postdiction modeling (a s opposed to prediction) While the models presented were based on a plausible, causal sequence grounded in previous theoretical research, we cannot assert that the models infer causality. Since the predictor s and outcomes were measured at the same time, the predictors cannot logically "predict" the behavior that has already happened becaus e they may be measuring currently held cognitions and beliefs ( Albarracin, Fishbein, & Middlestadt, 1998 ) Reporting may also be subject to retrospective errors or biases in recall As noted above, it may be that variables are related due to unmeasured third variables or are part of general constructs (e.g., "open and communicative relationships" featuring more likelihood to protect one another with condoms, have SSHCs, feel more comfortable about having these SSHCs, etc.)

PAGE 78

67 Nevertheless, research suggests that past behavior plays a k ey rol e in future behavioral decisions, and experimental studies have shown that past behaviors are actually exogenous variables that can "predict" current cognit ions, beliefs, and evaluations ( Albarracin & Wyer, 2000 ) P ostdiction modeling was necessary for this study, as it was not feasible to directly o bserve individuals engaging in a real time shared sexual hea lth conversation; especially since these conversations are typically private by nature Experimental manipulation of SSHCs was also not possible, although future investigations could assess the effects of interventions designed to increase the likelihood, depth, or ease of SSHCs on safe sex behaviors (akin to the studies in which measures the impact of disclosure intervention groups) I relied on the participants' ability to be reflective and discern what their experience was at a given time and provided numerous prompts throughout the survey to be clear about the target person and time in their relationship a question was referring to As recommended by Bryan Schmiege, and Broaddus ( 2007 ) I have attempted to be thoughtful with my interpretations and upfront with my readers about the use of postdiction models. Although it was not feasible in th e current study, temporal sequencing could be improved by collecting data longitudinally; that is, measuring participants' current attitudes, subjective norms, PBC, and intentions to engage in a SSHC during a period of sexual inactivity, and then assessing these constructs and participants' actual SSHC behavior when they enter a period of sexual activity. However, inferring causality would still remain problematic, as it is difficult to control for confounding variables outside of a rigorously controlled e nvironment. Additionally, as stated by Lei & Wu (2007) :

PAGE 79

68 A well fitting SEM model does not and cannot prove causal relations without satisfying the necessary conditions for causal inferencea selected well fitting model in SEM is like a retained null hypot hesis in conventional hypothesis testing. It remains plausible among perhaps many other models that are not tested but may produce the same or better level of fit (p. 40) The current study also relied heavily on the use of a self report survey, which can be subject to participant bias, memory distortion, or inaccurate interpretations of the questions. The sensitive nature of the questions also raises the issue of socially desirable responding Paulhus (1991) describes that social desirability is compris ed of two subtypes : self deceptive positivity (the tendency to give self reports that are honest but positively biased) and impression management (deliberate self presentation to an audience)" (p. 37). It i s possible that this study may have been vulner able to these biases For instance, in the general models which ask participants respond to TPB items using "any new sexual partner" as the referent, responses might reflect an unconscious positive bias that they would always hold certain intentions or be liefs to have the sexual health conversation (self deceptive positivity) On the other hand, in the partner specific models, which ask participants to respond to TPB items using their "last sexu al partner as the referent, responses might have be able to p rovide a more targeted measure of how the TPC constructs a ffect actual behavior because they were linked to a past situation. When participants answered questions about SSHC depth and ease with their la st partner, it is possible that the responses they pr ovided were based on preserving (or creating ) a more positive, social image (impression management).

PAGE 80

69 To try and protect against impression management bias we chose to offer the entire survey online and ensured that participants' responses were anonymous. Online data collection has been shown to be less prone to this type of social desirability bias compared to other methods of data collection (van Gelder, Bretveld, & Roeleveld, 2010) That is, i t is a suitable method for gathering data on sensitive resea rch topics, such as sexual behaviors and disease transmission, because it creates an environment that feels more anonymous and private (Kiesler et al., 1984) With regard to anonymity, r esearch suggests that allowing individuals to respond anonymously inc reases disclosure about socially undesirable behaviors (e.g., see Ong & Weiss 2000) ; therefore, it is plausible that greater anonymity may help reduce impr ession management responses W ithout a formal social desirability scale it is difficult to assert the extent to which social desirability is influencing our results; however, previous research using the TPB models found that social desirability had little effect on the relationships between the TPB components (Armitage & Conner, 1999) and only accounte d for 5% of the variance in intentions and behavior (Beck & Azjen, 1991). There were a number of measures that were created specifically for this study and thus have undergone limited validation ( e.g., no multi method design to discern method variance la ck of established measures of some constructs against which to validate the novel measures) This limi tation was inevitable given feasibility constraints and because some of the constructs of interest do not necessarily have any currently established meas urement Internal consistency and factor analytic methods were employed when appropriate and found to be adequate All measures also had face validity. Careful consideration and planning was used when creating the SSHC and TPB measures; the

PAGE 81

70 SSHC scale w as strongly based on an existing measure, the Health Protective Sexual Communication Scale (HPSC; Catania et al., 1992) and the TPB variables were created based on a widely used manualized guide for researchers (Francis et al., 2004), which provides very s pecific steps to create the TPB variables. This TPB guide has researchers supply the ir targeted behavior of interest and prescribes wording for attitudes, subjective norms, PBC, and intention s measures. The guide recommends that research ers create questi ons that are both positively keyed (items that are phrased so that an agreement with the item represents a relatively high level of the attribute being measured) and negatively keyed (items that are phrased so that an agreement with the item represents a r elatively low level of the attribute being measured), with the later requiring reverse scoring methods prior to data analysis. This mixing of positively and negatively keyed items is a common practice designed to minimize the risk of "response set", or a tendency to answer questionnaire items in the same way regardless of their content (Francis et al., 2004). Included in the online questionnaire were multiple c heckpoints that prevented participants from skipping entire blocks of questions without providi ng an answer about why they did not respond (e.g., "I didn't feel like answering the questions ,"). However, the questionnaire did not contain explicit validity check items targeting individuals who might simply not be reading and responding carefully suc h as adding in a question that where the intent is obvious and has a high likelihood of being answered correctly if the question was seen and appropriately read (e.g., "If you live in the U.S. select Strongly Agree"). The language that was used to first i ntroduce the concept of a shared sexual health conversation may have led participants to report more SSHC behavior, but we feel this is

PAGE 82

71 unlikely given later descriptions of SSHCs. That is, i n the beginning of the survey, where participants are first intr oduced to what an SSHC is, participants read a brief paragraph that said, "SSHCs refer to the conversations we have with others about condoms, birth control, STDs, drug use, history of sexual partners and experiences, and a sense of readiness to engage in sexual intimacy." The use of "we have" might have uninte ntionally influenced the data by essentially operating as a descriptive norm (by suggesting we all have such conversations) ; however, l ater presentations of the SSHC concept which were more proximal to participant responding about SSHCs, did not use this language. The paragraph that preceded the SSHC depth and ease questions read: "Here is a list of some things that people might talk about as they are considering/approaching sex with someone new. We call talking about these kinds of things a "Shared Sexual Health Conversation" or SSHC for short." Additionally, the fixed definition box that participants had access to throughout the online survey did not use the "we have" language (see Figure 8 ) Figure 8 Fixed S hared S exual H ealth Conversation Definition Textbox from Online S urvey. Note: This textbox was available to participants throughout the survey in the right hand margin of the screen.

PAGE 83

72 Future Directions The results from this stu dy have important implications for sexual health interventions. In particular, I hope that these findings will encourage clinicians and intervention developers to focus on how they can make sexual health conversations more comfortable for sexually active adults Part of this effort might be introducing potential discussion topics ( e.g., previous and current drug use, sexual experiences, risky behaviors, exposure to STDs, a sense of interpersonal safety regarding sex, and the use of preventative safe sex b ehaviors) and providing a forum to practice conversational skills in a low risk environment For instance, this could be integrated into a human sexuality course, in which students are broken up into group and asked to represent a "character" that presen ts with a spe cific sexual health history and a specified set of needs and wants they are looking for in their partner. One goal of this activity could be to have students figure out what they might say to another character to elicit information that would help their character make an informed sexual health decision. By allowing students to engage in this conversation in a small group setting, it could help normalize the proces s and would be relatively low risk for participants, since they are allowed to operate under the guise of a fictitious character Other goals this activity might fulfill are providing students with increased exposure to sexual health topics provid ing opportunities to get peer feedback or support on what they might say, increasing s ubjective norms by underscoring from peers the importance of these conversations, and assisting them in becom ing more comfortable talking about sexual health topics As previously noted by Cleary et al. (2002), young college women reported that they would be more likely to talk about sexual health if they felt their partner would be open and willing to engage in a

PAGE 84

73 conversation; by increasing the exposure to sexual health conversations, this may improve openness, willingn ess, confidence, and comfort talking about sex. Future research might also want to focus on moderating factors of SSHC depth and ease. A crucial ingredient in shared sexual health conversations is self disclosure; previous research suggests that self disclosure between partners is highly d ependent on the type of relationship an individ ual has with his or her partner Since g reater rates of disclosure occur in more intimate relationships ( Fife & Weeks, 2010 ; Keller, von Sadovszky, Pankratz, & Hermsen, 2000 ) it is plausible that the TPB model for conversational depth and ease could be moderated by relationship closeness. Individuals who are not inte rested in developing intimacy or commitment with their sexual partner might possess different patterns of attitudes, norms, perceived control, and intentions to engage in sexual health conversations than those who are more interested in developing a n intim ate connection with their partner, thus affecting conversational depth and ease outcomes. Greater rates of disclosure have also been seen among serodiscordant relationships ; that is, when one partner is HIV positive and the other is HIV negative ( Wolitski, Bailey, O'Leary, Gomez, & Parsons, 2003 ) Hi gh rates of disclosure among this population may be a result of how the disease manifests itself; without appropriate treatment, HIV can be fatal. However, in a population of college student s where relatively non fatal, sexually transmitted infections ar e mo re common (e.g., genital herpes, genital warts/HPV), individuals who have STDs might actually exhibit lower conversational depth and ease compared to those without STDs For example, imagine Ellie, a college female who has recently been diagnosed with HPV Ellie was initially

PAGE 85

74 worried about the diagnosis because she heard that certain strands of HPV have been linked to cervical and oral cancer. However she has also learned from her practitioner that 75% of college students have been exposed to HPV an d that it is highly probably that the virus will clear on its own within 90 days. Additionally, her practitioner tells her that males are often asymptomatic carriers of HPV and it is unclear when she may have contracted the virus because it can lie dorman t for months T he commonality of HPV its asymptomatic presentation, and its ability to remain undetected for months may be alluring enough to persuade Ellie forego disclosure in the short term and place others at risk for contraction. This study highl ight s the importance of how comfortable an individual believes they will be during a sexual health conversation (affective attitudes) on subsequent conversational depth and ease Interventions aimed at incre asing protective sexual health behavior may want to consider placing more emphasis on decreasing anxiety and fear beliefs, which may be associated with sexual health conversations as a means to improve comfor t R esearchers might also try to find variables that influence individuals a ffective experienc e s as a way to indirectly improve sexual health behavior.

PAGE 86

75 REFERENCES Albarracin, D., Fishbein, M., & Middlestadt, S. (1998). Generalizing behavioral findings across times, samples, and measures: A study of condom use. Journal of Appl ied Social Psychology, 28 (8), 657 674. Albarracin, D., Johnson, B. T., Fishbein, M., & Muellerleile, P. A. (2001). Theories of reasoned action and planned behavior as models of condom use: A meta analysis. [Review]. Psychological Bulletin, 127 (1), 142 1 61. doi: 10.1037//0033 2909.127.1.142 Albarracin, D., & Wyer, R. S. (2000). The cognitive impact of past behavior: Influences in beliefs, attitudes, and future behavioral decisions. Journal of Personality and Social Psychology, 79 5 22. Armitage, C. J., & Conner, M. (1999). Predictive validity of the theory of planned behaviour: The role of questionnaire format and social desirability. Journal of Community and Applied Social Psychology, 9 261 272. Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A meta analytic review. [Review]. British Journal of Social Psychology, 40 471 499. doi: 10.1348/014466601164939 Azjen, I. (1985). From intentions to actions: a theory of planned behavior. In J. Kuhl & J. Beckmann (Eds.) Action control: From Cognition to Behavior Heidelberg: Springer. Azjen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50 179 211. Azjen, I., & Madden, T. J. (1986). Prediction of Goal Directed Beh avior: Attitudes, Intentions, and Perceived Behavioral Control. [Article]. Journal of Experimental Social Psychology, 22 (5), 453 474. doi: 10.1016/0022 1031(86)90045 4 Bairan, A., Taylor, G. A. J., Blake, B. J., Akers, T., Sowell, R., & Mendiola, R. (200 7). A model of HIV disclosure: Disclosure and types of social relationships. [Article]. Journal of the American Academy of Nurse Practitioners, 19 (5), 242 250. doi: 10.1111/j.1745 7599.2007.00221.x Bandura, A. (1977). Self Efficacy: Toward a Unifying The ory of Behavior Change. Psychological Review, 84 (2), 191 215. Beck, L., & Ajzen, I. (1991). Predicting dishonest actions using the theory of planned behavior. Journal of Research in Personality, 25 285 301.

PAGE 87

76 Best, K. (2002). Counseling of couples fac ilitates HIV disclosure. Network, 21 (4), 25 27. Bryan, A. D., Schmiege, S. J., & Broaddus, M. R. (2007). Mediational analysis in HIV/AIDS research: Estimating multivariate path analytic models in a structural equation modeling framework. AIDS and Behavi or, 11 365 383. Carmack, C. C., & Lewis Moss, R. K. (2009). Examining the Theory of Planned Behavior Applied to Condom Use: The Effect Indicator vs. Causal Indicator Models. Journal of Primary Prevention, 30 (6), 659 676. doi: 10.1007/s10935 009 0199 3 Catania, J., Binson, D., & Stone, V. (1996). Relationship of sexual mixing across age and ethnic groups to herpes simplex virus 2 among unmarried heterosexual adults with multiple sexual partners. [Article]. Health Psychology, 15 (5), 362 370. Catania, J ., Coates, T., & Kegeles, S. (1994). A test of the AIDS Risk Reduction Model Psychosocial correlates of condom use in the AMEN cohort study. [Article]. Health Psychology, 13 (6), 548 555. Catania, J., Coates, T., Kegeles, S., Thompson Fullilove, M., Pet erson, J., Marin, B., . Hulley, S. (1992). Condom use in multi ethnic neighborhoods of San Francisco: The population based AMEN (AIDS in Multi Ethnic Neighborhoods) study. American Journal of Public Health, 82 284 287. Catania, J., Coates, T., Stal l, R., Turner, H., Peterson, J., Hearst, N., . Groves, R. (1992). Prevalence of AIDS related risk factors and condom use in the United States. Science, 258 1101 1106. Centers for Disease Control and Prevention. (2010). Sexually Transmitted Disease Surveillance 2009 Atlanta, GA: U.S. Department of Health and Human Services. Chandra, A., Mosher, W. D., Copen, C., & Sionean, C. (2011). Sexual behavior, sexual attraction, and sexual identity in the United States: Data from the 2006 2008 National Surv ey of Family Growth Retrieved from http://www.cdc.gov/nchs/data/nhsr/nhsr036.pdf. Cleary, J., Barhman, R., MacCormack, T., & Herold, E. (2002). Discussing sexual health with a partner: A qualitative study with young women. Canadian Journal of Human Sex uality, 11 117 132. Cohen, J. (1988). Statistical power analyses for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. Derlega, V., Greene, K., Petronio, S., & Gust, Y. (2003). Privacy and Disclosure of HIV in Interpersonal Relationships: A Sourcebook for Researchers and Practitioners Mahwah, NJ: Lawrence Erlbaum Associates.

PAGE 88

77 DeRosa, C., & Marks, G. (1998). Preventative counseling of HIV positive men and self disclosure of serostatus to sex partners: new opportunities for prevention. Health Psychology, 17 224 231. Desiderato, L. L., & Crawford, H. J. (1995). Risky sexual behavior in college students: Relationships between number of sexual partners, disclosure of previous risky behavior, and alcohol use. [Article]. Journal of Youth and A dolescence, 24 (1), 55 68. doi: 10.1007/bf01537560 Dolcini, M., Coates, T., Catania, J., Kegeles, S., & Hauck, W. (1995). Multiple sex partners and their psychosocial correlates: The popluation based AIDS in Multiethnic Neighborhoods (AMEN) study. Health Psychology, 14 1 10. Duncan, B., Hart, G., Scoular, A., & Bigrigg, A. (2001). Qualitative analysis of psychosocial impact of diagnosis of chlamydia trachomatic: implications for screening. British Journal of Medicine, 322 (7280), 195 199. Engers, C. K., & Bandalos, D. L. (2001). The relative performance of full information maximum likelihood estimation for missing data in structural equation models. Structural Equation Modeling: A Multidisciplinary Journal, 8 430 457. Fife, S. T., & Weeks, G. R. (2010). Barriers to recovering intimacy. In J. Carlson & L. Sperry (Eds.), Recovering intimacy in love relationships: A clinician's guide (pp. 157 179). New York: Routledge. Francis, J. F., Eccles, M. P., Johnston, M., Walker, A., Grimshaw, J., Foy, R., . Bonetti, D. (2004). Constructing questionnaires based on the theory of planned behaviors: A manual for health services researchers. Newcastle: Centre for Health Services Research. Frayley, S. (2002). Psychosocial outcomes in individuals living with genital herpes. Journal of Obstetics, Gynecologic, & Neonatal Nursing, 31 508 513. Friedman, A. L., & Bloodgood, B. (2010). "Something We'd Rather Not Talk About": Findings from CDC Exploratory Research on Sexually Transmitted Disease Communication wi th Girls and Women. [Article]. Journal of Womens Health, 19 (10), 1823 1831. doi: 10.1089/jwh.2010.1961 Geiser, C. (2013). Data analysis with Mplus New York, NY: Guilford Press. Green, J., Ferrier, S., Kocsis, A., Shadrick, J., Ukoumunne, O. C., Murphy S., & Hetherton, J. (2003). Determinants of disclosure of genital herpes to partners. [Article]. Sexually Transmitted Infections, 79 (1), 42 44. doi: 10.1136/sti.79.1.42

PAGE 89

78 Hernandez, B., Wilkens, L., Zhu, X., Thompson, P., McDuffie, K., Shvetsov, Y., . Goodman, M. (2008). Transmission of human papillomavirus in heterosexual couples. Emerging Infectious Diseases, 14 (6), 888 894. doi: doi:10.3201/eid1406.070616.2 Holt, R., Court, P., Vedhara, K., Nott, K., Holmes, J., & Snow, M. (1998). The role of di sclosure in coping with HIV infection. AIDS Care, 10 (1), 49 60. IBM Corp. (2012). IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY: IBM Corp. Jourard, S. (1971). Self Disclosure: The Experimental Investigation of the Transparent Self New York NY: Wiley. Kahn, J., Slap, G., Bernstein, D., Kollar, L., Tissot, A., Hillard, P., et al. (2005). Psychological, behavioral, and interpersonal impact of human papillomavirus and pap test results. Journal of Women's Health, 14 (7), 650 659. Retrieved fro m doi:10.1089/jwh.2005.14.650 Keller, M. L., von Sadovszky, V., Pankratz, B., & Hermsen, J. (2000). Self disclosure of HPV infection to sexual partners. [Article]. Western Journal of Nursing Research, 22 (3), 285 297. doi: 10.1177/01939450022044421 Kies ler, S., Siegel, J., & McGuire, T. W. (1984). Social psychological aspects of computer mediated communication. American Psychologist, 39 1123 1134. Kinsey, A. C., Pomeroy, W. B., & Martin, C. E. (1948). Sexual Behavior in the Human Male Philadelphia, P A: W. B. Saunders.First publication of Kinsey's Heterosexual Homosexual Rating Scale. Discusses Kinsey Scale, pp. 636 659. Kinsey, A. C., Pomeroy, W. B., Martin, C. E., & Gebhard, P. H. (1953). Sexual Behavior in the Human Female Philadelphia, PA: W. B. Saunders. Knight, K. R., Purcell, D., Dawson Rose, C., Halkitis, P. N., Gomez, C. A., & Team, S. (2005). Sexual risk taking among HIV positive injection drug users: Contexts, characteristics, and implication for prevention. AIDS Behavior, 4 (2), 147 158. Laska, M., Pasch, K., Lust, K., Story, M., & Ehlinger, E. (2009). Latent class analysis of lifestyle characteristics and health risk behaviors among college youth. Prevention science, 10 376 386. Lei, P. W. & Wu, Q. (2007). Introduction to structura l equation modeling: Issues and practical considerations. Educational Measurement: Issues and Practices (ITEMS module), 26 (3), 33 43.

PAGE 90

79 Masaro, C. L., Dahinten, V. S., Johnson, J., Ogilvie, G., & Patrick, D. M. (2008). Perceptions of sexual partner safety. [Article]. Sexually Transmitted Diseases, 35 (6), 566 571. doi: 10.1097/OLQ.0b013e3181660c43 McCaffery, K., Waller, J., Nazroo, J., & Wardle, J. (2006). Social and psychological impact of HPV testing in cervical screening: a qualitative study. [Article]. Sexually Transmitted Infections, 82 (2), 169 174. doi: 10.1136/sti.2005.016436 McEachan, R. R. C., Conner, M., Taylor, N. J., & Lawton, R. J. (2011). Prospective prediction of health related behaviors with the theory of planned behavior. Health Psychology Review, 5 97 144. Montgomery, K. A., Gonzalez, E. W., & Montgomery, O. C. (2008). Self disclosure of sexually transmitted diseases: an integrative review. [; Review]. Holist Nurs Pract, 22 (5), 268 279. Murray, S. & Miller, J. (2000). Birth control and condom usage among college students, CAHPERD Journal, 25 (1), 1 3. MuthÂŽn, L. K., & MuthÂŽn, B. O. (1998 2012). Mplus User's Guide (7th ed.). Los Angeles, CA: MuthÂŽn&MuthÂŽn. Noar, S.M. & Edgar, T. (2008). The role of partner communication in safer s exual behavior: A theoretical and empirical review. In: T. Edgar, S.M. Noar, & V.S. Freimuth (Eds.), Communication Perspectives on HIV/AIDS for the 21st Century (pp. 3 28). New York, NY: Lawrence Erlbaum Associates. Ong, A.D. & Weiss, D.J. (2000). Impact of Anonymity on Responses to Sensitive Questions, Journal of Applied Social Psychology 30 (8), 1691 1708. Paulhus, D. L. (1991). Measurement and Control of Response Bias. In: J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of Persona lity and S ocial Psychological Attitudes, V ol. 1, Measures of Social Psychological Attitudes Series (pp. 17 59) San Diego, CA: Academic Press. Perrin, K., Daley, E. M., Naoom, S., Packing Ebuen, J., Rayko, H., McFarlane, M., & al., e. (2006). Women's re action to HPV diagnosis: Insights from in depth interviews. Women's Health, 43 (2), 92 110. Ravert, R., Schwartz, S., Zamboanga, B., Kim, S., Weisskirch, R., & Bersamin, M. (2009). Sensation seeking and danger invulnerability: Paths to college student ri sk taking. Personality and Individual Differences, 47 763 768. Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art. Psychological Methods, 7 147 177.

PAGE 91

80 Seal, D. W. (1997). Interpartner concordance of self reported s exual behavior among college dating couples. [Proceedings Paper]. Journal of Sex Research, 34 (1), 39 55. SIECUS National Guidelines Task Force. (2004). Guidelines for Comprehensive Sexuality Education: Kindergarten 12th Grade (3rd ed.). New York, NY: Se xuality Information and Education Council of the United States. Simoni, J. M., Demas, P., Mason, H. R., Drossman, J. A., & Davis, M. L. (2005). HIV disclosure among women of African descent: Association with coping, social support, and psychological adap tation. AIDS Behavior, 4 (2), 147 158. Sullivan, K. M. (2005). Male self disclosure of HIV positive serostatus to sex partners: A review of the literature. Journal of the Association of Nurses in AIDS Care, 16 (6), 33. van der Straten, A., Catania, J. A ., & Pollack, L. (1998). Psychosocial Correlates of Health Protective Sexual Communication with New Sexual Partners: The National AIDS Behavioral Survey. Aids and Behavior, 2 (3), 213 227. van Gelder, M., Bretveld, R., & Roeleveld, N. (2010). Web based q uestionnaires: The future in epidemiology? American Journal of Epidemiology, 172 (11), 1292 1298, Waller, J., Marlow, L. A. V., & Wardle, J. (2007). The association between knowledge of HPV and feelings of stigma, shame and anxiety. [Article]. Sexually Tr ansmitted Infections, 83 (2), 155 159. doi: 10.1136/sti.2006.023333 Waller, J., McCaffery, K., Nazroo, J., & Wardle, J. (2005). Making sense of information about HPV in cervical screening: a qualitative study. [Article]. British Journal of Cancer, 92 (2), 265 270. doi: 10.1038/sj.bjc.6602312 Williams, S. S. (2001). Sexual Lying Among College Students in Close and Casual Relationships1. Journal of Applied Social Psychology, 31 (11), 2322 2338. Witte, S. S., El Bassel, N., Gilbert, L., Wu, E., & Chang, M. (2010). Lack of Awareness of Partner STD Risk Among Heterosexual Couples. [Article]. Perspectives on Sexual and Reproductive Health, 42 (1), 49 55. doi: 10.1363/4204910 Wolitski, R. J., Bailey, C. J., O'Leary, A., Gomez, A., & Parsons, J. T. ( 2003). Self perceived responsibility of HIV seropositive men who have sex with men for preventing HIV. AIDS Behavior, 7 (4), 363 372.

PAGE 92

81 APPENDIX A Participant Demographics Age: _______ Sex: Male Female Are you His panic or Latino? YES NO What best describes your race ? (Check all that apply) White Black or African American Native Hawaiian or Other Pacific Islander Asian American Indian or Alaska Native Other: ___________________________________________________________ Are you currently a student at the University of Colorado Denver and enrolled in a psycholog y class? YES NO How would you describe your family's economic status? Low income Lower middle income Higher middle income High inco me W hat is your sexual orientation ? Exclusively heterosexual with no homosexual Predominantly heterosexual, only incidentally homosexual Predominantly heterosexual, but more than incidentally hom osexual Equally heterosexual and homosexual Predominantly homosexual, but more than incidentally heterosexual Predominantly homosexual, only incidentally heterosexual Exclusively homosexual Have you ever been sexually active with another person ? That is, have you ever engaged in oral, vaginal, or anal sex with another individual? YES NO Have you ever considered/approached oral, vaginal, or anal sex with another individual? YES NO How many sexual partners have you had ? ______ *Note: Participants were provided with a definition box on the right hand side of their screen that provided the following def initions: Heterosexual = Sexually attracted to people of the opposite sex; Homosexual = Sexually attracted to people of one's own sex

PAGE 93

82 APPENDIX B General Theory of Planned Behavior (TPB) Questions The next questions relate to how you generally feel in r elationships with sexual/romantic partners (e.g., girlfriends, boyfriends, spouses, etc.) There are no right or wrong answers. We'd also like to introduce you to a new term we'll be using throughout the survey called a Shared Sexual Health Conversation, or "SSHC" for short. SSHCs refer to the conversations we have with others about condoms, birth control, STDs, drug use, history of sexual partners and experiences, and a sense of readiness to engage in sexual intimacy Throughout the survey in places wh ere we use the term "SSHC" we will place a box on the right hand side of your screen, which will contain this definition for you in case you forget. General I ntentions Rate how much you disagree or agree with the following statements about your general exp erience in relationships. Strongly Strongly Disagree Agree 1. I always expect to have a SSHC with any new sexual partner before oral, vaginal, or anal sex. 2. I always want to make sure that I have a SSHC with any new sexual partner before oral, vaginal, or anal sex. 3. I always have a SSHC with any new sexual partner before oral, vaginal, or anal sex. 4. I plan to have a SSHC with any new sexual partner before we engage in oral, vaginal, or anal sex 5. I will have a SSHC with any new sexual partner before we engage in oral, vaginal, or anal sex. 6. I am going to have a SSHC with any new sexual partner before we engage in oral, vaginal, or anal sex. 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 G eneral A ttitudes Next, you will read a prompt, followed by 7 pairs of words, with each word serving as an endpoint Place your response between each pair of words to describe whether you feel more strongly toward the word on the left o r the word on the right. For me having a SSHC with any new sexual partner before oral, anal or vaginal sex is generally: 1. Worthless 1 2 3 4 5 6 7 Useful ( I nstrumental attitudes) 2. Necessary 1 2 3 4 5 6 7 Unnecessary reverse score (I nstrumental attitudes) 3. U nimportant 1 2 3 4 5 6 7 Important (I nstrumental attitudes) 4. Anxiety provoking 1 2 3 4 5 6 7 Comfortable (A ffective attitudes) 5. Beneficial 1 2 3 4 5 6 7 Harmful reverse score (I nstrumental attitudes) 6. Unnatural 1 2 3 4 5 6 7 Natural ( A ffective attitudes) 7. Something that 1 2 3 4 5 6 7 Something that ruins the mood sets the mood (A ffective attitudes)

PAGE 94

83 General Subjective N orms Rate how much you disagree or agree with the following statements about your general experience in relationships. Strongly Strongly Disagree Agree 1. Most people who are important to me think that people should have a SSHC with any new sexual partner before oral, vaginal, or anal sex (Injunctive norm) 2. It is expected of me that I should have an SSHC with any new sexual partner before oral, vaginal, or anal sex (Injunctive norm) 3. Most people who are like me have an SS HC with any new sexual partner before oral, vaginal, or anal sex (Descriptive norm) 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 General Perceived B ehavioral C ontrol (PBC) Rate how much you disagree or agree with the following statements about your general experience in relationships. Easy Hard 1. In general, having an SSHC with any new sexual partner before o ral, vaginal, or anal sex is reverse score (Self efficacy) 1 2 3 4 5 6 7 Strongly Strongly d isagree a gree 2. I am generally confident that I can have a SSHC with any new sexual partner before oral, vaginal, or anal sex (Self efficacy) 3. I do not know the best way to have a SSHC with a new sexual partner before oral, vaginal, or anal sex reverse score (Self efficacy) 4. The decision to have a SSHC with an y new sexual partner before oral, vaginal, or anal sex is beyond my control reverse score (Controllability) 5. Whether or not I have a SSHC with any new sexual partner before oral, vaginal, or anal sex is entirely up to me (Controllability) 6. New sexual pa rtners make it hard to have a SSHC before oral, vaginal, or anal sex reverse score (Controllability) 7. New sexual partners won't let me have a SSHC before oral, vaginal, or anal sex reverse score (Controllability) 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7

PAGE 95

84 APPENDIX C Partner Demographics For the rest of the survey, we are interested in learning more about potential experiences people might have with another individual where oral, vaginal, or anal sex either occurred or was a real p ossibility This individual could have been a casual partner, like a hook up at a party, to someone more serious, such as someone you've recently dated (or are still dating) Essentially, we want you to think of anyone that you felt that there: 1) was a real chance that you would have oral, vaginal, or anal sex with, OR 2) someone that you actually did have oral, vaginal, or anal sex with. Have you ever had someone like this before? YES (Continue below) NO Peop le that selected "NO" were directed to an alternative questionnaire that asked similar questions, but was phrased entirely as a hypothetical (e.g., if you were to consider a partner for oral, vaginal, or anal sex, how old would you want them to be?); these participants were not included in this study IF YES: Now we'd like to get some information about your experiences with the LAST person where having oral, vaginal, or anal sex either happened or was a real possibility. We will be referring to this pers on throughout the survey as "this person", so keep that person in mind as you answer questions. How old was this person? About the same age as me More than two years younger than me More than tw o years older than me I don't know Was this person male or female? Male Female Was this person Hispanic or Latino? YES NO What best describes this person's ra ce ? (Check all that apply) White Black or African American Native Hawaiian or Other Pacific Islander Asian American Indian or Alaska Native Other: _____________________________________ Did you have or consider having any of the following with this person ? Check all that appl y. Oral sex Vaginal sex Anal sex None of the above What was the first sexual encounter you had with this person? Oral sex Vaginal sex Anal sex Oral and vaginal sex (in the same encounter)

PAGE 96

85 Oral a nd anal sex (in the same encounter) Vaginal and anal sex (in the same encounter) Oral, vaginal, and anal sex (all in the same encounter) *The next set of questions is presented to participants based on the sexual beha viors they reported considering or having with their last partner For instance, if a participant selected "oral sex" and "vaginal sex", they would receive 2 sets of questions below; the first set would be presented using "oral sex" as the referent, and t he second set would use "vaginal sex" as the referent Think back to the time right before you had [oral sex/vaginal sex/anal sex] with this person, or when you were seriously considering or approaching having [oral sex/vaginal sex/anal sex] with this per son (even if you did not end up actually having oral sex). How would you describe your relationship with this person at that particular time [refer ring to oral sex/vaginal sex/ anal sex, as appropriate] ? Check all that apply. A one nigh t stand Infatuated with each other lust and physical attraction were definite driving forces Relaxed we were in the process of getting to know each other, and not rushing into anything Casual friend s we were friends but never hooked up before Friend with benefits we have casually hooked up from time to time Dating we had already been on a few dates together Exclusive we decided that we wou ld only date each other Love birds we considered ourselves actively falling in love Committed we were in a serious, committed relationship Other: (please describe) ___________________________ We re you sexually active with anyone else at that time ? YES (Continue below) NO IF YES: Did this person know that you were sexually active with other people at that time? YES, my partner knew I ha d concurrent sexual partners at that time. NO my partner did not know I had other concurrent sexual partners at that time.

PAGE 97

86 APPENDIX D Shared Sexual Health Conversation (SSHC) Questions Here is a list of some things that people might talk about, as they are considering/approaching sex with someone new. We call talking about these kinds of things a "Shared Sexual Health Conversation" or SSHC for short. Before [oral sex/vaginal sex/anal sex] with your last partner, we'd like to know: How much did you discuss the topic together? Not a t all Ex tensively How easy was it to talk about this topic together ? *reverse score Did not discuss Easy Difficult 1. Shared feelings about condom use. 2. Shared feelings about using birth control, other than condoms. 3. Number of past sexual partners we each have had. 4. Sexual experiences we have had with the opposite sex. 5. Sexual experiences we have had with the same sex. 6. The most recent time tha t we have been tested for STDs/STIs. 7. The need for us both to get tested for STDs/STIs. 8. The most recent time that we have been tested for HIV/AIDS. 9. The need for us both to get tested for HIV/AIDS. 10. If we currently have any STDs or STIs. 11. If w e have ever had or been intimate with someone who, at the time, had a STD or STI. 12. If we currently have HIV/AIDs. 13. If we have ever had or been intimate with someone who, at the time, had HIV/AIDs. 14. Injection drug use/experiences we have had (e.g., heroin, cocaine, speed, methamphetamine). 15. Recreational drug use/experiences we have had (e.g., marijuana, prescription medication misuse, cocaine, salvia, ecstasy). 16. Shared feelings about unprotected sex. 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 (cont. on n ext page)

PAGE 98

87 Shared Sexual Health Conversation (SSHC) Questions (cont.) How much did you discuss the topic together? Not a t all Ex tensively How easy was it to talk about this topic together ? *reverse score Did not discuss Easy Difficult 17. Unprotected sex encounters we have had. 18. The importance of getting to know one another on an emotional level before go ing any further. 19. Waiting to have sex until we have known each other longer. 20. Things we are nervous about surrounding sex. 21. Things that make us feel physically safe in a relationship. 22. Things that make us feel emotionally safe in a relationship 23. What our comfort level is engaging in [oral/vaginal/anal sex]. 24. What our comfort level is engaging in other kinds of sexual intimacy. 25. Each other's expectations of what it would mean for our relationship if we decided to have [oral/vaginal/anal sex]. 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 0 1 2 3 4 Given the nature of SSHC topics above, do you think that you and your last partner most likely had some sort of SSHC before considering or approaching [oral sex/vaginal sex/anal sex] ? YES NO Did you have [oral sex/vaginal sex/anal sex] with your last partner? YES (Continue below) NO IF YES: Did you or your partner decide to use a condom or other barrier method of protection before [oral sex/vaginal sex/anal sex] ? YES NO

PAGE 99

88 APPENDIX E Last Partner Specific Theory of Planned Behavior (TPB) Questions We're curious about the SSHC experience you had with you r last partner Let's start w ith the some questions that focus on the period of time when you were considering or approaching having oral, vaginal, or anal sex with this person. Last P artner S pecific I ntentions Rate how much you disagree or agree with the following statements about y our specific experiences with your last partner. Strongly Strongly Disagree Agree 1. I expected that I would have a SSHC with this person before oral, vaginal, or anal sex. 2. I was s ure that I would have a SSHC with this person before oral, vaginal, or anal sex. 3. I fully intended to have a SSHC with this person before oral, vaginal, or anal sex. 4. I planned to have a SSHC with this person before oral, vaginal, or anal sex. 5. I meant to ha ve a SSHC with this person before oral, vaginal, or anal sex. 6. I was going to have a SSHC with this person before oral, vaginal, or anal sex. 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 Last P artner S pecific A ttitudes Next, you will read a prompt, fo llowed by 7 pairs of words, with each word serving as an endpoint. Place your response between each pair of words to describe whether you feel more strongly toward the word on the left or the word on the right. For me, I thought that having an SSHC w ith this person would be 1. Worthless 1 2 3 4 5 6 7 Useful (Instrumental attitudes) 2. Necessary 1 2 3 4 5 6 7 Unnecessary reverse score (Instrumental attitudes) 3. Unimportant 1 2 3 4 5 6 7 Important (Instrumental attitud es) 4. Anxiety provoking 1 2 3 4 5 6 7 Comfortable (Affective attitudes) 5. Beneficial 1 2 3 4 5 6 7 Harmful reverse score (Instrumental attitudes) 6. Unnatural 1 2 3 4 5 6 7 Natural (Affective attitudes) 7. Something that 1 2 3 4 5 6 7 Something that ruins the mood sets the mood (A ffective attitudes)

PAGE 100

89 Last P artner Specific Subjective N orms Rate how much you disagree or agree with the following statements about your specific experiences with your last partner. Stron gly S trongly Disagree Agree 1. Most people who are important to me probably think that I should have had an SSHC with this person before oral, vaginal, or anal sex with him or her (Injunctive norm) 2. In this situation, it would have been expected of me to have an SSHC before oral, vaginal, or anal sex with this person (Injunctive norm) 3. Most people who are like me to me would have had an SSHC with this person before oral, vaginal, or anal sex (Descriptive norm) 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 Last P artner S pecific Perceived Behavioral C ontrol (PBC) Rate how m uch you disagree or agree with the following statements about your specific experiences with your last partner. Easy Hard 1. Having an SSHC with this person* before oral, vaginal, or anal sex was reverse score (Self efficacy) 1 2 3 4 5 6 7 Strongly Strongly D isagree A gree 2. I was confident that I could have a SSHC with this person before oral, v aginal, or anal sex (Self efficacy) 3. I tried to have a SSHC with this person but I didn't know the best way to do it reverse score (Self efficacy) 4. The decision to have a SSHC with this person before oral, vaginal, or anal sex was beyond my control reverse score (Controllability) 5. Whether or not I had a SSHC with this person before oral, vaginal, or anal sex was entirely up to me (Controllability) 6. I tried to have a SSHC with this person before oral, vaginal, or anal sex, but they made it hard for me to do reverse score (Controllability) 7. I tried to have a SSHC with this person but they wouldn't let me go there reverse score (Controllability) 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7

PAGE 101

90 APPENDIX F Healthy Relationship Qualities (Befor e Sex) PROMPT FOR PARTICIPANTS THAT SAID THEY HAD A SSHC BEFORE SEX: After I had (or tried to have) the SSHC with my last partner, but before I actually had sex with him/her I felt... PROMPT FOR PARTICIPANTS THAT SAID THEY DID NOT HAVE A SSHC BEFORE SEX: Before I actually had oral, vaginal, and/or anal sex with my last partner, I felt..." Not at all Extremely t rue true 1. The relationship was fun (Relationship health) 2. The re lationship was intimate (Relationship health) 3. The relationship was exciting (Relationship health) 4. The relationship was healthy (Relationship health) 5. The relationship was enjoyable (Relationship health) 6. The relationship was comfortable (Relationsh ip health) 7. The relationship was not really worth it *reverse score (Relationship health) 8. I felt sexually safe with him or her (Relationship safety) 9. I felt physically safe with him or her (Relationship safety) 10. I felt emotionally safe with him or her ( Relationship safety) 11. I felt conflicted about the relationship *reverse score (Relationship certainty) 12. I felt worried about the relationship *reverse score (Relationship certainty) 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7

PAGE 102

91 APPENDIX G Interpersonal Connect ed ness (Before Sex) PROMPT FOR PARTICIPANTS THAT SAID THEY HAD A SSHC BEFORE SEX: After I had (or tried to have) the SSHC with my last partner, but before I actually had sex with him/her I felt... PROMPT FOR PARTICIPANTS THAT SAID THEY DID NOT HAVE A SSHC BEFORE SEX: Before I actually had oral, vaginal, and/or anal sex with my last partner, I felt..." Not at all Extremely true true 1. I felt close to this person (Positive feelings towards partner) 2. I knew this person very well (Positive feelings towards partner) 3. I was committed to this person (Positive feelings towards partner) 4. I was attra cted to this person (Positive feelings towards partner) 5. I was worried that this person would not love me *reverse score (Positive feelings from partner) 6. I was confident about the decisions I made thus far with this person (Positive feelings towards p artner) 7. I felt that this person was committed to me (Positive feelings from partner) 8. I felt that this person cared about me (Positive feelings from partner) 9. I felt that this person liked me (Positive feelings from partner) 10. I was ready to have sex wit h this person (Readiness & willingness to engage in sexual activity) 11. I was willing to have sex with this person (Readiness & willingness to engage in sexual activity) 12. I was nervous to have sex with this person *reverse score (Readiness & willingness to engage in sexual activity) 13. I felt pressure from this person to have sex *reverse score (Readiness & willingness to engage in sexual activity) 14. I felt stressed about having sex with this person *reverse score (Readiness & willingness to engage in sexu al activity) 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7

PAGE 103

92 APPENDIX H Healthy Relationship Qualities (Afte r Sex) After I had oral, vaginal, and/or anal sex with my last partner, I felt... Not at all Extremely true true 1. The relationship was fun (Relationship health) 2. The relatio nship was intimate (Relationship health) 3. The relationship was exciting (Relationship health) 4. The relationship was healthy (Relationship health) 5. The relationship was enjoyable (Relationship health) 6. The relationship was comfortable (Relationship he alth) 7. The relationship was not really worth it *reverse score (Relationship health) 8. I felt sexually safe with him or her (Relationship safety) 9. I felt physically safe with him or her (Relationship safety) 10. I felt emotionally safe with him or her (Relat ionship safety) 11. I felt conflicted about the relationship *reverse score (Relationship certainty) 12. I felt worried about the relationship *reverse score (Relationship certainty) 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7

PAGE 104

93 APPENDIX I Interpersonal Connectedness (After Sex) After I had oral, vaginal, and/or anal sex with my last partner Not at all Extremely true true 1. I felt cl ose to this person (Positive feelings towards partner) 2. I knew this person very well (Positive feelings towards partner) 3. I was committed to this person (Positive feelings towards partner) 4. I was attracted to this person (Positive feelings towards par tner) 5. I was worried that this person would not love me *reverse score (Positive feelings from partner) 6. I was confident about the decisions I made thus far with this person (Positive feelings towards partner) 7. I felt that this person was committed to me (Positive feelings from partner) 8. I felt that this person cared about me (Positive feelings from partner) 9. I felt that this person liked me (Positive feelings from partner) 10. I was ready to have sex with this person (Readiness & willingness to engage i n sexual activity) 11. I was willing to have sex with this person (Readiness & willingness to engage in sexual activity) 12. I was nervous to have sex with this person *reverse score (Readiness & willingness to engage in sexual activity) 13. I felt pressure from t his person to have sex *reverse score (Readiness & willingness to engage in sexual activity) 14. I felt stressed about having sex with this person *reverse score (Readiness & willingness to engage in sexual activity) 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7