USF Libraries
USF Digital Collections

Understanding the psychosocial aspects of waterpipe smoking among college students


Material Information

Understanding the psychosocial aspects of waterpipe smoking among college students
Physical Description:
Martinasek, Mary Pautler
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:


Subjects / Keywords:
Dissertations, Academic -- Health Sciences -- Doctoral -- USF   ( lcsh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


ABSTRACT: Waterpipe tobacco smoking has migrated from being a custom of some cultures to becoming a staple around college campuses. The social nature and flavored tobacco encourage initiation in this tobacco naïve age group. The study was a sequential mixed method design, employing primary data collection and analysis of a random sample of university students who live on campus at a single university. The study involved observations (N=6), intercept interviews with smokers and nonsmokers (N=63), three focus groups (N=31), and an online survey (N=288). Findings were centered on the constructs of the Theory of Reasoned Action by Ajzen and Fishbein (1975), who propose there are specific precursors that lead to intention to perform a behavior. The data suggested that attitudes were more positively correlated with intention to smoke waterpipe tobacco than subjective norm. Attitude is influenced by outcome expectancies. The data suggest that positive outcome expectancies are influenced primarily by the social nature of hookah smoking and other perceived positive benefits. Negative health effects and family/culture were also influential, albeit to a lesser extent. Development of interventions focused on an attitudinal shift may help to decrease uptake and continuation of waterpipe tobacco smoking in this population.
Disseration (Ph.D.)--University of South Florida, 2011.
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Mary Pautler Martinasek.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 153 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E14-SFE0004948
usfldc handle - e14.4948
System ID:

This item is only available as the following downloads:

Full Text


Understanding the Psychosocial Aspects of Waterpipe Smoking Among College Students by Mary P. Martinasek A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Community and Family Health College of Public Health University of South Florida Co-Major Professor: Carol Bryant, Ph.D. Co-Major Professor: Robert McDermott, Ph.D. Moya Alfonso, Ph.D., MSPH Eric Buhi, Ph.D. David Himmelgreen, Ph.D. Date of Approval: April 4, 2011 Keywords: hookah, shisha, tobacco, lungs, health Copyright 2011, Mary P. Martinasek


DEDICATION This dissertation is dedicated in memory of my fath er, Dr. E. E. Pautler, who would be so very proud and possibly surprised that I am following in his footsteps in the academic arena. He set a model example of what a p rofessor, educator, physician, and father should be.


ACKNOWLEDGMENTS This research would not have been possible without the support of several people. First, I would like to thank USF’s AHEC and the Ame rican Respiratory Care Foundation for funding this study. In particular, I would lik e to thank Leila Martini for her trust and confidence in the work that I proposed. Second, I would like to thank Dr. Carol Bryant and Dr. Robert McDermott, my co-chairs, who have me ntored me with unending encouragement throughout this process. I also cann ot thank Dr. Moya Alfonso enough for her unwavering willingness to provide support a nd feedback, even on the weekends! I thank Dr. Himmelgreen for his guidance in ethnogr aphic studies andDr. Buhifor providing me with research opportunities in a simil ar arena. I would also like to thank Dr. Hana Osman and Dr. Kay Perrin for their positiv e support and guidance over the past few years. I also would like to thank my fellow coh ort for all the hugs and study support. To my mother, who provides never-ending words of su pport and love. To my husband, Dushan, who is the perfect example of what every spouse of a doctoral student should be; understanding, supportive, patient, and loving. To, my boys and the joys of my life, Kyle and Tyler, may you also be motivated to reach your goals.


i TABLE OF CONTENTS LIST OF TABLES .................................... ................................................... ...................... iv LIST OF FIGURES ................................... ................................................... ..................... vi ABSTRACT .......................................... ................................................... ......................... vii CHAPTER 1: STATEMENT OF THE PROBLEM ............... .............................................1 Introduction ...................................... ................................................... .................... 1 Statement of the Problem .......................... ................................................... ........... 5 Purpose of the study .............................. ................................................... ............... 8 Need for the Study ................................ ................................................... ............... 9 Research Questions ................................ ................................................... .11 Delimitations ..................................... ................................................... ......12 Limitations ....................................... ................................................... .......12 Rationale ......................................... ................................................... ........13 Terms ............................................. ................................................... .........14 CHAPTER 2: LITERATURE REVIEW ..................... ................................................... ..17 Introduction ...................................... ................................................... .................. 17 Health Consequences ............................... ................................................... .......... 17 Growing Use ....................................... ................................................... ............... 24 History of use in Middle Eastern Countries ........ .......................................24 Dissemination to the United States ................ ............................................25 Factors Influencing Waterpipe Adoption in the U.S. ................................26 Adoption of practice among college students ....... .....................................30


ii Perceived Benefits ................................ ................................................... 30 Theoretical Framework ............................. ................................................... ......... 33 Theoretical Constructs ............................ ................................................... 34 Applicability of the Theory of Reasoned Action to W aterpipe Smoking ........................................... ................................................... .......36 Purpose of the Study .............................. ................................................... ............ 40 Significance....................................... ................................................... ......42 CHAPTER 3: METHODS ............................... ................................................... ..............44 Purpose of the Study .............................. ................................................... ............ 44 Study Design ...................................... ................................................... ................ 45 Study Population and Sample ....................... ................................................... ..... 46 Methodology ....................................... ................................................... ............... 48 Phases I and III Qualitative Research ............. ...........................................48 Participant Observations .......................... ..................................................4 9 Intercept Interviews .............................. ................................................... ..50 Demographic Characteristics of the Intercept Interv iews ......................... 51 Focus Groups ...................................... ................................................... ....52 Demographic Characteristics of the Focus Group Part icipants ................ 53 Verification Focus Group .......................... ................................................54 Online Survey ..................................... ................................................... ....57 Survey Instrument Development ..................... ......................................... 57 Response Rate ..................................... ................................................... ... 63 Survey Sample Characteristics ..................... ............................................ 65 Summary of Qualitative and Quantitative Research Sa mples .................. 67 Data Analysis ..................................... ................................................... ................ 68 Qualitative Analysis .............................. ................................................... ..68 Quantitative Data Analysis ........................ ................................................69 Univariate Analysis ............................... ................................................... ..69 Bivariate Analysis ................................ ................................................... ...70 Multivariable Analysis ............................ ................................................... 70


iii CHAPTER 4: RESULTS ............................... ................................................... ................72 Qualitative Study Findings – Phase I and Phase III .............................................. 72 Observations ...................................... ................................................... .....72 General Attitudes ................................. ................................................... .. 72 Behavioral Beliefs ................................ ................................................... .. 74 Normative Beliefs ................................. ................................................... 82 Motivation to comply .............................. .................................................. 84 Quantitative Study Findings – Phase 2 ............. ................................................... 86 Online Survey ..................................... ................................................... ....86 Prevalence of Waterpipe Tobacco Smoking ........... .................................. 86 Bivariate Analysis ................................ ................................................... .. 86 Tobacco Use Characteristics ....................... .............................................. 87 Multivariable Analysis ............................ .................................................. 97 CHAPTER 5: DISCUSSION ............................ ................................................... ...........102 Research Summary .................................. ................................................... ........ 102 Discussion of Results ............................. ................................................... .......... 103 Recommendations ................................... ................................................... ......... 109 Research Next Steps ............................... ................................................... ......... 110 Study Strengths and Limitations ................... ................................................... ... 111 Dissemination of Findings ......................... ................................................... ...... 114 Summary and Conclusion ............................ ................................................... .... 115 REFERENCES ........................................ ................................................... .....................117 APPENDICES ........................................ ................................................... ......................124 Appendix 1: Observation Guide .................... ................................................... 125 Appendix 2: Survey Instrument .................... ................................................... .. 136 Appendix 3: Survey Email ......................... ................................................... ..... 142


iv LIST OF TABLES Table 1: USF Housing Student Profile Spring 2011 .. ................................................... .....47 Table 2: Demographic Information for Focus Group P articipants ...................................55 Table 3: Theoretical Construct Measurements ....... ................................................... ........58 Table 4: Cronbach Alpha Scores for Factors ........ ................................................... ..........61 Table 5: Reliability of Constructs ................ ................................................... ...................61 Table 6: Demographic Characteristics of Sample .... ................................................... ......65 Table 7: Sequential Mixed Method Design ........... ................................................... .........67 Table 8: Survey Items Representing the Theoretical Constructs of Intention and Attitudes.. ....................................... ................................................... ..................88 Table 9: Constructs of Intention and Subjective Nor ms ................................................ ....89 Table 10: Constructs and Survey Items ............. ................................................... .............91 Table 11: Survey Items for Motivation to Comply and Subjective Norm .........................92 Table 12: Survey Items Representing the Constructs of Attitude and Outcome Expectancy ........................................ ................................................... .............93


v Table 13: Survey Items Utilized in the Analysis ... ................................................... .........95 Table 14: Factors, Cronbach Alpha, and Associated V ariables .........................................9 8 Table 15: Factors and Corresponding Correlation Coe fficients ......................................100


vi LIST OF FIGURES Figure 1: Waterpipe and Parts..................... ................................................... .................... 3 Figure 2: The Theory of Reasoned Action ........... ................................................... ......... 36 Figure 3: The University of South Florida ......... ................................................... ............ 46 Figure 4: Survey Case Deletions.................... ................................................... ................ 64 Figure 5: TRA model and Correlations .............. ................................................... ......... 101 Figure A1: Diagram of first two observation sites ................................................... ...... 129 Figure A2: Hookah bar layout ...................... ................................................... ............... 131 Figure A3: Layout of hookah bar.................... ................................................... ............. 133


vii ABSTRACT Waterpipe tobacco smoking has migrated from being a custom of some cultures to becoming a staple around college campuses. The soc ial nature and flavored tobacco encourage initiation in this tobacco nave age grou p. The study was a sequential mixed method design, emp loying primary data collection and analysis of a random sample of unive rsity students who live on campus at a single university. The study involved observation s (N=6), intercept interviews with smokers and nonsmokers (N=63), three focus groups ( N=31), and an online survey (N=288). Findings were centered on the constructs of the The ory of Reasoned Action by Ajzen and Fishbein (1975), who propose there are sp ecific precursors that lead to intention to perform a behavior. The data suggeste d that attitudes were more positively correlated with intention to smoke waterpipe tobacc o than subjective norm. Attitude is influenced by outcome expectancies. The data sugge st that positive outcome expectancies are influenced primarily by the social nature of hookah smoking and other perceived positive benefits. Negative health effec ts and family/culture were also influential, albeit to a lesser extent. Developmen t of interventions focused on an attitudinal shift may help to decrease uptake and c ontinuation of waterpipe tobacco smoking in this population.


CHAPTER 1: STATEMENT OF THE PROBLEM Introduction College life is associated with greater freedom and less parental constraints, allowing young adults to explore many types of beha viors in which they might not normally engage. Some of these behaviors have the potential to impose a health risk to the individual. More specifically, this life trans ition period correlates with initiation and escalation of current usage of tobacco product(Scha ne, Glantz, & Ling, 2009). Tobacco use is a common practice among college students (Cl arkin, Tisch, & Glicksman, 2008). The term social smoking is often used by college students to describe thei r tobacco smoking behavior as occurring more often with other s than in isolation (Moran, Wechsler, & Rigotti, 2004). With cigarette smoking, this concept of social smoking was found to be inversely associated with the intention to quit smoking in college students, indicating that although people characterize themse lves as social smokers, the practice may be occurring more frequently and perhaps in iso lation (Moran, et al., 2004). Tobacco use, in general, has the propensity to lead to lifelong nicotine dependence and to the development of health risks such as cardiovascu lar disease, cancers, and chronic obstructive pulmonary disease (Roskin & Aveyard, 20 09; Tavafian, Aghamolaei, & Zare, 2009). In the past decade, waterpipe tobacco smoking has b ecome a popular social event among college students. Campus activities and clubs have utilized waterpipe smoking to


encourage students to attend and participate in a v ariety of other social functions (DeCouteau, 2009; Hookahlover, 2008; Kinman, 2010; SouthSmoke, n.d.). Perceived as having a lower health risk than cigarette smoking, waterpipe smoking may be on the rise among college students (Eissenberg, Ward, Smith-Sim one, & Maziak, 2008; Primack, Walsh, Bryce, & Eissenberg, 2009).Waterpipe smoking involves a combination of tobacco, water, wood charcoal and a device known as a waterpipe, goza, shisha or hookah. A waterpipe can have one or multiple hoses attached to the body allowing multiple persons to smoke from the same source. Ch arcoal is utilized in the process of waterpipe smoking to heat the tobacco to temperatur es close to 450 C, whereas, cigarette tobacco is burned at temperatures of close to 900 C(Bacha, Salameh, & Waked, 2007). By creating negative pressure generated on inspirat ion through the hose, both the charcoal’s emitted chemicals and the tobacco smoke produced are passed through a water bowl component of the waterpipe where it is believe d by some that the nicotine and chemicals, such as arsenic, chromium and lead are f iltered (Grekin & Ayna, 2008). From the water bowl, the smoke continues through the hos e and into the mouth and lungs of the smoker. The typical course of smoke in a waterpipe during i nhalation is through a tube traveling from the top of the waterpipe, where the burning tobacco is located, descending into a bowl of water where it bubbles through the w ater after which it is inhaled through a tube attached to a mouthpiece. Figure 1.1 illustra tes a waterpipe device and parts.


Figure 1: Waterpipe and Parts Source: Picture courtesy of Elise Pautler Although many young adults are aware of the risks o f cigarette smoking, studies have found that young adults do not perceive waterp ipe smoking as being harmful or addictive when compared with cigarettes and other f orms of tobacco smoking (Eissenberg, et al., 2008; Smith, Curbow, & Stillma n, 2007). Research now confirms that waterpipe smoking carries risks as great, if not gr eater than cigarettes (Mohammad & Kakah, 2008; Monzer, Sepetdjian, Saliba, & Shihadeh 2008; Neergaard, Singh, Job, & Montgomery, 2007). This misperception of safety is believed to have contributed to the initiation and continued use among college students (Roskin & Aveyard, 2009). Waterpipe smoking has been a common cafe staple in other countries for many years, providing men the opportunity to socialize(M aziak, Rastam, Ibrahim, Ward, &


Eissenberg, 2008). Smoking, in general, is customa ry in many countries. In some case, it may exhibit more than just a personal habit, but rather an important aspect of culture.The social acceptability and togetherness t hat waterpipe smoking invites has extended beyond the boundaries of the Eastern Medit erranean Region (EMR) and on to college campuses and surrounding hookah bars (Lenne y & Enderby, 2008; Lyon, 2008; Primack, et al., 2008). Waterpipe smoking has been described by users as pr oviding a sense of sharing and belonging through passing the pipe from one per son to another (Prignot, Sasco, Poulet, Gupta, & Aditama, 2008). This social accep tance and the ubiquitous availability of waterpipe lounges/stores has contributed to its use in college students (Primack, et al., 2008). The thriving caf culture and media accepta nce of waterpipe tobacco is ahead of public health efforts to educate the public and det er the behavior. In addition to the social nature of waterpipe smoki ng, there are several other influential factors that contribute to the adoption of waterpipe tobacco smoking among college students. These factors include a mystical and exotic appeal, a perceived cultural experience, its affordability and popularity, the t aste of flavored tobacco, a novel experience, and the perceptions of being a less har mful mechanism of smoking. From a cultural perspective, Roskin et al. (2009) reported that students who were of Arabic origin found waterpipe smoking to be a natural expr ession of their heritage and students who were not Arabic viewed waterpipe tobacco smokin g as an alternative cultural view. Some of these non-Arabic students had discovered wa terpipe tobacco while visiting the EMR and sought a reminder of a culture different fr om their own. From the college student’s perspective, these factors mentioned abov e are all positive attributes of


waterpipe smoking and encourage continued use (Grek in & Ayna, 2008; Primack, et al., 2008; Rastam, Ward, Eissenberg, & Maziak, 2004; Ros kin & Aveyard, 2009). The factors that draw college students to waterpipe smo king appear to be both personal and social in nature. This dissertation study examined the degree to whic h personal factors (attitudes) and social factors (subjective norms) influence int ention to smoke waterpipe tobacco and the strength of these relationships. This study wa s conducted with college students from the University of South Florida. Understanding the se different factors of influence between waterpipe smokers and nonsmokers will help guide future interventions as well as add to the current body of literature on smoker’ s attitudes and beliefs about waterpipe smoking. Statement of the Problem Tobacco smoking is responsible for 440,000 deaths e ach year in the United States and 5.6 million years of potential life lost (ACS, 2010; Fromme, et al., 2009). Because tobacco is utilized in waterpipe smoking, the same carcinogens (polyaromatic hydrocarbons, and volatile aldehydes) that are in c igarette tobacco are also found in waterpipe tobacco (Monzer, et al., 2008). These ca rcinogens are the main components in mainstream smoke (smoke inhaled/exhaled by the smok er) and the side-stream smoke (smoke emitted into the atmosphere from burning cig arettes and tobacco) (Daher, et al., 2010). Additionally, these carcinogens are the pr ecursors to lung cancer and other respiratory ailments (Prignot, et al., 2008).


Use of tobacco products, in any form, has the abili ty to cause health problems and lead to dependency on nicotine. Marketing of tobac co products during college activities and events is common and college students are viewe d as an ideal target audience for tobacco companies to promote their products. Colle ge students constitute the youngest group that tobacco companies can legally target in their marketing strategies. Tobacco company documents indicate the company’s intention to promote smoking during this vulnerable life stage (Ling & Glantz, 2002). Along with the excitement and exploration that coll ege affords, college life produces stressors and uncertainties. Students uti lize and develop a myriad of tactics to deal with their individual stress: some are healthy and others are not. Common health risk behaviors adopted by college students to comba t the stressors that college invokes are tobacco smoking, alcohol consumption, and presc ription and illegal drug use (Naquin & Gilbert, 1996; Patterson, Lerman, Kaufmann, Neune r, & Audrain-McGovern, 2004). It has been estimated that one in four college stud ents smoke and of these, seventy-five percent will continue to smoke into adulthood (Von Ah, Ebert, Ngamvitroj, Park, & Kang, 2004). Physiologically, the biphasic qualiti es of nicotine (excitation and relaxation) drive the desire, need, and continuatio n of smoking. Because waterpipe smoking contains nicotine, the World Health Organiz ation (WHO) has declared it an emerging global public health problem (WHO, 2007). According to Maziak et al. (2004), the initiation o f waterpipe smoking in the U.S. occurs most often in college age students. Reasons cited for uptake have included the novelty of waterpipe, the ease of access, and the l ow cost of smoking. Eissenberg et al. (2008) reported that 20% of college students who re sponded to a questionnaire had tried


waterpipe tobacco in the past 30 days. Similarly, Smith et al. (2007) found nearly 16% of college freshmen participating in an Internet surve y responded to having smoked waterpipe tobacco in the past 30 days. The Global Youth tobacco survey indicated that ten to twenty percent of college students had smoke d hookah within the past month (Maziak, 2008). In addition to the potential for nicotine addiction waterpipe smoking has been shown to lead to infectious diseases and may be a g ateway to the usage of psychoactive substances, such as cannabis (Eissenberg, et al., 2 008; Prignot, et al., 2008). The health consequences of waterpipe smoking are becoming more evident as research is focusing on the health aspects and outcomes related to both casual and continual smokers. Transitioning from a social smoker to a lone smoker has been suggested as a sign of dependence (Maziak, Eissenberg, & Ward, 2005; Sa lameh, Waked, & Aoun, 2008). The lone smokers were noted to have a more intense smoking habit than a social smoker, suggesting that greater intensity may lead to addic tion (Eissenberg, et al., 2008; Maziak, Eissenberg, et al., 2005; Ward, et al., 2005). Asf ar et al. (2005) found that two-thirds of the waterpipe smokers in his study were willing to quit, but failed to do so. This failure to quit may represent a tobacco dependence issue. The addictive potential of waterpipe smoking is still being investigated and includes fa ctors such as duration and frequency of smoking, smoke chemical properties, type of tobacco type of charcoal and volume of inhaled smoke (Maziak, 2008).


Purpose of the study The purpose of this study was to identify the behav ioral determinants of intention to smoke waterpipe tobacco. Theoretically based re search exploring these constructs among waterpipe smokers in the U.S. and particularl y among college campuses is limited (Eissenberg, et al., 2008; Grekin & Ayna, 2008; Smi th-Simone, Curbow, & Stillman, 2008; Ward, et al., 2007). Furthermore, there is l ittle research on the specific social influences of waterpipe smoking as determinants of intention among U. S. public university students. In this study, Fishbein and Ajzen’s Theory of Reaso ned Action (TRA) (Fishbein & Ajzen, 1975) was used to investigate the impact o f intrapersonal and interpersonal factors on waterpipe smoking intention among colleg e students. Fishbein and Ajzen’s TRA (Fishbein & Ajzen, 1975) views a person’s inten tion as the immediate determinant of performing or not performing a behavior. The mo del also assumes that intentions will reflect rational and systematic decisions based on both social and personal influences. These determinants correspond with a person’s attit ude towards the behavior. Social determinants consist of an individual’s perception of surrounding significant people that influences the performance of a behavior and the in dividual’s motivation to comply with these perceptions. Social influences are known as subjective norms (Ajzen & Fishbein, 1980a). Personal determinants include individual’s beliefs and the personal evaluation of the behavior. Fishbein and Ajzen (1975) argue that the relationship of variables determines intention, and intention predicts behavi or.


n The TRA has been used to predict and understand a v ariety of health-related behaviors in university students, for example, AIDS prevention(Fisher, Fisher, & Rye, 1995), cigarette smoking(Rhodes & Ewoldsen, 2009), and drug and alcohol use. One study utilized this theory to understand the influe nces of waterpipe smoking among college students in the U.S. (Primack, et al., 2008 ). Primack et al. (2008)examined the association between harm perception, dependence, pe er acceptance and popularity as associated with the commonness of waterpipe smoking Although not using the TRA, Smith-Simone et al. (2008) explored psychosocial ri sk profiles of waterpipe smoking using attitudes and beliefs derived from cigarette smokers. The TRA was selected as the theoretical framework f or guiding this research based on the combination of intrapersonal construct s (e.g., attitudes) and interpersonal constructs (e.g., peer/other influencers), both of which influence the intention for waterpipe smoking, albeit in probable varying degre es. This study differed from Primack’s study in two important ways. First, this study provided a measure of the individual’s motivation to comply with significant others. Second, it provided more measures in assessing individual beliefs. Primack et al. (2008) did not provide a measure of individual’s motivation to comply and provided o nly one item to measure harm perception. Need for the Study Prevention of lung disease and promotion of lung he alth has been one of the primary overall goals of public health since its in ception in dealing with tuberculosis years ago. Because smoking is a preventable cause of death, efforts to deter young


people from smoking and educate the public on the d eleterious effects of secondhand smoke have charged both public health practitioners and research institutions to fund and conduct research to curb the morbidity and mortalit y related to smoking. The tobacco industry continues to develop and distribute a plet hora of nicotine products to maintain addiction in smokers. With the U.S. Food and Drug Administration (FDA) now having control over tobacco products(FDA, 2009) we may fin d ourselves in a position to help those with addictions as opposed to constantly reac ting to the marketing tactics of Big Tobacco. In the meantime, we must develop grassroots inter ventions based on research with the priority population. In a press release issued March 8, 2007, the Americ an Lung Association recommended “… increased research on all aspects of hookah use.” Despite this push to increase research in this tobacco arena, only a few articles have been published in the United States. The American Lung Association (ALA ) release also mentions that ”hookah bars are growing in popularity in the U.S. especially among 18-to 24-year-olds, becoming the first new tobacco use trend of the twe nty-first century” (Association, 2007). These national recommendations align with this rese arch study aimed at college students who partake in waterpipe smoking. A 2007 trend report(ALA, 2007) issued on the emergi ng deadly trend of waterpipe smoking stated: More research is needed into the health effects of waterpipe use, and the patterns and process of beginning to use wa terpipes amongst various populations. Since little data exis t on prevalence


of hookah use in the United States, national survey s on youth and adult tobacco use should consider adding a question on this topic. There also is virtually no research on the risks of secondhand smoke from waterpipe use. (p. 6) It is evident that waterpipe tobacco research is at the forefront of research topics for the public health practitioners. U.S. college students have a higher prevalence of w aterpipe smoking as compared to the general population(Smith-Simone, et al., 200 8). Given the increase in hookah lounges, advertisement, and percentage of waterpipe smokers around college campuses, college students are considered a vulnerable popula tion for increased marketing and promotion efforts aimed to increase waterpipe smoki ng (Moran, et al., 2004). To develop strategies that educate and promote healthy behavio ral alternatives and decrease waterpipe smoking, it is necessary to investigate t he personal and social influences of waterpipe smoking from the priority population. This research focused on the population of college students at the University of South Florida(USF) and included a survey that obtai ned a large sample of students from the university. In addition, qualitative data was collected through intercept interviews, focus groups, and observational studies. Research Questions Six research questions guided the proposed research : Research Question #1:


What is the relationship between attitudes and int ention to smoke waterpipe tobacco? Research Question #2: What is the relationship between subjective norms and the intention to smoke waterpipe tobacco? Research Question #3: What is the relationship between beliefs of importa nt others and subjective norms? Research Question #4: What is the relationship between motivation to comp ly and subjective norms? Research Question #5: What is the relationship between outcome expectanci es and attitudes? Research Question #6: What is the relationship between evaluation of outc ome expectancies and attitudes? Delimitations This research was conducted with students who resid e in university housing and were 18 years of age or older. Limitations There are known limitations to this research that d eserves mentioning. The study sample was drawn only from students who reside in u niversity housing at the University of South Florida thereby limiting its generalizabil ity. Of these individuals, only those


who were at the Marshall Center (student center) du ring recruitment for the qualitative portion were recruited. Those students who may be living within university housing, but are not at the Marshall Center or notice the focus group flyers did not have an opportunity to participate in the qualitative research. Additi onally, any other students who attend the university, such as graduate students, commuter stu dents, distance learning students, or students who attend the other university campuses w ere excluded from participating in the study. Other limitations to the study included students wh o may be unaware of waterpipe smoking. Of the participants who did choose to par ticipate, there remains the limitation of social desirability and self-reporting, both of which have the propensity to alter the results. Because this study is cross-sectional in design, it does not provide directionality or causality claims. Rationale The rationale for selecting students who live in Un iversity housing is two-fold. First, this sample of students is closer in age to the age of initiation found in the literature and may highlight whether initiation in this sample of college students is greater in high school years or college years. Barnett et al. (200 9) found in a sample of Florida high school students that 11% of students had first trie d waterpipe tobacco during their high school years. Grekin et al. (2008) found that 58% of the Michigan college sample they surveyed had initiated waterpipe smoking after the age of 17 years, suggesting college years as the time to begin experiencing waterpipe t obacco. Secondly, the study may


provide USF with more focused information to better equip the Student Health Services for campus-based prevention programs and health lit erature. Terms 1. Ajami (Ajamy) – Also known as tumbak plain tobacco made of moistened shredded leaves, soaked for hours in water before b eing squeezed and packed in the bowl of the hookah (Chaouachi, 2009). 2. Arghileh – A type of waterpipe used in Easter Mediterranean Region(Shihadeh, 2003) 3. Attitude – A person’s negative or positive judgment of a behavior(Ajzen & Fishbein, 1980a) 4. Behavioral beliefs – the beliefs that underlie a pe rson’s attitude (toward a behavior) (Ajzen & Fishbein, 1980a). 5. Carbon monoxide – An odorless, colorless, toxic gas found in tobacco smoke( An introduction to indoor air quality 2009). 6. Goza – An apparatus used to smoke tobacco, similar in s tructure to a waterpipe(Khater, Abd El-Aziz, Al-Sewaidan, & Chaou achi, 2008). 7. Hookah – An apparatus or ancient pipe traditionally used in Africa and Asia (Chaouachi, 2009). 8. Hubble Bubble – A method of smoking through a water pipe, whereby, the sound of the air flowing through the water bowl pro duces a sound described as such 9. Important (significant) others – People who influen ce a decision to behave a certain way (e.g., spouse, parents, friends)(Ajzen & Fishbein, 1980a).


10. Jurak – A mixture of about 30% tobacco and 70% molasses/ honey/glucose syrup and minced fruits. It does not contain glyce rin as in maassel (Chaouachi, 2009). 11. Maassel (Mo’assel ) – Also known as tobamel (tob stands for tobacco and mel for honey, in latin). A mixture of about 30% tobac co and 70% molasses/honey/glucose syrup plus glycerol and esse nces(Chaouachi, 2009). 12. Narghile – A Persian/Iranian and Turkish word to describe a water pipe or apparatus, typically based on a coconut as the vess el (Chaouachi, 2009). 13. Normative beliefs – the beliefs that underlie a per son’s subjective norm (Ajzen & Fishbein, 1980a). 14. Polyaromatic hydrocarbons (PAHs) –A group of over 1 00 different chemicals that are formed from incomplete combustion. Some o f which are known carcinogens ( ToxFAQs 2010) 15. Psychoactive substances –A drug that can produce mo od changes and distorted perceptions (FreeDictionary, 2010) 16. Shisha – Of Persian origin, bottle (recipient) of water m ade of glass with a typical flask/vial form (Chaouachi, 2009). 17. Subjective norm – a person’s belief that specific i ndividuals or groups think he should or should not perform (a particular behav ior) (Ajzen & Fishbein, 1980a). 18. Tumbak –Also known as ajami plain tobacco made of moistened shredded leaves, soaked for hours in water before being sque ezed and packed in the bowl of the hookah(Chaouachi, 2009).


19. Waterpipe – also referred to as hookah or shisha has a mouthpiece, hose, water bowl, body, and a head that is filled with to bacco and then heated with charcoal (Eissenberg, et al., 2008).


CHAPTER 2: LITERATURE REVIEW Introduction Waterpipe smoking has obtained popularity among col lege students in the U.S., partly due to its exotic appeal, social nature, and perceived harmlessness. Waterpipe smoking differs from other health risk behaviors am ong young adults in that it is novel, lacks policy control, lacks uniform health messagin g, and holds diverse variations in its practice. For example, smokers can add alternative liquids into the water bowl as well as additional substances into the tobacco for varying degrees of effects. More studies are confirming waterpipe smoking as a health risk behav ior needing immediate attention both from the realm of health messages and overarching p olicy; however, more information is still needed to determine how best to align policy and messages to achieve behavior reduction. Health Consequences As mentioned previously, tobacco smoking is respons ible for 440,000 deaths each year in the United States and 5.6 million years of potential life lost (ACS, 2010; Fromme, et al., 2009). It has been determined that smoking causes coronary artery disease, stroke, and lung disease, however, due to the highly addict ive nature of nicotine many people continue to smoke (Roskin & Aveyard, 2009; Tavafian et al., 2009). Waterpipe smoking is an emerging problem from a public health perspec tive. Although waterpipe smoking is


perceived as being less harmful than cigarettes, mo re evidence is suggesting that it contains more harmful agents and has similar addict ive potential as cigarettes (Bacha, et al., 2007; El-Nachef & Hammond, 2008; Neergaard, et al., 2007; Shihadeh, 2003). Unlike cigarette smoking, waterpipe smoking has bee n shown to lead to infectious diseases, low birth weight infants, and possibly th e use of psychoactive substances (Eissenberg, et al., 2008; Prignot, et al., 2008; T amim, et al., 2007). The health consequences of waterpipe smoking are becoming more evident as research is increasing. Because tobacco is utilized in waterpipe smoking, t he same carcinogens (polyaromatic hydrocarbons, and carbon monoxide) th at are in cigarette tobacco are also found in waterpipe tobacco (Barnett, Curbow, Soule, Tomar, & Thombs, 2011; Monzer, et al., 2008). These carcinogens are the main comp onents in mainstream smoke (smoke inhaled/exhaled by the smoker) and are the precurso rs to lung cancer and other respiratory ailments (Prignot, et al., 2008). Wate rpipe smoking requires larger inhaled respiratory volumes which exposes the smoker to mor e carcinogens than during cigarette smoking (Fromme, et al., 2009). In a single study looking at heavy metals, Shihadeh (2003) found that waterpipe smoke contained an incr eased amount of nickel, arsenic, and cobalt. Although waterpipes do not emit as much se cond hand smoke as cigarettes, the large volumes of mainstream smoke exhaled expose ot hers to these hazardous chemical components. Another concern surrounding waterpipe smoking in the U.S. is the use of plastic hoses versus the permeable leather hoses us ed in the Eastern Mediterranean Region (EMR). The plastic hoses lead to significan tly increased levels of toxins, such as carbon monoxide and particulate matter(Saleh & Shih adeh, 2008).


n Because tobacco leaf combustion is an incomplete pr ocess and produces both gaseous and particulate matter, the waterpipe tobac co smoking does not minimize the health effects of the components (Al Mutairi, Shiha b-Eldeen, Mojiminiyi, & Anwar, 2006). Waterpipe and cigarette smoke contain simil ar toxic agents and due to the longer inhalational puffs required to generate the smoke w ith waterpipe smoking it is believed that up to 100 times more smoke is inhaled with wat erpipe than with cigarettes (Eissenberg, et al., 2008). The belief that waterpipe smoking is less harmful t han cigarette smoking is shared by physicians as well as users. Waterpipe smoking is perceived to be less harmful and less addicting because water filtration is thought to deliver less nicotine than cigarette smoking (Chaaya, et al., 2004; Shihadeh, 2003; Ward et al., 2007). This view is based on the fact that nicotine is water soluble and the bel ief that not only nicotine but other toxic substances will be filtered out prior to the smoke being inhaled (Neergaard, et al., 2007). In contrast to this view, research has shown that o nly five percent or so of the nicotine is dissolved in the water and waterpipe smokers increa se the duration of smoking and the volume of puffs to titrate the necessary nicotine t o meet the pleasurable or dependent effects that they may need or desire (Ward, et al., 2007). Polyaromatic hydrocarbons (PAH) and carbon monoxide (CO) are the main components of the mainstream tobacco smoke that cau se cancers and lung health issues (Monzer, et al., 2008). El Nachef (2008) conducted carbon monoxide measurements during waterpipe smoking sessions. The carbon mono xide levels exceeded environmental protection standards (O'Rourke, Hatch er, & Stepanski) levels of greater than 35 parts per million averaged over an hour (El -Nachef & Hammond, 2008). Other


researchers found some of the more carcinogenic PAH s to be as much as 50 times greater in waterpipe smoke as compared to cigarette smoke ( Sepetdjian, Shihadeh, & Saliba, 2008). Comparing cigarette smoke carcinogens to th ose produced by hookah in side stream smoke, Daher et al. (2010) found that the am ount of toxicants and ambient carcinogens in a single session of waterpipe smokin g equals that produced by two to ten cigarette smokers. A more recent study by Barnett et al. (2011) found exhaled CO to be much higher in patrons exiting hookah bars compared to patrons exiting regular bars that allow cigarette smoking. It is important to note that CO has a 200-300 times greater affinity for hemoglobin than does oxygen. In the presence of high CO level s individuals are at risk for hypoxemia (low level of oxygen in the bloodstream) and the side effects of carbon monoxide poisoning (Pierson & Kacmarek, 1992). CO poisoning is witnessed by nausea, headaches and blurred vision initially and dependin g on the level of exposure; it can lead to coma. Tufts University reported a student who s moked hookah for two hours straight and then vomited immediately after she left the lou nge (Wolf, 2010). Incidences such as these are not well documented, and may be common in rooms with little or no ventilation. Additionally, the same article report ed a dormitory fire from an unattended waterpipe. Fires have the potential to lead to eve n greater levels of CO exposure. Carbon monoxide has a six hour half-life in room ai r, indicating the potential for greater deleterious effects on the human body (Walsh, Czerv inske, & DiBlas, 2010). Lim et al. (2010) reported a hospital case of a young man who fell and injured his head after smoking shisha (Hookahlover). His CO level in his bloodstream on hospital admission was nearly twenty-eight percent; normal levels are less than two percent. Two other


emergency room cases also reported similarly high C O levels in waterpipe smokers (Cavus, Rehber, Ozeke, & Ilkay, 2010; Uyamk, Arslan Akay, Ercelik, & Tez, 2009). CO results in cellular poisoning and is often diffi cult to diagnose without a patient history of some type of exposure. Because tobacco leaf combustion is an incomplete pr ocess, it yields both gas and particulate matter. The gas composition consists of nitrogen, carbon dioxide, carbon monoxide, nitrosamine, acetaldehyde, formaldehyde, hydrocarbons and hydrogen cyanide. The particulate matter is composed of tar and nicotine (Al Mutairi, et al., 2006). It is the nicotine that leads to the cigarette addi ction and there is emerging evidence that waterpipe smoking also may lead to dependence (Jack son & Aveyard, 2008). Ward et al. (2007) found a strong association between cigarette smoking and waterpipe smoking suggesting that cigarette smoking may be a gateway to the use of waterpipe smoking and also that frequent use of waterpipe smoking may lea d to cigarette consumption. Asfar et al. (2005) found that men who routinely smoked wate rpipe tobacco in a social setting became lone smokers as time passed. Transitioning from a social smoker to a lone smoker has been suggested as a sign of dependence (Maziak, Eissenberg, et al., 2005; Sa lameh, et al., 2008). The lone smokers were noted to have a more intense smoking h abit than a social smoker, suggesting that greater intensity may lead to addic tion (Eissenberg, et al., 2008; Maziak, Eissenberg, et al., 2005; Ward, et al., 2005). Asf ar et al. (2005) found that two-thirds of the smokers in his study were willing to quit, but failed to do so. This inability to quit may represent a tobacco dependence issue. In a U. S. study of two campuses, thirteen percent of students said stated they were “hooked o n hookah” (S. Smith-Simone, Maziak,


Ward, & Eissenberg, 2008). The addictive potential is still being investigated and includes factors such as duration and frequency of smoking, smoke chemical properties, type of tobacco, type of charcoal and volume of inh aled smoke (Maziak, 2008). According to DiClemente (2010), addiction begins wi th experimentation, followed by casual use, regular use, abuse and then dependence (DiClemente, 2010). Experimentation with tobacco products is common amo ng young adults (Moran, et al., 2004). The college years provide a prime opportuni ty to implement programs focused on both helping individuals quit smoking and prevent t hose who are experimenters from becoming regular smokers (Wechsler, Rigotti, Gledhi ll-Hoyt, & Lee, 1998). However, before programs can be implemented effectively, a m ore complete understanding of the drivers and the behavior from the perspective of th e consumer is needed. The potential for acquiring an infectious disease i s another public health concern with waterpipe smoking. Prignot et al. (2008) note d that waterpipe smoking was a contributor to the spread of tuberculosis by infect ed persons who shared a mouthpiece with non-infected individuals during a smoking sess ion. This discovery highlights the concern over viruses and bacteria that may be trans mitted through oral secretions. Due to lack of public health oversight, poor sanitation an d cleaning procedures of waterpipes raise concern of infectious disease spread. In Ind ia, if a person does not share the mouthpiece it is considered an offense (Maziak, War d, Afifi Soweid, & Eissenberg, 2004). Untreated infectious diseases can lead to chronic d iseases, and thus, pose a long term public health problem. An early sign of chron ic lung disease can be noted in


pulmonary function testing, specifically the forced expiratory volume in one second (FEV1). Two separate studies conducted pulmonary functi on tests on waterpipe smokers and found not only their FEV1 was decreased, but also that they showed decreased values in their peak flow rates and the forced expiratory flow (Kiter, Ucan, Ceylan, & Kilinc, 2000; Mohammad & Kakah, 2008). These measurements reflect stricture in bronchial diameter resulting from either inflammatory respons es or immune responses. Associated drug use is yet another public health co ncern with the epidemic of waterpipe smoking. In a qualitative study conducte d with Palestinian youth, Makhoul and Nakkash (2007) found that youth added crushed h allucinogenic pills to their waterpipe tobacco and smoked the combination. Ther e are concerning anecdotal reports of waterpipe smokers adding wine and other alcoholi c beverages to the water bowl of the pipe to experiment for different effects and flavor s. Waterpipes have been used for smoking hashish and marijuana (Maziak, Ward, Afifi Soweid, & Eissenberg, 2005). Waterpipe smoking has been shown to predispose indi viduals to additional negative health consequences that are of public hea lth concern. Tavafian et al. (2009) conducted a survey among waterpipe smokers and foun d that waterpipe smoking was associated with a lower quality of life, poor healt h conditions, and increased physical limitations. These individuals generally reported suffering more depression and anxiety than nonsmokers (Tavafian, et al., 2009). These fi ndings correlate with the literature of depression associated with smoking cigarettes (Schl eicher, Harris, Catley, & Nazir, 2009). College students in the U.S. gave symptoms of depression as a reason they choose to smoke waterpipe (Grekin & Ayna, 2008). Depressi on and the risk of suicide in college students should be a major mental health and public health concern. Al Mutari et al.


(2006) found that waterpipe smokers have a higher i ncidence of chronic bronchitis than did cigarette smokers and that waterpipe smoking ca n lead to oral cancers. El-Hakim et al. (1999) reported two initial cases of lip carcin oma associated with waterpipe smoking. Waterpipe smoking has been linked to bronchial canc ers, atherosclerosis, and low birth weight infants (Ashmawi, 2003; Gupta, Boffetta, Gab orieau, & Jindal, 2001; Nuwayhid, Yamout, Azar, & Kambris, 1998). Growing Use Worldwide there are estimated to be 100 million dai ly waterpipe smokers (Ward, et al., 2005). Depending on the country of origin, waterpipe smoking may be referred to as shisha, boory or goza (Egypt and Saudi Arabia) hookah (Africa and the In dian subcontinent), hubble bubble (many regions), narghile, argileh or nargile (Israel, Jordan, Lebanon and Syria) (Chaaya, et al., 2004; Neergaard et al., 2007). Current studies demonstrate that twenty percent of the population i n the Eastern Mediterranean Region (EMR) smoke waterpipe tobacco (Tavafian, et al., 20 09). In Israel, approximately twenty-two percent of youth 12 to 18 years of age s moke waterpipe every weekend (Tavafian, et al., 2009). Some individuals in the EMR do not even consider waterpipe smoking a form of tobacco smoking, consider it natu ral, and harmless (WHO, 2006). History of use in Middle Eastern Countries Waterpipe smoking has been around for centuries wit h one of the earliest recordings of its use in 1616 in India (Goodman, 19 93). Identifying the actual country from which waterpipes were invented is less clear, however, most research agreement centers on the Persians for invention and the Musli ms for the spread of its use (Goodman, 1993). Even in the early years of waterpipe usage for tobacco consumption, waterpipe


tobacco smoking was the centerpiece of social inter action among coffeehouses (Goodman, 1993). Its use in coffeehouses has trans cended many centuries and remains customary with elder men in the Eastern Mediterrane an Region (EMR). These coffeehouses are still prevalent in the EMR and con tinue to encourage socialization. The Middle East experienced a resurgence of waterpi pe smoking in the 1990s with the introduction of flavored tobacco, known as maassel (Maziak, 2008; Rastam, et al., 2004). The practice of waterpipe tobacco smoki ng has spread to women and children in the EMR (Salameh, et al., 2008). In the EMR, ci garette smoking in women is considered taboo, however, waterpipe smoking is not (Maziak, Hammal, et al., 2004). This acceptance has prompted women to take up the p ractice at an alarming rate partly due to the misperception that it is safe and suppor ted by increased family tolerance (Al Mutairi, et al., 2006; Asfar, Ward, Eissenberg, & M aziak, 2005). Waterpipe smoking by some women in Syria has been described as “a harmle ss toy” (Mohammad & Kakah, 2008). In 2005, the Lebanon Global Youth Tobacco S urvey (GYTS) indicated that 33.9% of school children were current waterpipe smo kers surpassing the number of cigarette smokers (Khalil, Heath, Nakkash, & Afifi, 2009; Maziak, 2008). It is not uncommon for a child to smoke with their parents in the Middle East. Dissemination to the United States The spread of waterpipe lounges/restaurants and oth er social places to smoke are beginning to take hold in the U.S., especially amon g college students. It is not certain if this smoking trend reflects students’ desires to ex perience practices from another culture or just an opportunity to relax and socialize.


The introduction of flavored tobacco is believed to have made waterpipe smoking more popular among youth. Called maassel flavored tobacco was introduced in the 1990s and currently accounts for ninety percent of waterpipe tobacco sales. Maassel provides tobacco with a fruity flavor. It also has less nicotine-rich tobacco due to added stems and glycerin to aid in fermentation. W hen burned, flavored tobacco produces a caramelized smell similar to cotton cand y (Shihadeh, 2003). The lure of fruit flavored smoke, the associated social aspect of wat erpipe smoking, and added misperception of water filtering properties of a wa terpipe have led smokers to believe that waterpipe smoking is a safer alternative to to bacco consumption than traditional cigarette smoking, cigars, and chewing tobacco (Maz iak, 2008; Smith, et al., 2007). Although Primack et al. (2008) reported that nearly 300 cafes opened in the U.S. between 1999 and 2004, investigation into the socio -demographic characteristics of individuals who frequent hookah bars and the derive d benefits/barriers need to be explored. Nationwide, it is estimated that on aver age five new hookah bars open each month ("Hookah bars," 2009). These cafes are sprea ding into malls, hotels, and even into residential neighborhoods (Neergaard, et al., 2007) This increased availability of waterpipe tobacco, through both bars and Internet s ites, has been an added contribution to its increased use among college students, aside fro m the perceived psychological and physiological benefits cited in the literature (Maz iak, et al., 2008). Factors Influencing Waterpipe Adoption in the U.S. Reasons for waterpipe smoking’s adoption in the Uni ted States can be attributed to several possible factors including immigration, marketing, ease of access, lack of regulations, and appeal among youth.


Throughout history, the U.S. has become culturally enriched with traditions and values from around the world. Despite terrorist ev ents from 2001, the United States has seen an increase in persons obtaining their legal p ermanent residence during the past several years (Security, 2009). Immigrants from S yria, India, Egypt, Iran, Iraq, Israel, Lebanon, Saudi Arabia, and Turkey have introduced M iddle Eastern cultural traditions such as waterpipe smoking to the U.S. With continu ed influx of persons from the EMR, the prevalence of waterpipe smoking is likely to in crease. According to a study by the Center for Immigration Studies, Middle Easterners r epresent one of the fastest growing segment to the US, with one-third the total living in California, New York, and Michigan (Camarota, 2007). Weglicki et al. (2008) conducted a study among a community sample of Arab-American and non-Arab-American high school students. The study concluded that the Arab-American youth had a higher prevalenc e of ever waterpipe smoking and current waterpipe smoking than non-Arab-Americans, respectively (38% vs. 21% ever smokers; 17% vs. 11% current smokers) (Weglicki, Te mplin, Rice, Jamil, & Hammad, 2008). This may be attributed to waterpipe tobacco smoking being more customary in families of Middle Eastern descent. Jamil et al. ( 2011) did find a positive correlation between having a father, mother, or sibling smoking waterpipe tobacco at home and an individual smoking waterpipe tobacco (OR = 9.5, p < 0.01). This study suggests that familial social norms and customs of Arab-Americans may contribute to initiation and continuation of smoking waterpipe tobacco in younge r adult males (Jamil, et al., 2011). Along with the practice of waterpipe smoking, colle ge students may have acquired their perception of the habit as safe from Middle Eastern Immigrants. These beliefs, knowledge, and attitudes are often shaped by cultural attributes and may vary


among residents as the U.S. becomes more ethnically diverse (Asfar, et al., 2005). Research among U.S. college students has shown that they also share the inaccurate belief that waterpipe smoking is less harmful than cigarettes (Grekin & Ayna, 2008; Roskin & Aveyard, 2009; Smith, et al., 2007). Anot her factor contributing to the increased popularity of waterpipe smoking is the av ailability of hookah bars. Currently there are five hookah bars/restaurants around the U niversity of South Florida (USF) located in Tampa, Florida and visible on the World Wide Web. Access to information about waterpipe smoking on th e World Wide Web may also fuel interest. In the last decade, there has been a steady increase in the number of social internet sites, blog sites, and commercial s ites related to waterpipe smoking. Most recently, a blog was posted regarding how to smoke hookah (waterpipe) in a dormitory without being caught (Whokah333, 2010). These read ily available forms of communication have led to the spread of waterpipe s moking to the U.S. along with the proliferation of waterpipe cafes that have opened i n the past decade (Maziak, 2008). Social technology has drawn an increasingly large n umber of high school and college students, enabling individuals to share the ir day to day lives with others. These information highways also enable individuals to sha re their experiences with waterpipe smoking, as well. Hookah blog sites and websites a llow discussions of favorite flavors and favorite bars. In addition, the ease of orderi ng waterpipes and flavored tobacco products online has led to increased waterpipe smok ing on college campuses (Parna, Usin, & Ringmets, 2008; S. Smith-Simone, et al., 2 008). Most recently, Tampa has experienced a new channeling of information leading to the popularity of waterpipe smoking. A shopping website called “Group On” adv ertised the cultural event of


n hookah smoking as a venue to experience the culture of the Tampa Bay area (Reuters, 2009). In 2008, students from the University of So uth Florida started a Facebook interest group surrounding hookah. The “USF mUndays” site h as current wall posts advertising hookah lounges. The current friend list is over 10 00 persons (Facebook, 2010). Increased marketing efforts have been shown to lea d to initiation of smoking (Gilpin & Pierce, 1997). Marketing of waterpipe sm oking through sponsorship of college activities such as music events, advertising in col lege papers and local bars, and provision of free samples has the propensity to lead to incre ased usage of waterpipes. Jewish student organizations on college campuses sponsor “ Hookah in the Sukkah” events during annual Sukkot harvest festival is one exampl e of how waterpipe has been incorporated into the college life (Kelly, 2009; Le win, 2006). Students also use waterpipe smoking as a draw to get other students to join the ir organization (DeCouteau, 2009). Additionally, through commercialization, the Intern et has made waterpipe smoking appear “cool” to youth and young adults (Ra stam, et al., 2004). The Internet has been suggested as the primary source for home smoke r’s purchases (Maziak, 2008). Smith-Simone et al. (2008) found statistically sign ificant increases in waterpipe smoking in college students due to peer influence, appearan ce of being “cool”, and the attractiveness of the product through marketing sou rces. Eissenberg et al. (2008) reported that twenty percent of college students wh o responded to a questionnaire had tried waterpipe tobacco in the past 30 days. Simil arly, Smith et al. (2007) found sixteen percent of college freshmen participating in an int ernet survey responded to having smoked waterpipe tobacco in the past 30 days.


Lax regulations and no required health warnings may also contribute to the spread of waterpipe tobacco smoking. Currently there are no national tobacco control policies that address waterpipe smoking packaging or distrib ution (Maziak, 2008). This lax control has several effects, for instance it leads individuals to believing that lack of regulations must be an indication of safeness and i t also leads to easy access through internet sales (Primack, et al., 2009; S. Smith-Si mone, et al., 2008). Legislation that stymies cigarette smoking in public places often do es not include waterpipe smoking (Primack, et al., 2009). Hookah bar owners argue t hat they are exempt from smoking laws because the laws state “the use or possession of a lighted cigarette, pipe or tobacco product” and their argument is that you do not lig ht hookah tobacco ("Some hookah bars fighting new smoking ban," 2010). The World Health Organization Framework Convention on Tobacco Control (WHO FCTC) has called upon countries to regulate package labeling, health warnings, content, and emi ssions related to waterpipe tobacco smoking (WHO, 2006). On September 22, 2009 the FDA placed a ban on cigarettes containing flavored tobacco aimed at luring youth t o smoke (FDA, 2009). However, because these controls do not include waterpipe smo king, it potentially gives a false sense of safety in the use of waterpipe tobacco. Adoption of practice among college students Perceived Benefits Waterpipe tobacco smoking also offers college stude nts a variety of benefits. The waterpipe’s use in providing a form of socializatio n and relaxation among peers, the mystical and exotic appeal, its availability and po pularity, the flavored tobacco attraction,


the novel experience, and the perceptions of being a less harmful mechanism of tobacco consumption have all contributed in some form to it s use in college students. Waterpipe smoking has been described by users as pr oviding a sense of sharing and belonging through passing the pipe from one per son to another (Prignot, et al., 2008). These benefits have been appreciated for centuries by elder men in the Eastern Mediterranean Region. Social acceptance of waterpi pe smoking and the opportunity it provides to socialize appears to have contributed t o its use in college students in the U.S. as well (Primack, et al., 2008). It is not uncommon for waterpipe bars to also provi de belly dancers and to burn incense both of which create a rather mystical atmo sphere for smokers. Currently in Tampa, one hookah bar offers belly dancers and a ma sseuse (Hayes, 2005). When surveyed, Canadian and English students reported th at they liked the exotic appeal that waterpipe smoking affords (Roskin & Aveyard, 2009). Along with the supernatural appeal, (Grekin & Ayna, 2008) found that both the l ow cost and the extensive availability facilitated the use of waterpipe smoking among coll ege students. Waterpipe smoking has created a sense of smoking culture among college st udents, as it differs from tobacco consumption patterns of their parents (Prignot, et al., 2008). Waterpipe smoking has also given students a sense of being popular, yet anothe r draw to incoming freshmen (Primack, et al., 2008). Popularity along with per ceived peer acceptance are strong predictors of waterpipe smoking among college stude nts (Primack, et al., 2008). The popularity of waterpipe smoking as viewed by colleg e students is through its stress reducing and relaxing effects (Maziak, Eissenberg, et al., 2004).


As noted above, m aassel the fruit flavored tobacco, has contributed to th e increased prevalence of waterpipe smoking in both t he EMR and the U.S. (Rastam, et al., 2004). Maassel is tobacco that has been combined with dried fruit s, glycerin, and flavorings to deliver a fruit flavored smoke which waterpipe smokers both taste and inhale. Maassel flavors range from apple, bubble gum, chocolate, f rappaccino, mint, orange soda, root beer to melon (Primack, et al., 2 009). The formulation and distribution of maassel is believed to have contributed to the resurgence in the Middle East and the increased incidence of waterpipe smoking in the U.S (Shihadeh, 2003; Ward, et al., 2005). The idea of tobacco being fruit flavored ha s led to its acceptance as a safer form of tobacco use among college students (Grekin & Ay na, 2008). Fruit being touted as “good for you” has led waterpipe smokers to view maassel tobacco smoking as having natural qualities, providing a false sense of healt hiness (Prignot, et al., 2008). The perception that the water is a filtering mechan ism of the harmful chemicals has led some college students to believe it is less harmful (Grekin & Ayna, 2008). The adoption of waterpipe smoking can be attributed to the harm reduction misperception (Maziak, Ward, et al., 2004). Roskin (2009) found that among Canadian and English students smoking was perceived as less harmful due to the smoothness of the smoke compared to cigarette smoking and, in addition, the majority of students felt they could quit at any time. This sense of ability to quit gi ves the impression that waterpipe tobacco may be seen as less addictive than the tobacco used in cigarettes. Ward (2006) surveyed U.S. Military recruits and found that they also bel ieved waterpipe smoking to be safer than other methods of tobacco smoking (Ward, Vander Weg, Relyea, Debon, & Klesges, 2006). Eissenberg et al. (2008) also found that am ong cigarette smokers in a U.S.


college, students believed it to be less harmful th an cigarette smoking (Eissenberg, et al., 2008). Some of the students even believed that wat erpipe smoking was an avenue for cigarette smoking cessation (Makhoul & Nakkash, 200 9; Roskin & Aveyard, 2009). In a study of college students at Johns Hopkins Universi ty, some students believed that nicotine replacement therapies were as harmful as c igarettes, and that waterpipe smoking was less harmful than both (Smith, et al., 2007). The proposed research will look at the applicability of using the Theory of Reasoned Actio n (TRA) to predict waterpipe smoking and to understand the cognitions that under lie the behavior to help design interventions designed to change the behavior. Theoretical Framework The Theory of Reasoned Action (TRA) (Figure 2.1) is a theoretical framework that is used to predict behavior under the assumpti on that humans are rational individuals and use information in an organized fashion to info rm behavioral decisions (Ajzen & Fishbein, 1980b). The TRA infers that humans consi der the consequences and implications of their behavior prior to action to p erform or not perform a particular behavior (Ajzen & Fishbein, 1980b). The theory ass umes that behavior is volitional and intention leads directly to behavior, thereby sugge sting that knowing the intention is a good predictor of the behavior occurring (Ajzen & F ishbein, 1980b). Although this intention to behavior concept seems quite simplisti c in form, the theory further suggests the researcher must understand the effect the deter minants have on intention if they are to inform developers of interventions. More specifica lly, if researchers can understand and predict the influencers on behavior then it is reas onable to think that these influences will


be a leverage point in programs and health messages to deter the behavior in the particular study sample. Theoretical Constructs The TRA(Glaser & Strauss, 2009) was developed in 19 75 by Martin Fishbein and Icek Ajzen to examine the relationship between atti tude and behavior. The TRA posits that an individual’s intention to engage in or refr ain from a behavior is the single best predictor that behavior. In the case of smoking, t he theory predicts that an individual who has some intention to start smoking should be m ore likely to engage in smoking than someone who has no intention to smoke. According to the TRA, behavioral intention (main predictor of the behavior) in turn, is influe nced by personal and social determinants. The personal determinant of intentio n is an individual’s attitude toward the behavior. This attitude is influenced by two facto rs: (1) a person’s beliefs about the behavioral outcome and (2) a person’s evaluation of the relative costs and benefits of the anticipated outcome. According to the theory, atti tudes are a function of personal beliefs. For example, a person who holds positive beliefs ab out smoking are more likely to have a positive attitude towards smoking and those with ne gative beliefs will more likely have negative attitudes. Under the umbrella of attitudes the personal beliefs are referred to as behavioral beliefs. The social determinant of intention is termed subje ctive norm. Subjective norm constitutes the social influences of intention. Su bjective norm is influenced by two factors; (1)an individual’s belief about whether im portant others will approve or disapprove of the behavior, and (2) the individual’ s motivation to comply with these beliefs. Subjective norms are also a function of b eliefs, but rather than personal beliefs,


they are related to the social aspects. More speci fically, beliefs about whether important others think he/she should or should not perform a particular behavior. For example, if important others believe that the person should smo ke then the individual will feel greater social pressure to smoke. Underlying a person’s su bjective norms, these beliefs are referred to as normative beliefs(Ajzen & Fishbein, 1980a). In review, the two main predictors of behavioral in tention are subjective norms and attitude. Subjective norms and attitude can ex ert varying degrees of influence on behavioral intention(Ajzen & Fishbein, 1980b). Whe n a positive attitude corresponds to a positive subjective norm then the direction of in tention is more easily determined because positive predictors on both personal and so cial factors equal positive behavioral intention. For example, an individual will have st rong intentions to smoke if they evaluate smoking favorably and if they believe impo rtant others think they should smoke. However, if attitude andsubjective norms exert comp eting influences (e.g., positive personal predictors and negative social predictors) then relative importance or weight of attitudesand subjective norms will determine behavi oral intention. For example, if an individual evaluates smoking favorably, yet importa nt others think they should not smoke then the intention to smoke is not as great. There fore when assessing influences on behavioral intention, it is important to understand and determine the weight of influence of both attitudes and subjective norms especially i n the case of developing interventions.


Figure 2: The Theory of Reasoned Action Adapted from ((Ajzen & Fishbein, 1980b, p. 8) While the authors of the TRA note that external var iables, such as demographic characteristics and personality traits may be relat ed to the behavior, they comment that these external variables are not an essential part of this theory. The external variables have no direct influence on behavior, but rather pl ay a relational role with determinants of attitudinal and subjective norm factors (Ajzen & Fishbein, 1980b). The TRA is best used for understanding behaviors that are voluntary such as alcohol consumption and tobacco use. Applicability of the Theory of Reasoned Action to W aterpipe Smoking Since the introduction of the TRA in the mid-1970s, many studies have been conducted to determine its applicability in various populations and health related behaviors. Behavioral outcome expectancies and the evaluation of these expectancies Beliefs of Important others and the motivation to comply with these beliefs Attitudes towards the behavior Subjective norm Intention Behavior


This section will discuss several research studies previously conducted utilizing the TRA and describe the samples that were utilized the methods of data collection, and major conclusions. Sneed et al. (1998) applied the TRA to understand c ondom use among Filipina commercial sex workers. Their research question wa s to determine whether attitudes or subjective norms contribute the most to the behavio r as mediated through behavioral intention. Participants (N=1394) were recruited fr om four sites that were geographically dispersed in the southern Philippines. More specifi cally, the participants were females employed in entertainment establishments. The stud y method included interviews, demographic information collection, as well as info rmation related to sex work. Overall, the results showed that individuals with more posit ive attitudes were more likely to report condom use, however, subjective norms were found to be a better predictor of behaviors as mediated through intention. Because the norms w ere operationalized through managers, a rationale was provided for targeting ma nagers of these organizations with the interventions. Sable et al. (2004) evaluated the intention of phys icians to prescribe emergency contraception as guided by the Theory of Reasoned A ction. The sample included 96 faculty physicians from four various universities. The methods included a cross-sectional 14-item survey distributed at faculty meetings. Ad ditionally they gathered six demographical responses. The study found that attit udes and subjective norms toward prescribing contraception strongly predicted intent ion to do so.


In another study examining adolescent’s intention t o use performance enhancing substances, Dodge et al. (2008) sampled 241 adolesc ent athletes from two high schools. Study methods included a cross sectional 14-item qu estionnaire distributed by the researchers. Similar to the previous studies, demo graphic information was gathered on the participants. Study findings showed that attitu des and subjective norms predicted both using legal substances and abstaining from legal su bstances. As related to smoking, Morrison et al. (2002) utili zed the Theory of Reasoned Action as a model to examine marijuana use in highrisk young women. The study participants consisted of 230 participants in the f irst wave and 235 participants in the second wave. The data analyzed consisted of partic ipants who had completed wave one and wave two (N=170). Participants were unmarried p regnant adolescents less than 17 years of age. They were recruited from three diffe rent clinics/agencies in three counties. The data collection method consisted of one-on-one interviews. Results showed that attitudes were more strongly related to intentions than were perceived norms. The authors comment that the authors of the theory stat e that relative weights of attitudes and subjective norms should vary depending on the behav ior being studied. Intention was found to be a good predictor of marijuana use in th is sample. Kaplan et al. (2001) examined the role of socio-env ironmental and personal factors on smoking acquisition among adolescents an d young Latinas. The sample included 1411 participants from two federally funde d family planning clinics. Participants were randomly assigned to either faceto-face interviews or telephone interviews. The 105 item questionnaire was guided by previous focus groups, interviews, and existing surveys. Intention to smoke was the s trongest predictor of experimentation


n and regular smoking. Peer smoking behavior predict ed intention to smoke as well as general risk attitudes. The transition from experi mental to regular smoking was associated with peer smoking. In another study examining smoking behavior among a dolescents, Qian Guo et al. (2007) used the TRA and the Theory of Planned Behav ior to predict adolescent smoking in China. The secondary data set was part of the C hina Seven Cities Study. Participants in the study were recruited from 147 schools in the seven cities. The data was retrieved from the initial survey administered at the various schools. Overall, the mediation effects of intention to smoke were supported by both models The results supported the hypothesis that attitudes and perceived norms on sm oking behavior were mediated by intention. The proportion of variance in smoking ex plained by attitudes, subjective norm, and perceived behavioral control were found to be s ubstantial. Primack et al. (2008) conducted a study to assess t he prevalence of and associations with waterpipe smoking among U.S. univ ersity students. The purpose of their study was to determine prevalence rates in a random sample of students along with determining associations between outcome variables, socio-demographic variables, and predictors (based on the TRA). The study involved a cross-sectional online survey of students at a large urban university. The survey w as the American College Health Association’s National College Health Assessment (N ehl, et al.), which is conducted each semester at selected universities. Items for this study were added to the survey for a fee. The sample population included 3600 students random ly selected using email addresses and key demographic data accessed through the unive rsity, of which 660 completed the


questionnaire. Survey items that were added to the NCHA included behavioral questions with three dichotomous smoking behavior responses. Based on the TRA, two items were included to measur e students’ expectancies and relative harm and addictiveness. Normative bel iefs were measured with two items. First, among your peers, how socially acceptable is it to smoke tobacco from a waterpipe? Second, what percentage of college stud ents do you think has ever smoked tobacco form a waterpipe? Findings from the study suggest that perceived harm was less strongly related to waterpipe use than perceived ad dictiveness. Primack et al. (2008) suggest further research on these two personal fact ors is warranted in determining a student’s decision to smoke waterpipe. Perceived p eer acceptability and perceived popularity were found to be strong predictors of us e. Using the TRA for explaining waterpipe use among co llege students is a viable theoretical framework given previous studies on smo king and other high risk behaviors have found it to be a reliable guide. Waterpipe sm oking, like alcohol intake and cigarette smoking, is a volitional behavior, further supporti ng the use of the theory(Guo, et al., 2007; Lafflin, Moore-Hirschi, Weis, & Hayes, 1994). Purpose of the Study Because of the newness of waterpipe smoking in the U.S. there are only a handful of published reports, however, the research and the se reports may increase in the future as funding agencies are becoming aware this new phenom enon and the impact waterpipe smoking may have among young people.The literature that is available on waterpipe smoking among U.S. college students lacks a qualita tive approach to understanding the


meaning behind the behavior and the context or sett ing of the behavior among college students. This research employs both qualitative i nquiry and observations of the behavior of interest. This qualitative information was used to reinforce what the quantitative data reports. Furthermore, these qualitative data have helped to inform the quantitative survey and ensure that the survey adequately assessed the behavioral constructs in terms that are understood by the research participants. As noted previously, college age is a vulnerable ti me in a young person’s life. More specifically, health risk behaviors are common during this time period. To encourage more immediate action from decision maker s, college authorities, and funding agencies to the waterpipe phenomenon, researchers h ave cited the need for prevalence rates(Barnett, Curbow, Weitz, Johnson, & Smith-Simo ne, 2009). Although this research is college specific, the findings may encourage oth er universities to utilize the survey or a similar survey to assess the prevalence in their in dividual colleges, giving a broader understanding of the problem from a statewide persp ective. This information may then prompt current tobacco control activities to includ e waterpipe smoking. This study included an online survey with students housed at U SF and includes socio-demographic characteristics of both those who smoke waterpipe t obacco and those who do not. Primack et al. (2008) cite in their research in the U.S. that greater sociodemographic information is needed. This informatio n would help to determine differences between smokers and nonsmokers of water pipe tobacco. To the best of my knowledge, there are no interventions, and the need for social marketing or other programs is being called for in the literature (Cob b, Ward, Maziak, Shihadeh, &


Eissenberg, 2010). With the information gained, th is research may set the stage for a social marketing campaign at USF. Additionally, Primack et al. (2008) cited the need for more information on belief factors. The proposed research is guided by the TR A, which evaluates both the personal and social beliefs associated with intention to per form a behavior. The TRA has demonstrated appropriate applicability to predictin g and understanding marijuana use among college students (Morrison, Golder, Keller, & Gillmore, 2002) and smoking among teenagers (Hanson, 2005) This research employed the TRA to understand the de terminants of subjective norm, attitude, and intention and to better underst and the relationship of the theoretical constructs specific to waterpipe tobacco smoking. Significance Waterpipe smoking differs greatly from cigarette sm oking in many aspects and has not been in the spotlight of policy and public health messaging. Because it is an entirely different mechanism of smoking tobacco, th ere needs to be selectively developed programs aimed at curbing its use. Garnering a bet ter understanding of predictors of waterpipe smoking from the priority population will aid in developing and tailoring programs and health messages. Gathering qualitativ e data to improve understanding of this phenomenon could help to determine what draws individuals or groups of young people to partake in smoking. It is clear that a b etter understanding of the characteristics of smokers may help to both educate public health w orkers, but also USF authorities.


To the best of my knowledge, this study is the firs t study to use the TRAin its entirety, measuring all constructs, thereby, both a dding to previously studied constructs and providing new insight to other measurable const ructs. Additionally, the survey has allowed for comparisons between attitudinal and sub jective norm measurements to assess their independent influence on behavioral intention As mentioned previously, waterpipe smoking has alre ady been deemed a public health problem, and this study may sharpen the focu s on potential interventions to reduce its growth in popularity based on the information t hat is gained. This study has the ability to provide insight into designing a USF bas ed program aimed at decreasing waterpipe smoking among college students living in housing at the University of South Florida. Finally, the information gained in this s tudy could illuminate the perceived benefits and barriers to waterpipe smoking and add to the current limited body of literature in this arena. Researchers have indicated that college age student s are more vulnerable to waterpipe smoking behaviors (Barnett, et al., 2009) Barnett et al. have conducted waterpipe studies among Florida youth and seek more prevalence rates state-wide (Barnett, et al., 2009). Understanding the prevale nce rates will help to elucidate the seriousness of this problem and better equip public health workers to seek local changes and statewide policy change.


CHAPTER 3: METHODS This chapter describes the methods that were used i n this study. The chapter is organized into six sections: (a) the purpose of th e study, (b) the research questions, (c) study design, (d) study population and sample, (e) method ology,and (f) data analysis. Purpose of the Study This study was designed to provide an understanding of the factors that influence waterpipe smoking intention among college students. The purpose of the study was to gather quantitative data to explore six research qu estions related to waterpipe smoking and to provide contextual rich qualitative data to allow for greater depth and elaboration in responses. Although knowledge of the prevalence of waterpipe smoking among young adults has recently been gathered, relatively littl e is known about the factors that influence university students’ smoking decisions or intention. This study explored the relationship between predictors of waterpipe smokin g intention in a sample of students at the University of South Florida (USF) who have prim ary residence in university housing. The Theory of Reasoned Action (TRA) served as a gui de to the investigation. The research questions are outlined below. Research Question #1: What is the relationship between attitudes and int ention to smoke waterpipe tobacco?


Research Question #2: What is the relationship between subjective norms and the intention to smoke waterpipe tobacco? Research Question #3: What is the relationship between beliefs of importa nt others and subjective norms? Research Question #4: What is the relationship between motivation to comp ly and subjective norms? Research Question #5: What is the relationship between outcome expectanci es and attitudes? Research Question #6: What is the relationship between evaluation of outc ome expectancies and attitudes? Study Design This study employed a sequential, mixed method desi gn to investigate the determinants of waterpipe use intention among colle ge students. Research was conducted in three phases. The initial qualitative phase was used to explore waterpipe smokers’ characteristics, and the beliefs, attitude s, and social factors that influence waterpipe smoking intention. The second, quantitati ve phaseexplored and examined bivariate relationships between waterpipe smoking i ntention, attitudes,outcome expectancies, normative beliefs, and socio-demograp hic characteristics while controlling


for threats to internal validity. A final qualitat ive phase was conducted to verify findings and clarify any unanticipated quantitative findings This triangulation of data served to increase the study’s validity. Study Population and Sample This study was conducted with college students atte nding an urban university institution – USF – in Tampa, Florida. College stu dents were selected as participants for this study for several reasons. First, as noted in two U.S. studies, waterpipe smoking was most frequently initiated by youth during their lat e teens and early twenties (Barnett, et al., 2009; Grekin & Ayna, 2008). Second, while pra cticed by people in many age groups, waterpipe smoking appears to be increasing rapidly among students at universities and colleges. Third, it was determined that more infor mation was needed to understand the variation in influence of predictors of waterpipe s moking initiation among young people(Primack, et al., 2008). Lastly, the study s ample and triangulation of data will better informprevention measures to decrease the po tential morbidity associated with waterpipe smoking among university housed college s tudents at this particular university. USF is situated in Hillsborough County on the west central coast of Florida. Its location can be seen in Figure 3. In Fall of 2010, 30,963 students were enrolled in undergraduate programs on the Tampa campus(USF, 201 0). Figure 3: The University of South Florida


The sample for this study was drawn from undergradu ate students who were living in USF housing on the Tampa campus during th e Spring 2011 semester. The USF housing student profile for the Spring 2011 semeste r can be seen in Table 1. Table 1: USF Housing Student Profile Spring 2011 Description Number % Non-resident alien 251 4.7 Race/Ethnicity unknown 70 1.3 Hispanic/Latino/Spanish 848 15.9 American Indian/Alaskan Native 17 0.3 Asian/Pacific Islander 287 5.4 Black, non-Hispanic 687 12.9 Native Hawaiian/Pacific Islander 8 0.1 White, non-Hispanic 3057 57.2 Two or more races 116 2.2 Total 5341 100% Inclusion criteria for the study samples of undergr aduate students included: (1) current residence in USF owned housing, and (2) 18 years of age or older.Five thousand three hundred and forty-one undergraduates living i n USF housing were selected randomly as participantsfor this study because they were considered to be more representative of younger students engaged in campu s activities, a better reflection of students impacted by college-based promotional even ts for waterpipe smoking, and a more accessible sample for college-based prevention efforts. The survey sample was identified and a recruitment email was dispersed th roughthe university’s Registrars’ office. Purposive student samples meeting these inclusion c riteria were drawn for the qualitative phases of the study. Students were rec ruited for individual interviews and focus groups using a central location intercept met hod conducted at the Marshall Student


Center, in addition to flyers for the focus groups. This Center attracts a large number of students living on campus and is believed to be fre quented less often by students who commute from the surrounding area. Students come t o the Center for a variety of activities (eating, socializing, studying, and shop ping) that afford them time to talk with a researcher about the project, participate in a shor t interview, and/or schedule participation in a focus group discussion. From the focus group and intercept interview participants, those interested in participating in the study’s ve rification stage were recruited. Preliminary observations were conducted in five of the waterpipe lounges/restaurants surrounding the USF Tampa campu s. However, at the time of data collection, one lounge had closed and one was no lo nger offering patio hookah and consisted mainly of Middle Eastern middle aged men inside the lounge. A minimum of two observations were conducted at each of the thre e remaining sites during the observational qualitative phase. Methodology Phases I and III Qualitative Research The initial qualitative phase included intercept in terviews, focus groups, and observations of local waterpipe lounges/restaurants to explore the factors that influence waterpipe smoking. Qualitative methods are the pre ferred method for exploring people’s perceptions of the factors that influence health be haviors and understanding the context in which choices are made. Observation is especially helpful in identifying differences between people’s self-reported behaviors and what t hey actually do in a given situation. Qualitative methodologies were used in the initial exploratory phases to gain insights into students’ perceptions of water pipe smoking related to their normative and behavioral


n beliefs. The Office of Research, Division of Resea rch Compliance Institutional Board at USF approved the study protocol (IRB #108637) and d ata collection instruments. Participant Observations The study initiated with observations of participan ts at waterpipe lounges/restaurants surrounding the USF campus. Ho okah bars were observed six times during the Fall 2010 and Spring 2011 semesters.Obse rvations were conducted on Thursday, Friday, or Saturday evenings for a two ho ur time period between 8 and 11:30 p.m. Observation reports at each site were guided by an observation guide, both which can be found in Appendix 1. Observations began priorto intercept interviewing a nd continued into the survey distribution phase.Based on preliminary observation s, it was determined that the best time to observe and be able to secure a seat was be tween eight and eleven p.m.As a participant observer, purchasing and smoking of a w aterpipe was undertaken at each observations site to fit in socially in the group a nd avoid appearing as a researcher and questioning from the establishment. An observation guide was developed and reviewed by two anthropologists prior to observation conduction. Due to the ubiquity of cell phones, the researcher employed a notepad on the cell phone for observational note ta king. Observational data consisted of detailed notes of the physical layout of the lounge the composition of people, a description of the types of waterpipe devices/mouth pieces, other activities that are taking place concurrently, and any conversation that can b e retrieved for terminology associated


with waterpipe smoking. Of special interest was th e composition of groups that smoke together (number of participants,gender, relative a ge, and ethnicity). This form of participant observations allowed the r esearcher to observe participants without interviewing them and enabled the researcher to validate interview data and take notes during the observation. Observa tional information chronicled the researcher’s own perceptions (reflexivity) and feel ings along with the details of the observation as outlined in an observational guide b ased on the interviews and focus groups. Additionally, the observations were interp reted to provide tentative conclusions, and develop new points to observe in the next obser vation session. Schensul et al. (1999) suggest that observers begin by observing the setti ngs, tracking event sequence, counting, and ethnographic mapping. Spending time in the fie ld and using rich descriptions to convey the observations improves the validity of ob servations (Creswell, 2009). The researcher transcribed the observational notes into a Microsoft Word document immediately following the observation to ensure rel iability of the results. Intercept Interviews Intercept interviews were also conducted during the initial qualitative phase. An interview guide was developed to explore normative and behavioral beliefs about waterpipe smoking, identify important others, and i nquire about how important these others are to the participant’s intention to smoke waterpipe, knowledge of perceived norms related to waterpipe smoking, and perception of benefits and barriers of waterpipe smoking. The guide was reviewed by two dissertatio n committee members. Interviews were conducted in the Marshall Center over a period of two weeks. A kiosk was secured


in the middle of the Marshall Center for recruitmen t, consent signing, and interviewing. A hookah device was set up on the kiosk for display Students approached the interviewer and inquired about the hookah. If the student lived in University owned housing and was at least 18 years of age, the inter viewer explained the study and inquired about their interest in voluntary participation. T hose who agreed were given an informed consent document, which was reviewed, ensuring the interviewee of confidentiality and the voluntary nature of the interview. The intervi ewee was asked their permission to digitally record the interview. All but one partic ipant agreed to be recorded. All interviewees received a $10 gift card incentive for participation. Initially, twenty interviews (10 smokers and 20 nonsmokers)of approxi mately 15 to 20 minutes long were proposed. Because saturation was not reached with t his sample size, the researcher conducted a total of 17 individual and 6 dyadic int erviews with smokers and also conducted a total of 29 individual and 2 dyadic int erviews with nonsmokers. Demographic Characteristics of the Intercept Interv iews Background information was limited to year of study and major for students who participated in the intercept interviews. Data wer e gathered from 11 female and 18 male smokers with areas of study ranging from Biomedical Science to Psychology. Nonsmoker data were gathered from 18 females and 16 males with areas of studies for these students representing a variety of discipline s (Business, Psychology, Public Health, and German). Results of these interviews were used to inform the survey response options.


Focus Groups Focus groups can stimulate a rich discussion of the factors that influence waterpipe smoking and clarify differences between s tudents’ perceptions of their behavior and practices observed in hookah bars (Deb us, n.d.). Additionally, focus groups are a good precursor to the quantitative component of the research because they reveal the words students use to discuss waterpipe smoking and its determinants (Neumeier, 2006). Focus groups were conducted in a reserved conferenc e room at the Marshall Student Center after completion of theintercept int erviews. All focus groups were conducted by a moderator and a note taker (an anthr opology doctoral candidate). Three focus groups wereconducted with students who acknow ledged having smoked waterpipe tobacco, currently lived in University housing and were 18 years of age or older. Development of the focus group guide was based on t he literature, the TRA, and the intercept interview findings.A minimum of two focus groups, lasting two hours, were proposed for this study; however, an additional foc us group was conducted to ensure the achievement oftheoretical saturation (Glaser & Stra uss, 2009). Students were recruited for the focus groups using flyers posted around the campus in areas that were approved for student rese arch. The researcher posted flyers in Psychology, the Marshall Center, the College of Pub lic Health, and campus residencies that were identified through the intercept intervie ws as having a large number of waterpipe smokers. Additionally, students were rec ruited during the intercept interviews at the student center. Names and contact informati on from students indicating their interest in participating in the focus groups was s tored in a file on a password protected


computer. Additionally, the researcher gave these individuals a card with the focus groups dates and the researcher’s email address on it in the event they had any additional questions. Some of the recruited students contacte d the researcher via email to inquire about other friends that wished to participate in t he focus group. These individuals were asked to contact the researcher to determine if the y met the inclusion criteria. All participants were sent a reminder email one week pr ior to the focus group and again the prior day. These same participants were sent a tex t message the day of the focus group as a final reminder. All focus groups lasted no more than two hours and were scheduled within a two week period of time. Prior to conduction of the fo cus groups, the study was explained, the consent forms were reviewed and signed, and a d emographic form was provided to each participant. Consent forms were placed in a se aled envelope and stored in a locked file cabinet after the focus group. A copy of the consent form was provided to any participant who requested one. Two recorders were p laced at each end of the table to ensure good sound quality for transcription. Record ings were stored in a password protected personal computer.Focus group participant s were provided with a $30 gift card and sandwiches for participation. Demographic Characteristics of the Focus Group Part icipants The initial focus group was attended by a predomina te number of males. Participants ranged in age from 19 to 24 years of a ge and represented four countries of origin. The second and third focus group included a more even distribution of males and females; however, participants in the third focus g roup were comprised mostly of


students from India, where waterpipe tobacco smokin g is very prevalent. Demographic information and smoking characteristics were obtain ed only for the focus group participants and can be found in Table 2. Verification Focus Group A group discussion was conducted during the third and final phase of the research to obtain students’ assessment of study co nclusions and clarify any unanticipated survey findings (Bernard, 2006). Fou r students who currently smoke hookah and three students who do not smoke hookah a greed to participate in the group discussion. Five males and two female students par ticipated in the verification group discussion. A summary PowerPoint presentation of th e research findings was provided during the focus group. As each slide was presented students were asked to provide their views on the accuracy of the conclusions reached.


Table 2: Demographic Information for Focus Group P articipants Gender Age (n) Country of Origin (n) Times smoked in past 30 days Bowls smoked in typical session Lone Smokers (n) How often do they smoke alone FG #1 12 males 0 females 3 – 19 years 2 – 20 years 3 – 21 years 1 – 22 years 1 – 23 years 1 – 24 years Canada = 2 India = 4 Egypt = 1 US = 5 Avg. = 2.17 Max = 8 Avg. = 2 2 1-2 x year Rarely FG #2 4 males 5 females 2 – 19 years 2 – 20 years 1 – 21 years 3 – 22 years 1 no response Bangladesh = 1 India = 1 US = 5 Thailand = 1 Avg. = 5.22 Max = 15 Avg. = 1.9 4 Used to a lot 2 x month 1 x month no response FG #3 5 males 5 females 1 – 19 years 2 – 20 years 3 – 21 years India = 8 Germany = 1 England = 1 Avg. = 19 Max = 80 Avg. = 1.4 4 1 x week rarely 1 x week


1 – 22 years 2 – 23 years Every a.m. Totals 21 males 10 females 6 – 19 years 6 – 20 years 7 – 21 years 5 – 22 years 3 – 23 years 1 – 24 years 1 no response Canada = 1 India = 13 Egypt = 1 US = 10 Thailand = 1 England = 1 20 Rarely = 2 Used to a lot = 1 1x month = 1 2 x month = 1 Every day = 1 No response = 1


Phase II Quantitative Research Online Survey In the second phase of the research, an online surv ey was administered to a random sample of students living in USF housing. T he survey was administered after the intercept interviews and focus groups were complete d. For efficiency and uniformity, the survey was administered online. This mode of surve y delivery was free to the researcher and provided a high speed of returned responses. Al l students in the sample have access to computers. The survey was designed to protect s tudent privacy, allowing them to complete the survey when and where they chose, and allowed for confidentiality in a voluntary capacity. College students are online re gularly and at ease communicating via this medium with 86% of college students having gon e online and 85% owning their own computer(Jones, 2002). Survey Instrument Development Establishing Face and Content Validity Instrument development was guided by surveys that h ave been used to explore waterpipe smoking attitudes and believes in other s tudies (Chaaya, et al., 2004; Eissenberg, et al., 2008; Grekin & Ayna, 2008; Mazi ak, Eissenberg, et al., 2004; Primack, et al., 2008; Roskin & Aveyard, 2009; S. S mith-Simone, et al., 2008), the intercept interviews and focus group findings, obse rvational findings, and topic-specific literature. This information and guides such as the National Cancer Institute’s adolescent smoking consequences questionnaire (NCI, n.d.) and the Fishbein-Ajzen-Hanson


questionnaire (Fishbein, Azjen, & Hanson, n.d.)prov ided further guidance into question structure and response options. Additionally, Dr. Kenneth D. Ward (University of Me mphis) and Dr. Brian A. Primack (University of Pittsburgh),experts in the f ield of waterpipe smoking research, reviewed the survey instrument. They provided detai led feedback that helped refine important items and add other questions appropriate for the college population. Doctoral dissertation committee members alsoprovided guidanc e and input. After the survey was reviewed by the experts, it wa s pretested with 12 college students who reported living in university housing and being a hookah smoker. The purpose of the pretest was to ensure readability, c omprehension, and to estimate the time required for completion. Students from the populat ion under study were recruited via email to complete the survey and provide written fe edback. An incentive of a $10 Amazon gift card was provided. The average time to complete the survey was six minutes. The final survey was assessed for grade l evel readability using the FleschKincaid online tool readability test, resulting in the estimate of grade 5.9 and can be found in Appendix 2. Theoretical construct measure ments can be seen in Table 3. Table 3: Theoretical Construct Measurements Variable Measure Scale Intention Intention to smoke tobacco using a waterpipe within the next few months One 7-point Likert scale from extremely unlikely to extremely likely Attitude Attitude toward If I smoke hookah, this behavior is…. or My smoking hookah Four 7-point Likert scales including bad/good, awful/nice, not fun/fun, and


n is… unpleasant/pleasant Subjective Norm If I smoke hookah most people who are important to me would…… One 7-point Likert scale ranging from disapprove/approve Behavioral Beliefs and Outcome Evaluation Modal Beliefs about hookah smoking and the corresponding evaluation of each of the beliefs 36 7-point Likert scales; 18 for modal beliefs ranging from unlikely to likely and 18 for outcome evaluation ranging from unimportant to important. Normative Beliefs and motivation to comply Beliefs about approval or disapproval of smoking from four significant others (e.g., parents, friends) and the corresponding motivation to comply 8 7-point Likert scales; 4 related to perceptions of others ranging from disapprove to approve and 4 for motivation to comply ranging from disagree to agree Establishing Internal Consistency and Stability Test-retest reliability (stability) of items was es timated using correlational analysis and item percent agreement, depending on t he nature of the specific items. The sample completing the instrument consisted of six s tudents from the sample population and six students who were nonsmokers, recruited fro m the College of Public Health (the home college of the researcher). The testing timef rame was one week. This testing timeframe was based on the work of Marx, Menezes, H orovitz, Jones, and Warren (Marx, Menezes, Horovitz, Jones, & Warren, 2003), who show ed no difference in reliability


between 2 days and 2 weeks for a health related qua lity of life instrument. An Amazon gift card of $10 was provided to students who compl eted both the “test” and the “retest.” Dichotomous items produced an average 95.3% agreeme nt over time. The Likert scaled response options were evaluated both as indi vidual items, scaled items, and as sum score grouped data. Spearman correlational analysi s using sum scores revealed a total survey correlation of 0.920. Spearman correlation w as used due to the Likert scale being considered ranked or ordered data. After the test and retest, formulated scales and it ems to operationalize constructs and provide evident of construct validity using fac tor analysis. The main purpose of this factor analysis was to understand the structure of latent factors that exert influence on the observed variables. The survey was developed and e ntered into a free online survey tool for students called Checkbox. The tool provides a URL that links participants to the survey and collects the data in a spreadsheet forma t for analysis. The URL was made available to five of the undergraduate public healt h classes at the college, each consisting of approximately 40 students. Two drawings of $25 Amazon gift cards per class was the incentive for participation. Fifty-five students vo luntarily completed the survey. Using SPSS (version 19), factor analysis with var imax rotation was conducted on the hookah beliefs and opinions items to develop scales. Using Kaiser’s criterion of retaining factors with an eigenvalue of greater tha n 1.0 and visualization of the scree plot(Fields, 2005), it was determined that four fac tors could reliably be extracted, these were labeled: Social; Focused; Health; and Family. An additionalquestion related to physical activity was removed from the survey based on its failure to load on one of the four identified factors. The lowest value for loadi ng on a factor was 0.494. The four


factors were then evaluated for internal consistenc y reliability using Cronbach’s alpha. Cronbach alpha scores can be seen in Table 4. Table 4: Cronbach Alpha Scores for Factors Factors Cronbach Alpha Scores Factor 1 – Social 0.905 Factor 2 – Focused 0.808 Factor 3 – Health 0.809 Factor 4 – Family 0.747 Reliability was assessed on other constructs within the survey instrument as well as inter-item correlations.Two items under the cons truct of beliefs (relax and relieve stress) provided near perfect correlation (.957), t herefore, they were merged into one item: If I smoke hookah, it will help me relax and reliev e stress. Table 5 provides a summary of the Cronbach alpha scores of the key the oretical constructs. Table 5: Reliability of Constructs Construct Cronbach’s Alpha Beliefs about hookah smoking 0.846 Importance of beliefs 0.911 Attitude regarding hookah smoking 0.912 Important others 0.826 Motivation to comply with important others 0.728


After development and approval of the survey, it wa s manually entered into the USF Health Checkbox survey tool(USF, 2009a). It en ables researchers to deliver online surveys through an external link and does not impos e severe limits in terms of number of cases and variables. Dillman et al. (2009) web survey tailored designed principles were utilized as a guide to decrease the total survey errors (e.g., co verage error, sampling error, nonresponse error, measurement error). More specifica lly, for web surveys, Dillman et al. (2009) recommended paying close attention to contac t details, taking precautionary steps to ensure delivery of web surveys, providing clear instructions, and being prepared for issues such as bounced and inquiry emails. The USF Institutional Review Board approved the sur vey instrument and distribution plan. Because the sample was limited t o students 18 years of age and over, a waiver of documentation of consent was approved f or the survey. Survey Administration The survey was disseminated by the University’s Reg istrar’s office over the course of two weeks during February 2011. The time period for dissemination was carefully chosen to fall after the spring semester began and before spring break to make sure the 30 day recall question regarding past use of waterpipe tobacco did not include a holiday period. Whereas having the survey disseminated through the Registrar’s Office made it possible to send the survey to a current list of st udents enrolled at USF, it also placed control of the process outside the investigator’s i mmediate control. Unfortunately, the initial sampling frame included students having “ev er lived” in University housing


instead of those currently living in university hou sing. After these students responded and the error noticed, the Registrar’s Office recre ated the sampling frame using the correct inclusion criteria and distributed the surv ey to 1500 students selected randomly. When the respondent rate did not meet the desired r ate based on the power calculation, an additional 1500 students were selected at random an d recruited by email through the Registrar’s Office. Response Rate Only 8% of students recruited for the study complet ed a usable survey. Initially, 350 surveyswere submitted. Of these responses, two surveys were discarded because they failed to respond to the truthfulness question; sev en surveys were eliminated because respondents completed less than 75% of the question s; one was deleted for non-response to a key question, Have you ever smoked waterpipe tobacco even one or two puffs?; 51 surveys were discardedbecause the respondents repor ted no longer living in Universityowned housing; and one final survey was deleted due to filling out the survey twice. Thus, only 288 cases were included in the final ana lysis. Figure 4 depicts the survey case deletion process.


Figure 4: Survey Case Deletions Recruitment When inviting selected students to participate in t he survey, the Registrars’ office used a subject heading (Short USF student survey_Fr ee Song Download and drawing entry) and invitation in the body of the email desc ribing the project, estimated length of time to complete the survey, confidentiality inform ation, voluntary nature of the survey, eligibility requirements, link to the survey, and a description of a gift provided for completing the survey. Survey incentive consisted of one downloadable song from Hip Digital Media(Digital, 2009) and a chance to win on e of ten $50 Amazon gift cards. Incentives were recommended by Dillman et al. (2009 ) to increase response rates. The FINAL SURVEY CASES 288 1 Discarded for not answering key question regarding ever smoked 51 Discarded for not living in University-owned housing 7 Discarded for answering less than 75% of questions 2 Discarded for not answering Truthfulness Question 1 Discarded for duplication 350 Initial Survey Cases


email body sent to eligible participants can be fou nd in Appendix 3. To safeguard against students under the age of 18 years taking the surve y, the survey was programmed to terminate after students entered their age. Howeve r, it should be noted that limiting repeat survey attempts was not an available option in the survey tool. When the survey was completed, participants were instructed to ente r their email address in which the code for the downloadable song and instructions were sen t. Their name was placed on a list for the drawing at the end of survey data collection.Em ail addresses and song download codes were kept in a password protected file separa te from the data. Survey Sample Characteristics Demographic variables measured included age, gender ethnicity, citizenship status, and religious affiliation (Table 6). Appro ximately 70.5% of students reported being 18 and 19 years of age, with the mean age of students completing the survey being 19.2 years of age ( s = 1.2). Most students completing the survey were W hite (66.0%), a distribution comparable to their representation in the population. Additionally, most of the students were female (71.5%). Table 6: Demographic Characteristics of Samplea Demographics Sample, n=288 Population n (%) n (%) Gender unable to attain Female 206 (71.5%) Male 79 (27.4%) Age 18 years 93 (32.3%) 19 years 110 (38.2%)


20 years 50 (17.4%) 21 years 18 (6.3%) 22 years 12 (4.2%) 23 years or older 5 (1.7%) Race/Ethnicity White 190 (66.0%) 3057 (57.2%) Black 31 (10.8%) 687 (12.9%) Asian/Pacific Islander 17 (5.9%) 295 (5. 5%) American Indian/Alaskan Native 2 (0.7%) 17 (0.3%) Other 46 (16.0%) 1285 (24%) Hispanic/Latin origin Yes 54 (18.8%) 848 (26%) No 233 (80.9%) 2603 (74%) US Citizenship Yes 277 (96.2%) No 8 (2.8%) Religion No religion 76 (26.4%) Catholic 69 (24.0%) Protestant 30 (10.4%) Other Christian 84 (29.2%) Jewish 5 (1.7%) Muslim 3 (1.0%) Buddhist 3 (1.0%) Other 15 (5.2%) a Missing data not shown; percentages may not equal 1 00% due to rounding


Summary of Qualitative and Quantitative Research Sa mples In sum, this study employed a three phase, sequenti al mixed methods design (See Table7) to identify the bivariate relationships bet ween the primary determinant constructs of the Theory of Reasoned Action as well as their r elationship to waterpipe smoking intentions. Sample sizes for each phase are provi ded in Table 7 below. Table 7: Sequential Mixed Method Design Phases Design Methods Participants I Qualitative 17 Smoker Interviews 6 Smoker Dyadic Interviews 11 females/18 males 29 Nonsmoker Interviews 2 Nonsmoker Dyadic Interviews 18 females/16 males 3 Focus Groups 10 females/21 males 6 observations Varied II Quantitative Online Survey(N=288) 206 females/ 7 9 males III Qualitative 1 focus group 4 smokers/3 nonsmokers 2 females/5 males As seen in Table 7, many more males than femalespar ticipated in thequalitative phase of the research and more females than males p articipated in the quantitative phase. Although both females and males were invited to par ticipate in the focus groups, it was surprising to find almost double the attendance onc e the sessions began. Two females came to the first focus group, but when they saw th at it was primarily males, they chose not to stay and decided to attend another one of th e other sessions. The intercept interviews were conducted in the Marshall Center on several different occasions within a two week time period in an attempt to collect diver se students. The research was not designed to capture gender differences; however, it should be noted that there was inequality in participation both in the qualitative inquiry and survey responses.


Data Analysis Qualitative Analysis Intercept interview data were reviewed for inaccura cies and key findings. Recordings of the intercept interviews and the focu s groups were transcribed by an outside professional service. Observational field notes were typed into a word document and the notetaker provided a summary for each focus group. Verbatim transcripts of the intercept interviews and of the focus group discuss ions were entered into NVivo version 8.0. Two codebooks were developed with apriori codes based on the Theory of Reasoned Action and research questions; one codeboo k was designed for the smokers and another one was designed for the nonsmokers. T hree researchers coded the qualitative data using the constant comparative met hod (Glaser & Strauss, 2009). Data analysis was done iteratively, with coding beginnin g as soon as information was available from the intercept interview. Three researchers cod ed each transcript and compared codes for inter-rater reliability adjusting how categorie s/codes were assigned. Researchers compared notes, using Google Documents and NVivo an notations throughout the interview document, on each topic for consistency a nd agreed on code definitions and appropriate statements. The coding began with smok er intercept interviews followed by the first smoker focus group which was hand coded t ogether as a group, allowing the researchers to discuss codes and come to agreement prior to subsequent focus group coding. The final focus groups were coded by each individual coder using the Google Documents and NVivo annotations, along with email, for any necessary communication. After the smoker data was coded, the researchers me t to discuss emerging themes.


n The nonsmoker intercept interviews were coded in a similar fashion, ensuring consistency and reliability among the coders. Foll owing the nonsmoker coding, the researchers met and discussed the emerging themes. Quantitative Data Analysis An apriori power calculation for multivariable analysis was c onducted using two online calculators to determine the necessary sampl e size using a confidence level of 95% and an alpha of .05 (Biostatistics, n.d.; Raosoft, 2007). Necessary sample sizes calculated consisted of 328 and 381 necessary cases for multivariable analysis, such as logistic regression. These samples were based on a margin of error of 5%, a confidence level of 95% and a population size of 4700 (student s in USF housing during Fall 2010). The Raosoft calculator is specific for web surveys (Raosoft, 2 007). This number exceeded the minimum sample size (100) needed to co nductexploratory factor analysis(MacCallum, Widaman, Zhang, & Hong, 1999). Because online survey response rates for online surveys vary from 5 to 20 % (Fowler, 2009), the number invited to participate was increased to 3000. The survey da ta were transferred from the Checkbox survey tool into an IBM SPSS software version 19.0 for analysis. Univariate Analysis Prior to data analysis, demographic data were scree ned for outliers and continuous data were screened for non-normality using Shapiro-Wilk’ s statistic for normality. Univariate procedures included frequency distributions, basic descriptive statistics of variables ( gender, age, ethnicity, citizenship and religion ) Frequencies and proportions, indicating prevalence in the sample, waterpipe use in the past 30 days and ever having used


waterpipe tobacco were generated with the socio-dem ographic variables to discern any potential differences between groups of individuals such as gender, ethnicity, and age. Bivariate Analysis Bivariate analyses consisted of relationship and co rrelational analyses and constituted the main analyses to answer the propose d research questions. The survey questions consisted of ordinal response categories. These response categories lend themselves to Spearman rank-order correlations with non-normal data. Additional analysis, although not to directly answer the resea rch questions, assessed relationships between nominal data, such as demographic informati on, using Chi-Square tests of independence.Odds ratios were calculated on all 2 x 2 tables to improve interpretation of the relationships. Because the proposed research responses were Likert -type, these response categories were analyzed as ranked data. Interrela tionships of survey items representing constructs were evaluated for multicollinearity and singularity. To answer the research questions, correlations were conducted between the unidirectional constructs as displayed in the theoretical model. More specifically, corre lations were calculated between outcome expectancy and evaluation of outcome expect ancy with attitudes; motivation to comply and beliefs of important others with subject ive norm: subjective norm and intention; and attitude and intention. Multivariable Analysis


Factor analysis wasused to assess and/or understand the underlying structure of a set of observed variables used to measure outcome e xpectancy in the final survey analysis. The purpose was to understand latent stru ctures of the attitudinal construct in the final sample of 288 students.Cronbach’s alpha w as assessed to determine score reliability. To improve interpretation, varimax ro tation was employed. Correlational matrices were generated during factor analysis to e nsure that multicollinearity would not a problem for the data analysis. Additionally, Kai ser-Meyer-Olkin (KMO) and Barlett’s test of Sphericity with anti-imaging was used to en sure that the sampling is adequate. KMO and Barlett's test are included in the analytic al output for factor analysis.


CHAPTER 4: RESULTS The primary purpose of this study was to determine if there were relationships between key constructs of the Theory of Reasoned Ac tion and college students’ intention to smoke waterpipe tobacco. The study employed a se quential mixed method design: in Phase 1, focus groups, intercept interviews, and ob servations were conducted. Results informed Phase 2, which consisted of an online surv ey, followed by Phase 3, verification of results. This chapter is divided into three part s: qualitative research findings from Phase 1 and Phase 3, followed byquantitative findin gs from Phase 2.Qualitative Study Findings – Phase I and Phase III Observations The first section summarizes the general attitudes and behavioral beliefs about hookah expressed by USF students in individual inte rviews and focus group discussions with notations from the observations and the verifi cation focus group. Students also discuss the people who have influenced their decisi ons regarding hookah smoking. General Attitudes Among current smokers, hookah is viewed primarily a s a social activity. Given the importance of socialization among college stude nts, it is not surprising that hookah’s ability to be shared with friends is highly valued. As one young man said, it is “a great “icebreaker.”


“It encourages a lot of socialization because you t end to have three or four people who are fixed to a pipe. You c an't really be bouncing around. So it facilitates a lot of convers ation.” (male) “You could go insane if you never talk to nobody bu t when you have an event, you could get out and relax and actu ally smoke hookah with a bunch of people and socialize. It kin d of brings you back to the routine now instead of being so distrau ght if you were just by yourself.” (male) Only on one occasion during the observations was a person smoking by himself. During the majority of the observations, there were two to eight people smoking together, often sharing a pipe or two. The hookah lounges we re arranged to encourage socialization and group interaction with large couc hes either rounded or facing each other. During the verification focus group, smokers agreed that this social aspect is the primary reason they chose to smoke hookah. They de scribed the on campus smoking as providing the same group interaction as seen in the hookah bars. A lone smoker was viewed as a deviation from the norm and more of a b ehavior exhibited by a chronic smoker or an individual who doesnot want to study a lone and desires the companionship of the hookah. In contrast, students who were not regular smokers were more likely to describe hookah in negative terms, such as “gross” and “disg usting”. Many non-smokers viewed hookah as a drug, in large part because of the use a bong-like device and its similarity to cigarette smoking. For these students, hookah is a form of deviance that they reject. “I don’t really know too much about it, but I usual ly stay away from smoking and drugs and all that.” (male)


“At first I didn’t even know what it was. [My] ini tial impression was, you know, people who are gangsters or like pro stitutes, things like that – they just – they do this kind of things .” (female) During the verification stage of the research, nons mokers reconfirmed their desire not to participate but were less likely to call it a “drug.” Smokers offered that beliefs of hookah as being a drug may be a misconception that nonsmokers have because of not being exposed to hookah until they enter college an d because of smoke production may create an association of hookah with cigarettes or marijuana. Behavioral Beliefs Students shared many positive and negative beliefs about hookah smoking. Whereas awareness of these beliefs was widespread, smokers tended to report positive beliefs about hookah’s benefits and reject ones abo ut its negative effects; conversely, students who considered themselves to be nonsmokers believed that hookah has largely negative outcomes and offered few benefits. As noted above, smokers described hookah as a socia l activity that enables them to make new friends or bond and reconnect with frie nds they already have. Several described it as a time to “chill” with their friend s. During the verification group discussion, one student agreed and called it a “kin ship” or “brotherhood” in their smoking circle. “Well, it’s kind of a social thing. I've met a lot of close friends here, just by going outside and smoking, and meetin g people through that.” (male)


Several smokers described hookah as exhibiting othe r positive social qualities. “Socially the way I think sounds like, if I smoke c igarettes people look at you the wrong the way and say, if I smoke h ookah they say, "This guy's got taste." or something like that.” (male) “Yes, but it’s the same party we both went to. We saw somebody smoking [hookah] out there. It was the first time for us, so we thought it was really cool.” (male) Whereas none of the smokers called hookah smoking c ool, many nonsmokers felt others were smoking so they would look more mature and fit in. “I think it’s wrong. I think it’s just another way to look cool.” (female) “It’s just sad. People just do it just to feel cool and I just think it’s just a foolish reason.” (male)“I guess it makes them look cool. Social factor. Th ey feel older.” (male) Although viewed primarily as a social experience, r eports of smoking alone were not uncommon. As one male student explained, “ cigarettes are like 5 minutes and hookah is like 45 minutes.” During the verification phase, this point of view was confirmed, and because of the duration of time invo lved in smoking hookah, it was often used during individual study time or while watching television. Participants associated it with smoking a cigarette in the sense of if you are a cigarette smoker you will go outside and have a smoke by yourself. One smoker said, “ Time flies when you are having fun.” (male) “Smoking alone is a once in a blue moon thing when you just want to relax by yourself, and 95% of the time, it’s wit h friends.” (male) During the verification dialogue, nonsmokers though t it would be “weird” to see a person smoking hookah alone.


Students also reported smoking hookah as a form of recreation. It was described as a fun and entertaining way to avoid boredom. On e student referred to it as a “hobby.” “I mean there are more things you can do with it, l ike blow smoke rings or exhale really large puffs through a light and it looks pretty interesting.” (male) “Just a fun atmosphere. Just things you can do with the hookah, like blowing rings, and making jokes and having fun .” (female) “It consumes time when you're bored nothing to do .” (male ) During the observations at the hookah bars, partici pants seemed to be having fun given the laughter and game-playing which took plac e. Verification discussions revealed that it is a hobby in the sense that some students collect devices to display around their rooms. Nonsmokers viewed smokers as having too little to d o and suggested smoking hookah offered smokers a means to fighting boredom. Nonsmokers in the verification group expressed that they felt that hookah smoking was what smokers choose to do as opposed to lacking anything to do. “Well, when we hang out with this person, they ten d to be, they tend to seem a little bored and when they get bored they are like, “Hey, do you want to smoke hookah?” (female) Hookah’s flavor is also seen as a valuable benefit. As observed in the bars, smokers have over 40 flavors from which to choosean d are allowed to mix flavors to enhance the experience. The flavor aspect was conf irmed as a positive attribute in the verification process.


“I like the flavor. I like the – I guess you would call it like the rush, it’s oxygen stuff.” (male) “My friends like anybody – usually my friends who l ike “hey you want to try like maybe this guava tasting tobacco. We got great tasting tobaccos in different flavored tobaccos.” (male) Relaxation was mentioned frequently as a reason to smoke hookah. It is seen as a way to escape from the pressures of school and a me chanism to cope with other frustrations of daily life. At one hookah bar obse rvation, a young man was lying on the couch smoking his hookah while watching the televis ion mounted on the wall above him. During the verification, several students said that just lying down and relaxing is nonproductive and that smoking hookah is a “proacti ve” way to relax. “It’s comfortable, it’s at the house. You just rel ax.” (male) “It’s the same thing as like eating in or dining ou t, and I think you’d rather eat in, you save money, and it’s more comfortable. So relaxed, like home food.” (male) “[It’s] a way to meditate, I guess, and blow out yo ur frustration.” (male ) Students also reported other physiological response s as desirable. These perceived physiologic responses, as noted in the qu otes below, were confirmed during the verification, but one student commented that it dep ends on the device and how the shisha is packed and the size of the device. “Like you get a little bit of a buzz, I mean I get lightheaded sometimes and I like that. It’s nice.” (female) In contrast to students who enjoyed how hookah made them feel, others reported they were disappointed with their initial trial and were dissuaded from doing it again


because it failed to deliver the pleasant feelings they expected. Confirmed in the verification is that some people mistakenly think h ookah is similar to marijuana in its properties. “Actually, I met my friend one time at a hookah pla ce. That’s where I tried that at one time but then I realized – hey, it’s just basically like cigarette smoking. So I’m not a smo ker.” (male) “I kind of think, I thought originally it was kind of like the marijuana but then when I actually did try it, it w as – I didn’t get any feeling so apparently it’s just like a cigarett e.” (male) “I thought originally it was kind of like the marij uana but then when I actually did try it, it was – I didn’t get a ny feeling so apparently it’s just like a cigarette. That it was something similar to marijuana but a legalized version.” (female) Some smokers believe that hookah helps them focus w hen studying and improves their ability to concentrate. A few even reported doing homework at the hookah bars or while smoking in their dorms. These reports are co nsistent with observations of students working on laptops in hookah bars. “…but a lot of times usually I would go there with just one friend and we’d just go with our homework and do our homew ork there, and even though it usually is more of a social thin g for me, sometimes I would do that and I’d get a lot done.” (male) Health issues were mentioned most often as a reason not to smoke; however, other disadvantages also were mentioned by smokers and non-smokers alike. The most common problems attributed to hookah were shortness of breath, coughing, and headaches. During the last two obser vations, many participants werecoughing, and the researcher experienced a head ache during one of the smokiest nights spent observing activities at a hookah loung e. When reviewing the verification


n tape, students were coughing frequently. Additiona l health risks mentioned were inability to be good at sports do to the effects on the lungs. “Definitely the lung effects; decreased lung capaci ty, damage to your lung tissue itself; you can probably hurt y our throat if you do it wrong or if it burns.” (male) “Yes, a little more forceful cough. It was like a dry cough.” (female) Although some smokers reported negative physiologic al effects from hookah, these problems did not seem to deter them from smok ing it. For example, a student who smoked hookah previously experienced these effects and quit as a result. “Over time I started to see that it started messing up my respiratory. Actually I used to smoke a lot, I had quit about two years ago. I totally quit, so anything that’s smok ing related. So but I feel like I started seeing my neurons going d own and my stamina going down.” (male) Many students believe hookah smoking could result i n more serious health outcomes, such as cancer and exposure to contagious diseases spread by sharing hookah pipes. “Besides the fact that you can get lung cancer, thr oat cancer, other things you can catch from other people that y ou don’t know if they’re infected when you’re sharing the mouthpi ece and stuff like coxsackievirus or other things such as that.” (male) “It’s like when it’s not ventilated, there’s just s moke everywhere and that’s recycling the smoke. I don’t think that’ s good for my lungs.” (male) Many nonsmokers explained their decision not to use hookah as a way to minimize the risk of these health problems and to a void the possibility of addiction to smoking hookah.


“I feel like the first-hand smoke is going to be da maging to my lungs. My parents smoke so I don’t want to have any more damage to my lungs and I guess that’s it.” (female ) “Because I grew up with parents who smoke. The idea of just inhaling like a puff of tar, it’s kind of gross.” (female) “Also, the instrument used to inhale the smoke [tha t’s] passed on, maybe it’s something that’s contagious.” (female) In general, nonsmokers disregard what smokers have told them about hookah’s relative safety compared to cigarettes. “I think a lot of people see the benefit as they th ink it’s healthier than cigarette smoking, and they don’t think it’s anywhe re near as addictive.” (female) “I would think anyone who sees hookah as an alterna tive to cigarette smoking. They’d be more like, “Oh yes, yo u should do it. It’s natural. It goes through water. You’re perfect ly fine.” (female) Some smokers acknowledged that hookah could create these health problems; however, they believe they will remain safe as long as they do not smoke it frequently. This belief was confirmed in the verification phase “Yes, because like you get the cancer when you pret ty much smoke it like every single day.” (male) “Yes, but I know that you don’t smoke it as much [a s cigarettes]. If you don’t smoke it as much it’s fine. You won’t ever really go to get addicted…” (male) In contrast to smokers who reported being able to c oncentrate better when smoking hookah, nonsmokers believe that hookah has the opposite effect on concentration and that students who smoke hookah su ffer academically. However, in the verification stage, it was felt that there are othe r mediators that contribute to poor performance and that it would be difficult to prove hookah smoking as the sole or primary cause. The students felt that freshmen com e to college to party and are not


serious about academic performance, a factor consid ered to be a more likely explanation for poor academic performance. “..but on my floor and they just like their school kind of goes down. Their grades. Their focus. “Oh, we’re going to go s moke.” Just negative.” (male) “You can probably get addicted maybe not chemically but mentally. It’ll affect your school life.” (female) Other negative outcomes acknowledged by smokers inc lude hassles associated with pipe maintenance, ashes, and accidents. Howev er, during verification it was determined that set-up was for anything that you wa nt to do to have fun, not just hookah. They felt that all devices take time to set up. Du ring one of the observations, a participant was carrying the hookah device up to th e counter for a refill and dropped the device on the ground. The glass shattered and char coal and tobacco spilled on the floor. The owner seemed a little perturbed. “It could burn holes through things or kind of ashe s all over the place and things like that.” (male) “Well one time that it came off and fell to the cou ch and the couch kind of got a hole burning through it.” (male) Nonsmokers also find hookah’s smell unappealing, bu t not necessarily bad. “I have family, like extended family who smoke and is not great for their teeth. They have [body] smell all the time.” (male) Another factor influencing the use of hookah is its cost. The cost of smoking in the hookah bars or restaurants observed for this st udy varied from $5 for a 45-minute session to $12 for an unlimited session. To avoid these charges, some students purchase a pipe so they can smoke at locations convenient to them. During the verification focus


group, one student equated it with drinking alcohol and said that it is cheaper to buy a bottle and go home to drink than to go to a club or bar and drink alcohol. “I figured it was a good investment, instead of dis hing out money every day [to go to a bar].” (male) “Because it’s too expensive to go smoke in a hookah lounge, just so you save money.” (male) Normative Beliefs According to the profile sheets obtained during the focus groups and the interview guides, the age at which students began smoking hoo kah ranged from 10 years to 25 years, with the majority initiating during their fr eshman year at college. Students participating in the verification focus group claim ed that beginning at younger ages was an outlier phenomenon, and that the range 17 to 24 years of age more accurately reflects waterpipe smoking initiation. Many smokers reported that hookah has become normative and that smoking is now socially accepted, or even expected on college campuses. The hookah bars were more likely to have a younger popu lation than the hookah restaurants, as noted during the observations. “Just like the people that we hung out with. It wa s sort of like you had to do it.” (male) “Kind of it’s like everyone else does it as if it’s not – it’s just an everyday thing, so it’s like you don’t – you feel m ore accepted to try it. I guess it’s more [unintelligible] because it’s not like you’re smoking cigarettes.” (female) Many smokers reported that older peers were a major factor in their decision to start. In fact, for some, hookah smoking is a norm or a valued tradition passed from older to younger students.


“When you go to high school and you get older frien ds and like grades above you, you just chill with them and that ’s what they do so you just smoke with them.” (male) “It was like that’s when I guess it was [introduced ] because you have older friends when you’re younger, but now yo u’re the oldest like seventeen and all your friends are older so th ey can show you other things that you probably can’t get on your ow n.” (male) “You like pass it on. Like from the older genera tion to the younger generation around the high school.” (male) Hookah smoking also was described by smokers as a s taple or norm at student parties. “So it’s something to do, just like holding a cup a t a party, like it’s just something there.” (male) Some students reported that hookah smoking is even accepted by their parents, but this point may reflect the students’ cultural t raditions rather than a more widespread or universal norm as many students were reportedly from India and other countries where waterpipe tobacco smoking is considered customary. “My parents they like to smoke hookah, and they’re not saying go ahead smoke as much as you want, they’ll say mostly it’s we just be careful with it. So I know my parents used to d o a lot of stuff when they were younger,[laughter] so they don’t hav e a lot of negative attitude about hookah.” (male) Even my parents actually mentioned it to me like, “ Oh guess what we just tried smoking hookah” and I’m like “Wait wh at?” How weird is that, but that’s because my cousin took th em somewhere and so but it was just like, “Oh so we tried it” an d it was like they didn’t think it was a big deal at all.” (female) However, most smokers said their parents are unawar e of their hookah smoking practice or disapprove of it. During verification, students said that because it is not easily detected (no clothing stains and or bad smell) that parents are unaware of their smoking.


“I grew from a very strict family, so anything out of our culture we can’t.” (male) “My mom doesn’t know [laughter] and no one in my fa mily knows. Some of my friends are not okay with it, friends wh o don’t do it are hesitant towards it.” (female) Siblings are another source of normative beliefs, with many smokers reporting that their family members influenced them to smoke for the first time. “Definitely my family. I grew up around it.” (male) “My sister took me to my first Hookah bar.” (male) Most nonsmokers described hookah as popular activit y among college students and reported that they have friends, roommates, and neighbors who smoke. “A lot of my friends, because quite a few of my fri ends actually do smoke hookah, so they always talk about it.” (female) A few nonsmokers felt that hookah was strange or no ted that the practice was imported from other cultures. “I thought it was weird that everyone was smoking o ut of the same [unintelligible] little thing – using the same pipe I guess.” (female) “I also have some Arabic friends and they tend to b e primarily Hookah smokers.” (male) “I’d say just neighbors from the dorm and a friend did too, and their parents but it’s religious.” (male) Motivation to comply When discussing people who had influenced their dec ision to smoke hookah, many smokers mentioned their parents, siblings, and friends as having a significant impact.


“Yes, because I was at a family friend’s house and we were all gathering around the hookah. I didn’t know. Nobod y knew what it was and my mom didn’t know. She was like, “What is this?” and she tried it, and she’s like, “Hey, [name] take it” and she made me smoke it.” (male) “…it’s probably very important because these are t he people unlike my friends, I am from, I am a part of them. So if they have a suggestion for me, I have to kind of take it at fac e value. It’s not like an advice. It’s kind of like mandatory like th is is my family.” (male) Other smokers also acknowledged the opinions of oth ers, but were careful to note that the choice to smoke was ultimately their own. “It’s the same for me. I know what they want me to do ever since I was little and they’ve taught me important values a nd stuff to stick to that, but overall it’s my life so it’s like a 70 /30.” (female) “It’s important, but it’s more of a personal decisi on that I don't think they should be concerned about.” (female) A few smokers denied that they had been influenced by others at all. “Not important at all. I respect how they feel abo ut it, but it’s not going to affect my choice on who I am because I’m g oing to do what’s going to make me happy.” (female) Most nonsmokers also denied being influenced by oth ers, saying that the decision not to smoke has been theirs alone. They described themselves as “independent thinkers” capable of resisting encouragement or temptation fr om others around them. “I really don’t feel I should smoke it. If they’re smoking it, then they can smoke it, even if it’s around me, but I’m not going to do it.” (female) “I know all the times the risks for a common goal that we have and just then, whatever. For something like this that I already know, that is not going to help me in any type of way. I could care less with what they want me to do.” (male)


“In some cases I feel like that’s important, but in other cases like if they’re like, “Oh you should smoke some hookah.” I’ m not going to do it because my family tells me to do it.” (female) Quantitative Study Findings – Phase 2 Online Survey Prevalence of Waterpipe Tobacco Smoking To assess the prevalence of waterpipe tobacco smoki ng, respondents were asked: Have you ever tried smoking tobacco from a hookah e ven 1-2 puffs. Almost half (42.7%) of students reported having ever smoked waterpipe t obacco (“waterpipe ever smokers”). Current waterpipe tobacco use (past 30 days) was re ported by 43(14.9%) students (“waterpipe current smokers”). Intention to smoke waterpipe tobacco in the next few months was assessed by collapsing the likert scale response into two variables: likely and unlikely. Sixty-nine percent of the respondents re ported they were unlikely to smoke waterpipe tobacco and 22% were likely to smoke wate rpipe tobacco in the next few months. Bivariate Analysis The primary analysis to answer the research questio ns consisted of bivariate correlations. More specifically, the chi-square te st of independence was used to understand the relationships between nominal variab les, and Spearman rank-order correlation was used to understand the relationship s between ordinal variables.


Tobacco Use Characteristics Chi-square tests of independence were used to exami ne relationships between demographic characteristics and tobacco use variabl es. The primary variables related to tobacco use were waterpipe ever smokers and waterpipe current smokers Results of the bivariate analysis indicated that ha ving ever tried hookah was not statistically related to gender, age, or religion; however, there was a statistically significant relationship between being a waterpipe ever smoker and ethnicity ( c2=19.170,df=4, p= 0.001). In particular, compared w ith Blacks, Whites had greater odds of reporting ever waterpipe use than Blacks (O R= 8.2, p < 0.001), and had slightly greater odds of reporting ever waterpipe use than A sians/Pacific Islanders (OR = 1.6, p = 0.35). Asians/Pacific Islanders had greater odds o f reporting ever waterpipe use than Blacks (OR = 5.1, p < 0.05). Being a waterpipe cur rent smoker (past 30 day use) was not statistically correlated with age, gender, ethnicit y, or religion. Bivariate Relationships to examine the Research Que stions in the Study Bivariate correlations of all survey items for each TRA construct revealed no multicollinearity or singularity. The primary biva riate analysis used was the Spearman rank-order correlation coefficient (alpha of 0.05). Constructs that were represented by more than one variable were averagedand then correl ated. All items were assessed with 7-point Likert-type scales with endpoints indicated in each of the tables below, accompanied by a “neither/nor” mid-scale response o ption.Due to non-normality of the data, Spearman’s rhowas calculated to examine corre lations among constructs in the theoretical model as guided by the research questio ns.


Research Question #1: What is the relationship between attitudes and int ention to smoke waterpipe tobacco? The intention construct of the theoretical model wa s evaluated from one item assessing the respondents’ intention to smoke water pipe tobacco within the next few months. The construct for attitude was assessed by averaging four survey items assessing the respondents’ negative or positive evaluation re lated to hookah smoking. These items were averaged based on the theoretical implication that attitudinal variables are not weighted for their importance (Ajzen & Fishbein, 19 80b). Table 8 represents the survey items associated with the theoretical constructs of intention and attitude. Table 8: Survey Items Representing the Theoretical constructs of Intention and Attitudes Construct Reliability Survey Item Mean Intention I intend to smoke hookah within the next few months 2.51 Attitude 0.950 -If I smoke hookah, this behavior is: bad/good -For me, smoking hookah is: awful/nice -For me, smoking hookah is: not fun/fun -For me, smoking hookah is: unpleasant/pleasant 3.01 3.17 3.27 3.22 After averaging the attitudinal construct survey it ems, the relationship between attitude and intention was examined using the Spear man’s rho correlation coefficient. The results (N=276) indicated a statistical signifi cant relationship (p < 0.001) between the attitude construct and intention to smoke waterpipe tobacco within the next few months. The correlation coefficient ( r = 0.804) represents a strong positive relationship between


n attitude and intention indicating that as people’s evaluation of hookah becomes more positive, their intention to smoke it within the ne xt few months increases. Research Question #2: What is the relationship between subjective norms and the intention to smoke waterpipe tobacco? The second immediate precursor to intention to perf orm a behavior is subjective norm. Subjective norm is an evaluation of normativ e beliefs about performing a behavior. In other words, it is represented by peo ple’s beliefs that important others think they should or should not perform a behavior. For this study the subjective norm was evaluated on a scale ranging from extremely disappr ove to extremely approve. The constructs of intention and subjective norms were e valuated with two individual survey questions (Table 9). Table 9: Constructs of Intention and Subjective Nor ms Construct Survey Item Mean Intention I intend to smoke hookah within the next few months 2.51 Subjective Norm -If I smoke hookah, most of the people who are important to me would: disapprove/approve 2.79 Subjective norm and intention (N=285) were analyzed using Spearman’s rho and produced a statistically significant relationship ( p < 0.001). Spearman’s rho correlation coefficient ( r = 0.512) indicates a moderate relationship between subjective norm and intention to smoke waterpipe tobacco within the nex t few months.

PAGE 100

n At this point in assessing the first two research q uestions for this study, the two primary determinants of intention as indicated by t he TRA were both statistically significant, however, attitude represented a strong er correlation ( r = 0.804) than subjective norm ( r = 0.512) for intention to smoke waterpipe tobacco in the next few months. Research Question #3: What is the relationship between beliefs of importa nt others and subjective norms? The next step in the analysis was to examinekey con structs embedded in the TRA. In particular, the relationship between subjective norm and beliefs of important others was examined. The construct of beliefs of importan t others was represented by four items indicated by important others that were ident ified during the qualitative phase of the research. More specifically, parents, siblings friends, and boyfriends/girlfriends were indicated as being important others who influence h ookah use. The constructs of subjective norm and beliefs of important others and their respective survey items are noted in Table 10.

PAGE 101

n Table 10: Constructs and Survey Items Construct Reliability Survey Item Mean Subjective Norm If I smoke hookah, most of the people who are important to me would: disapprove/approve 2.79 Beliefs of Important Others 0.829 If I smoke hookah, my parents would: disapprove/approve -If I smoke hookah, my friends would: disapprove/approve -If I smoke hookah my boyfriend/girlfriend would: disapprove/approve -If I smoke hookah, my siblings would: disapprove/approve 2.23 3.89 3.40 2.99 Spearman’s rho correlation coefficient was used to examine the relationship between the two theoretical constructs, beliefs of important others and subjective norm (N=283). The constructs were statistically signifi cantly related (p < 0.001) with a moderately strong correlation coefficient ( r = 0.699), indicating that as a person believes important others approve of hookah smoking the pers on’s normative beliefs of the behavior increase. Research Question #4: What is the relationship between motivation to comp ly and subjective norms? According to the TRA, motivation to comply is also related to subjective norm, thereby influencing a person’s normative beliefs ab out hookah smoking. The motivation to comply items were drawn from the same important others (parents, friends, siblings, boyfriend/girlfriend), but assessed the degree to w hich the respondent intends to perform

PAGE 102

n a behavior that they believe their important others think they should perform. Table 11 represents the survey items for motivation to compl y and subjective norm. Table 11: Survey Items for Motivation to Comply and Subjective Norm Construct Reliability Survey Item Mean Motivation to Comply 0.766 Most of the time when my parents think I should do something, I go along with it: disagree/agree -Most of the time when my friends think I should do something, I go along with it: disagree/agree -Most of the time when my boyfriend/girlfriend think I should do something, I go along with it: disagree/agree -Most of the time when my siblings think I should do something, I go along with it: disagree/agree 4.65 3.71 3.95 3.79 Subjective Norm -If I smoke hookah, most of the people who are important to me would: disapprove/approve 2.79 Using Spearman’s rank-order correlation (N=283), mo tivation to comply was not statistically significantly related to subjective n orm ( r = 0.046, p=0.221). Research questions #3 and #4 examined the social in fluences to smoke or not to smoke waterpipe tobacco. The results suggest that in this population, students are more influenced by their perception of important others’ desire for them to smoke or not to smoke hookah versus the student’s general motivatio n to comply with important others.

PAGE 103

n The last two research questions focus on the direct relationship between attitude and the precursors (outcome expectancies and evalua tion of outcome expectancies). Research Question #5: What is the relationship between outcome expectanci es and attitudes? With respect to the TRA, two influences on attitude are outcome expectancies and evaluation of outcome expectancies. Outcome expect ancies were evaluated using 21 items to reflect the diverse components identified in the qualitative phase of the research. Table 12 represents the survey items representing t he constructs of attitude and outcome expectancy that were averaged to determine the stat istical relationship. Table 12: Survey Items Representing the Constructs of Attitude and Outcome Expectancy Construct Reliability Survey Item Mean Attitude 0.950 -If I smoke hookah, this behavior is: bad/good -For me, smoking hookah is: awful/nice -For me, smoking hookah is: not fun/fun -For me, smoking hookah is: unpleasant/pleasant 3.01 3.17 3.27 3.22 Outcome Expectancies 0.887 Read the statements and select an answer that best describes your beliefs and opinions whether you have smoked hookah or not: unlikely/likely If I smoke hookah, it will help me relax and relieve my stress -If I smoke hookah, it will help me stay more focused If I smoke hookah, it will help me to meet a 2.89 2.09 2.03

PAGE 104

n potential date -If I smoke hookah, I will be more social -If I smoke hookah, it will give me a headache If I smoke hookah, it will help me to make new friends If I smoke hookah, it reinforces my culture If I smoke hookah, I will have fun -If I smoke hookah, I will feel more intellectual If I smoke hookah, it will bring my family together -If I smoke hookah, it will help me to think more clearly If I smoke hookah, I am being safer than if I smoke cigarettes If I smoke hookah, it will help to pass the time -If I smoke hookah, it will give me a buzz -If I smoke hookah, it gives me a “legal high” If I smoke hookah, it makes me dizzy If I smoke hookah, it makes me short of breath If I smoke hookah, it makes my chest hurt If I smoke hookah, it makes me cough -If I smoke hookah, I will become addicted If I smoke hookah, I will get lung cancer 2.71 3.93 2.71 1.73 3.43 1.96 1.40 1.79 2.64 2.92 3.06 2.70 3.75 3.47 3.38 4.03 2.56 4.00 Analysis (N=252) of attitude and outcome expectancy consisted of averaging the variables that measured each construct and then exa mination using Spearman’s rho.

PAGE 105

n Outcome expectancy was statistically significantly related to attitude about hookah smoking (p<0.001) with a corresponding moderate cor relation of 0.409 indicating that outcome expectancy is related to attitudes but only to a moderate degree. Therefore, as outcome expectancies increase, so does the positive attitude regarding hookah smoking. Research Question #6: What is the relationship between evaluation of outc ome expectancies and attitudes? The final research question focused on evaluation o f outcome expectancies as related to attitude about hookah smoking. Evaluati on of outcome expectancies assessed whether the previous outcome expectancies were view ed by the respondents as being important. Table 13 represents the survey items u tilized in the analysis to determine the relationship. Table 13: Survey Items Utilized in the Analysis Construct Reliability Survey Item Mean Attitude 0.950 If I smoke hookah, this behavior is: bad/good -For me, smoking hookah is: awful/nice -For me, smoking hookah is: not fun/fun -For me, smoking hookah is: unpleasant/pleasant 3.01 3.17 3.27 3.22 Evaluation of Outcome Expectancies 0.928 In deciding to smoke hookah, how important are each of the following reasons: unimportant/important The lightheadedness that results from smoking The stress relief that results from smoking hookah -The social activity that results from smoking 3.63 3.29 3.68

PAGE 106

n hookah -The ability to make new friends by smoking hookah Meeting potential dates by smoking hookah -The tradition of smoking hookah The “buzz” I get from smoking hookah The headaches that result from smoking hookah The shortness of breath that results from smoking hookah -The chest pains I experience the next day from smoking hookah -The ability to stay focused when smoking hookah -The ability to pass time when smoking hookah -The relaxed feeling from smoking hookah -The ability to be athletic after smoking hookah -That hookah is less harmful than cigarettes -The group dynamics when smoking hookah -The “fun” that occurs when smoking hookah -The bad health effects from smoking hookah 3.00 2.36 2.07 2.76 3.65 3.76 4.10 3.16 3.08 3.55 3.17 3.45 3.36 3.40 4.55 Evaluation of the relationship between evaluation o f outcome expectancies (N=254) and attitude revealed a statistically signi ficant relationship (p < 0.001). Spearman’s rho correlation between evaluation of ou tcome expectancies and attitude was

PAGE 107

n low ( r = 0.289) indicating that there is only a slight in crease in attitude as outcomes are viewed as more important. Research questions #5 and #6 assess the attitudinal component related to waterpipe tobacco smoking. The results indicated t hat students in this population have a more positive attitude about hookah smoking based o n the attributed beliefs about hookah smoking versus how important these beliefs are to t hem. Multivariable Analysis Exploratory factor analysis was conducted with the items representing outcome expectancies to identify latent variable structure. Using SPSS version 19, factor extraction was performed with the 18 variables note d in Table 17 representing outcome expectancy. To ensure adequate sampling Kaiser-Meye r-Olkin (KMO) measure of sampling adequacy was performed and an acceptable v alue of 0.908 was identified. Additionally, Bartlett’s test of sphericity was sta tistically significant (p < 0.001) indicating that there are some relationships between the varia bles that are included in the analysis making factor analysis an appropriate test to condu ct. Factor extraction was based on Kaiser’s criterion of retention of a factor with an eigenvalue of 1.0 or greater and examination of the scree plot. To improve interpre tation of factors, orthogonal rotation using varimax was employed. Through examination of the rotated component matrix and the communalities of the variables, three factors were extracted and can be found in Table 14. Factor 1 consisted of ten items: relax and relieve stress, stay focused, meet potent ial date, be more social, make new friends, have fun, safer than ciga rettes, pass the time, give me a “buzz,”

PAGE 108

n and give me a “legal high.” Factor 1 represents the benefits that hookah smokin g provides to students. Cronbach’s alpha for Factor 1 was 0.926. Factor 1 accounted for 37% of the total variance. Examination of Cronbach’ s alpha if items are deleted indicated that all items were appropriately extracted into th is factor. Factor 2 consisted of seven items: gives me a headache, makes me dizzy, makes me short of breath, makes my chest hurt, makes me cough, I will become addicted, and I will get lung cancer. Factor 2 represents the negative health effects. Cronbach’s alpha for Factor 2 was 0.888. Factor 2 accounted for 20.5% of the total variance. Examina tion of Cronbach’s alpha if items were deleted indicated all items were appropriately extracted into this factor. Factor 3 consisted of four items: reinforces my culture, I will feel more intellectua l, brings my family together and helps me to think more clearly. Factor 3 represents meeting expectations. Cronbach’s alpha score for Factor 3 was 0.824. Factor 3 accounted for 6% of the total variance. Examination of Cronbach’s al pha it items were deleted indicated that all items were appropriately extracted into th is factor. The three factors together accounted for a cumulative 63.5% of the total varia nce. Table 14: Factors, Cronbach Alpha, and Associated V ariables Factors Cronbach’s alpha Variables (factor loa ding) Factor 1 0.926 If I smoke hookah it will help me relax and relieve my stress (0.787) “Benefits” If I smoke hookah it will help me stay more focused (0.632) If I smoke hookah, it will help me meet a potentia l date (0.647) If I smoke hookah, I will be more social(0 .733) If I smoke hookah, it will help me to make new friends (0.752)

PAGE 109

nn If I smoke hookah, I will have fun (0.827) If I smoke hookah, I am being safer than if I smoke cigarettes (0.608) If I smoke hookah, it will help to pass the time (0.806) If I smoke hookah, it will give me a buzz(0.801) If I smoke hookah, it gives me a “legal high” (0.669) Factor 2 0.888 If I smoke hookah, it will give me a headache (0.622) “Negative Health Effects” If I smoke hookah it makes me dizzy (0.733) If I smoke hookah, it makes me short of breath (0.866) If I smoke hookah, it makes my chest h urt (0.887) If I smoke hookah it makes me cough (0.847) If I smoke hookah I will become addict ed (0.711) If I smoke hookah I will get lung canc er (0.654) Factor 3 0.824 If I smoke hookah it reinforces my culture (0.644) “Meeting Expectations” If I smoke hookah I will feel more intellectual (0.652) If I smoke hookah it will bring my family together (0.784) If I smoke hookah it will help me to think more clearly (0.697) Further investigation into the correlation of facto rs associated with the identified outcome expectancy and general attitude revealed st atistically significant results for each factor and the corresponding general attitude. More specifically, analysis using Spearman’s rho produced a correlation coefficient o f 0.674 (p < 0.001) for Factor 1 and attitude, a coefficient of -0.353 (p < 0.001) for Factor 2, and a coefficient of 0.636

PAGE 110

(p < 0.001) for Factor 3. The relationship of Factor 1 ( benefits) with attitude represents a positive correlation indicating that as a person’s perceived benefits of smoking waterpipe tobacco increase then their attitude about waterpip e tobacco smoking also increases or becomes more favorable. The relationship of Factor 2 (negative health effects) was negatively correlated with attitude. This relation ship indicates that as a person’s perceives waterpipe tobacco as being more negativel y related to health their attitude about waterpipe tobacco smoking decreases or become s less favorable. Factor 3 (meeting expectations) was positively related to attitude in dicating that as a person perceives family as being a benefit of waterpipe tobacco smok ing then their attitude towards smoking becomes more favorable. These extracted f actors and their corresponding correlation coefficients with the attitude construc t can be seen in Table 15. Table 15: Extracted Factors and Correlations with A ttitude Factor Correlation Coefficient P-value N Factor 1 (benefits) 0.674 <0.001 269 Factor 2 (negative health) -0.353 <0.001 268 Factor 3 (family values) 0.636 <0.001 26 9 In summary, a person’s attitude regarding waterpipe tobacco smoking is a more important determinant of intention to smoke waterpi pe tobacco than subjective norms in this sample population as seen in Figure 5. On eva luation of the precursors to attitude, behavioral beliefs exert a stronger relationship th an evaluation of behavioral beliefs, however, they both positively influence attitudes t o waterpipe tobacco smoking. On evaluation of the determinants of subjective norm, beliefs of important others provides

PAGE 111

the most influence. This analysis suggests that pe rsonal factors are of greater importance when determining facto rs that influence intention to smoke waterpipe toba cco. behavioral beliefs that influence attitudes toward waterpipe tobacco smoking, two factors (benefits and family values) were strongly correlat ed, suggesting that as these beliefs are viewed as outcomes of waterpipe tobacco smoking then the a ttitude toward waterpipe tobacco smoking becomes more favorable. Factor 2 ( negative health) is negatively correlated, suggesting that a person believes that negative health effects are the outcome to wate rpipe tobacco smoking then their attitude toward wa terpipe tobacco smoking becomes more unfavorable. Figure 5 : TRA model and C the most influence. This analysis suggests that pe rsonal factors are of greater importance rs that influence intention to smoke waterpipe toba cco. behavioral beliefs that influence attitudes toward waterpipe tobacco smoking, two factors (benefits and family values) were strongly correlat ed, suggesting that as these beliefs are as outcomes of waterpipe tobacco smoking then the a ttitude toward waterpipe tobacco smoking becomes more favorable. Factor 2 ( negative health) is negatively correlated, suggesting that a person believes that negative health effects are the outcome rpipe tobacco smoking then their attitude toward wa terpipe tobacco smoking becomes more unfavorable. : TRA model and C orrelations the most influence. This analysis suggests that pe rsonal factors are of greater importance rs that influence intention to smoke waterpipe toba cco. And of the behavioral beliefs that influence attitudes toward waterpipe tobacco smoking, two factors (benefits and family values) were strongly correlat ed, suggesting that as these beliefs are as outcomes of waterpipe tobacco smoking then the a ttitude toward waterpipe tobacco smoking becomes more favorable. Factor 2 ( negative health) is negatively correlated, suggesting that a person believes that negative health effects are the outcome rpipe tobacco smoking then their attitude toward wa terpipe tobacco smoking

PAGE 112

CHAPTER 5: DISCUSSION This chapter includes a discussion of the results r elative to the reviewed literature and the proposed research questions. It is organiz ed into the following sections: research summary, discussion of the results, recommendations next steps in the research process, strengths and limitations of the study, implication s for public health practice, suggestions for dissemination of findings, and a summary and co nclusion. Research Summary The prevalence of waterpipe tobacco smoking among c ollege students in the US is increasing (Primack, et al., 2008; Smith-Simone, et al., 2008; Sutfin, et al., 2011). Researchers have noted the importance of monitoring waterpipe smoking and developing responsive interventions to minimize its health imp act; thus, items that provide insights about these specific behaviors are often included i n general health surveys. College students may be a particularly vulnerable populatio n when it comes to experimentation with waterpipe smoking because of its perceived soc ial and contextual value. Literature reviews and this research show that hookah smoking is a valuable social component of their college experience. It serves as an avenue t o help them build relationships and to meet new people. The TRA promotes understanding of the antecedents o f intention to perform a given behavior. According to the TRA, intention is the central determinant of a behavior; however, as people formulate their intention they c onsider other specific and independent constructs. In particular, the theory posits that individuals take into account favorable or

PAGE 113

unfavorable attitudes in adopting the behavior, as well as perceived social pressure (subjective norm). The primary purpose of this mixed method study was to examine the relationship between constructs of the TRA and waterpipe smoking A final study objective was to illuminate the relationships of variables that infl uence attitudes toward waterpipe smoking by identifying factors that contribute to o utcome expectancies. The study included sequential primary data collection emanati ng from naturalistic observations, one-on-one intercept interviews, focus group interv iews, and an online survey of randomly selected students living in University-own ed housing. The survey sought demographic data, information about waterpipe smoki ng behaviors, and responses to items specific to the TRA that were developed based on the qualitative data. Items were subjected to extensive pilot testing prior to disse mination. Exploratory factor analysis was the primary multivariable analysis to simplify the outcome expectancy construct and provide items for future research to confirm or ref ute the underlying factor structures. Discussion of Results In the study, current waterpipe tobacco use (past 3 0 days) was nearly 15%. Other US university-based studies report similar results. For example, Primack et al. (2008) estimated current use to be 9.5% in a random sample of undergraduate and graduate students; Eisenberg et al. (2008) reported current use to be 20.4% in their study of firstyear psychology students; Smith et al. (2006) repo rted a 15.3% rate in a study of freshman at a private university; and most recently Sutfin et al. (2011) reported a 17% rate in a random sample of 3770 students across eig ht universities in one US state as can be seen in Table 16.

PAGE 114

Table 16: US College Studies The proportion of students who reported that they h ad ever tried waterpipe tobacco smoking was 42.7% in this study, comparable to 40% to 50% range found in other US studies (Eissenberg, et al., 2008; Primack et al., 2008). These results suggest that the prevalence of waterpipe smoking among coll ege students is not a waning fad. Author State N Waterpipe smoking Sutfin, et al. 2011 N.C. 3770 College drinking survey 17% past 30 days Primack, et al. 2010 8 colleges in US 8745 NCHA 7.2% past 30 days 30% ever tried Eissenberg, et al. 2008 Virginia 744 VCU psychology students 20% past 30 days Primack, et al. 2008 Pennsylvania 647 U of Pitt NCHA 9.5%past 30 days 41% ever smokers Smith-Simone, et al. 2006/2008 D.C. 411 Johns Hopkins freshmen 15% past 30 days Grekin, et al. 2008 Michigan 602 psychology students 15% ever tried Ward, et al. 2007 Va/Tenn 143 caf and restaurant patrons 22% weekly

PAGE 115

In the present sample, Whites were more likely to h ave tried waterpipe tobacco than Blacks or Asian/Pacific Islanders, a finding t hat also is consistent with other university-based studies that examine hookah smokin g (Primack, Fertman, Rice, AdachiMejia, & Fine, 2010; Primack, et al., 2008; Smith, Curbow, & Stillman, 2006; Sutfin, et al., 2011). Whereas these findings suggest that Wh ites should be given greatest priority in anti-hookah interventions, current adoption may be slower among other racial or ethnic groups, making it important to monitor their waterp ipe use and to note if it begins to show an uptake pattern similarto that observed amon g Whites. The TRA has been applied mostly to cigarette use w ith the constructs being reliable predictors of intention to smoke cigarette s. In the area of waterpipe tobacco use, the TRA has been applied only partially, thus yield ing limited construct appraisal. This study took into account each theoretical construct of the TRA. Consistent with the theory, this study revealed that attitude and subje ctive norms were positively correlated with intention to smoke waterpipe tobacco. Holding positive attitudes toward waterpipe smoking was, however, more positively associated wi th intention than was subjective norm. This relationship was consistent with qualit ative findings, suggesting that students made their own decisions about waterpipe tobacco sm oking. Furthermore, the results of this study suggest that when outcome expectancies latent constructs were extracted, the social aspect s and meeting expectations aspects of waterpipe smoking were significantly related to int ention to smoke waterpipe tobacco. Students find waterpipe smoking a way to seek pleas ure ( getting a buzz or legal high, finding it a relaxing and stress-free way to pass t he time or socialize with friends ). Students also seek meeting expectations related to bringing family together and reinforce

PAGE 116

their culture along with staying focused. However, this finding requires further explication as to how why waterpipe tobacco smoking is related to expectations. This study was consistent with previous literature examining perceived benefits of waterpipe smoking among college students. Primack e t al. (2008) examined a random number of US undergraduate and graduate students in one university and found that social acceptability and popularity of smoking hook ah were strong predictors of use. In a study of US college freshman,Smith-Simone et al. (2 008) reported that students perceived waterpipe tobacco smoking as the most soc ially acceptable form of tobacco smoking among their peers. Other international stu dies, have reported similar perceived benefits. Roskin et al. (2009) surveyed British co llege students and found they viewed waterpipe tobacco smoking as a relaxing and afforda ble way to enjoy the fruit flavors. Chaaya et al. (2003) studied college students in Le banon and found that students perceived waterpipe tobacco smoking as a way to pro mote family gatherings, a way to relax and freely communicate with others, and a tim e to think. In this research, the social nature and relaxation benefits were strongly correlated with positive attitudes towards waterpipe tobacco s moking. Additionally, students perceived waterpipe tobacco smoking as being less h armful than other forms of tobacco use. Some students did not believe that the tobacc o used in waterpipe contained nicotine and some students did not believe that the product smoked was related to tobacco. Whereas other students understood that it was indee d tobacco in the waterpipe, they believed that the water served as a filtration mech anism that offered protection from harmful chemicals.

PAGE 117

Similar to previous research, this study found that smokers believed it to be a safer alternative than other forms of tobacco produ cts such as cigarettes and chew tobacco. For example, Smith-Simone et al. (2006) r eported that US college freshmen perceived waterpipe tobacco smoking to be less harm ful than cigarettes. Primack et al. (2008) found similar perceptions, but also reported that students believed waterpipe tobacco to be less addictive. Sutfin et al. (2011) found that college students in eight different universities in one US state were more li kely to be current waterpipe tobacco smokers if they viewed it as less harmful than ciga rettes. These same misperceptions were found in international studies of college stud ents. Roskin et al. (2009) reported that the majority of waterpipe tobacco smokers felt that it was less harmful than cigarettes and some even felt it way to wean themselves from cigar ette smoking. Chaaya et al. (2003) reported that Lebanese university students believed waterpipe tobacco smoking to be less harmful than cigarettes. This research also supports previous studies indica ting that students associate popularity with waterpipe tobacco smoking. Qualita tive inquiry and verification discussions supported a sense of “fitting in” and a feeling of maturity when smoking hookah. Students viewed waterpipe tobacco smoking as enabling them to make new friends and meet potential dates. Primack et al. ( 2008) found perceived popularity a strong predictor of waterpipe tobacco smoking in hi s random sample of college students. In contrast to findings reported by Roskin et al. ( 2009), an “exotic appeal” of waterpipe tobacco smoking emerged neither from qual itative data, nor in the verification focus group. Furthermore, students in the sample f ailed to report that waterpipe tobacco smoking was a gateway activity to cigarette smoking as Ward et al (2006) suggest.

PAGE 118

Rather, this research supports the potential of wat erpipe tobacco smoking to be a gateway to poly-pharmacy including alcohol and illicit drug s used in combination, both on and off campus as qualitative data indicated. Similar to o ther research conducted with college students, this sample also smoked hashish or mariju ana in combination with the waterpipe tobacco (Sutfin, et al., 2011). Having depressed affect has been cited by some auth ors as a reason for students to smoke waterpipe tobacco (Grekin & Ayna, 2008). Alt hough depression did not emerge in this study as an antecedent of waterpipe tobacco smoking, there was a strong correlation with the need for escapism from the dai ly pressures of school and work. This form of escapism through hookah smoking may be rela ted to depression, but warrants further investigation. The affordability of smoking waterpipe tobacco was discussed in the qualitative research. Students who smoke hookah more often fou nd that the bars/restaurants were becoming too expensive and elected to purchase thei r own pipe and tobacco. With occasional smokers, the bar scene was seen as an af fordable venue, but for the routine smoker, the benefits and affordability associated w ith owning a pipe was viewed as a more attractive and cost-effective alternative. Som e students even referred to hookah pipe ownership as an investment. Although this study did not investigate dependence and waterpipe tobacco, it was interesting to note that in the qualitative inquiry students found lone smokers to be acceptable, at least from the smoker’s perspective. Lone smoking was not viewed as a form of dependence. The literature comparing socia l versus lone waterpipe tobacco smoking does imply a dependence issue. Both Salame h et al. (2008) and Maziak et al.

PAGE 119

n (2004) found that intensity of smoking was related to lone smoking patterns and that waterpipe smoking can produce dependence. Recommendations The relatively high rates of waterpipe tobacco smok ing found in this sample and other studies (Primack, et al., 2008; Smith-Simone, et al., 2008; Sutfin, et al., 2011) suggest that a need for a preventive intervention t argeting undergraduate students exists. Of special concern are widely held misperceptions t hat hookah is safe to smoke, and unlike cigarettes, is safe from the risk of addicti on or serious health problems. Although there are influencers of waterpipe smoking, student s report they are the ultimate decision makers related to hookah smoking. Therefore, unders tanding their wants and needs, particularly those met by smoking waterpipes, is a priority. We have learned from this study that personal facto rs weigh heavier on intention than do the social factors. One strategy to deter uptake and decrease current use would be to reinforce the negative health effects and edu cate students on the true facts regarding waterpipe tobacco smoking as they enter college so that they can make informed decisions about smoking. Consideration should also be taken to inform their caregivers, as this study may suggest that caregivers are impor tant to college students and they too may be misinformed about the safety of waterpipe to bacco smoking. It also became apparent during this research that University perso nnel, who are responsible for students, are also misinformed about waterpipe tobacco smokin g. Educating USF Housing Resident Assistants about the harmful effects provi des a readily accessible resource for students. During the course of this research, adve rtisements for local hookah cafes were

PAGE 120

seen on campus and in the Oracle (campus paper). C ollege administrators should not allow such advertisements and perhaps AHEC and othe r college organizations can utilize these modes of information dissemination to educate readers. From a policy perspective, there are several concer ns. The first concern is the status of regulation from the federal level. Water pipe tobacco is currently excluded in general tobacco legislation because of the fact tha t the waterpipe tobacco is not burned directly, but rather, is combusted from the charcoa l. Including waterpipe tobacco under the same regulations as cigarettes would increase t axation and possibly make waterpipe tobacco a less affordable behavior choice. Secondl y, there is anecdotal discussion of making the USF campus smoke-free. This discussion has both positive and negative implications. Banning smoking on campus may decrea se the smoking circles and cast a more negative light on hookah smoking, similar to c igarettes. The ban would need to be specific as to the type of smoking that is not allo wed on campus so as to avoid the loopholes currently acknowledged in federal legisla tion. The negative implications include the possibility that students may migrate f rom smoking outdoors to smoking indoors. This transition is concerning given the a mount of carbon monoxide that hookah produces and reports of fires in dormitories. The potential for adverse health effects would put students at risk. Research Next Steps Additional research can expand understanding of fac tors that can be used to prevent college students from smoking hookah. Impr oved understanding of the motivational factors for initiating waterpipe tobac co smoking might be obtained through

PAGE 121

use of a larger sample. In turn, these findings mi ght benefit development of a countermarketing plan or campaign. The current study examined hookah smoking based on constructs derived from the TRA. Although the TRA has been useful in impro ving understanding of cigarette smoking, its value may not transfer directly to the practice of waterpipe tobacco smoking. Thus, consideration of alternative behavioral model s and frameworks seem warranted, not only for anticipating who will or will not grav itate toward waterpipe tobacco smoking, but for creating responsive interventions that dissuade experimentation and uptake. Additional next research steps would be to test the TRA model using confirmatory factor analysis or structural equation modeling. Study Strengths and Limitations This research study capitalized on the benefits of a mixed method design in which qualitative methods were used to identify attitudes behavioral beliefs, and other factors that influence college students’ decisions to smoke hookah. Results were used to create an online survey for exploring the relationship bet ween key constructs in the TRA and smoking intention. Moreover, the qualitative metho ds aided in contextualizing waterpipe smoking and added to the validity of the quantitati ve data (Creswell, 2009). The results have the potential to heighten awareness of the iss ue of waterpipe smoking and inform the development of waterpipe smoking prevention eff orts among college students. This study has several important limitations. Firs t, with the exception of observational studies, the study relied on self-rep orted data to identify the determinants of intention to smoke waterpipe. Self-report data has the potential to alter the estimates of true relationships due to social desirability. Soc ial desirability response bias may have

PAGE 122

influenced this study’s results as found in aggrega te estimates of alcohol consumption, whereby respondents may be concerned about social n orms (Fowler, 2009). Also, people in waterpipe lounges/restaurants may change their b ehavior if they realize they are being observed (Gittelsohn, Shankar, West, Ram, & Gynwali 1997). To help minimize reactivity, the researcher smoked in almost every l ounge/restaurant and avoided obvious note taking and other activities that might have re vealed research conduction. However, these efforts to minimize reactivity do not negate investigator bias during the observational periods. Despite application of random sampling techniques, the study is delimited to the actual respondents, thereby creating response bias. Furthermore, participants included only a modest fraction of the students solicited. A low response rate threatens internal validity. Primack et al. (2008) reported the respo nse rates of online surveys at between 10 and 30 percent. The present study had just an 8 % response rate despite the incorporation of participant incentives and multipl e attempts to reach students. Additional efforts to increase response were includ ed in the survey design elements, such as increasing convenience and accessibility of the survey by placing it online, and allowing respondents to stop and resume as needed ( Dillman, Smyth, & Christian, 2009). Finally, was not possible to compare characteristic s of responders and non-responders for insights into bias that may have been introduced in to the final study sample do to an inability to obtain population characteristics othe r than ethnicity. More men than women attended the focus groups. Sev eral factors may have accounted for this difference. The recruitment for focus groups took place in the Marshall Center. During the verification phase, st udents commented that men are more

PAGE 123

willing to declare their behaviors than are women. If this belief is valid, it may have accounted for the unequal gender response. Overall however, the focus groups were well attended. The interviews with smokers predominantly involved men. Although this distribution may indicate that more men than women are likely to be hookah smokers, a definitive conclusion would require a sample balanc ed more equitably by sex. The nonsmoker interviews reflected no gender bias. Despite the efforts to decrease response bias, more women took the survey than did men and more men than women participated in the smoker interviews and smoker focus groups. Sutfin et al. (2011) had a 63% femal e response rate to the online survey across eight universities and Primack et al. (2008) had a 66% female response rate, indicating this phenomenon is not unusual. One reco mmendation would be to better understand appropriate incentives for male students In this particular research, it was discovered in the verification focus group that a f ree song download was not an incentive given that students pirate free songs. It is possi ble that had the survey been made available for a longer period of time and more remi nders or prompts had been included, a better response rate might have been obtained. The institution from which the sample was obtained represents yet another sampling limitation. Although USF is a public univ ersity, it may not be representative of other public universities, especially given the non response to the survey which threatens the validity of the conclusions. Therefore, conclu sions are delimited to students living in University-owned housing at this single university.

PAGE 124

Due to the cross-sectional nature of the study, dir ectionality and causality cannot be determined. This study did not explore the TRA in its entirety, as the goal of the TRA is to predict and understand a behavior. To fully understand the model and determine the relationship between intention and behavior, a long itudinal study would need to be employed. This study looked only at the correlatio n between individual constructs in the TRA. Additionally, despite the widespread use of t he TRA to predict and understand behavioral intention, this theory has been criticiz ed for not explaining behavioral change, a matter that is key in alcohol and drug education prevention. Another limitation of using TRA is the assumption that human beings act rationa lly. Evolving literature on behavioral economics indicates that humans are not always rational or predictable in their behavior (Ariely, 2008). Dissemination of Findings There are many audiences that would benefit from th ese findings, including professionals in public health, particularly health education, as well as other health practitioners. Potentially, student services and s tudent health services personnel at colleges and other educational institutions could b enefit from these findings by including factual data about waterpipe smoking in new student orientations. Further beneficiaries may be persons working in tobacco use, including pe rsonnel from voluntary health organizations such as the American Lung Association the American Cancer Society, the American Heart Association, and others. Finally, s ocial and behavioral science researchers who are beginning to explore the waterp ipe smoking phenomenon in greater detail can continue to advance understanding of the reasons why it is attractive to persons of college age.

PAGE 125

One mode of dissemination would consist of journal articles to include the American Journal of Public Health Tobacco Control Journal of American College Health Respiratory Care Journal American Journal of Health Behavior the College Student Journal and American Journal of Emergency Medicine For those organizations that would benefit from more direct communication s uch as college health departments and the American Lung Association local chapter, an informal summary of findings in a presentation format will be offered. Another mode of dissemination would be through formal oral presentations at conferences such as Am erican Association of Respiratory Care’s National Conference and the American Public Health Association’s national conference. Finally, due to the population from wh ich the samples were drawn, an executive summary will be provided to the Universit y of South Florida and the campus health department for informing and assisting with potential future waterpipe interventions. For funding this study, the Area He alth Education Center at USF will be provided with a summary of findings and a formal or al presentation. Currently AHEC funds local talks on waterpipe tobacco smoking in t he Tampa Bay community. Summary and Conclusion Educating students who are new to college, as well as their parents, about waterpipe tobacco smoking may help to inform their decisions whether or not to smoke hookah. As noted in this study, some parents are u naware of the dangers of the behavior or at least provide a counter-marketing point of vi ew.. The results of this study suggest that prevention programs, possibly including social marketing campaigns, should focus on decreasing favorable attitudes associated with w aterpipe smoking through counter-

PAGE 126

marketing by clarifying its addictive nature, as we ll as the associated communicable diseases and other negative health effects. Associ ating waterpipe smoking to cigarette smoking may help to deter adoption of the behavior as borne out in the qualitative data that showed nonusers of hookah associated it with c igarettes, whereas users disassociated the two behaviors. To the extent that current find ings might be duplicated elsewhere, waterpipe tobacco smoking is prevalent on or near c ollege campuses. Understanding the theoretical constructs of intention to smoke waterp ipe tobacco among college students helps to tailor interventions. Despite its limitat ions, this study provides guidance for developing targeted prevention programs focused on waterpipe tobacco use. These interventions may help to decrease the prevalence o f waterpipe smoking and potential nicotine addiction.

PAGE 127

REFERENCES American Cancer Society. (2010). Smoking. Retriev ed January 29, 2010, from _cigarette_smoking_and_c ancer.asp Ajzen, I., & Fishbein, M. (1980a). Understanding attitudes and predicting social behavior (Pbk. ed.). Englewood Cliffs, N.J.: Prentice-Hall. Ajzen, I., & Fishbein, M. (1980b). Understanding attitudes and predicting social behavior : Prentice hall. Al Mutairi, S. S., Shihab-Eldeen, A. A., Mojiminiyi O. A., & Anwar, A. A. (2006). Comparative analysis of the effects of hubble-bubbl e (Sheesha) and cigarette smoking on respiratory and metabolic parameters in hubble-bubble and cigarette smokers. Respirology, 11 (4), 449-455. American Lung Association. (2007). An emerging deadly trend: Waterpipe tobacco use Washington, D.C.: American Lung Association. American Lung Association (2010). Hookah use: 1st n ew tobacco trend of the 21st century. Retrieved March 12, 2010, from Ariely, D. (2008). Predictably irrational : The hid den forces that shape our decisions Retrieved from WorldCat database Available from Asfar, T., Ward, K. D., Eissenberg, T., & Maziak, W (2005). Comparison of patterns of use, beliefs, and attitudes related to waterpipe be tween beginning and established smokers. BioMed Central Public Health, 5 19. Ashmawi, M. (2003). Some predictive markers of athe rosclerosis among smokers. Ain Shans Medical Journal, 44 633-639. Bacha, Z. A., Salameh, P., & Waked, M. (2007). Sali va cotinine and exhaled carbon monoxide levels in natural environment waterpipe sm okers. Inhalation Toxicology, 19 (9), 771-777. Barnett, T., Curbow, B., Soule, E., Tomar, S., & Th ombs, D. (2011). Carbon monoxide levels among patrons of hookah cafes. American Journal of Preventative Medicine, 40 (3), 324-328. Barnett, T., Curbow, B., Weitz, J., Johnson, T., & Smith-Simone, S. (2009). Water pipe smoking among middle and high school students. American Journal of Public Health, 99 (11). Bernard, H. (2006). Research methods in anthropolog y : Qualitative and quantitative approaches Retrieved from WorldCat database Availab le from tml Biostatistics. (n.d.). Biostatistics at Darmouth. Retrieved March 12, 2010, from

PAGE 128

Camarota, S. (2007). Immigrants in the United State s, 2007: A profile of America's foreign-born population. Retrieved from Cavus, U., Rehber, Z., Ozeke, O., & Ilkay, E. (2010 ). Carbon monoxide poisoning associated with narghile use. [casebook]. Emergency Medical Journal, 27 (5), 406. Chaaya, M., El-Roueiheb, Z., Chemaitelly, H., Azar, G., Nasr, J., & Al-Sahab, B. (2004). Argileh smoking among university students: A new to bacco epidemic. Nicotine Tobacco Resesarch, 6 (3), 457-463. Chaouachi, K. (2009). Hookah (Shisha, Narghile) Smo king and Environmental Tobacco Smoke (ETS). A critical review of the relevant lite rature and the public health consequences. International Journal of Environmental Respiratory Public Health, 6 (2), 798-843. Clarkin, P., Tisch, L., & Glicksman, A. (2008). Soc ioeconomic correlates of current and regular smoking among college students in Rhode Isl and. Journal of American College Health, 57 (2), 183-190. Cobb, C., Ward, K., Maziak, W., Shihadeh, A., & Eis senberg, T. (2010). Waterpipe tobacco smoking: An emerging health crisis in the U nited States. American Journal of Health Behavior, 34 (3), 275-285. Creswell, J. (2009). Research design : Qualitative, quantitative, and mi xed methods approaches : Sage Publications. Daher, N., Saleh, R., Jaroudi, E., Sheheitli, H., B adr, T., Sepetdjian, E., et al. (2010). Comparison of carcinogen, carbon monoxide, and ultr afine particle emissions from narghile waterpipe and cigarette smoking: Side stream smoke measurements and assessment of second-hand smoke emission factor s. Atmospheric Environment, 44 (1), 8-14. Debus, M. (n.d.). Methodological review: A handbook for excellence in focus group research. Retrieved from _excellence_in_focus_grou p_research_-full_text.pdf DeCouteau, E. (2009). Freebie frenzy Vanguard DiClemente, C. (2010). The challenge of changing addictive behaviors: Unde rstanding the journey into and out of addiction University of South Florida. Eissenberg, T., Ward, K. D., Smith-Simone, S., & Ma ziak, W. (2008). Waterpipe tobacco smoking on a U.S. College campus: Prevalence and co rrelates. Journal of Adolescent Health, 42 (5), 526-529. El-Nachef, W. N., & Hammond, S. K. (2008). Exhaled carbon monoxide with waterpipe use in US students. Journal of the American Medical Association, 299 (1), 36-38. Facebook. (2010). USF hookah club interest group. Retrieved March 20, 2010, from Food and Drug Administration. (2009). Tobacco Produ cts. Retrieved October 15, 2009, from Fields, A. (2005). Discovering statistics using spss (2nd ed.): Sage Publishing. Fishbein, Azjen, & Hanson. (n.d.). Fishbein/Ajzen-H anson questionnaire. Retrieved from ure/Fishbein-AjzenHanson_Questionnaire.pdf

PAGE 129

n Fisher, W., Fisher, J., & Rye, B. (1995). Understan ding and promoting AIDS-preventive behavior: Insights from the theory of reasoned acti on. health Psychology, 14 (3). Fowler, F. J. (2009). Survey research methods, Appl ied social research methods series Retrieved from WorldCat database Available from 5.html FreeDictionary. (2010). The Free Dictionary. Retr ieved March 20, 2010, from ance Fromme, H., Dietrich, S., Heitmann, D., Dressel, H. Diemer, J., Schulz, T., et al. (2009). Indoor air contamination during a waterpipe (narghi le) smoking session. Food and Chemical Toxicology, 47 (7), 1636-1641. Gilpin, E., & Pierce, J. (1997). Trends in adolesce nt smoking initiation in the United States: Is tobacco marketing an influence? Tobacco Control, 6 122-127. Glaser, B., & Strauss, A. (2009). The discovery of grounded theory: Strategies for qualitative research London: Aldine Transaction. Goodman, J. (1993). Tobacco in history : The cultures of dependence London: New York. Grekin, E. R., & Ayna, D. (2008). Argileh use among college students in the United States: an emerging trend. Journal of Studies in Alcohol and Drugs, 69 (3), 472475. Guo, Q., Johnson, C., Unger, J., Lee, L., Zie, B., Chou, C., et al. (2007). Utility of the theory of reasoned action and theory of planned beh avior for predicting Chinese adolescent smoking. Addictive Behaviors, 32 1066-1081. Gupta, D., Boffetta, P., Gaborieau, V., & Jindal, S (2001). Risk factors of lung cancer in Chandigarh, India. Indian Journal of Medical Research, 113 142-150. Hanson, M. (2005). An examination of ethnic differe nces in cigarette smoking intention among female teenagers. Journal of American Academic Nurse Practice, 17 149155. Hayes, S. (2005). Students take a hazy break at hoo kah bar St. Petersburg Times Hip Digital. (2009). Music Everywhere. Retrieved March 25, 2011, from Hookah bars. (2009). from Hookahlover. (2008, March 20). Hookah stories Fra t house hookah. Environmental Protection Association. An introduction to indoor air quality (2009). Retrieved from Jackson, D., & Aveyard, P. (2008). Waterpipe smokin g in students: prevalence, risk factors, symptoms of addiction, and smoke intake. E vidence from one British university. BioMed Central Public Health, 8 174. Jamil, H., Janisse, J., Eisouhag, D., Fakhouri, M., Arnetz, J., & Arnetz, B. (2011). Do household smoking behaviors constitute a risk facto r for hookah smoking? Nicotine and Tobacco Research, Epub ahead of pring (Jan 31). Jones, S. (2002). The Internet Goes to College. Pew Internet and American Life Retrieved from 2/PIP_College_Report.pdf. pdf

PAGE 130

Kelly, J. (2009). Students build sukkah at UM as pa rt of Jewish holiday, noting its 'a good place to chill' Missouliam Khalil, J., Heath, R., Nakkash, R., & Afifi, R. (20 09). The tobacco health nexus? Health messages in narghile advertisements. Tobacco Control, 18 (5), 420-421. Khater, A. E., Abd El-Aziz, N. S., Al-Sewaidan, H. A., & Chaouachi, K. (2008). Radiological hazards of Narghile (hookah, shisha, g oza) smoking: activity concentrations and dose assessment. Journal of Environmental Radioactivity, 99 (12), 1808-1814. Kinman, A. (2010). Have a nargilah. Campus Retrieved from Kiter, G., Ucan, E. S., Ceylan, E., & Kilinc, O. (2 000). Water-pipe smoking and pulmonary functions. Respiratory Medicine, 94 (9), 891-894. Lafflin, M., Moore-Hirschi, S., Weis, D., & Hayes, B. (1994). Use of the theory of reasoned action to predict drug and alcohol use. Substance Use and Misuse, 29 (7), 927-940. Lenney, W., & Enderby, B. (2008). "Blowing in the w ind": a review of teenage smoking. Archives of Diseases in Children, 93 (1), 72-75. Lewin, T. (2006). Collegians smoking hookah....fill ed with tobacco The New York Times Ling, P., & Glantz, S. (2002). Why and how the toba cco industry sells cigarettes to young adults: Evidence from industry documents. American Journal of Public Health, 92 (6), 908-916. Lyon, L. (2008). The rising allure and danger of ho okah. U.S. News and World Report Retrieved from /the-risingallure--and-danger--of-hookah.html MacCallum, R., Widaman, K., Zhang, S., & Hong, S. ( 1999). Sample size in factor analysis. Psychological Methods, 4 (1), 84-99. Makhoul, J., & Nakkash, R. (2009). Understanding yo uth: Using qualitative methods to verify quantitative community indicators. Health Promotion and Practice, 10 (1), 128-135. Marx, R., Menezes, A., Horovitz, L., Jones, E., & W arren, R. (2003). A comparison of two time intervals for test-retest reliability of h ealth status instruments. Journal of Clinical Epidemiology, 56 730-735. Maziak, W. (2008). The waterpipe: time for action. Addiction, 103 (11), 1763-1767. Maziak, W., Eissenberg, T., Rastam, S., Hammal, F., Asfar, T., Bachir, M. E., et al. (2004). Beliefs and attitudes related to narghile ( waterpipe) smoking among university students in Syria. Annals of Epidemiology, 14 (9), 646-654. Maziak, W., Eissenberg, T., & Ward, K. D. (2005). P atterns of waterpipe use and dependence: Implications for intervention developme nt. Pharmacology and Biochemical Behavior, 80 (1), 173-179. Maziak, W., Hammal, F., Rastam, S., Asfar, T., Eiss enberg, T., Bachir, M. E., et al. (2004). Characteristics of cigarette smoking and qu itting among university students in Syria. Preventive Medicine, 39 (2), 330-336. Maziak, W., Rastam, S., Ibrahim, I., Ward, K. D., & Eissenberg, T. (2008). Waterpipeassociated particulate matter emissions. Nicotine Tobacco Research, 10 (3), 519523.

PAGE 131

Maziak, W., Ward, K. D., Afifi Soweid, R. A., & Eis senberg, T. (2004). Tobacco smoking using a waterpipe: A re-emerging strain in a global epidemic. Tobacco Control, 13 (4), 327-333. Maziak, W., Ward, K. D., Afifi Soweid, R. A., & Eis senberg, T. (2005). Standardizing questionnaire items for the assessment of waterpipe tobacco use in epidemiological studies. Public Health, 119 (5), 400-404. Mohammad, Y., & Kakah, M. (2008). Chronic respirato ry effect of narguileh smoking compared with cigarette smoking in women from the E ast Mediterranean region. International Journal of Chronic Obstructive Pulmon ary Disease, 3 (3), 405-414. Monzer, B., Sepetdjian, E., Saliba, N., & Shihadeh, A. (2008). Charcoal emissions as a source of CO and carcinogenic PAH in mainstream nar ghile waterpipe smoke. Food and Chemical Toxicology, 46 (9), 2991-2995. Moran, S., Wechsler, H., & Rigotti, N. (2004). Soci al smoking among US college students. Pediatrics, 114 1028-1034. Morrison, D., Golder, S., Keller, T., & Gillmore, M (2002). The theory of reasoned action as a model of marijuana use: Tests of implic it assumptions and applicability to high-risk young women. Psychology of Addictive Behaviors, 16 (3), 212-224. Naquin, M., & Gilbert, G. (1996). College students' smoking behavior, perceived stress, and coping styles. Journal of Drug Education, 26 (4), 367-376. National Cancer Institute. (n.d.). NCI measures gui de for youth tobacco research. Tobacco Control Research Retrieved from Neergaard, J., Singh, P., Job, J., & Montgomery, S. (2007). Waterpipe smoking and nicotine exposure: a review of the current evidence Nicotine Tobacco Research, 9 (10), 987-994. Nehl, E. J., Blanchard, C. M., Peng, C. Y., Rhodes, R. E., Kupperman, J., Sparling, P. B., et al. (2009). Understanding nonsmoking in African American and Caucasian college students: An application of the theory of p lanned behavior. Behavioral Medicine, 35 (1), 23-29. Neumeier, M. (2006). The brand gap : How to bridge the distance between business strategy and design : a whiteboard overview : New Riders. Nuwayhid, I. A., Yamout, B., Azar, G., & Kambris, M A. (1998). Narghile (hubblebubble) smoking, low birth weight, and other pregna ncy outcomes. American Journal of Epidemiology, 148 (4), 375-383. O'Rourke, N., Hatcher, L., & Stepanski, E. (2005). A step-by-step approach to using SAS for univariate and multivariate statistics : Wiley InterScience. Parna, K., Usin, J., & Ringmets, I. (2008). Cigaret te and waterpipe smoking among adolescents in Estonia: HBSC survey results, 1994-2 006. BioMed Central Public Health, 8 392. Patterson, F., Lerman, C., Kaufmann, V., Neuner, G. & Audrain-McGovern, J. (2004). Cigarette smoking practices among American college students: review and future directions. Journal of American College Health, 52 (5), 203-210. Pierson, D., & Kacmarek, R. (1992). Foundations of respiratory care New York: Churchill Livingstone.

PAGE 132

Prignot, J. J., Sasco, A. J., Poulet, E., Gupta, P. C., & Aditama, T. Y. (2008). Alternative forms of tobacco use. Int Journal of Tuberculosis and Lung Disease, 12 (7), 718727. Primack, B. A., Sidani, J., Agarwal, A. A., Shadel, W. G., Donny, E. C., & Eissenberg, T. E. (2008). Prevalence of and associations with wate rpipe tobacco smoking among U.S. university students. Annals of Behavioral Medicine, 36 (1), 81-86. Primack, B. A., Walsh, M., Bryce, C., & Eissenberg, T. (2009). Water-pipe tobacco smoking among middle and high school students in Ar izona. Pediatrics, 123 (2), e282-288. Raosoft. (2007). Database web survey software for g athering information. Retrieved March 12, 2010, from Rastam, S., Ward, K. D., Eissenberg, T., & Maziak, W. (2004). Estimating the beginning of the waterpipe epidemic in Syria. BioMed Central Public Health, 4 32. Reuters. (2009). GroupOn Launches in Tampa. Retri eved October 15, 2009, from 79+21-Sep2009+BW20090921 Rhodes, N., & Ewoldsen, D. (2009). Attitude and nor m accessibility and cigarette smoking. Journal of Applied Social Psychology, 39 (10), 2355-2372. Roskin, J., & Aveyard, P. (2009). Canadian and Engl ish students' beliefs about waterpipe smoking: a qualitative study. BioMed Central Public Health, 9 10. Salameh, P., Waked, M., & Aoun, Z. (2008). Waterpip e smoking: construction and validation of the Lebanon Waterpipe Dependence Scal e (LWDS-11). Nicotine Tobacco Research, 10 (1), 149-158. Saleh, R., & Shihadeh, A. (2008). Elevated toxicant yields with narghile waterpipes smoked using a plastic hose. Food and Chemical Toxicology, 46 (5), 1461-1466. Schane, R., Glantz, S., & Ling, P. (2009). Nondaily and social smoking: An Increasingly prevalent pattern. Archives of Internal Medicine, 169 (19), 1742-1744. Schleicher, H. E., Harris, K. J., Catley, D., & Naz ir, N. (2009). The role of depression and negative affect regulation expectancies in toba cco smoking among college students. Journal of American College Health, 57 (5), 507-512. Security, H. (2009). 2008 Yearbook of Immigration S tatistics. Retrieved from ook/2008/ois_yb_2008.pdf Sepetdjian, E., Shihadeh, A., & Saliba, N. A. (2008 ). Measurement of 16 polycyclic aromatic hydrocarbons in narghile waterpipe tobacco smoke. Food and Chemical Toxicology, 46 (5), 1582-1590. Shihadeh, A. (2003). Investigation of mainstream sm oke aerosol of the argileh water pipe. Food and Chemical Toxicology, 41 (1), 143-152. Smith-Simone, S., Curbow, B., & Stillman, F. (2008) Differing psychosocial risk profiles of college freshmen waterpipe, cigar, and cigarette smokers. Addict Behav, 33 (12), 1619-1624. Smith-Simone, S., Maziak, W., Ward, K., & Eissenber g, T. (2008). Waterpipe tobacco smoking: Knowledge, attitudes, beliefs, and behavio rs in two U.S. samples. Nicotine Tobacco Research, 10 (2), 393-398. Smith, S. Y., Curbow, B., & Stillman, F. A. (2007). Harm perception of nicotine products in college freshmen. Nicotine Tobacco Research, 9 (9), 977-982. Some hookah bars fighting new smoking ban. (2010).

PAGE 133

SouthSmoke. (n.d.). Hookah clubs. Retrieved from Tamim, H., Al-Sahab, B., Akkary, G., Ghanem, M., Ta mim, N., El Roueiheb, Z., et al. (2007). Cigarette and nargileh smoking practices am ong school students in Beirut, Lebanon. Am Journal of Health Behavior, 31 (1), 56-63. Tavafian, S. S., Aghamolaei, T., & Zare, S. (2009). Water pipe smoking and healthrelated quality of life: A population-based study. Archives of Iranian Medicine, 12 (3), 232-237. ToxFAQs (2010). University of South Florida. (2009a). Checkbox (Ver sion 4.6): Checkbox survey Solutions, Inc. University of South Florida. (2009b). Pocket Facts. Retrieved November 14, 2009, from University of South Florida. (2010). Pocket Facts. Retrieved January 22, 2011, from Uyamk, B., Arslan, E., Akay, H., Ercelik, E., & Tez M. (2009). Narghile (hookah) smoking and carboxyhemoglobin levels. [Letter to th e editor]. The Journal of Emergency Medicine Von Ah, D., Ebert, S., Ngamvitroj, A., Park, N., & Kang, D. (2004). Predictors of health behaviours in college students. Journal of Advanced Nursing, 48 (5), 463-474. Walsh, B., Czervinske, M., & DiBlas, i. R. (2010). Perinatal and pediatric respiratory care (3rd ed.). St. Louis, Mo.: Saunders/Elsevier. Ward, K. D., Eissenberg, T., Gray, J. N., Srinivas, V., Wilson, N., & Maziak, W. (2007). Characteristics of U.S. waterpipe users: A prelimin ary report. Nicotine Tobacco Research, 9 (12), 1339-1346. Ward, K. D., Hammal, F., VanderWeg, M. W., Eissenbe rg, T., Asfar, T., Rastam, S., et al. (2005). Are waterpipe users interested in quitt ing? Nicotine Tobacco Research, 7 (1), 149-156. Ward, K. D., Vander Weg, M. W., Relyea, G., Debon, M., & Klesges, R. C. (2006). Waterpipe smoking among American military recruits. Prev Med, 43 (2), 92-97. Wechsler, H., Rigotti, N. A., Gledhill-Hoyt, J., & Lee, H. (1998). Increased levels of cigarette use among college students: a cause for n ational concern. Journal of the American Medical Association, 280 (19), 1673-1678. Weglicki, L., Templin, T., Rice, V., Jamil, H., & H ammad, A. (2008). Comparison of cigarette and water-pipe smoking by Arab and non-Ar ab-American youth. American Journal of Preventative Medicine, 35 (4), 334-339. World Health Organization. (2006). Tobacco use in shisha: Studies on waterpipe smoking in Egypt World Health Organization. (2007). The scientific basis of tobacco product regulation : Report of a WHO Study Group ; [the third meeting of the WHO Study Group on Tobacco Product Regulation (TobReg) was held in Kob e, Japan from 28 to 30 June 2006]. Conf Author(s): Meeting of the WHO Stud y Group on Tobacco Product Regulation (TobReg) ; (3 :; 2006.06.28-30 : ; Kobe, Japan) : Geneva. Whokah333. (2010, February 20, 2010). Smoking a col lege dorm room. Wolf, B. (2010). Pros and woes from smoking through a hose. The Tufts Daily

PAGE 134


PAGE 135

Appendix 1: Observation Guide Date: Observer: Location: Start Time of Session: ______________ End Time of S ession: _________________ Total Number of Participants at Start of Session: Are all participants smoking Hookah? Yes No If Not, Describe Group Dynamics: (Where are smokers sitting compared to nonsmokers) Gender of Participants: Number of Males _____; Numb er of Females _____ Site Description: (What is the room like? How are t ables arranged?) Description of Ambiance: (Is music playing? Are dr inks and food being served?) Description of Hookah devices (Are there multiple h oses, mouthpieces, tall, short) Description of Smoking: (How are participants shari ng the hookah) Describe the Overall events: (Are people eating, dr inking? What is the conversation?) AdicionalObservacin Notes: Observation #1 June 18, 2010 Location: Hookah restaurant/bar Mary Martinasek (ethnographer) Location: We entered the front of the restaurant and requeste d seating in the hookah area on the patio. The time was 2115. The patio, which seats around 30 people, was nearly full. Tables lined each side of the long patio with a nar row walking area in between. The patio was situated against the side of the restaura nt building. The building wall had two flat screen televisions mounted on them. Speakers were mounted on the wall, as well. We were seated at the far end of the patio, close t o the hookah bar that had several empty bar stools and many empty hookahs waiting to be ord ered. The patio tables were arranged so that 2, 4, or 6 people could sit comfor tably. The patio had an overhang roof

PAGE 136

Appendix 1 (continued) to keep the rain off and a rolled up plastic barrie r on the side facing the parking lot in the event of inclement weather. Atmosphere: Music was piped through the speakers and consisted of reggae to dance hall type genres in addition to Egyptian style music. At 2230 the m usic got slightly louder. A one-man band was setting up a sound system around 2300. Dr inks and food were being served by scurrying wait staff. The lighting was relatively bright for an evening outdoors. Hookahs were placed on tables and on the floors beside the tables. The noise and music were so loud at times it was difficult to converse with peo ple at my own table. Later in the evening a belly dancer came out onto the patio. Pe ople were letting out bird like shrills and placing dollars in the waist of her pantaloons. Participants: When we arrived there were five tables occupied. T he first table had 4 females and 2 males most of whom appeared to be in their 20s to l ow 30s, however one female appeared to be in her 50s and a male in his 40s. A ll appeared to be Caucasian. There was one hookah in the middle of the table and every one shared the mouthpiece to smoke. This table smoked hookah the entire time that we we re there. The second table had four females and three males all whom appeared to be in their mid-20s and 30s. The two females at the end of the table were the primary ho okah smokers. At the third table was seated one female and three males who all appeared to be in their 20s. The fourth table had two females and two males in the 20s. The one female was smoking the hookah rapidly and she was the primary smoker of the group The fifth table had three females in their mid-20s. There were no smokers and no alcoho l was being consumed in this group. One table (Table 2) changed composition during the observation with two people in their 30s joining about midway through. During the obser vation period, the people that left the patio were mostly couples. Activities: Activities consisted of drinking, eating, smoking l aughing, cursing, and cheering. Hookah Devices: Hookahs were about 2.5 feet tall with a glass vase and filled with water. There was a single hose with multiple mouthpiece tips provided in plastic sleeves. Most of the participants shared the mouthpiece. The hose was o f a plastic feel and the permanent tip was made of metal and the handle was made of wood. Ethnographer observation: What struck me the most about this experience was a s we were leaving for the evening, around 2315 and walking down the long aisle to leav e the patio, there was one long table with at least eight people sitting, eating, drinkin g and smoking hookah. In the middle of

PAGE 137

Appendix 1 (continued) the table, seated, was a little girl who was 8-10 y ears of age. She was being exposed to the hookah smoke, drinking atmosphere, and conversa tions. Observation #2 July 3, 2010 Location: Hookah Restaurant Mary Martinasek (ethnographer) Location: This location was the same as observation #1. The hookah bar was located outdoors under a roof adjacent to a restaurant. There were fans blowing on us as it was summer time and flat screen televisions adorned the outsid e wall of the restaurant. Tables were arranged in groups of 2 8, some being square and others oblong. Because of the narrow length of the area, it was difficult to observe peo ple at the other end. We arrived at 2100 and the only seating available was near the entranc e to the patio. Because of the rain earlier in the day, there were plastic scroll down coverings on the patio to block potential rainfall from reaching the patron’s tables. About 2230, the plastic was rolled up to allow an open-air feel. Prior to this open air exposure, the only opening to the outside was at the entrance to the patio. At the entrance was a c ourtyard area with a gate around it and a fountain situated in the middle. Couches made of r attan and other seats were configured to face the fountain. This area was not covered an d was saturated from the earlier rain, therefore no one sat in this section of the patio. Atmosphere: Sports were being played on the flat screen TVs, bu t there was no volume coming from the source. Around 2230, about the same time they l ifted the patio plastic roll down covering, music began to play through the outdoor s peakers. Because of the loudness of the sound, it was difficult to carry on a conversat ion. Participants: Table 1 closest to us had three middle aged Middle Eastern looking individuals (2 males and 1 female) sitting around it. Only the eldest ma n and the woman smoked the hookah which was situated on the floor between their seats The other person at the table was smoking cigarettes. Table 2 had four under-aged looking individuals dri nking water and socializing. There were 2 males and 2 females. They sat male facing fe male. All smoked hookah except for one of the guys. They left the table around 2130 as a group. Table 3 arrived around 2130 and they appeared to be of Middle Eastern decent. A man

PAGE 138

Appendix 1 (continued) and young adult shared a hookah and mouthpiece and sat next to each other at the corner of a long table. The young adult appeared to be ner vous as he was tapping his foot and inhaling large quantities of smoke when the pipe fi rst was delivered to their table. A woman sat with them but did not smoke until two oth er women joined their table. The one woman who joined the table was wearing a burca. The three women ordered a separate hookah and shared it amongst themselves. They also shared the permanent mouthpiece. Table 4 had three Middle Eastern looking men who ap peared to be in their 30s. Two more men joined them after about 20 minutes. Three of the men at the end of the table smoked hookah, but the two that joined did not. Th e men used the disposable mouthpieces provided. They drank beer, ate, and co nversed. It was too loud to hear conversation topics. Table 5 arrived about 2152. There were four people (2 males and 2 females) The females sat opposite the males at the square table. They ap peared to be in their 30s. Only one man smoked the hookah and he smoked it frequently a nd with large inhalations during the observation period. Table 6 had two females and one male who all appear ed to be in their 20s. One hookah sat in the middle of their table. One of the female s seemed to smoke the hookah more passionately and frequently than the others. The se cond female was passed the pipe on occasion, but took a small puff and quickly passed it on. She appeared to be participating just to fit in and did not seem to enjoy it as much as the other female. They all shared the permanent mouthpiece. After a while another guy jo ined the table and also joined in the smoking. Table 7 at the far end of the patio sat two females and a male. They were smoking hookah, but it was too difficult to observe them fr om the distance. At Table 8 there were two guys who each had their o wn hookah device placed on the floor next to them. Activities: Because it was a restaurant and full bar, patrons w ere eating and drinking while enjoying their friends, conversations and sharing hookah. Hookah Devices: Hookah devices were placed on table tops and on the floor next to the tables, Most were about 2.5 feet tall with a glass vase filled with w hat appeared to be tap water. They were single hose devices and plastic mouthpieces were pr ovided for the table patrons. It appeared that most people did not change out the ti p, but rather shared the permanent mouthpiece. The hose had a plastic feel and the per manent mouthpiece was made of metal. The handle was made of wood. Servers wander ed around the patio providing fresh charcoal to the top of the hookahs. The charcoal w as a round wood charcoal.

PAGE 139

n Appendix 1 (continued) Ethnographer observation: Towards the end of the observation period, a table of eight came into the patio. They all appeared to be in their mid-30s. They had a very s mall infant in the stroller with them. Although they did not order hookah during the short time we were there, there was plenty of smoke being blown around the patio by the fans a nd the patrons. A diagram of the first two observation sites can be found in Figure A-1. Figure A1: Diagram of first two observation sites Observation #3 December 3, 2010 Location: Hookah Bar Mary Martinasek (ethnographer) Location: The hookah bar was situated at the end of a strip c enter. There were two glass doors at the entryway. As we entered, immediately in front of us was a tall desk and a cashier who proceeded to ask for our identification and col lect money for the entrance into the hookah bar. The cost was discounted if a USF ID wa s provided. The cost of $11 was for Hookah Bar Prep Area Restaurant on this side of wall Couch Couch Patio AreaUnderCover Open Air Patio Area Entrance fountain

PAGE 140

Appendix 1 (continued) unlimited hookah. The bar closed at 3 a.m. so that would allot us 6 hours of smoking if we elected to stay until closing. The bar was long and relatively narrow. Couches we re arranged in semi-circles lining each side of the bar. They sat up to about eight pe ople comfortably. The couches were a plush blue velvet material. Centered in the middle was a wooden table for the hookah device and ordering menu. The walls were painted a dark red, as was the ceiling. The walls were adorned with very large pictures portray ing somberness such as a noose above two people’s heads wrapped in a sheet. There was a stage for a band, but no one was set up at this time. There was no alcohol served at the bar, just a refrigerator of beverages at the entrance where patrons selected their choices a nd paid at the cash register. A diagram of the bar can be found in figure A-2. Atmosphere: The lighting was dim and a down tempo music played through the speaker system. Streams of hookah smoke could be seen billowing up towards the ceiling. Participants: Because of the seating, it was very difficult to ob serve anyone else except for the people adjacent to our couch. When we arrived there were a bout ten people there (7 males and 3 females). During the observation period, two girls arrived together and 2 females and a male left at the same time. It seemed that most of the people were college students. Except for the three people who left, everyone else stayed there the two hours that we were there. The male working the front was also sm oking hookah, very heavily and very passionately. Of the people that were observable, a ll seemed to be using the disposable mouthpieces. The males seemed to be smoking more o ften than the females. Activities: People were talking, drinking soda, blowing smoke r ings, playing chess, and active on their laptops. The bar had free Wi-Fi. Hookah Devices: The hookah devices were about 2 feet tall and had a decorative colored glass vase. The hose was plastic with a wooden handle and a metal p ermanent tip. Plastic tips were provided by the young female servers. The hookah t op was wrapped in foil and the server said it keeps the charcoal from burning the carpet or couches if they were to get knocked over.

PAGE 141

Appendix 1 (continued) Figure A2: Hookah bar layout Observation #4 December 11, 2010 Location: Hookah Bar Mary Martinasek (ethnographer) Location: This hookah bar was situated in the middle of a sma ll strip center. The entrance to the bar consisted of two glass doors. Upon entering, w e were faced with a tall desk and a male attendant who asked to see our identification. When asked if we could get a student discount, he said that they don’t give discounts th ere. We paid $10 for a hookah, which came with one free refill. This bar was divided in to two areas separated by a five-foot wooden partition. We sat on the left side of the p artition. We arrived at 2045 and were the only smokers at the start of the observation an d by the time we left at 2245 there were three groups of smokers. The room had red velour co uches that seat two people. There were two couches facing each other with an oblong t able in the middle for the hookah device or to place drinks, etc. There were about f our to five of these couch clusters in HookahPrepArea IDPlease Band Stage bathroom Couch withCoffee table

PAGE 142

Appendix 1 (continued) each of the two areas. The lighting was dim with te ardrop glass lights hanging over each table. A TV was mounted at the ceiling level in the back corners of the bar. The bar sold a variety of beers and nonalcoholic beverages. Ther e was a set up on the side we were on for an acoustic one-man band. Random pictures rang ing from postcard size to larger adorned the walls. It appeared they were on consig nment as there were prices next to each picture. They looked like student artwork and varied from cityscapes to animals. A diagram of the layout can be found in Figure A-3. Atmosphere: There was stereo music playing in the background. It was light enough that we could hear other conversations. The music ranged from reg gae to light rock sounds. Activities: People were drinking both beer and nonalcoholic bev erages. One group was playing checkers. One group was eating a bag of Airheads. Participants: Most people stayed within their groups and didn’t s ocialize with other groups. First group arrived around 2040 and consisted of five males and one female. They departed at 2210. Everyone in this group smoked hookah and seemed to range in age from 19-22 years. The female sat in between two males on one couch and th e other two males sat on the opposite couch. They ordered a total of three hooka hs. Their discussion ranged from ninja turtles to school. Hookah Device: The hookah was about 2.5 feet tall and similar to o ther bars in that it was a glass vase on the bottom, single plastic hose with a wooden handl e and metal permanent mouthpiece. The bar did offer plastic tips. Ethnographer Observation: Although not viewed as a typical pick up location, one of the guys from the other group came over to our couches and tried to initiate a co nversation. He asked us where we were from and what other hookah bars we had attended. H e was from Lakeland and said that he and his buddies often come to Tampa because they like to hookah bars. Along with the talking and laughing that took place throughout the observation, it was noticable that many people were coughing. This may be more noticable to me because of my clinical background.

PAGE 143

Appendix 1 (continued) Figure A3: Layout of hookah bar Observation #5 January 28, 2011 Location: Hookah Bar Mary Martinasek (ethnographer) Location: This bar was the same hookah bar as the previous lo cation #4. We arrived at the location at 2100 and left at 2310. We decided to sit on the right side of the partition this time. The walls were noted to be brown in color. There was a hookah menu provided at each table allowing us to pick our flavors. I counted 20 tea hookah choices and 48 regular hookah choices. The attendant said that the tea hookah doe s not contain nicotine. Participants: When we arrived at the lounge there were two colleg e-aged males each with their own hookah. One of the guys was lying on the couch smok ing his hookah. No conversation took place between them. They were drinking sodas. The other guy was trying to blow smoke rings. They each received one hookah refill d uring the observation period and then departed together around 2200. couches couch couch StageBand Bathroom Entrance IDplease

PAGE 144

Appendix 1 (continued) An Asian couple arrived around 2145 and shared a ho okah and mouthpiece. They were there for the duration of the observation. They cu ddled often and were playing a game called Connect 4. An African American female and Hispanic male arrive d around 2200. They each ordered their own hookah and did not share. The each sat on their own couch facing each other. The conversation was not discernable. On the other side of the partition were three groups that came in at different times. At one poi nt, one of the groups moved to our side and sat watching the television, which was playing the news. Atmosphere: It was a relaxing atmosphere with light music playi ng in the background. Ethnographers ObservationThere was more smoking occurring tonight compared t o the last observation here. My eyes were burning by the time I left. Observation #6 February 12, 2011 Location: Hookah Bar Mary Martinasek (ethnographer) Location: This observation took place in the same location at observation #3. We arrived at this location at 2100. When we arrived there were many c ouches already filled with what appeared to be mostly college-aged young adults. B ecause of the layout of the room it was difficult to observe others beyond a few couche s. The stage for a band had guitars in their stands, but no one was playing when we arrive d. The stereo music played a variety of music from light rock to more contemporary easy listening music. The patrons were singing along with the song, “Ain’t no way to hide your lying eyes”, which is from my era! I ordered my usual hookah, but decided to get mango flavored this time. An acoustic one-man band started playing at 2130. He sang and played the guitar while people watched. People clapped after each song and cheered him on to sing more songs. It appeared to be “open mic” night as he encouraged others to come up and sign or play a tune. Hookah: I was reviewing the hookah menu in more detail than the previous observation and noted that they offered 34 flavors and the waitress said I could mix up to three flavors. They also had supreme flavors for $5 extra. They were a different brand than the other 34 flavors. At the bottom of the menu was a Surgeon G eneral’s warning that if pregnant

PAGE 145

Appendix 1 (continued) may cause premature birth, low birth weight or feta l injury. There were no other warnings noted. I asked the waitress if I could smoke cigar ettes in the bar and she said “unfortunately not”. My hookah got a little strong at one point and I asked the waitress if she could change the charcoal and so she offered to repack the shisha but I just asked her to change the charcoal. After she did she took the hose and sucked on my mouthpiece to see if it was better. I was a little disgusted with the fact that she shared my mouthpiece and didn’t put a tip on it. Participants: Because we were seated by the door it was easy to w atch people coming and going. At 2140, two guys and a girl left together. At 2030 tw o males and two females entered the bar. I did notice one single guy in the very back of the bar sitting by himself smoking a hookah. He left just before we did. At 2035, two males and two females left together. Activities: People were laughing, talking, and I noticed there were a lot more people coughing than in previous observations. Activities consisted of card games, laptop work, and general conversations. At the couch adjacent to ours were two guys blowing hookah bubbles using some device. I asked the waitress what it wa s and she said they offer cut off plastic bottles and a container of soap so that patrons can blow bubbles for fun. Ethnographer Observation: One male who entered later in the evening came in s moking an electronic cigarette. There was a lot of discussion at his table but unfo rtunately I could not hear any of it. There was a lot of smoke in the bar tonight. On oc casion the owners would open the front door to air out the place. My eyes were burn ing and I was starting to get a headache. When I got home I checked my exhaled car bon monoxide level and it was at 13 ppm, with 0 ppm being normal. This increase in CO may be why I had a headache. By the morning my level was down to 6 ppm and by 3 p.m. it was 3 ppm. My headache lingered throughout the day.

PAGE 146

Appendix 2: Survey Instrument 1. Do you live on campus in university-owned housing s uch as campus residencies, fraternity houses, sorority houses or apartments on campus? [ ]Yes 1 Go to Q2 [ ]No 2 Terminate survey 2. How old are you? [ ] 16 years or younger 8 Terminate survey [ ] 17 years 7 Terminate survey [ ] 18 years 6 Go to Q3 [ ] 19 years 5 Go to Q3 [ ] 20 years 4 Go to Q3 [ ] 21 years 3 Go to Q3 [ ] 22 years 1 Go to Q3 [ ] 23 years or older 1 Go to Q3 Smoking Status Questions] These questions ask about hookah smoking. Hookah s moking also is known to some people as shisha smoking, narghile/arghile smoking and waterpipe tobacco smoking. The following questions refer to smoking tobacco only such as shisha in these devices (hookah/waterpipe), unless otherwise specified. 3. Have you ever tried smoking tobacco from a hookah e ven 1-2 puffs/draws? [ ] Yes 1 (Go to Q6) [ ] No 2 (Continue to Q5) 4. If you were to consider smoking hookah in the futur e, which individuals or groups might encourage you to smoke? [ Options of level o f influence; not at all, somewhat influential, very influential] [] Friends [] Family [] Hookah lounge owners [] Sorority/Fraternity

PAGE 147

Appendix 2 (continued) (Go to Q11) 5, About what age were you when tried smoking hook ah for the first time? 6. The first time you used hookah, with whom did you smoke? [ ] I smoked it alone [ ] one or both of my parents [ ] one or more adults other than parents [ ] one or more of my siblings [ ] one or more of my friends [ ] Other ___________________ 7. During the past 30 days have you smoked hookah, even one or two puffs? [ ] Yes 1 [ ] No 2 8. When you smoke hookah, about how often do you share the mouthpiece with other people? [ ] All or most of the time [ ] Some of the time [ ] Never or almost never 9. Where do you smoke hookah most often? Select all that apply: [] Dorm [] Apartment [] Hookah bar [] House Other (please specify_______) [Intention Question] 10. I intend to smoke hookah within the next few months [Probably: probably not] with response options of e xtremely, quite, slightly neither on each side of the 7 pt. scale

PAGE 148

Appendix 2 (continued) [Behavioral Belief Questions] Below are some statements about hookah smoking. Pl ease read the statement and select an answer for each one that best describes y our beliefs and opinion. Likely to unlikely – 7 pt. scale 11. If I smoke hookah, it will help me relax and reliev e my stress 12. If I smoke hookah, it will help me stay focused on my schoolwork. 13. If I smoke hookah, I will be more social. 14. If I smoke hookah, it will give me a headache. 15. If I smoke hookah, it will help me to make new frie nds. 16. If I smoke hookah, it will help me to meet others o f the opposite sex. 17. If I smoke hookah it reinforces my culture. 18. If I smoke hookah, it gives me a “legal high.” 19. If I smoke hookah, it makes me dizzy. 20. If I smoke hookah, it makes me short of breath. 21. If I smoke hookah, it makes my chest hurt the next day. 22. If I smoke hookah it makes me cough. 23. If I smoke hookah, I will become addicted. 24. If I smoke hookah, I will get cancer. 25. If I smoke hookah, I will have fun. 26. If I smoke hookah, I will feel more intellectual. 27. If I smoke hookah, it will bring my family together 28. If I smoke hookah, it will help me think. 29. If I smoke hookah, I feel safer than if I smoked ci garettes. 30. If I smoke hookah, it will help to pass the time. [Questions related to evaluation of Behavioral beli efs] 31. In deciding to smoke hookah, how important is each of the following reasons for smoking hookah? 1=extremely unimportant, 2=somewhat unimportant, 3= neither important nor unimportant, 4=somewhat important, 5=extremely impo rtant 35. The lightheadedness that results from smoking hooka h. 36. The stress relief that results from smoking hookah. 37. The social activity that results from smoking hooka h. 38. The ability to make new friends by smoking hookah. 39. Meeting people of the opposite sex by smoking hooka h. 40. The traditional importance of smoking hookah 41. The “high” I get from smoking hookah. 42. The headaches that result from smoking hookah 43. The shortness of breath that results from smoking h ookah. 44. The chest pains I experience the next day from smok ing hookah. 45. The ability to stay focused when smoking hookah.

PAGE 149

n Appendix 2 (continued) 46. The ability to pass time while smoking hookah. 47. The relaxed feeling from smoking hookah. 48. The ability to be athletic 49. That hookah is less harmful than cigarettes 50. The group dynamics 51. The “fun” that occurs from smoking hookah 52. The bad health effects from hookah [Attitude Questions ] Response options on 7 pt scale [ Extremely, quite, slightly, neither] Please read the following four statements and selec t an option that best describes your attitude regarding hookah smoking. 53. For me, smoking hookah results in a behav ior that is: __Extremely Bad: __Quite Bad:___Slightly Bad:___Nei ther bad nor good:___Slightly good:___Quite Good:___Extemely Good 54. For me, smoking hookah is: [Pleasant : unpleasant] 55. For me, smoking hookah is: [Nice: awful] 56. For me, smoking hookah is: [A lot of fun: not fun] [Questions about Important Others] 57. If I smoke hookah, my parents would: Approve: disapprove with choice options extremely; quite; slightly; neither on each side of the scale. 58. If I smoke hookah, my friends would: Approve; disapprove with choice options extremely; quite; slightly; neither on each side of the scale. 59. If I smoke hookah, my boyfriend/girlfriend would: Approve; disapprove with choice options extremely; quite; slightly; neither on each side of the scale. Question about Motivation to comply with Important others 60. Most of the time, when my parents think I should do something, I go along with it. Agree/Disagree with choice options extremely; quite ; slightly; neither on each side of the scale.

PAGE 150

Appendix 2 (continued) 61. Most of the time, when my friends think I should do something, I go along with it. Agree/Disagree with choice options extremely; quite ; slightly; neither on each side of the scale. 62. Most of the time, when my boyfriend/girlfriend thin ks I should do something, I go along with it. Agree/Disagree with choice options extremely; quite ; slightly; neither on each side of the scale. [Question about subjective norms] 63. If I smoke hookah, most of the people who are impo rtant to me would: Disapprove to Approve with choice options extremely ; quite; slightly; neither on each side of the scale. [Question about cigarette use] 64. Within the past 30 days, on how many days did you s moke regular cigarettes, even a puff? [] None [] 1-2 days [] 3-5 days [] 6-10 days [] 11-20 days [] 21-30 days [ Demographic Questions] 65. In what year were you born? Drop down choice 66. What is your sex? [] Male 2 [] Female 1 67. How do you usually describe your race? Please mark that which applies the best. [] White [] Black []Asian/Pacific Islander []American Indian/Alaskan Native []Other______________

PAGE 151

Appendix 2 (continued) 68. Are you of Hispanic/Latino origin? [] Yes [] No 69. In what country were you born? __________ (open end ed) 70. Are you a U.S. citizen? [] Yes 1 [] No 2 71. What is your religion? [] No chosen religion [] Atheist/Agnostic [] Catholic [] Protestant [] Other Christian [] Jewish [] Muslim [] Buddhist [] Other _________________ 72. Have you noticed any flyers around campus with info rmation about the health effects of hookah smoking? [] Yes 1 [] No 2 73. Please enter your email address for your free song download information to be sent.

PAGE 152

Appendix 3: Survey Email Dear USF Students (Smokers and NONsmokers): You have been randomly selected to take an importan t online survey being conducted by Mary Martinasek, a graduate student in the College of Public Health at the University of South Florida. This survey seeks information regar ding how you feel about hookah tobacco smoking (waterpipe smoking/narghile smoking ) and asks you about your perception. This survey pertains to tobacco smokin g with a waterpipe NOT smoking marijuana or other substances. Please read the following important detailed inform ation before beginning the survey What will be done: Participate in an online survey that will take less than 10 minutes to complete. Benefits of this study: By taking this survey, you will be helping me under stand some of the reasons why college students like you decide to smoke (or not s moke) hookah. For agreeing to participate in this survey, you will receive a Free Song Download and a chance to win one of ten $50 Amazon gift cards To prevent multiple responses from the same respondent, a unique identification number (I.D.) h as been assigned to your email address. Risks or discomforts : No risks or discomforts are anticipated from taking part in this study. If you feel uncomfortable with any question, you can skip that question or withdraw from the study altogether. If you decide to stop before you have c ompleted the survey, your responses will not be discarded. Confidentiality: Your responses will be confidential. I will not know your IP address when you respond to the Internet survey. I will ask you to provide a n email address when you complete the survey so that I can send you the free song downloa d. Only I will see your individual survey responses. After I have finished data collec tion and sent out the song download information, I will destroy all participants’ email addresses from my files.

PAGE 153

Appendix 3 (continued) Decision to stop taking the survey at any time: Your participation is voluntary; you are free to wi thdraw your participation from this study at any time. If you do not want to continue, you can simply exit the website. How the findings will be used: The results of the study will be used for scholarly purposes only. The results from the study will be presented in educational settings and at professional conferences, and the results might be published in a professional journa l in the field of public health. Neither your name nor any identifying information about you will be included in any of these reports. The survey results will be used is to prov ide health and safety information to the USF student community. Contact information: If you have concerns or questions about this study, please contact me -Mary Martinasek at (813) 493-4546. This research study has been app roved by the Institutional Review Board of USF IRB # 108637. The distribution of this survey has been approved in accordance with the USF system Policy Number 0-520. If you have any IRB questions, you can contact the IRB at (813) 974-5741. By beginning the survey, you acknowledge that you h ave read this information and agree to participate in this research. This survey may be accessed with the link below: x?surveyid=5496

xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2011 flu ob 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004948
XX9999 (Online)
1 100
Martinasek, Mary Pautler.
0 245
Understanding the psychosocial aspects of waterpipe smoking among college students
h [electronic resource] /
by Mary Pautler Martinasek.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 153 pages.
Includes vita.
(Ph.D.)--University of South Florida, 2011.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
ABSTRACT: Waterpipe tobacco smoking has migrated from being a custom of some cultures to becoming a staple around college campuses. The social nature and flavored tobacco encourage initiation in this tobacco nave age group. The study was a sequential mixed method design, employing primary data collection and analysis of a random sample of university students who live on campus at a single university. The study involved observations (N=6), intercept interviews with smokers and nonsmokers (N=63), three focus groups (N=31), and an online survey (N=288). Findings were centered on the constructs of the Theory of Reasoned Action by Ajzen and Fishbein (1975), who propose there are specific precursors that lead to intention to perform a behavior. The data suggested that attitudes were more positively correlated with intention to smoke waterpipe tobacco than subjective norm. Attitude is influenced by outcome expectancies. The data suggest that positive outcome expectancies are influenced primarily by the social nature of hookah smoking and other perceived positive benefits. Negative health effects and family/culture were also influential, albeit to a lesser extent. Development of interventions focused on an attitudinal shift may help to decrease uptake and continuation of waterpipe tobacco smoking in this population.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Bryant McDermott, Carol Robert .
Bryant McDermott, Carol Robert .
Dissertations, Academic
x Health Sciences
t USF Electronic Theses and Dissertations.
4 856