Citation
Health Information Seeking on the Internet by College Men in Latino Fraternities

Material Information

Title:
Health Information Seeking on the Internet by College Men in Latino Fraternities
Creator:
Chavarria, Enmanuel A
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (189 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Health and Human Performance
Health Education and Behavior
Committee Chair:
CHANEY,ELIZABETH H
Committee Co-Chair:
STELLEFSON,MICHAEL L
Committee Members:
CHANEY,JERRY DON,JR
DODD,VIRGINIA JONES
Graduation Date:
5/3/2014

Subjects

Subjects / Keywords:
College students ( jstor )
Focus groups ( jstor )
Fraternities ( jstor )
Health care services ( jstor )
Health information ( jstor )
Hispanics ( jstor )
Information search ( jstor )
Internet ( jstor )
Online searching ( jstor )
Questionnaires ( jstor )
Health Education and Behavior -- Dissertations, Academic -- UF
college -- fraternities -- health -- latino
City of Miami ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Health and Human Performance thesis, Ph.D.

Notes

Abstract:
Latinos are now the fastest growing and largest ethnic group in the U.S. According to U.S. Census Bureau estimates, 1 in 3 Americans will be of Latino descent by the year 2050. Despite this shift in the demographics of the US, we know relatively little about the Latino population's health profile and their health status. Recent research indicates that 66% of Latino Internet users look online for health information. This mixed-methods study sought to provide further insight on health information seeking on the Internet by college men in Latino fraternities. Data analysis procedures in this study included descriptive statistics, multiple regression, and qualitative thematic analyses of focus group transcripts and think-aloud protocol interviews; to investigate the variables attributes of online health information (source, message, and content); frequency of online health information seeking, level of social support, self-efficacy to engage in health behavior and seek out healthcare services among college men in Latino fraternities. Qualitative findings indicated that college men in Latino fraternities are mostly searching for online health information for persons other than themselves, such as family members. Findings show that this unique sub-group did not self-report searching for online health information on safe sexual practices, or sexually transmitted diseases. Concerns related to monetary income were associated with an increase in self-reported self-diagnoses and self-treatment. Quantitative findings found that Frequency of online use for health information seeking (Several Times a Week), Source (Somewhat Non-Influential), Content (Somewhat Non-Influential), and Age (18-20) were statistically significant predictors of self-efficacy to engage in health behavior. Message (Neither Influential nor Non-Influential), and Age (21-23) were found to be statistically significant predictors of self-efficacy to seek out health care services. Taken together, this is the first study to explore online health information seeking among college men in Latino fraternities. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: CHANEY,ELIZABETH H.
Local:
Co-adviser: STELLEFSON,MICHAEL L.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-05-31
Statement of Responsibility:
by Enmanuel A Chavarria.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
5/31/2015
Resource Identifier:
907379243 ( OCLC )
Classification:
LD1780 2014 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

1 HEALTH INFORMATION SE EKING ON THE INTERNET BY COLLEGE MEN IN LATINO FRATERNITIES By ENMANUEL ANTONIO CHAVARRIA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQ UIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2014

PAGE 2

2 2014 Enmanuel Antonio Chavarria

PAGE 3

3 To God, my w ife, and f amily

PAGE 4

4 ACKNOWLEDGMENTS This dissertation would not have been completed successfully without the support, en couragement, and guidance of many people. I especially wish to extend a special tha nks to my committee chair, Dr. Elizab eth H. Chaney, for her constant support, optimistic spirit, and her exemplary leadership t hrough the often confusing and disorienting se ries of hoops that occurred during my doctoral program and throughout my dissertation research. I would like to thank Dr. J Don Chaney for his continued encourage ment and guidance since day one. I would like to thank him for the opportunities he has shar ed with m e to expand my academic skills, and always making sure my ultra technological ideas and presentations would seamlessly come to live. I would like to thank Dr. Michael L. Stel lefson for his words of wisdom and for providing me with endless opportun ities to expand and perfect my skills set. His attention to details and work ethic are definitely a model of professionalism to follow. I would like to thank Dr. Virginia Dodd for her valuable feedback, her caring nature and always making sure each commit tee m eeting included a bit of humor to ease any nervousness. I would like to also thank Dr. Hui Bian, for her suggestions on statistical methods to pursue to allow for better interpretation of results. A special thanks to my parents, Manuel Chavarria and Luisa Chavarria for their continued support, encouragement and prayers. Their resilient and perseverant nature has definitely been inherited. I would like to also thank Dr. Rev. Raulston Nembhard and Mrs. Heather Nembhard for their continued prayers and su pport. Finally, I would like to thank my wife, Dr. Nikita Chavarria, for her endless encouragement through the roughest of times, for her endless patience, and endless

PAGE 5

5 support through any and all obstacles. No matter how dark a day might be h er smiles, gi ggles and hugs always e nsure that any day ends bright.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 10 LIST OF FIGURES ................................ ................................ ................................ ........ 11 LIST OF OBJECTS ................................ ................................ ................................ ....... 12 LIST OF ABBREVIATIONS ................................ ................................ ........................... 13 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 16 Background ................................ ................................ ................................ ............. 16 Statement of the Problem ................................ ................................ ....................... 23 Conceptual Underpinnings for the Study ................................ ................................ 24 Purpose of the Study ................................ ................................ .............................. 25 Believed Interrelationships ................................ ................................ ............... 25 Research Questions ................................ ................................ ......................... 26 Key Variables ................................ ................................ ................................ ... 26 Independent variables ................................ ................................ ................ 26 Dependent variables ................................ ................................ .................. 27 Definition of Key Terms ................................ ................................ ........................... 27 Limitations, Assumptions, and Design Controls ................................ ...................... 28 Delimitations of the Study ................................ ................................ ....................... 29 Significance of the Study ................................ ................................ ........................ 29 Summary ................................ ................................ ................................ ................ 30 2 REVIEW OF RELATED LITERATURE ................................ ................................ ... 32 Introduction ................................ ................................ ................................ ............. 32 Literature Review Methods ................................ ................................ ..................... 32 Search Procedures ................................ ................................ ........................... 32 Selection Criteria ................................ ................................ .............................. 33 Results ................................ ................................ ................................ .................... 34 Discussion ................................ ................................ ................................ .............. 35 Health Care Utilization and Seeking by Latino College Males .......................... 36 Health Behavior Management Exhibited by Latino College Males ................... 36 OHI Seeking Amongst Latino College Males ................................ .................... 36 Self Efficacy and Health Pertaining to Latino College Males ............................ 37 Issues of Measurem ent ................................ ................................ .................... 37

PAGE 7

7 Conceptual Underpinnings of Previous Research ................................ ............ 37 Self Efficacy Theory ................................ ................................ ......................... 38 Summary ................................ ................................ ................................ ................ 41 3 RESEARCH DESIGN AND METHODOLOGY ................................ ....................... 44 Introduction ................................ ................................ ................................ ............. 44 Problem and Purpose Overview ................................ ................................ ............. 44 Research Questions ................................ ................................ ............................... 45 Setting and Population ................................ ................................ ............................ 45 Setting ................................ ................................ ................................ .............. 45 Participants ................................ ................................ ................................ ....... 48 Data Collection Procedures ................................ ................................ .................... 48 Focus Group Sessions ................................ ................................ ..................... 48 Questionnaire Pre testing via Cognitive Interviews ................................ .......... 56 Survey Implementation ................................ ................................ ..................... 61 Instrumentation ................................ ................................ ................................ ....... 62 Extent of Online Use ................................ ................................ ........................ 63 Types and Factors ................................ ................................ ............................ 64 Access of information ................................ ................................ ....................... 66 Level of Influence (Source, Message, and Content) of Accessed OHI ............. 67 Self Efficacy to Seek Healthcare Services and Engage in Health behavior ..... 68 Duke Social Support Index (DSSI) and Social Support ................................ .... 70 Demographic Items ................................ ................................ .......................... 71 Reliability and Validity of Study ................................ ................................ ............... 72 Reliability for Quantitative Aspects ................................ ................................ ... 73 Validity for Quantitative Aspects ................................ ................................ ....... 73 Credibility for Qualitative Aspects ................................ ................................ ..... 74 Transferability for Qualitative Aspects ................................ .............................. 74 Dependability and Confirmability, of Qualitative Aspects ................................ 75 s Subjectivity Statement ................................ ................................ ...... 76 Who I am in Relation to the Research ................................ .............................. 76 My Experience, Training, and Theoretical Perspectives ................................ ... 76 Personal Bias and how it was Addressed ................................ ......................... 76 Data Analysis ................................ ................................ ................................ .......... 77 Particip ant characteristics ................................ ................................ ................. 77 Think aloud protocol interviews ................................ ................................ ........ 78 RQ1: How often do CMLF use of the Internet to locate health informatio n? .... 79 RQ2: What types of health information do CMLF look for online? .................... 80 RQ3: What are factors associated with OHI seeking among CMLF? ............... 81 RQ4: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social support, age, and education predict self efficacy to engage in health behavio r among CMLF? ................................ ... 82 RQ5: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social support, age, and education predict self efficacy to seek out health care services among CMLF? .............................. 89 Apriori Analysis ................................ ................................ ................................ 90

PAGE 8

8 Summary ................................ ................................ ................................ ................ 91 4 FI NDINGS ................................ ................................ ................................ ............... 95 Organization of Findings ................................ ................................ ......................... 95 Participant Characteristics ................................ ................................ ...................... 95 Participant Characteristics for Focus Groups ................................ ................... 95 Participant Characteristics for Think aloud Protocols ................................ ....... 96 Participant Charac teristics for Survey Respondents ................................ ........ 97 Think aloud Protocol Interviews ................................ ................................ .............. 98 Errors in the Questionnaire ................................ ................................ ............... 99 Misunderstanding or Misrepresentation of Items ................................ .............. 99 Suggestions for Improvements ................................ ................................ ....... 101 Interna l Consistency Reliability Analysis ................................ ............................... 102 Research Questions ................................ ................................ ............................. 103 RQ1: How often do CMLF use of the Internet to locate health inform ation? .. 103 RQ2: What types of health information do CMLF look for online? .................. 103 Informational needs of others ................................ ................................ ... 1 04 Symptoms ................................ ................................ ................................ 105 Diagnoses ................................ ................................ ................................ 106 Physical appearance and weight concerns ................................ .............. 107 Natural living ................................ ................................ ............................ 107 Treatments for conditions or diseases ................................ ..................... 108 Monetary iss ues ................................ ................................ ....................... 109 Immigration ................................ ................................ .............................. 109 Home remedies ................................ ................................ ........................ 109 Curiosity and clarific ation through social media ................................ ....... 110 Quantitative findings of types of OHI sought ................................ ............ 110 RQ3: What are factors associated with OHI seekin g among CMLF? ............. 112 Family and loved ones ................................ ................................ ............. 112 Monetary issues ................................ ................................ ....................... 113 Fear of condition, disease or disorder ................................ ...................... 114 Physical appearance ................................ ................................ ................ 115 Quantitative findings of factors relating to OHI seeking amon g participants. ................................ ................................ .......................... 116 RQ4: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social support, age, and education predict self efficacy to engage in hea lth behavior among CMLF? ................................ 118 RQ5: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social support, age, and education predict self efficacy to seek out health care services among CMLF? ............................ 120 Summary ................................ ................................ ................................ .............. 122 5 DISCUSSION AND CONCLUSION ................................ ................................ ...... 135 Introduction ................................ ................................ ................................ ........... 135 Discussion of Findings ................................ ................................ .......................... 135

PAGE 9

9 Demographics ................................ ................................ ................................ 135 Pretesting and Internal Consistency Reliability Analysis ................................ 136 Frequency of Internet Use to Locate Health Information ................................ 137 Types of Searched OHI ................................ ................................ .................. 138 Factors Influencing OHI Seeking ................................ ................................ .... 138 Self Efficacy to Engage in Health behavior ................................ .................... 139 Self Efficacy to Seek Out Health Care Services ................................ ............. 141 Implications ................................ ................................ ................................ ........... 142 Theo retical Implications ................................ ................................ .................. 142 Implications for Health Educators ................................ ................................ ... 143 Recommendations for Future Research ................................ ............................... 143 Limitations of the Study ................................ ................................ ......................... 145 Conclusion ................................ ................................ ................................ ............ 146 APPENDIX A RECRUITMENT EMAIL FOR UNIVERISTY OF FLORIDA AND FLORIDA INTERNATIONAL UNIVERSITY STUDENTS (FOCUS GROUP) ........................ 148 B INFORMED CONSENT FORM FOR FOCUS GROUP PARTICIPANTS .............. 149 C ................................ ........................... 151 D RECRUITMENT EMAIL FOR UNIVERSITY OF MIAMI STUDENTS (THINK ALOUD PROTOCOL INTERVIEW) ................................ ................................ ...... 153 E INFORMED CONSENT FOR THINK ALOUD PROTOCOL PARTICIPANTS ....... 154 F INTERVIEWER QUESTION GUIDE FOR THINK ALOUD PROTOCOLS ............ 156 G RECR UITMENT EMAIL FOR STUDENTS OF BARRY UNIVERSITY, EMBRY RIDDLE AERONAUTICAL UNIVERSITY, FLORIDA ATLANTIC UNIVERSITY, FLORIDA STATE UNIVERSITY, NOVA SOUTHEASTERN UNIVERSITY, UNIVERSITY OF CENTRAL FLORIDA, AND THE UNIVERSITY OF SOUTH FLORIDA ................................ ................................ ................................ .............. 157 H INFORMED CONSENT FORM FOR SURVEY PARTICIPANTS ......................... 158 I AD HOC QUESTIONNAIRE ................................ ................................ ................. 160 LIST OF REFERENCES ................................ ................................ ............................. 171 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 189

PAGE 10

10 LIST OF TABLES Table page 1 1 Indep endent and dependent variables of study. ................................ ................. 31 3 1 List of college campuses investigated in Florida, location, student population and percentage of Latino students. ................................ ................................ .... 92 3 2 List of Lambda Theta Phi Chapters in Florida and their related sector affiliation. ................................ ................................ ................................ ............ 92 3 3 Summary of data analyses used in answering each research q uestion. ............ 93 4 1 Focus group demographic characteristics. ................................ ....................... 123 4 2 Think s. ...................... 124 4 3 ................................ .......... 125 4 4 Categories, codes and associated definitions of think alou d protocol analysis. 127 4 5 Internal Consistency Reliability Analysis ................................ .......................... 127 4 6 Extent of online use for health information see king. ................................ ......... 127 4 7 Device most used when accessing the internet. ................................ ............... 128 4 8 Place respondent mostly find themselves while searching on line. ................... 128 4 9 Frequency and percentage of types of online health sought by participants. ... 129 4 10 Factors that influence OHI s eeking. ................................ ................................ .. 130 4 11 Resources visited for health information. ................................ .......................... 132 4 12 Resource most frequently visited by participants. ................................ ............. 132 4 13 Multiple linear regression predicting self efficacy to engage in health behavior. ................................ ................................ ................................ ........... 133 4 14 Multiple linear regression predict ing self efficacy to seek health services. ....... 134

PAGE 11

11 LIST OF FIGURES Figure page 2 1 Stem tree illustrating manuscript selection process using various s earch databases and combinations of controlled vocabulary. ................................ ...... 43

PAGE 12

12 LIST OF OBJECTS Object page 4 1 Definition of codes (.pdf file 436 KB) ................................ ................................ 103 4 2 Codes with Associated Quotations (.pdf file 349 KB) ................................ ....... 104 4 3 Themes with Associated Codes (.pdf file 57.3 KB) ................................ ........... 104

PAGE 13

13 LIST OF ABBREVIATION S C MLF The abbreviated term for college men in Latino fraternities. D SSI Duke Social Support Index is a scale that measures perceived social support. The scale has been used previously in a myriad of study populations. The scales has been empirically shown to be both a reliable and valid method of assessing social support. O HI The abbreviated term for online health information. N ALFO The National Association of Latino Fraternal Organizations, Inc., (N ALFO) was established in 1998 to promote the advancement of Latino/a fraternities and sororities. Today, NALFO's 20 fraternities and sororities are bound by a shared commitment to fraternal unity, family values and empowering Latino and underserved communi ties. NALFO's purpose is to promote and foster positive inter fraternal relations, communication, and development of Latino/a Fraternal organizations through mutual respect, leadership, honesty, professionalism and education.

PAGE 14

14 Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy HEALTH INFORMATION SE EKING ON THE INTERNET BY COLLEGE MEN IN LATINO FRATERNITIES By Enmanuel An tonio Chavarria May 2014 Chair: Elizabeth Hensleigh Chaney Major: Health and Human Performance Latinos are now the fastest growing and largest ethnic group in the U.S A ccording to U.S. Census Bureau estimates, 1 in 3 Americans will be of Latino descent by the year 2050. Despite this shift in the demographics of the US we know relatively little about rofile and their health status. Recent research indicates that 66% of Latino Internet users look onli ne for health informat ion T h is mixed methods study sought to provide further insight on health information seeking on the Internet by college men in Latino fraternities D ata analysis procedures in this study included descriptive statistics, multiple regression, and qualitati ve thematic analyses of focus group transcripts and think aloud protocol interviews ; to inves tigate the variables attributes of online health information (source, message, and content); frequency of online health information seeking, l evel of social suppor t, self efficacy to engage in health behavior and seek out healthcare services among college men in Latino fraternities. Qualitative findings indicated that college men in Latino fraternities are mostly searching for online health information for persons o ther than themselves such as family members. Findings show that t his unique sub group did not self report searching for online health information on safe sex ual practices, or sexually transmitted

PAGE 15

15 diseases. Concerns related to monetary income w ere associat ed with an increase in self reported self diagnoses and self treatment. Quantitative findings found that Frequency of online use for health information seeking (Several Times a Week), Source (Somewhat Non Influential), Content (Somewhat Non Influential), a nd Age (18 20) were statistically significant predictors of self efficacy to engage in health behavior Mess age (Neither Influential nor Non Influential) and Age (21 23) were found to be statistically significant predictors of self efficacy to seek out he alth care services. Taken together, this is the first study to explore online health information seeking among college men in Latino fraternities.

PAGE 16

16 CHAPTER 1 INTRODUCTION Background A ccording to the U.S. Census Bureau the Latino population is expected to nearly triple in the next 35 years, from 46.7 million in 2008 to 132.8 million by 2050 ( U.S. Census Bureau, 2008 ). If these projections hold true the share of the Latino population will increase from 15% to 30% and one in three U.S. residents will be o f Latino origin ( U.S. Census Bureau, 2008 ). Moreover, Latinos are now the fastest growing and largest ethnic group in the U.S. ( Martin & Fogel, 2006 ). T rends now show that an increasing birth rate rather than immigration infl ow is the leading growth engi ne for the Latino population in the U.S. ( Martinez & Ariosto, 2011 ). If the population of Latinos in the United States were considered a country, Latinos would constitute the second largest Spanish speaking country in the world (National Museum of the Amer ican Latino Commission [NMAL], 2011). Latinos in the United States make up 1 out of every 6 people, and 1 out of every 4 babies born each year (NMAL, 2011). Today, Latinos are 50 million strong and growing by 1 million every year (NMAL, 2011). Despite the growth in the Latino population within the U.S., we know very little about the Latino health profile and their health status ( LaVange et al. 2010; Sorlie et al. 2010 ). Unfortunately, Latinos have been underrepresented in research studies over the past two decades ( LaVange et al., 2010; Sorlie et al., 2010 ). Lack of Latino inclusion in research studies have led to an overwhelming number of gaps in knowledge about Latino health ( Harris et al., 2003; Sorlie et al., 2010 ). These gaps includ e knowledge gaps in Hispanics and mental health (

PAGE 17

17 dental health profile ( Crozier, 2011 ), and research in chronic disease s such as tuberculosis ( Boulter, Moran, & Cole, 2011 ) to name a few. It is important for re searchers to begin to incorporate Latinos in national health studies to fill these existing knowledge gaps. In particular, there is a need to differentiate whether difference s exist pertaining to prevalence of conditions, risk factors, diseases and disorde rs among subgroups of Latino populations ( De la Rosa, 1989; Betancourt, Carillo, Green & Maina, 2004; LaVange et al., 2010; Sorlie et al., 2010 ). In 2006, the National Institutes of Health (NIH) led the Hispanic Community Health Study, Study of Latinos (SO L), the largest longitudinal epidemiological study of health and disease in Latino populations living in the United States ( LaVange et al., 2010 ). R esults from th is study are still several years away from being made public yet bridging knowledge gaps in Latino health has already commenced. R esearch by the U.S. Centers for Disease Control and Prevention (CDC) has shown that Latinos are twice as likely as non Latino Blacks and three times more likely than non Latino Whites to be without aprimary health ca re provider ( Pleis & Lethbridge ejku 2 007 ). The Pew Hispanic Center/Robert Wood Johnson Foundation Latino Health survey found that younger, less educated Latino men without health insurance are least likely to have a usual health care provider ( Livings ton, Minushkin & Cohn, 2008 ). Interestingly, however, 50 % of Latinos without a primary healthcare provider are high school graduates with health insurance born in the United States ( Livingston, Minushkin & Cohn, 2008 ). T he primary reason survey respondent s gave for not having a regular health care provider was not related to the cost of health care but rather that they seldom fe e l sick ( Livingston, Minushkin & Cohn, 2008 ). A bout seven in ten Latinos

PAGE 18

18 (71 %) surveyed report ed obtaining health information th rough their social networks including family, friends, churches, and community groups. Over eighty percent (83%) indicate d that they obtain health information from some branch of the media, including the Internet ( Livingston, Minushkin & Cohn, 2008 ). The survey also found that over 3/4 ths of survey respondents (79%) acted on the health information obtained from the media ( Livingston, Minushkin & Cohn, 2008 ). H ealth information obtained by Latinos via various media sources may be inaccurate, however, which can lead to acting on false or misleading health information ( Spraggins, 2009 ). S urvey results highlighted the power and potential of alternative sources of media to disseminate health information to disparate segments of the Latino population ( Livingston Minushkin & Cohn, 2008 ). Pe a Purcell (2008) found that health information obtained online by Latinos led to increased understanding of medical conditions and treatments as well as increased confidence for engag ing in conversation with physicians about health concerns during medical visits. Hence, obtaining health information from media sources, such as the I nternet, can have several benefits but if false or misleading health information online is found, such can lead to harmful implications. According to Young (2001) because of low health literacy that may exist among the general Hispanic population, Latinos are at greater risk to obtain harmful or misinterpret health information on the Internet. Undoubtedly, empirical research is needed to determine whether health information obtained online by Latinos might be associated with use of available health resources, engage in healthy behaviors, health behavior and self efficacy to seek out health care services

PAGE 19

19 Today, digital media play s a critical role in consumer health, led by the Internet as the primary medium for dissemination of health information (Atkinson & Gold, 2002; Bush, Bowen, Wooldridge, Ludwig, Meischke & Robbins, 2004; Madden & Fox, 2006). Notwithstanding concerns related to the quality of online health information (OHI) i t is estimated that more than 113 million American adults access the Internet annually and are influenced by nearly 70,000 health related websites (Pagliari, Sloan, Gregor, Sullivan, Detmer, Kahan, Oortwijn & MacGillivray 2005; Fox, 2006). In a recent study conducted by the Pew Internet and American Life Project, it was estimated that e ight in ten I nternet users look online for health information (Fox, 2011). Other findings from the Pew Internet and American Life Project include: Individuals the Internet than The likelihood of searching for health information on the Internet increased with more education; and There was no difference s in searching for health information online across Blacks, Whites, and Latinos nearly 60% of each population group searched for health information on the Internet ( Fox, 2006 ). Given that Latinos are searching for health in formation online at a rate proportion al to other population groups, it is important to explore reasons why underrepresented Latinos are searching the Internet for health information at relatively high rates, and whether the pursuit of OHI by Latinos is ass ociated with relevant health outcomes. R esearch examining the effects of OHI search es and its implications on the general public is not a new area of inquiry (Burton, 2005; Easaw, 2010; Escoffery, Miner, Adame, Butler, McCormick, & Mendell, 2005 ; Hanik & S tellefson, 2011; Kim, Park & Bozeman, 2011; Miller & Bell, 2011;

PAGE 20

20 Norman & Skinner, 2006; Nustad, Adams & Moore, 2008; Pea Purcell, 2008; Zuckerman, 2009). In fact, OHI seeking has been researched since the turn of the century (Morahan Martin & Anderson, 2000). However, this research often underrepresent s Latino college student s due mostly to the particular demographics of the schools where the studies were conducted ( Easaw, 2010; Escoffery et al., 2005; Hanik & Stellefs Norman & Skinner, 2006 ). Therefore, research investigating OHI seeking among Latino college students is critical to understand whether this sup population is able to seek, find, understand, and appraise healt h information from electronic sources and apply the knowledge gained to addressing or solving a health problem ( Hanik & Stellefson, 2011 ; & Casey, 2006; Morahan Martin & Anderson, 2000; Norman & Skinner, 2006; Spraggins, 2009 ). This co nstruct is known as eHealth literacy A study by Cline and Haynes (2001) found that little research had been conducted on the effects of seeking hea lth information on the Internet among diverse populations. A review of the literature indicates that studies looking at the variables attributes of OHI (source, message, and content); frequency of OHI seeking, l evel of social support, self efficacy to engage in health behavior and self efficacy to seek out healthcare services in conjuncture are yet to be publish ed. Until as recently as 2007 studies investigating college student health had been nearly non existent Studies published prior to 2007 focused mostly on alcohol and tobacco use and were limited to individu al four year institutions (Lust et al., 2007) The University of Minnesota system wide student health report published by Lust and

PAGE 21

21 colleagues (2007) was the first of its kind to investigate the comprehensive health of college students by examining health risk behaviors other than alcohol, tobacco and d rug use. Even though this study represent ed a step in the right direction for illuminating the health needs of an often overlooked segment of the general population (college students) the percentage of Latino/Hispanic students surveyed was less than 2% ac ross all campuses included in the study (Lust et al., 2007) The underrepresentation of Latinos in college health research is necessary to help understand the comprehensive health of Latinos in college (CDC, 2011; Harris et al., 2003 ). With the growth of t he Latino population, more Latinos are enrolling in colleges nationwide (Fry, 2012). Latinos age d 18 24 now outnumber African American students of the same age on college campuses across the US A ccording U.S. Cen sus Bureau data (analyzed by the Pew Founda tion) n early 350,000 more Latinos were enrolled in college in 2010 than in 2009 B y the year 2015, Latino enrollment in college is expected to increase by 73 % more than 3 times the rate of African Americans and 15 times the rat e of non Latino Whites (Fry, 2012; Menendez, n.d. ). Today, an un expected disparity ex ists where Latinos constitute 18.3 % of the traditional college aged population, yet they make up only 11 % of total enroll ees in colleges and universities across the US (Fr y, 2012; Menendez, n.d.). This means that even with record breaking enrollment rates, there is still room for improvement in the number of Latinos enrolling in collge. Accordingly Latinos earn fewer bachelor degrees than other ethnic and racial groups. On ly 13 % of Latino s 25 to 29 years of age earn a % of Asians, 39 % of non Latino W hites and 19 % of African America ns in the same age group ( Thus, there is

PAGE 22

22 also room for improvement concerning Latin os completing college and not merely enrolling in college. Of interest h owever is the fact that m ore Latino college enrollees have led to more Latino males becoming members of fraternal organizations than ever before (Fry, 2012; Helem, 2004; Miranda, 1999 ). Whether higher membership of Latinos in fraternal organizations have led to improved college completion among college men in fraternities remains to be investigated. Misguided s ocietal preconceptions have led researchers to overlook and often ignore th e health needs of Latino subpopulations with higher levels of SES due to the belief that lower SES Latino populations have greater health needs ( V ega, Rodriguez, & Gruskin, 2009). Yet, little research supports the aforesaid claim ( Goldman, Kimbro, Turra & Pebley, 2006 ). In fact, from the limited available research a paradox seems evident F or several Latino sub populations the opposite holds true ; lower SES Latino sub populations seem to enjoy lower mortality rates than other segments of the population ( Vega, Rodriguez, & Gruskin, 2009; Wei et a l., 1996 ). R esearch suggests that overall mortality advantages are observed more frequently in Latinos with low SES, with little or no health advantage s existing for Latinos at higher SES levels (Vega, Rodriguez & Gruskin, 2009; Wei et al., 1996). Clearly, the role of SES in health and mortality among Latinos is complex both within and across populations in the Americas (Goldman, Kimbro, Turra & Pebley, 2006; Vega, Rodriguez, & Gruskin, 2009; Wei et al., 1996). Historically, research has associated wealth with membership in a fraternal organization (Levine & membership in a fraternal organization is dynamically more complex due to the myriad of college student

PAGE 23

23 subpopulations e ntering college (Gohn & Albin, 2006) Thus it is not uncommon for members of fraternal organizations to be from high and low income backgrounds (Gohn & Albin, 2006). T he literature describing any aspect of health for Latino college males, who are also me mbers of a fraternal organization, is nearly non existent J ust a handful of studies have been conducted in which Latino college men comprise a small p ercentage of cohorts of primarily White or African American fraternal organization s ( Guardia & Evans, 2 008 ; Scott Sheldon, Carey & Carey, 2008 ). Most of the health studies that do mention Latino members of larger fraternal organizations describe either alcohol consumption and its implications (Barry, 2007; Borsari & Carey, 1 999 ; Cashin, Presley & Meilman, 19 98; Kuh & Arnold, 1993; Larime r, Anderson, Baer & Marlatt, 2000; Larimer, Irvine, Kilmer & Marlatt, 1997; Lo & Globetti, 1995 ; Workman, 2001) or look to describe sexual be haviors and risk factors among f raternal organizations at large ( Foubert, 2000; Guar dia & Evans, 2008; Scott Sheldon, Carey & Carey, 2008 ). O ver 2 million Latino students entered college in 2011, and more and more of these students are joining Greek organizations on campus ( Helem, 2004; Miranda, 1999 ). The National Association of Latino Fraternal Organizations Inc. (NALFO) estimates that as man y as 4 0,000 Latino students and graduates are Greeks, which represents a fourfold increase since the mid '90s ( Helem, 2004 ). S tudies focusing exclusively on Latino Greek Letter Organizations will h elp to bridge the knowledge gap that exists describing the health of this unique sub population Statement of the Problem Latinos are the fastest growing segment of the US po pulation ( Nielsen, 2013; U.S. Census Bureau, 2008). Although Latinos may soon bec ome the largest minority

PAGE 24

24 group in the United States, their inclusion in national health studies has historically been near ly non existent (LaVange et al., 2010; Sorlie et al., 2010). Even more understudied is the use of the I nternet to seek health informat ion by Latinos ( Pea Purcell 2008 ) More specifically, comprehensive health research on Latino college male members of fraternal organizations is scarcely available and limited to studies investigating alcohol use sexual health and drug abuse (Barry, 20 07; Borsari & Carey, 1999; Cashin, Presley & Meilman, 1998; Foubert, 2000; Guardia & Evans, 2008; Kuh & Arnold, 1993; Larimer, Anderson, Baer & Marlatt, 2000; Larimer, Irvine, Kilmer & Marlatt, 1997; Lo & Globetti, 1995; Scott Sheldon, Carey & Carey, 2008; Workman, 2001). Furthermore, many mistakenly believe that Latino males in college, who belong to a fraternal organization, also belong to a higher socioeconomic segment of the population, and as such, may experience fewer health concerns than other Latin os (Ingelmo, 2012) This misconception may lead to few investigation s of the comprehensive health of this particular segment of the population (Ponce & Comer, 2003). Therefore, this study aimed to bridge an informational gap concerning the use of OHI among fraternal Latino males in colleges/universities. Conceptual Underpinnings for the Study Previous studies on OHI seeking, looking at the variables of social support, fraternal membership, self efficacy to seek out health care services and self efficacy to engage in health behavior were guided by the Health Belief Model (HBM) and social constructs of HBM, as well as Social Cognitive Theory (SCT) ( Easaw, 2010; Eysenbach, 2008; Glanz, Rimer & Lewis, 2002; Kim, 2012; Lu, 2006; Tamim, 2012; Wright & Bell, 2003 ). However, Self efficacy theory presents the most applicable theory for the goals of this study (Bass, Ruzek, Gordon, Fleisher, McKeown Conn & Moore,

PAGE 25

25 2006; Oh & Lee, 2012; Smith McLallen, Fishbein, & Hornik, 2011 ) since it closely parallel s the aims of t he study at hand and is more applicable to answering the research questi ons posed in this investigation. Realizing the value that theory guided research Self efficacy Theory into certain aspects of questionnaire development and the efficacy Theory is highlighted in Chapter 2. Purpose of the Study The purpose of this study is to bridge existing knowledge gaps for college men in Latino fraternities (CMLF) by examining the frequency of health information searches the types of OHI searched for factors that influence OHI seeking among Latino men belonging to a fraternity and what predictor variables explain the most variance in self efficacy to engage in health behavior and self efficacy to seek out health care services. Believed Interrelationships The research literature suggests that individuals coping with stressful health challenges such as a recent illness diagnosis or chronic disease management were strongly motivated to engage in OHI seeking behaviors (Weaver et al., 2010). Research also suggests that increased social support can lead to increased health information seeking and in turn higher levels of sel f efficacy among African American women (Warren et al., 2010), but remains to be investigated among CMLF While evidence of relations among several variables pertaining to self efficacy to seek out health care services in relation to OHI seeking exists, It is unclear whether these relationships will be observed in Latino men belonging to a college/university fraternity.

PAGE 26

26 Research Questions T his study seeks to investigate the following research questions: 1. How often do college men in Latino fraternities use o f the Internet to locate health information? 2. What types of health information do college men in Latino fraternities look for online? 3. What are factors associated with online health information seeking among college men in Latino fraternities ? 4. To what ext ent do attributes of online health information (source, message, and content); frequency of online health information seeking, social support, age, and education predict self efficacy to engage in health behavior among college men in Latino fraternities ? 5. To what extent do attributes of online health information ( source, message, and content ) ; frequency of online health information seeking, social support, age, and education predict self efficacy to seek out health care services among college men in Latino fraternities ? Key Variables I t is of great interest and importance to bridg e the knowledge gap on Latino health related to OHI seeking to investigate how psychosocial variables such as self efficacy to seek health care services and self efficacy to eng age in health behavior may be associated with OHI seeking among Latino men attending a college/university T o best answer the 5 research question s posed in this study a mixed methods approach will be used with a formative qualitative phase to set the st ate for a quantitative study Ind ependent variables L evel of influence to take action based on source, message, and content of OHI (very influential, influential, somewhat influential, neither influential nor non influential, somewhat non influential, non influential, non influential, very non influential)

PAGE 27

27 Frequency of Internet use to locate health information (i.e. every day, several times a week, several times a month, every few months, or less often). L evel of social support as measured by the Duke So cial Support Index ( low, moderate, high ). Age (18 26). Education (Freshman, Sophomore, Junior, Senior Graduate, Professional ). The manner in which the independent variables are included within each research question and the level of scale is summarized i n Table 1 1. D ependent variables Frequency of Internet use to locate health information (i.e. every day, several times a week, several times a month, every few months, or less ofte n[than every few months] ). Ty p es of health information sought ( e.g. A speci fic disease or medical problem, A certain medical treatment or procedure, Nutrition, Exercise, Prescription drugs etc. ) Reasons for OHI seeking (i.e. my own health status, the health status of a family member, the health status of a friend the health st atus of a significant other, monetary issues, concerns over weight or physical appearance ) and their respective level of influence (i.e. very influential, influential, somewhat influential, neither influential nor non influential, somewhat non influential non influential, non influential, very non influential). S elf efficacy to engage in health behavior ( low confidence, moderate confidence high confidence ). S elf efficacy to see k out health care services ( low confidence, moderate confidence, high confiden ce ). The exact means by which each dependent variable is utilized for each research question, along with its level of scale can be found in Table 1 1. Definition of Key Terms For the purposes of this study several definitions of key terms have been provi ded below

PAGE 28

28 S OCIAL S UPPORT The presence of people who can offer a listening ear, companionship that leads to a sense of belonging, or material aid such as goods and services (Cohen, Mermelstein, Kamarck, & Hoberman, 1985). S ELF E FFICACY T he measure of one's own ability to complete tasks and reach goals (Ormrod, 2006) O ne's belief in one's ability to succeed in specific situations (Luszcynska, & Schwarzer, 2005). F RATERNAL O RGANIZATION The term fraternal organization is from the Latin frater, meaning brother. A fraternal organization is a brotherhood or a type of social organization whose members freely associate for a mutually beneficial purpose such as for social, professional or honorary principles ( Fraternal Organization, n.d. ). H EALTH B EHAVIOR A n action taken by a person to maintain, attain, or regain good health and to prevent illness. Health behavior ref lects a person's health beliefs ( Health Behavior, n.d. ). O NLINE H EALTH I NFORMATION S EEKING T he intentional, active efforts to obtain specific health information on the Internet, above and beyond the normal patterns of media exposure and use of interpersonal sources ( Miller & Bell, 2011 ). Limitations, Assumptions, and Design Controls As with any empirical endeavor this study is not without cert ain limitations Due to limitations in time and funding we are unable to survey and conduct focus groups with all 7 nationally recognized Latino fraternal organizations and their membership who belong to NALFO (National Association of Latino Fraternal Org anizations, Inc. [NALFO], n.d.). Also, it is important to understand that not all Latino Greek Letter Organizations are members of NALFO or represented by the association. Several assumptions are also taken into consideration in this study. It is assumed t hat members of the fraternal organization surveyed have at least some level of access to computers and I nternet either on campus or through personal means due to student requirements at higher education institutions. It is also assumed that, although mem bers are mostly of Latino origin all are fluent in the English langua ge. Hence, all focus group s and think aloud protocol interviews were conducted in the English language. The questionnaire was

PAGE 29

29 also written solely in the English language. Several measure s have also been taken to ensure robustness of reliability and validity for this study. Such measures include formative research to ensure validity of questionnaire items such as conducting think aloud protocols to pretest the questionnaire gu ide implemen tation steps, and highlight nuances overlooked during planning. Delimitations of the Study Participation in this study was delimited to a ) males b ) enrolled in Florida universities and colleges (listed in C hapter 3) c ) between the ages of 1 8 26 years and d ) members of Lambda Theta Phi Participants not meeting th ese criteria were excluded from the study. The study was delimited to examin ing the independent and dependent variables previously stated. Relations hips surfacing from variables other than th os e listed were not considered in data analyses Variables were only measured with assessment ins truments and procedures listed in Chapter 3 The respondents in this study agreed to voluntarily participate and may not be representative of those who chose not to participate. Significance of the Study The study is unique in that it focuses on a specific subgroup of Latinos. The study contributes to a better understanding of purposes for OHI seeking from actual health information seekers (Latino men belonging to a college/university fraternity) The findings of from this study will identify the type s of OHI sought by users F indings also highlight how type of health information sought might affect self efficacy to seek out health care services and self efficacy t o engage in health behavior This study provides important information to d esigners of consumer health information systems on the Web, and health behavior specialist s by providing important insights into the types of health

PAGE 30

30 information sought by Latino mal e consumers and how information sought may relate to health behaviors. Clearly, the study also provides important implications for consumers themselves and health providers by providing data on the extent to which OHI seeking is related to self efficacy fo r seek ing out health care services and engaging in health behavior Undoubtedly, the study ultimately serve s to shed light on motivators of health behaviors that can impr ove the health of Latino males. Summary Chapter 1 introduced the research topic by des cribing that Latinos are becoming the largest minority in the United States and yet, their health as a whole has not been thoroughly investigated. With I nternet users seeking health information online more than ever and with more Latinos in college beco ming members of a fraternal organization, it is important to bridge the knowledge gap between OHI seeking in Latino men who are college students and self efficacy to seek out health care services and engage in health behavior The literature review and th e research questions provided direction for the focus of this study. This dissertation is organized into five chapters. Chapter 1 present ed the background of the study and t he research problem. Chapter 2 annotate s the findings of relevant literature studyi ng OHI seeking (among Latino fraternity males in college) and social support, self efficacy to seek out health care services and self efficacy to engage in health behavior (among other variables) Chapter 2 further presents and explains the theoretical und erpinning s which guides th is dissertation study. Chapter 3 describe s the study setting, recruit ment of study participants participant characteristics and overall met hodology of the research study. Chapter 4 discuss es the f indings of the study.

PAGE 31

31 Finally, C hapter 5 provides a discussion stating the limitations encountered in this study and offering recommendations as well as implications of the findings for future studies. Table 1 1. Independent and dependent variables of study. Research Question Independent Variable (s) (level of scale) Dependent Variable (level of scale) RQ 1: How often do CMLF use of the Internet to locate health information? Ethnicity [Latino] (Discrete) Extent of use of the World Wide Web [WWW] {Internet} to locate health information ( Continuous) RQ 2: What types of health information do CMLF look for online? Ethnicity [Latino] (Discrete) Types of health information searched online ( Collected qualitatively through focus groups Con tinuous ) RQ 3: What are factors associated with OHI seeking among CMLF ? Ethnicity [Latino] (Discrete) Factors that influence OHI seeking ( Collected qualitatively through focus groups Conti nuous) RQ 4: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social su pport, age, and education predict self efficacy to engage in health behavior among CMLF ? Level of influence o n a ttribut es of OHI sou rce (Continuous) message (Continuous), content (Continuous) ; frequency of OHI seeking ( Continuous ); level of social supp ort ( Continuous ); age ( Continuous ); education (Continuous) Self efficacy to engage in health behavior (Continuous ) RQ 5: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social support, age, and education pred ict self efficacy to seek out health care services among CMLF ? Level of influence on attributes of OHI source (Continuous), message (Continuous), content (Continuous); frequency of OHI seeking (Continuous); level of social support (Continuous); age (Con tinuous); education (Continuous) Self efficacy to seek out health care services (Continuous)

PAGE 32

32 CHAPTER 2 REVIEW OF RELATED LI TERATURE Introduction The purpose and scope of Chapter 2 is to explain the findings of relevant literature related to OHI seeking, among CMLF by comparing, contrasting, analyzing, and discussing the limited available literature on OHI seeking among CMLF C hapter 2 presents an overview of the search process and the results of each step of application of inclusion/ exclusion criteria. Ch apter 2 concludes with a d i scussion of the research gaps foun d during the literature review. Literature Review Methods Search Procedures O nly manuscripts published in English from January 2000 to December 2012 were considered. The searched databases includ ed: ERIC, PsychINFO, PubMed, Academic Search Premiere, CINAHL Plus, Applied Social Sciences Index and Abstracts and ProQuest. T he following key terms were entered in various combinations with the Boolean op erator AND: College students, Latino men, Hispani c men, fraternal organization, online health information search, consumer hea lth information seeking on the I nternet, se arching for health information o nline, health behavior, health care utilization, utilization of health services, social support, self ef ficacy, and benefits Additional relevant articles were identified through hand searches after scanning the reference section of each database identified article in the efforts to enhance the breadth of the examination

PAGE 33

33 Selection Criteria Inclusion criter ia specified that studies include Latino male part icipants, ages 18 26, enrolled in college, searching for OHI Each included study necessitated to describe the following: C haracteristics of participants (i.e. demographics, age, gender) ; R eport of conte nt of health information searched; R es ; R eport of associations between health information search es and health behavior outcome ; R eport of health care utilizat ion (rates and factors predicting ut ilization ); Report of the association between self efficacy and health behavior ; Report of the association between self efficacy and seeking health care services; Report of the association between health information searches and perceived soci al support Al l study designs were considered (e.g. Random Control Trials, Pilot studies, qualitative studies, quasi experimental, and observational study designs) Exclusion Criteria: S tudies were excluded if there was insufficient detail related to study attributes, in OHI search does not exist) as required under Insufficient detail on demographics; Partici pants not between the ages of 18 26; Study includes a non col lege population; Editorial, book chapter, commentary or other non peer reviewed study ; Studies where a Hispanic/Latino population was not researched; Studies where members of a fraternal organization is not researched; Studies that do not have exclusively Latino male participants.

PAGE 34

34 After accounting for inclusion and excl usion criteria the matrix method proposed by G a rrard (2010) was used to systematically organize findings from the reviewed studies. Results Figure 2 1, illustrates the process used to select articles for inclusion in the literature review The following describes the stages of application of inclusion/exclusion criteria. Ei ghteen available articles remained following the database search, and removal of duplicates and non relevant articles. Si xteen of those articles came directly from the search database and t wo of the articles resulted from a hand search of the reference lists of the sixteen articles. The eighteen articles covered four main categories : health behavior management ( n = 1) (Chiso lm, 2010) health care seeking and utilization ( n = 2 ) ( Ai, No l, Appel, Huang & Hefley, 2013; Livingston, Minushkin, & Cohn, 2008 ) OHI seeking ( n = 1 2 ) ( Buhi, Daley, Fuhrmann, & Smith, 2009; Easaw, 2010; Fogel, Fajiram, & Morgan, 2010; Escoffery et al., 2005 ; Ghaddar, Val erio, Garc ia, & Hansen, 2012; Gray, Klein, Noyce, Sesselberg, & Cantrill, 2005; Hanik & Stellefson, 2011; Morahan Martin & Anderson, 2000; Nustad, Adams, & Moore, 2008; Pe a Purcell, 2008; Percheski, & Hargittai, 2011; Shanahan, 2009 ) an d articles on social support and self efficacy ( n = 3) ( Dobransky & Hargitti, 2012; Freeman, Barker, & Pistrang, 2008 ; Zuckerman, 2009 ). Application of primary exclusion resulted in 2 articles excluded due to insufficient detail on demographics (Shanahan, 2009; Zuckerman, 2009), two articles excluded because they did not meet the age restriction ( Ghaddar, Val erio, Garc ia, & Hansen, 2012; Gray, Klein, Noyce, Sesselberg, & Cantrill, 2005 ), five articles were excluded because studies surveyed a non college pop ulation ( Ai, No l, Appel, Huang & Hefley,

PAGE 35

35 2013 ; Chisolm, 2010; Easaw, 2010; Livingston, Minushkin, & Cohn, 2008; Pe a Purcell, 2008 ), and one article was a commentary and not a research article (Morahan Martin & Anderson, 2000). Such exclusion resulted in only two remaining categories, OHI searching ( n=6 ) (Buhi, Daley, Fuhrmann, & Smith, 2009; Fogel, Fajiram, & Morgan, 2010; Escoffery et al., 2005 ; Hanik & Stellefson, 2011; Nustad, Adams, & Moore, 2008; Percheski, & Hargittai, 2011; Shanahan, 2009) and soc ial support and self efficacy ( n=2 ) ( Dobransky & Hargitti, 2012; Freeman, Barker, & Pistrang, 2008 ). Of those remaining articles a secondary exclusion of the criteria Hispanic/Latino population not researched resulted in 3 articles being excluded (Fogel, Fajiram, & Morgan, 2010; Freeman, Barker, & Pistrang, 2008; Hanik & Stellefson, 2011). Five articles remained for further review (Buhi, Daley, Fuhrmann, & Smith, 2009; Escoffery et al., 2005 ; Nustad, Adams, & Moore, 2008; Percheski, & Hargittai, 2011; Dob ransky & Hargitti, 2012) Final exclusion eliminated studies that did not include participants who are members of a fraternal organization ( n= 4) (Buhi, Daley, Fuhrmann, & Smith, 2009; Escoffery et al., 2005 ; Percheski, & Hargittai, 2011; Dobransky & Hargit ti, 2012) and studies that do not exclusively have Latino male participants ( n= 1) (Nustad, Adams, & Moore, 2008). Hence, the search resulted in no manuscripts of o nline information seeking, among Latino fraternity males in college However, the eighteen re maining articles, after the database vocabulary search, removal of duplicates and removal of non relevant articles; were used to help conceptualize the study direction for the project at hand. Discussion Realizing that no studies could be found on the top ic of interest, the current research topic is unique and yet to be studied In order to set a foundation for this

PAGE 36

36 dissertation study, the discussion will focus on the latter eighteen articles found (but excluded) during the literature search. Although no single article encompassed all of the attributes sought during the literature review the articles covered four main categories which are discussed below. Health Care Utilization and Seeking b y Latino College M ales S ome research on Latinos indicates tha t 27% to 34% of Latino Americans nationwide do not have health insurance or access to a usual source of health care ( Livingston, Minushkin, & Cohn, 2008 ). Foreign born Hispanics in the United States are less likely to have prescription medications ; however, individuals who have recently immigrated to the US are less likely to visit ambulatory care or emergency room settings ( Ai, Nol, Appel, Huang, & Hefley, 2013 ). Health B ehavior Management Exhibited b y Lati no College M ales Chisolm (2010) has advanced the idea that OHI searching in and of itself m ay be considered a health behavior whereby d ifferent types of searches are associated with different patient characteristics. According to Chisolm (2010), health information search is not a monolithic behavior. regular use of the Internet were the most consistent predictors of use. Chisolm (2010) concluded that health Internet behaviors can successfully be described using models designed for traditional heal th behaviors; however, different health information seeking behaviors have different user profiles. OHI S eeking Amongst Latino College M ales R ecent research on health information seeking among the U.S adult population finds that age, education, and gender are among the most important predictors of searching for health information online (Baker, Wagner, Singer, & Bundorf, 2003; Cotten

PAGE 37

37 & Gupta, 2004; Hassani, 2006; Hesse, Nelson, Kreps, et al., 2005). However, few studies have examined additional predictors beyond these basic demographic characteristics. Furthermore, limited previous work has explored whether OHI supplements other sources or health information or if it is providing health information to people who have access to few other information sources Self E fficacy and H ealth Pertaining t o Latino College M ales D iscussing health information with peers in similar situations is more common among females (73%) than males (63%) (Rideout, 200 1 ). Rather than just discussing the information, African Americans (52%) are more likely to change their health behaviors based on the information they learn from their online queries, unlike Hispanics (42%) and non Hispanic whites (37%) (Rideout, 2001) who are more reluctant to act of the health information they locate Issues of Measurement A n analytical tool to assess all of the measures in this study among the stated population was not found during the literature review. Hence, it was necessary to create a questionnaire to assess the measures of this study. Easaw (20 10) offers a 25 item questionnaire th at assesses OHI seeking in a college female population Given the similarity in populations (college students), this questionnaire was utilized and adapted for the purpose of this study Fur ther details on uestionnaire and its application can be found in C hapter 3. Conceptual Underpinnings of Previous R esearch Previous studies on OHI seeking, looking at the variables social support, fraternal membership, self efficacy to seek out health care services and se lf efficacy to engage in health behavior have been guided by the Health Belief Model (HBM) and social

PAGE 38

38 constructs of HBM, as well as Social Cognitive Theory (SCT) ( Easaw, 2010; Eysenbach, 2008; Glanz, Rim er & Lewis, 2002; Kim, 2012; Lu, 2006; Tamim, 2 012; Wright & Bell, 2003). Social Cognitive Theory has also been used in several studies relating to self efficacy and health information seeking (Bandura, 1977; Bandura, 2004; Bass, Ruzek, Gordon, Fleisher, McKeown Conn, & Moore, 2006 ; Lee, Hwang, Hawkins, & P ingree, 2008; Lorig et al., 1999 ). However, Self E fficacy T heory also utilized in several similar studies (Bass, Ruzek, Gordon, Fleisher, McKeown Conn & Moore, 2006; Oh & Lee, 2012; Smith McLallen, Fishbein & Hornik, 2011 ) presents the best applicable t heory for the goals of this study In understanding that those studies utilizing Self efficacy theory more closely parallel the research aims of the current study Self E fficacy T heory is a logical conceptual underpinning for the purposes of the study. A summary of the Self efficacy Theory used for the current study is described below. Self E fficacy T heory Self E ffi cacy T heory is derived from Social Cognitive T heory (Bandura, 1986) As a central construct oci al Cognitive T heory (199 7) s elf efficacy is particular behavior when faced with a variety of challenges Self Efficacy T heory proposes that self confidence is a precursor to behavior change and skill development ( Bandura, 2006). performance, feelings, choices, and motivation regarding behavior change are determined, in part, by how effective they believe they can be (Bandura, 1982) A key principle of Self Efficacy Theory is that individuals are more likely to engage and put forth more effort and persistence in activities that they have higher feelings of efficacy for and less likely to engage in those activities for which they have les s feelings of

PAGE 39

39 efficacy ( Van der Bijl & Shortridge Baggett, 2002). Research has shown that self efficacy has an indirect positive relationship to overall and condition specific health outcomes (Sue, 2012) P atients with higher self efficacy have better prob lem solving skills and exhibit better self care (Sue, 2012). Regarding Internet use, high self efficacy is positively associated with (a) willingness to choose and participate in computer based activities, (b) expectations of success of computer use (c) pe rseverance when faced with computer use difficulties and (d) computer based performance (Eachus & Cassidy, 2006). As was the case in a study by Selsky, Luta, Noone, Huerta and Mandelblatt (2013), young Latinos with higher self efficacy had higher odds of i ntend ing to use the Internet for cancer information Bandura (1977) identified four sources of self efficacy. The sources are : (1) mastery experience, (2) vicarious experience or modeling, (3) emotional or physiological arousal, and (4) verbal persuasion ( Bandura, 1997). The most influential source of self efficacy comes from mastery experience. Mastery experience refers to when an individual succeed s through a series of gradual steps that solicit the performance of desired behaviors often through increment al goal setting ( McAlister, Perry & Parcel, 2008). When specific tasks are completed successfully, perceived self efficacy is ; however when repeated failure occurs, feelings of mastery are diminished An ex ample of self efficacy through mastery experience in the Latino population is evidenced in Salabarra Pe a Trout, Gill, et al. (2001), where Latino youth had high mastery, self esteem and self efficacy for completing their Tuberculosis treatment regimen, due to enhanced social support from

PAGE 40

40 parents and family and by taking small steps daily in adhering to their medication regimen The second source of efficacy, vicarious experience or social modeling is less influential than mastery experience (Chu, Huber Mastel Smith, et al., 2009). Vicarious experience occurs when learning is achieved through observing behavior performed by others. According to Bandura (1994) s eeing people similar to oneself succeed by they too possess the capabilities to master p.80 ). On the other hand, i f individuals see themselves as very different from the models they are observing, then they will not be influenced to a high degree by the m McAlister, Fernandez Esquer, Ramirez and colleagues (1995), studied this phenomenon in the Latino community via designing a unique communication strategy to promote cancer screening where community role models were showcase d getting screened; the effort resulted in more adherence in the experimental community when compared to the control community Thus, f or Latino male college students it may be necessary to have their peers also engage in healthy behaviors and in seeking health resources for them to also engage in similar activities or express that they too have engaged in such activities Emotional or physiological arousal is another source of self efficacy gathered through improving physical and emotional states. This i ncludes responses such as anxiety, stress and fatigue. Bandura (1977) theorizes that p.82 ). Barkley and Burns (2000), in their study regarding co ndom use and self efficacy among

PAGE 41

41 arousal is related to self efficacy by noting that fear may influence condom use As it pertains to the study at hand, emotional or physiolog ical arousal may very well influence Latino male college students to search for OHI regarding their personal health The fourth source of self efficacy is acquired through verba l persuasion. Verbal persuasion includes suggestions, positive appraisal and social encouragement Feedback should be corrective and framed in a positive way to create higher self efficacy (Bandura, 1977) Verbal persuasion is more than positive appraisals ; it also includes structuring situations in ways that are set up for success to avoid situations that are likely to result in failure (Bandura, 1994). In a study by Nguyen, Carson, Parris and Place (2003), verbal persuasion was noted to be important when educati ng Hispanic adolescents about healthy practices, available health care services, and support resources. I ncreasing self efficacy for health behaviors and self efficacy to seek health care resources among Latino male college students may involve the use of verbal persuasion from peers. Summary Chapter 2 compared, contrasted, and analyzed the findings of relevant literature related to online information seeking among Latino fraternity males in college, including how this behavior is related to self efficacy to seek out health care services and self efficacy to engage in health behavior Chapter 2 also described the procedures followed to produce the findings of the literature review. Chapter 2 revealed that no prior studies have examined the variables considered in this dissertation study In addition, Chapte r 2 brought to light the need for questionnaire develop ment to address the

PAGE 42

42 purpose s Finally, Chapter 2 detailed the conceptual underpinnings that set the foundation for the current study

PAGE 43

43 Figure 2 1 Stem tree illustrating manuscript selection process using vario us search databases and combinations of controlled vocabulary.

PAGE 44

44 CHAPTER 3 RESEARCH DESIGN AND METHODOLOGY Introduction The purpose of Chapter 3 is to describe the data collection and data analysis methods used to study OHI seeking among CMLF Due to the ex ploratory nature of the study a mixed methods approach was employed taking place in two phases Phase one employed qualitative focus groups and think aloud protocols Phase tw o involved a quantitative survey The q ualitative methods in phase one include d conducting a series of f our focus groups, along with think aloud protocols to to answer research questions 2 and 3 and inform questionnaire development. Quantitative methods for the study involve d a dministering a survey to answer the remaining research q uestions 1, 4, and 5 Statistical methods involved in answering the research questions are explained in further detail later in C hapter 3 Problem and Purpose Overview A lthough Latinos may soon become the largest minority group in the United States, thei r inclusion in national health studies has historically been low (LaVange et al., 2010; Sorlie et al., 2010). Today, 93 million Americans or approximately eight in ten Internet users have searched for a health related topic online, according to the Pew Int ernet & American Life Project (Fox, 2011). With the growth of the Latino population, more Latinos are enrolling in colleges nationwide (Fry, 2012). More college enrollees have led to more Latino males becoming members of fraternal organizations (Fry, 2012; Helem, 2004; Miranda, 1999). Li mited research has been conducted on the comprehensive health of this unique sub group of college students (Lust et al., 2007)

PAGE 45

45 with studies limited to examinations of alcohol and drug use ( Barry, 2007; Borsari & Carey, 1 99 9 ; Cashin, Presley & Meilman, 1998; Kuh & Arnold, 1993; Larime r, Anderson, Baer & Marlatt, 2000; Larimer, Irvine, Kilmer & Marlatt, 1997; Lo & Globetti, 1995 ; Workman, 2001 ) and sexual health ( Foubert, 2000; Guardia & Evans, 2008; Scott Sheldon, Carey & Carey, 2008 ). T he purpose of this study is to extend health research among CMLF by examining the research question s described below Research Questions 1. How often do CMLF use of the Internet to locate health information? 2. What types of health information d o CMLF look for online? 3. What are factors associated with OHI seeking among CMLF ? 4. To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social support, age, and education predict self efficacy to engage in health be havior among CMLF ? 5. To what extent do attributes of OHI ( source, message, and content ) ; frequency of OHI seeking, social support, age, and education predict self efficacy to seek out health care services among CMLF ? Setting and Population Setting The stud University of Florida, the Institute of Hispanic Conference Room, housed at 1504 West University Avenue, Gainesville, Florida, 32603. Florida International Universit SW 8th Street, Modesto A. Maidique Campus, Miami, Florida, 33199. 33146 The University of Florida and Florida International University served as the setting for the focus group sessions. The University of Miami was the setting for the think aloud

PAGE 46

46 protocols and interviews Students from UF and FIU took part in the focus group sessions while students from UM completed think aloud protocol interviews. The remainder of the schools participated in the survey The University of Florida is proudly home to the Institute of Hispanic Latino Cultures. The location is affectionately named La Casita and is the preferred meeting place for mo based organizations. This location presents the ideal setting to conduct a focus group session due to its proximity to the campus, its private and quiet resources, and the no cost option for individuals affiliated with the Uni versity of Florida. Reservations for La Casita are made hassle free by filling out an online request form at http://www.multicultural.ufl.edu/reservations/ibc_space_reservations. Florida offers a v ariety of flexible event spaces and private meeting rooms available for reservation to individuals looking to conduct academic related events. Reservations were made available via their online reservation page, http://gcevents.fiu.edu/VirtualEms/. Room GC3 14 was secured for the purposes of this study. The Otto G. Richter Library is the main library at the University of Miami, located in the center of the Coral Gables campus. Group study room 125 provided for a private, quiet, and convenient setup to allow f or intimate conversation with think aloud protocol participants. Reservations were made available via their online reservation page, http://libcal.miami.edu/booking/richter study. employed the administration of an online sur vey which involved participants from the University of Central Florida, University of South

PAGE 47

47 Florida, Nova Southeastern University, Embry Riddle Aeronautical University, Florida State University, Florida Atlantic University, and Barry University. Participa nts who participated in the qualitative portion of the study were not eligible to complete the online survey. The following five reasons provide the rationale for including the total of ten schools listed in Table 3 1, for both phases of this investigati on : Florida International University, the University of Miami, NOVA Southeastern University and Florida State University have been ranked as part of the top 25 colleges in the nation for Latinos, based on factors such as tuition, percentage of enrolled Lat ino students, offering of tailored programs and organizations, as well as selectivity (Aragones, n.d.). Schools, such as the University of Florida and Florida International University, have been considered as top schools for Latinos in the Science, Techno logy, Mathematics, and Engineering fields (STEM) by the associated press and organizations such as Huffington Post (Gamboa, 2012). Many of these institutions, such as Barry University and the University of Central Florida, have membership or associate mem bership with the Hispanic Association of Colleges and Universities (HACU) (Hispanic Association of Colleges and Universities [HACU], n.d.). Florida is among the top states for increased levels of Latino enrollment on college campuses (Liu, 2011; Lilley, 20 Large a vailability of Latino based fraternal organization on campus es in Florida ( NALFO n.d.). Students from UF and FIU took part in the focus group sessions while students from UM completed think aloud protocol interviews. The remai nder of the schools participated in the survey. These schools have populations ranging between 4,597 and 50,968 students with the Latino student population percenta ge ranging between 10% ( Embry Riddle Aeronautical University ) and 66% (Florida International University). Table 3 1. The schools are located as far south as Miami, FL and as far north as Tallahassee, FL.

PAGE 48

48 Participants Individuals belonging to a fraternity with a majority of Latino members (Lambda Theta Phi Latin Fraternity, Inc.) was recruited fo r the purposes of this study. Lambda Theta Phi was the first fraternity to be identified as Latin, and was founded on December 1, 1975, at Kean College in Union, New Jersey. Lambda Theta Phi Latin Fraternity, Inc., currently has 123 undergraduate chapters and 12 alumni chapters in 24 U.S. states. Lambda Theta Phi takes pride in currently having 10 fully functional undergraduate chapters in the State of Florida (Table 3 2). The 10 chapters are divided into 2 sectors: Florida Sector I and Florida Sector II (T able 3 2). Each ch apter maintains an average of 31 active brothers per semester. Active members are member s who have paid all financial dues to the fraternity and are in good academic standing with their respective universit y The study recruited college males who are members of Lamba Theta Phi between the ages of 18 and 26, enrolled at several Florida college campuses enumerated in Table 3 1. A full description of participant characteristics can be found in Chapter 4. Data Collection Procedures Prior to data collection, approval from the University of Florida Institutional Review Board (IRB) was secured on June 5 th 2013. The process took longer than usual due to the mixed methodology design of the study protocol and the need to secure an umbrella IRB app roval f rom multiple institutions within the state of Florida Focus Group Sessions Focus groups were chosen as a method to identify and explore issues such as types of health information sought and factors influencing OHI searches that could not be studie d in depth through a survey alone Focus groups allow copious amounts of

PAGE 49

49 information to be collected from numerous participants in a relatively short span of time (Morgan, 1997). Usually, when participants hear someone else articulate an idea, they are som etimes stimulated to express a different view. Hence, focus groups allow differences in phenomena to emerge quickly (Morgan, 1997). The focus group is a type of data collection in which the researcher determines the topic, and the data are collected throug h group interactions (Morgan, 1997). According to Morgan (1997), the main advantage of focus groups as a means of qualitative research is the opportunity to observe a large group of individuals in a limited amount of time. For example, two eight person foc us groups can produce as many significant ideas as ten individual interviews (Fernz, 1982). With this, typically a focus group should include anywhere between 8 12 participants and at minimum 2 groups should be formed until data saturation is reached ( Grud ens Schuck Allen, & Larson, 2004 ; Oates, 2002). Even though participant observation has several advantages, such as a greater variety of interactions and more access t o settings in which a substantial set of observations can be collected on the study individuals in a more natural social setting than in one on one interviews (Marshal l & Rossman, 1999; Morgan, 1997). With the ability to observe participants in social groups, unexpected issues can arise and add to the pool of knowledge e specially when complicated topics are being discussed F ocus groups can also uncover factors that in fluence opinions, behaviors, and/or motivators of a homogeneous group by creating a permissive environment that encourages participation (Krueger & Casey, 2000). Focus

PAGE 50

50 deta Online information searching is a habitual behavior, and identifying and explaining factors influencing the behavior is easier in a moderated conversation with several participants who are able to respond to and elaborat e on each Due to the fact that many ethnic student organizations, including Lambda Theta Phi do not have physical mailing addresses available ( due to numerous factors ) conventional recruitment methods as explained by Dillm an, Smyth and Christian (2008) were not possible. I n order to enroll the maximum number of participants from the University of Florida and Florida International University for the focus group s the researcher established a working relationship with each re b oard to notify and remind brothers of the opportunity to help progress Latino health by participating in this research study. LambdaNET (https://www.thelambdas.net/) is an online resource for Lambda Theta Phi members that main tains an up to date directory of every active and non active member as well as a listserv email address for every fraternity chapter in Florida Hence, LambdaNET was utilized to send recruitment emails to the respective chapters and their members. Fratern ity members attending the University of Florida were first approached with informed consent via secure recruitment emails that were sent on June 24 2013. Due to low initial enrollment into the study another email was sent on July 8, 2013 to motivate mor e participants to register for the first planned focus group on July 22, 2013. The date of the first focus group was planned with the cooperation of the respective chapter president, who notified the researcher of days and times chapter meetings were held which allowed for fraternity members to conveniently attend the focus group

PAGE 51

51 session prior to the ir fraternity group meeting. The emails sent included the risks and benefits associated with participation, privacy methods, compensation, and autonomy The em ails also included information on the location, time and address of the scheduled session, as well as a link to the scheduling page: http://latinosonlinehealth.wordpress.com/focus groups/ The scheduling page allowed for participants to read over informed consent and input their names for the scheduled session Participant names were kept confidential and were only visible only to the researcher A limit for registration was also imposed for the session, capp ing registration at 1 5 participants (in the hope s of having at least 12 or 80% attend ) A limit was put in place to keep groups small and to keep expenses at a minimum. An example of the recruitment email is provided in Appendix A. An example of the informed consent form attached to emails is provided in Appendix B Participants who attended the first focus group session were given $5 Amazon.com gift cards as compensation for their time and effort Refreshments and snacks were also made available during the focus group session. Fraternity members atten ding Florida International University (FIU) were first approached with informed consent via secure recruitment emails sent out on July 23, 2013 Reminder emails were sent on July 31, 2013 to remind participants of the scheduled focus group session on Augus t 5, 2013 Secure recruitment emails were again sent out a week prior to each of the planned focus groups on August 12, 2013 and August 19, 2013 Emails for subsequent sessions encouraged participants not to reschedule for sessions, if they previously atte nded a planned session. The date s were planned with the cooperation of the respective chapter president, who notified the

PAGE 52

52 researcher of days and times chapter meetings were held, which allowed for the careful planning of dates on which to hold focus group session s that would enabled fraternity members to conveniently attend the focus group session prior to their usual meeting. The methodology followed for recruitment of FIU focus group participants differed from the methodology followed for recruitment of U F focus group participants only in the aspects described above. Overall, the focus group sessions were conducted following the methods published by Grudens Schuck Allen, and Larson (2004). Such focus group fundamentals include d the form ation of homogeneo us groups where participants f e l t free and open to discuss their ideas in the presence of people who do not differ in the terms of status, power, job, income, education or personal characteristics. Previous research suggests that men are more forthcoming o n discussing issues when in groups of men similar to themselves (Garfield, 2010).Thus, t o allow for participants to feel comfortable and free to share their thoughts and ideas, the focus groups were limited to only Latino male college students who are memb ers of Lambda Theta Phi. T he goal of focus groups is to produce good conversation on a given topic. Good conversation question, disagree, contradict themselves, and interrupt. However, the researcher must foc us group moderator wants both natural features of conversation as well as focused discussion in the course of the given time. To protect against experimenter bias and to avoid inadvertently introducing experimental error, t he principal investigator served as the trained moderator conducting focus group sessions following the focus group

PAGE 53

53 moderator guide (A ppendix C ) which was developed using guidelines synthesized from several sources following the hourglass format ( funnel approach ) where two broad questio ns guide the conversation and transition questions serve to direct the conversation onto the next topic ( Krueger & Casey, 2000 ) ; in the funnel approach analysis of data from prior previous focus group session s helps to create broad questions for successive focus group s until saturation of topics occurs (Hesse Biber & Leavy, 2003; Hesse Biber & Leavy, 2008; Krueger & Casey, 2000; Krueger, 1998; Morgan, 1997; Morgan & Krueger, 1998). Random selection of focus group participants was not possible due to limite d sample size; thus, convenience sampling was used. The researcher oversampled participants given the possibility for attrition to occur. A total of 41 participants were recruited to participate in the four focus groups The first focus group had 11 partic ipants. Focus group II had 9 attendees, focus group III had 10 attendees, and focus group IV had 11 attendees. The recruited participants provided a sample size that achieved data saturation, and confirmed to group sizes larger than 8 persons each as sugge sted by Grudens Schuck, Allen, and Larson (2004). The focus group sessions for this study ran between 1 and 1.5 hours, which is characteristic of focus groups ( Onwuegbuzie Dickinson Leech, & Zoran, 2009). Prior to commencing focus group sessions, partic ipants were given ample time to read and sign informed consent forms (Appendix B) if they agreed to participate in the study. As discussed in further detail below, ice breaker questions were included in the troduce themselves. The first question

PAGE 54

54 their names, age, year in college and major. The first of primary focus group questions T typically influenced or affected your expe riences of going online to search for health The following presents how the four sources of self efficacy were incorporated into the moderator guide (Appendix C) and used during focus group sessions. T he construct of m astery experience was in corporated into the the focus group sessions by allowing participants to introduce themselves to begin to assert themselves into the conversation In allowing each participant to introduce themselves, also came into play since it wa s understood that participants all share similar backgrounds and by witnessing each other express confidence in answering the questions participants themselves will hold higher confidence Although the previously stated may not be proven research by (Garf ield, 2010) suggests that males are more forthcoming when they are in groups of other males with similar backgrounds. The moderator guide also had cues and prompts for reminding encouraging r espondents to answer confidently. This represents the self efficacy theory construct of verbal persuasion Hosting the focus group sessions in private, yet easily assessable locations, along with handing out refreshments was done in order to make partici pants comfortable and alleviate stress. Thus, th emotional or physiological arousal also utilized during focus group sessions by helping to increase efficacy towards answering questions by making a positive mood

PAGE 55

55 possib le. The use of subsequent rounds of focus groups, as suggested in the hourglass approach discussed earlier efficacy through mastery experience with each subsequent conducted focus group session S uccess during the previous sessions allowed for the moderator to feel more confidence and ease in moderating focus group sessions Self efficacy theory also influenced questions in the moderator guide (Appendix C) hat was challenging about confidence participants had while searching for OHI. T he focus group sessions were a udio record ed and follow ed established anonymity and confiden tiality guidelines such as reminding participants to keep their experiences confidential. Participants were also made aware that complete confidentiality of their experiences could not be fully guaranteed since complete Once sessions commenced the moderator followed the moderator guide (Appendix C) as closely as possible. Immediately after each focus group session audio recording s were transcribed and deleted not more tha n 48 hours after the focus group concluded which was done to be compliant with UF IRB requirements put into place to preserve anonymity and confidentiality Transcription s of the focus group discussion were carefully documented, with any and all identif ying information removed from the final record The software InqScribe (2.2) was utilized to facilitate transcription with the use of c ustom shortcuts for audio control. The completed transcriptions were saved onto a secure, encrypted and password protect ed computer only accessible to the principal

PAGE 56

56 investigator. The transcripts were loaded to the software ATLAS.ti (7.1.6) which facilitated data analysis. In order to proceed with survey implementation, the focus group data analysis was completed prior to c onducting think aloud protocols Questionnaire Pre testing via Cognitive Interviews For details on questionnaire development for pretest and implementation purposes, incl uding how items were selected a nd informed by focus group data, please Questionnaire p re testing was important t o ensure proper implementation of the survey, full understanding of questionnaire items by participants, as well as accurate portrayal and proper analysis of survey data G iven that t he goal of the think aloud protocol interview is to examine decision heuristics pertaining to the given questionnaire, the researcher need ed to description and explanation of the experience (Ericsson & Simon, 1993). O bservation alo ne w ould not have reveal ed the internal processes of decision making while answering items on the questionnaire (Lindlof & Taylor, 2010) Think aloud protocols have several strengths, including being the closest of the three methods used in this study to p roviding naturalistic observation like data (Lindlof & Taylor, 2010). In the think aloud protocol, talking while doing (as opposed to talking about the experience after the fact), according to Ericsson and Simon (1993), may prompt more structured informati on solicitation and process description. The actions of taking the questionnaire and explaining how participants arrived at their answer is designed to allow the participant to articulate parts of the process in real time that the participant mig ht not be able to recall after having taken the survey (Ericsson & Simon, 1993). G uidelines suggested by Ericsson and Simon (1993) were followed, including participants that resemble d actual users as closely as possible in age, preparing a

PAGE 57

57 sound interviewer question guide with prompts and clear explanation as well as task warm ups, and correcting highlighted errors promptly to red uce redundancy. As discussed earlier, conventional recruitment methods such as those stated by Dillman, Smyth and Christian (2008) could not be implemented due to a lack of a physical mailing address for Lambda Theta Phi Latin Fraternity on the University of Miami campus. Thus, with email addresses found on LambdaNET, a convenience sample of f raternity members attending the University of Mi ami were approached with informed consent through secure recruit ment emails on August 19, 2013 requesting participation in think aloud protocol interviews on August 31, 2013. Twelve half hour time slots were made available on August 31, 2013 starting at 9 :00 AM, breaking for lunch at 12:00PM and continuing at 1:30PM with the interviews ending at 3:00PM Of the 12 time slots 10 were f illed with willing participants O nly 8 of these participants attended their scheduled appointments even though r eminder ema ils were sent a few days prior to the scheduled think aloud protocol interviews. An example of the recruitment email is provided in Appendix D An example of the attached informed consent form can be seen in Appendix E Participants who completed think alo ud protocols were given compensation for their time and effort in the form of a $5 Amazon gift card. Prior to conducting the think aloud protocol interviews, an interviewer que stion guide was developed per suggestions from Ericsson and Simon (1993), and va n Someren, Barnard, and Sandberg (1994). According to van Someren, Barnard, and Sandberg (1994) warm up tasks are crucial for the participant to understand the

PAGE 58

58 process of the think aloud protocol, thus per the formed question guide participants in this s tudy were asked: Imagine we are at the front of your home, please walk me through each room in your home and count the number of windows in each room. As you do this, please describe to me in detail what you are seeing. As suggested by Someren and colleagu es (1994) the question guide provided for participants to be reassured that any mistakes were the researchers fault and that there were no right or wrong answers. As per Someren and colleagues (1994) participants were also probed to continue speaking whi le they worked through the questionnaire. Prompts suggested in the formed question guide include: The question guide a s per Ericsson and Simon (1993), also provid ed reminders to debrief the participant and to prepare for the next participant by resetting equipment and correcting errors in the questionnaire prior to the start of the next participant. The question guide used in this study can be seen in Appendix F Web based surveys present an array of advantages for social behavioral scientists (Chaney, Barry, Chaney, Stellefson, & Webb, 2012). However, web based surveys benefit from the qualitative process of cognitive interviewing which helps the researcher in cla rifying issues with the web based surveys prior to mass dissemination. Cognitive interviews reduce survey error and lend credence to the data collected before subsequent analysis is performed. Prominent survey methodologists such as Chan ey, Barry, Chaney, Stellefson, and Webb ( 2012 ) suggest the use of Camtasia software or similar computer software which offer screen video capture to aide and inform cognitive

PAGE 59

59 interviewing for formulating valid and reliable web based surveys. Camtasia allows researchers to ca pture all computer screen movements and activities ( as text inputs, video output, cursor movement, and screen changes ) of research participants while they attempt to complete web based surveys. usefulness as a data collect ion tool, essentially allowing the software to serve as non are conducted with at least 8 12 interviewees that match the social and demographic characteristics of the intended audience (Kerwin & Willi s, 2011). Overall, use of Camtasia i n the cognitive interviewing process represents a novel technological application that can aide survey developers in producing high quality instruments. T he use of Camtasia soft ware was essential for collecting and analyzing data in the current study. Such methodology and analysis scheme was also utilized by DeMaio, Rothgeb, and Hess (1998) for improving survey quality for the U.S. Census questionnaire. C ognitive interviews, usin g think aloud protocols with Camtasia software were conducted in Rm. 125 of the UM Richter library which provided relaxed about the questions in front of them The researcher acted as the sole interviewer in each of the think aloud protocol sessions to avoid inadvertently introducing experimental error. During the think aloud protocols, the participants were first provided with ample time to read informed consent forms and con tinue with the study only if they provided electronic the informed consent. Participants were reassured of their privacy and anonymity and told that identifying information would not be published, per IRB guidelines.

PAGE 60

60 A total of 8 participants from UM, too s think aloud protocol sessions as per recommendations from Chaney et al. ( 2012 ) which states that about 8 12 cognitive interviews are sufficient to reach data saturation and give the researcher an idea as to whether the developed quest ionnaire is clear ly written and understood. A fter each interview was completed, minor errors in the questionnaire identified by the participants such as spelling errors, were corrected prior to the start of the next interview. Once all think aloud protoco ls were completed the audio was transcribed using InqScribe (2.2) which facilitated use of custom keyboard commands for toggling through the recordings. Once transcribed, recordings were deleted to adhere to IRB guidelines, in an effort to ensure privacy and anonymity of participants. The completed transcriptions along with notes on each think aloud protocol session were then saved onto a secure, encrypted and password protected computer only accessible to the principal investigator. The transcripts wer e loaded to the software ATLAS.ti (7.1.6) which facilitated analysis. T ra nscriptions were then coded to highlight errors in the questionnaire, misunderstanding or misinterpretation of items, and suggestion s for improvements described by participants. This level of detail about the interpretation of the questions, response categories, and format of the questionnaire is invaluable for providing information regarding ways in which the questions may not be achieving the goals of the questionnaire and/or yieldi ng valid and reliable data. Once all corrections were made to the questionnaire the second phase of the study, the online survey was ready to commence.

PAGE 61

61 Survey I mplementation Phase two of th e study, the quantitative phase of this study, involved survey imp lementation of the developed questionnaire, described later in Chapter 3. To allow for maximum recruitment fraternity chapter presidents of the schools in this study (Table 3 1), were all addressed via conference call on July 12, 2013 during their s tate meeting. During the meeting support for this study was solicited. Fraternity chapter presidents were told of the significance of the study and of the forthcoming survey. Thus, a working relationship was established which would prove to be invaluable in t he course of the survey implementation. LambdaNET the fraternity email listserv directory was once again used to send secure recruitment emails to fraternity members from Barry University, Embry Riddle Aeronautical University, Florida Atlantic University Florida State University, NOVA Southeastern University, the University of Central Florida, and the University of South Florida on September 14, 2013. M embers of Lambda Theta Phi, Latin Fraternity, Inc. attending Barry University, Embry Riddle Aeronautic al University, Florida Atlantic University, Florida State University, NOVA Southeastern University, the University of Central Florida, and the University of South Florida were approached with informed consent via secure recruitment email sent to each chapt on September 14, 2013 with attached informed consent forms detailing anonymity, confidentiality, and privacy as per UF IRB regulations The survey remained open from September 14, 2013 through October 31, 2013. Reminder emails were sent out o nce every 2 weeks, encouraging participants to take the questionnaire and thanking participants who had already completed the questionnaire as well as discouraging participants from taking the questionnaire more than once. Communicating with the fraternit y chapter presidents allowed for an increase

PAGE 62

62 in the number of completed questionnaires between the months of September and October. Please see Appendix G for a sample of the sent recruitment emails and see Appendix H for the electronic informed consent at tached to the cover of the questionnaire. Participants were only allowed to complete the survey if they agreed to the informed consent. From apriori analysis a quota of 200 participants was set to allow for adequate compensation of participants completing questionnaires Participants who completed the questionnaire were given the option of inputting their email address at the end of the questionnaire to receive compensation for their efforts in the form of an electronic $2 Amazon.com gift card sent to thei r emails. The survey had 153 participants respond and 127 participants completed each item in the questionnaire ( response rate = 59.6% ) when calculated based on the number of active members from the respective schools included in the survey ( n =305) detailed in Table 3 2. The email addresses were deleted once gift cards were emailed, to comply with UF IRB regulations. Data stored on Qualtrics is protected with high end firewall systems and Transport Layer Security (TLS) encryption ( Qualtrics, n.d. ). Thus, the data remained secure while on the Qualtrics server. Once the survey data was ready for analysis, the data was exported onto a secure, encrypted and password protected computer only accessible to the principal investigator. The data was then uploaded to I BM SPSS Statistics (version 2 2 ) for statistical analysis. Instrumentation Results from the think aloud protocols were taken to form a completed and refined questionnaire for survey implementation which was also uploaded to Qualtrics. The corresponding lin k to the completed questionnaire was included in distributed emails as

PAGE 63

63 described above, for ease of access for survey respondents Each section of the instrum ent, the source the aspect it measures along with how focus group data informed item selection a nd the use of self efficacy theory where applicable is described below. Please refer to Appendix I for the complete ad hoc questionnaire. Extent of Online Use Arising from focus group data w ere the overarching question s h ow frequently what locat ions and ith what devices do most CMLF search the Internet for OHI Due to the numerical nature of these questions, they were best answered through quantitative methods ; as was the case in the research performed by Hasebrink and Hanna Domeyer ( 2012 ) in which frequency of media use was best empirically tested through quantitative means. resp Rice (2006), readies the participant for the question concerning extent of use. T hus, t o analyze the extent of online use for health information of CMLF an item assessing fre quency of use, developed and used by Rice (2006), was appropriate to running of regressions, as in Rice (2006), where this item is rated on a continuous scale of 1 through 5,

PAGE 64

64 This item was also adapted from Rice (2006), and was used in this study to validate later items also assessing Internet use for health information seeking. The it nternet, do you do that MOSTLY using nation. This item gives clearer insight into which devices CMLF are utilizing to access location CMLF mostly find themselves when seeking for health information. Types and Factor s To assess types of OHI searched by CMLF the study conducted by Easaw (2010), where college women where sampled and investigated on their OHI searches, was referenced. Items assessing types of OHI searched and factors influencing OHI searches used in Eas topics include: A specific disease or medical problem A certain medical treatment or procedure

PAGE 65

65 Nutrition Exercise Prescription drugs Over the counter drugs Substance abuse (alcohol or drugs) Mental health issues (i.e. stress, anxiety, and depression) Sexually transmitted diseases H ealth insurance options and concerns available response selections. analysis efficient and because topics added from focus gr oup results would cover a greater realm of response options. However, future research should definitely consider The findings from the conducted focus groups in this study led to the addition of the following topics to the list above: Health information for others (e.g. family, friends, or significant other) Concern over symptom(s) (e.g. pain, bruising, discomfort) A recent diagnosis Concerns on weight or physical appearance Home remed ies (e.g. splints, sprains) Natural living (e.g. organic food and products) Information on immigration health concerns Exploring online posts, tweets, or videos on health Family planning (e.g. condoms, birth control, safe sexual practices or pregnancy of s ignificant other) For a complete view of the topics in the item please refer to the full questionnaire in Appendix I To assess factors influencing OHI searches, Easaw (2010) was again referenced. The item used for assessing the influence of such factors believe influenced you to search online for health information? Please select the level of

PAGE 66

66 includes a 7 what Non conducted focus groups sessions in this study. The added factors are as follows: The health status of a significant other (i.e. girlfriend) Monetary issues Concerns over weight or physical appearance The components were given the same 7 point Likert scale as the original components. Access of information To differentiate between online access to health information and other sourc es of health information, the following two items were sourced from Fox and Duggan (2013) and informed from focus group data information, do you typically go to any of the following to access that infor list of options includes: Health care provider (nurse or doctor) Pharmacist Friends/Peers Online search engine (e.g. Google, Yahoo, Bing) Online video site (e.g. YouTube, Vimeo) Mobile health app(s) Family member(s) Online forum (e.g. blog f or FAQ) Social Networking site (e.g. Facebook, Twitter, Pinterest) Each option above for the respondent to indicate whether the option is used. Focus group sessions helped to corroborate the selections spelled ou t in Easaw (2010).

PAGE 67

67 the previous question, which of these do you consult the MOST for health information? le options to which the respondent can select the most frequented source. Level of Influence (Source, Message, and Content) of Accessed OHI Focus group data helped to inform factors that influence CMLF to search for OHI However, quantitative methods were needed to shed light on whether the source, message or content of the gathered health information motivate d the participants to take action Although qualitative methods provide a starting point for level of influence having respondents answer on a 7 poi nt Likert scale eases analysis and interpretation by enabling numerical calculations Creswell ( 2014). To assess the level of influence of accessed OHI to spur action as such relates to source, message or co ntent the following item was sourced and slightly modified from OHI The item used in this study was modified by adding the component message (the information communicated to the seeker) as a response opt ion, while maintaining other aspects of the item true to Eastin (2001). The item primes the respondent and prepares them for the upcoming question

PAGE 68

68 on prompting you to take action in terms of either seeking out health services or managing your health 7 words, the credibility of the site, author (i.e. physician or health professional), or agency where the cond point Likert tes, point Likert selec tion. Please refer to Appendix I for a full view of the item. Self Efficacy to Seek Healthcare Services and Engage in Health behavior Self Efficacy Theory was applied to this study by assessing self efficacy to engage in health behavior among Latino men in college belonging to a fraternity and self efficacy to seek out he alth care services. Thus self efficacy was assessed by adapting several items on the questionnaire by means and methods described in Bandura tablish perceived self efficacy and not self esteem, locus of control, or outcome expectancies. This was done since self efficacy is concerned with stating each item on th e response scale on a 100 point scale, ranging in 10 unit

PAGE 69

69 instructions should establish the appropriate mindset that participants should have when rating the strength of belief in their personal efficacy to engage in health behavior and self efficacy to seek out h ealth care services have established the appropriate mindset in getting to participants to think about their perceived ability to search for OHI For example, the item on self efficacy to seek out healthcare services, prepares the This section is designed to help us gain a better understanding of how the health information you search online affects your confidence in seeking out health care services. Please rate how certain you are that you can do the things discussed below by writi The very same is done for the item assessing self efficacy to engage in health behavior us gain a better understanding of how the health information you search online affects your confidence in managing health. Please rate how certain you are that you can do For the item assessing self efficacy to seek healthcare services, items used in Bandura (2006) and from the freely available health utilization scale from the Stanford Patient Education Research Center (n.d.) were adapted to use in this study The resulting item is composed of 5 units, each on a sliding scale of 0 100, where the sum of the scores can range from 0 through 500, d C omposite scores of 0 334 where the higher the sel ected the number the higher the confidence level of the respondent.

PAGE 70

70 For the item assessing self efficacy to engage in health behavior items were efficacy item construction (2006), the resulting item is composed of 7 units. The composite score for this item rang es from 0 engage in health behavior A score of 0 466 700 signifies that the respondent feels re can be accessed in Appendix I Duke Social Support Index (DSSI) and Social Support Focus group sessions shed light on aspects such as family, friends, significant others, monetary issues and fear s of health condition s To add depth to nuances on social support and to make statistically supported statements on social support a valid and reliable assessment instrument was incoporated into the questionnaire The abbreviated 11 item Duke Soc ial Support Index (DSSI) has been previously utilized in a host of different population groups, including the general population, women, older adults, and persons suffering with chronic diseases (Koenig, Westlund, George, Hughers, Blazer, & Hybels, 1993; P owers, Goodger, & Byles, 2004; Wardian, Robbins, Wolfersteig, J ohnson, & Dustman, 2012). The established use of the instrument and its abbreviated DSSI can be found in App endix I Example of items of the DSSI include, How many times during the past week did you spend time with someone who does not live with you, that is, you went to see them or they came to visit you or you went out together? How satisfied are you with the kinds of

PAGE 71

71 relationships you have with your family an d friends? number 1, 2, or 3 respectively. Thus, the sum total of the instrument ranges from 11 to 33, where higher scores signify the r To establish a relationship among DSSI and fraternity membership, the following 2 thinking about the questions on social support, did you think about people who share membership with you in your fraternal excluded from the study. Demographic Items Quali tative data alone was not sufficient to give detailed numbers on different demographic aspects of the sampled participants. Thus, the questionnaire also includes select demographics items sourced from the U.S. Census Bureau (2009) and Easaw (2010). Items a Hispanic, Latino or Spanish origin Please type origin, for example, Argentinean, Colombian, Dominican, Nicara Response choices were kept the same as the form and slightly modified to remove an offensive race response term ch oice. The term was identified during cognitive interviews

PAGE 72

72 and a review of the literature to confirm the decision to remove the term found that only Census (Anderson & Fienb erg, 2000; Assefa, 2014; Fama, 2013; Yen, 2013). The final selection of response choices for this item can be found in Appendix I Easaw (2010). The response choices were amended, and include only the surveyed universities and colleges in this study. The response choices for this item can be found in Appendix I Items on academic classification, category of major/program of study, relationship status and living situation were all a on health information internet searches performed by college women. The items and response choices can also be found in Appendix I Since the study was only interested in college students age 18 continuous response options 18 26. Reliability and Validity of Study In all research contexts, concerns exist about the truth value, applicability, consistency and neutrality of the research design (Lincoln & Guba, 1985). In quantitative research, these concepts are talked about using the terminology of validity and reliability Much of the work in ensuring validity and reliability in quantitative methods is front loaded, focusing on testing instruments and sample selectio ns. However, in qualitative research, researchers are focused on participants (the sample) who know the most about the phenomena being studied, not on whether the participants have an equal chance of being selected from a population under study. Qualitativ e researchers have developed separate concepts that address issues of truth value, applicability, consistency and neutrality (Lincoln & Guba, 1985): credibility (an

PAGE 73

73 equivalent to internal validity), transferability (external validity), dependability (relia bility), and confirmability (objectivity). The following addresses how both research paradigms (quantitative and qualitative) were applied to this study. Reliability for Quantitative Aspects Internal consistency reliability of items within subscales of th e adapted questionnaire were reliability. Specifically, reliability measures were conducted for the following subscales included in the formed instrument: Duke Social Support Index (DSSI) Self Effi cacy to Seek Health Care Self Efficacy to Engage in Health behavior Level of Influence associated with Attributes of Online Health Information (Source, Message, Content) Nunnally and Bernstein (1994) of .70 and greater is an acceptable reliability statistic for a scale that is newly developed. The respective reliability coeffici ents calculated in this study are reported in Chapter 4 Validity for Quantitative Aspects In applying Nunnally and Bernstei validity, and i n the effort to achieve construct validity for the study at hand, each section of the ad hoc questionnaire was adapted from previously tested scale items which were p ublished in peer reviewed work s; these efforts are described in detail later in Chapter 3 under instrumentation. Validity threats such as experimenter expectancie s and bias were also addressed and are described later in C hapter 3

PAGE 74

74 Credibility for Qualitative Aspects Credibility is conc erned with the truth value of the research (Lincoln & Guba, 1985). Because it does not consider reality to be objective, qualitative research asks if the research accurately reflects the reality of the participants as reality constructors (Lincoln & Guba, 1985). Lincoln and Guba (1985) suggest two avenues and several techniques for producing credible research. The first set of techniques is concerned with collecting data in such a way that it makes it more likely that the data and subsequent theorizing will techniques is concerned with checking research findings against the perceptions of the participants; confirming the representations/results with the constructors of the reality. The data collection system for this study included persistent observation and triangulation (Lincoln & Guba, 1985). Persistent observation is related to salience making sure the researcher is describing the most important aspects and the most valued aspects of t he phenomena (Lincoln & Guba, 1985). Triangulation of sources and methods also helps with credibility by drawing similar descriptions from multiple participants using multiple research methods (in this case focus groups and think aloud protocols) (Lincoln & Guba, 1985). At the end of each focus group session, member checking was done to confirm the initial interpretations of the moderator (i.e., basic descriptive triangulation) and enhance the descriptive and interpretive validity of the data analysis ( Leec h & Onwuegbuzie, 2007; Maxwell, 2005). This was done by asking Transferability for Qualitative Aspects Qualitative researchers describe the applicability of research results to othe r research contexts in terms of transferability (Lincoln & Guba, 1985). Reality is context

PAGE 75

75 researchers do not work to prove that X will always affect Y under enumerated condi tions. They assume context will change the relationship between Y and X and even the natures of Y and X. In qualitative research, demonstrating the extent to which (Li ncoln & Guba, 1985). The second researcher must collect empirical data and then (Lincoln & Guba, 1985). research with enough rigor and detail that subsequent researchers can make accurate comparisons and determinations of transferability (Lincoln & Guba, 1985). Thus, this study was written as detailed as possible to allow for replicable results. Dependabilit y and Confirmability of Qualitative Aspects Dependability is the qualitative term used when describing concepts of consistency, which correspond to reliability and replicability. While quantitative researchers are concerned with whether the instrument mea sures a concept reliably, the qualitative researcher recognizes that different contexts and research settings always will produce variations in the data. Confirmability is the qualitative measure of neutrality, corresponding with the quantitative notion of objectivity. Neutrality is concerned with the extent to which the research results reflect the data or the researchers (Lincoln & Guba, 1985). Before data collection, qualitative researchers often write out their cultural categories assumptions about th eir personal views of the research topics in order to identify stereotypes, biases and prejudices (Ahern, 1999; McCracken, 1988). Acknowledging these preconceptions and biases is the first step in keeping them from distorting the data analysis (Pyett, 20 03). Indeed, many researchers

PAGE 76

76 say that once their assumptions have been explicated, it makes the research process easier by giving them a place from which to make comparisons to participant realities (Ahern, 1999; McCracken, 1988; Pyett, 2003). Hence, plea subjectivity statement below. Who I am in R elation to th e Re search My interest in conducting research to explain often overlooked aspects of Latino health comes from multiple perspectives. From a young age my first passion has always been to study inequalities and aim to provide solutions, such as educating others, to lessen t he occurrence of injustices. As a brother of Lambda Theta Phi, Inc ., I have experienced recent health situations face d by brothers who pledged alongside me. These experiences have elicited thought provoking questions as to whether our brotherhood and constant communication, as well as the sharing of health content we seek and find online with each other, impacts our health. My Experience, Training, and T heoretica l P erspectives Prior to pursuing a Doctor of Philosophy in Health and Human Performance health education and behavior Having a background in educating others on the import ance of health prevention has allowed me to recognize the importance of health communication in creating health behavior change I feel that my training and experiences give me multiple perspectives a nd knowledge from which to draw for the purposes of thi s and future research. Personal Bi as an d h ow it was A ddressed From personal experiences I wish to explore the possible relationship between social support, health information seeking and self efficacy to engage in health behavior

PAGE 77

77 and self efficacy to see k out health care services Implications of positive results from this study include helping designers of consumer health information systems on the Web, and health behavior specialist s with important insights into the types of health information sought by Latino male consumers and how information sought might relate to health behaviors. The study can also serve to inform health officials rendering health service resources, as to the types of servic es most needed by Latino males, among other benefits. In o rder to be completely objective, a ll focus group sessions were audio recorded, transcribed a nd analyzed a s occurred The act of philosophy was avoided by implementing s trategies such as member checks during focus group ses sions and peer review for study design and methodology Implementing such strategies prov e d invaluable to ensure validity and reliability of the study. Data Analysis IBM SPSS Statistics (version 2 2 ) was utilized for quantitative data analysis. The software ATLAS.ti (7.1.6) was used to facilitate the analysis of qualitative data. The data analysis performed in this stud y is detailed below Participant characteristics D emographics were gathered f rom f ocus groups based on participants answers to ice breaker questions which asked participants to say their names, age, major, year in school and whether they considered themselves of Latino origin This method of collecting sensitive demographic data during focus group sessions via the means of ice breakers is not a new concept and ha s been suggested by researcher Krueger ( 1998 ) when formulating focus group questions For ex ample, Owsley, McGwin, Scilley, Girkin, Phillips, and Searcey ( 2006) have utilized this method of using ice breaker questions to

PAGE 78

78 secure demogra ph ic information for their study on attitudes among older African Americans pertaining to their eye care providers. Researchers Meade, Calvo, Rivera, and Baer ( 2003 ) have also utilized ice breakers during focus group sessions to draw demographic data from participants during their study on prostate cancer screenings among Hispanic farmworkers and African American men. For think aloud protocol participants, the answered demographic items of the questionnaire (prior to refinement) were used in formulating a t able of frequencies and percentages with accordance to the the information was also post refinement Qualtric s served as the online database for the demographic items for both the survey respondents and the think aloud protocol participants. The data, for survey respondents and think aloud protocol participants w ere downloaded from Qualtrics and tables were forme d using IBM SPSS Statistics (version 22). Think aloud protocol interviews M ethods described by Van Somere n et al. (1994) and schemes followed by DeMaio et al. (1998) w ere utilized for an alyzing the interview protocols. M ethods include the recording and tr anscription of think aloud protocol interviews, as explained earlier and the coding of nuances to highlight errors in the questionnaire, misunderstanding or misinterpretation of items, and suggestions for improvem ents described by participants. A codebook with code definitions and the category for which the codes fell into was constructed. The three categories codes were placed into were as follows: E rrors in the questio nnaire M isunderstandin g or misinterpretation of items S uggestions for improvem ents

PAGE 79

79 Hig hlighted nuances easy to address, such as spelling errors and grammatical errors were quickly corrected between scheduled participants to keep from having redundan cy in highlighted nuances. Coded transcripts were used to track for those variations that we re harder and more time pressing to address, such as s uggestions for removal of items, clarification of items or rewording of questions. Chapter 4 describes in depth the findings from interview protocol analyses. RQ1: How often do CMLF use of the Internet to locate health information? To answer research question 1, a frequency table was developed depicting each of categor ies every day, several times a week, several times a month, every few months or less often The percentage of each category was also calculated. cell phone, desktop, laptop, tablet computer, smart TV, e book reader, gaming console (e.g. XBOX 360, PLAYSTATION 3, Wii, etc.), and another web enabled device was analyzed and organized int o a chart displaying frequency and percentage at home, at work, at school, or somewhere else was also analyzed where frequency and percentage of each category was organized into a chart. The data analysis methods described above mirror the methods undertaken by Rice (2 006) and Fox and Duggan (2013) by reporting descriptive statistics pertaining to frequency and percentage of both frequency of online use for health information seeking and the location most frequented when accessing OHI as well as the device most used to access OHI

PAGE 80

80 RQ2: What types of health information do CMLF look for online? To answer research question 2, two steps were taken. The first step required the analysis of focus group data to arrive at the types of health information searched by CMLF The seco nd step analyzed the survey data for created for each of the types listed in the item. The focus group sessions were analyzed using th e method described by Casey (199 8) where transcriptions are first made, followed by a simplified analysis were themes lead to development of questions that may be clarified in subsequent focus also incorporates in Chapter 3 As in Casey (199 8) once all four of the focus groups were completed they were all analyzed (in conglomerate) systematically coding different nuances and later combed through for patt erns trends and themes that aro se. A codebook was developed as suggested by Morgan (1998) where codes where grouped into families and families into resulting themes. The resul ting themes, when applicable, w ere then listed as a type of health information sought. The resulting codebook and codes with associated quotations can be found under findings in Chapter 4. The 12 types of health information sought by college students listed in the study by Easaw (2010) were introduced to the questionnaire along with types of health information sought resulting from the focus group analysis. As stated earlier, from here the collective types of health information sought were analyzed from survey data and a frequency and percentage table was developed. Such methods of c ombining qualitative and quantitative methods for an in depth analysis has been recommended by Lincoln

PAGE 81

81 and Guba (1985) forerunners in qualitative research ; and further descri bed by Trainor and Graue (2012), experts in the field of mixed methodology researc h. B oth sets of researchers claim that the addition of qualitative inquiry adds dim ension to results of the study. RQ3: What are factors associated with OHI seeking among CMLF ? Research question 3, similar to research question 2, was answered using both qualitative and quantitative methods in a series of two steps. The first step required the analysis of focus group data to arrive at additional factors that influence OHI seeking by CMLF The resulting factors were added to the item elieve The second step analyzed the survey data from the resulting formed item in the questionnaire. Frequency and an swers. This portion of the analysis mirrors those steps taken by Easaw (2010), Pe a Purcell (2008) and Livingston, Minushkin, and Cohn (2008) where factors influencing OHI seeking were researched and frequencies and percentages of factors were calculated and reported. A dditional factors were added to the item Focus group data were analyzed following the steps detailed by Casey (1998) where all the focus groups were coll ectively analyzed while systematically coding different nuances and later comb ing through for arising patterns trends and themes. A codebook was developed as suggested by Morgan (1998) where codes where grouped into families and families into resulting themes. The resulting themes, when applicable, were then listed as a factor affecting OHI seeking. The resulting codebook and codes with associated quotations can be

PAGE 82

82 found under findings in Chapter 4. Thus, as recommended by Lincoln and Guba (1985) and fu rther described by Trainor and Graue (2012) qualitative and quantitative methods were combined to form an in depth analysis of factors influencing OHI seeking and to lend additional dimension to the results for the line of inquiry. To add to the findings of research question 3, descriptive statistics were calculated for the item, Frequency and percentages of each of the selected resources us ed by respondents to access health information were tabulated. From the resources listed in the previous item, the item infor mation? Please se yzed via descriptive statistics. A frequency and percentage table was created to portray the findings of most frequented resource used by participants to access health information. This portion of the data analysis mirrored the analysis steps taken by Fox and Duggan (2013), and Percheski and Hargiatti (2011) where descriptive results were calculated and reported based on sources of health information used by survey respondents. RQ4: To what extent do attributes of OHI (source, me ssage, and content); frequency of OHI seeking, social support, age, and education predict self efficacy to engage in health behavior among CMLF ? To address the research question, f irst the data was dummy coded using the methodology described by Hardy (1993 ), where t he number of dummy variables created for the categorical variable s was one less than the number of its categories. Thus, the variables level of influence associated with attributes of OHI (source, message, and content), frequency of OHI seeking, DSSI sum age, and education (year in college) were all categorized and dummy coded to allow for multiple regression in IBM SPSS

PAGE 83

83 Statistics (version 22). Since, the dependent variable for this particular research question is continuous, the best approach t o answering the research question was to run a multiple regression with the independent variables stated above along with the dependent variable self efficacy for health behavior The variables level of influence associated with attributes of OHI in terms of source, message and content are in place to determine which of the attributes of OHI the source of OHI the message of OHI or the content (the manner and media) of OHI plays a bigger role in prompting respondents to take action. Each attribute was input independently into the regression. Each of the attributes has a 7 point Likert type what Non The data w ere transformed for each of the three attributes and t he number o f dummy variables to be created for any categorical variable is one less than the number of its categories. Thus, the number of dummy codes for each of the three attributes was 6, resulting in 18 variables inputted into the multiple regression to assess th is construct. The variable frequenc y of OHI seeking has a 5 point Likert type response option, the variable has 5 categ ories, the DSSI sum is the variable based on the composite scores of the items in the DSSI where the sum total of the instrument ranges from 11 to 33 ; higher scores signif y the

PAGE 84

84 categories where score 11 18 were low social support, scores of 19 26 were moderate social support, and scores of 27 33 were high social support. Since the variable now had 3 categories, guidelines suggested by Hardy (1993) could be applied and resulting in 2 categories, where high social supp ort was the reference category or the category not coded to prevent redundancy. The continuous variable age was transformed into a categorical variable to address the data normality. The variable was categorized into 3 categories to help with normalizing the data ages 18 20, ages 21 23, and ages 24 26; where the category ages 24 26 se rved as the reference category. The variable Year in College variables as stated by Hardy (1993), the construct resul ted in 5 dummy variables where The dependent variable Self efficacy to Engage in Health behavior was kept continuous. As explained earlier the variable is the sum of the 7 units in the construct, where each is ba sed on a continuous scale of 0 100. The sum of the scores can range from 0 700. Where the range of 0 a score range of 234 467 signifies the respondents feel they a score range of 468 700 signifies the resp ondent feel they are Although the variable can technically be categorized the true nature of the variable is continuous, and was kept as such for multiple regression analysis.

PAGE 85

85 Thus, following methods explained by Agresti and Fi nlay (2009) and Cohen, Cohen, West and Aiken (2002) to allow for multiple regression various assumptions were checked for in the data: Independence of errors (residuals). A linear relationship between the predictor variables (and composite) and the depend ent variable. Homoscedasticity of residuals (equal error variances). No multicollinearity. No significant outliers or influential points. Errors (residuals) are normally distributed. These assumptions allow for one to (1) provide information on the accurac y of the resulting predictions, (2) test how well the regression model fits the data, and (3) determine the variation in the dependent variable explained by the independent variables Independence of residuals was assessed via the use of the Durbin Watson statistic, where a value of approximate to 2 indicates that there is no correlation between residuals. To normalize the data and allow for Durbin Watson statistics closer to two t he bootstrapping technique along with the dummy coding of categorical and c ontinuous independent variables w ere conducted The dummy coding took place as explained above. The bootstrapping function was added to the analysis, where several simulated sample sizes were tested (1000, 3000, 5000 and no bootstrapping) and the most favo rable method was selected. For this particular case, the bootstrapping function with a 3,000 simulated sample size was selected.

PAGE 86

86 The dependent variable, Self Efficacy for Health behavior was transformed to address slightly negative skew ed ness in the data. SEBehManLGTRANS Efficacy for Health behavior was c alculated on SPSS by applying the following formula for each of the recorded responses Efficacy for Health behavior Log 10 ( 701 Self Efficacy For Health behavior ). In the previous formula Efficacy for Health behavior the number 1 to the response. The data is normalized by applying the above formula at eac Efficacy for Health behavior Th e previous function served to transform the data along the variable Self Efficacy for Health behavior These nuances bootstrapping and transforming the data, were addressed following the guidelines by Lund and Lund ( 2013 ). An assumption of multiple linear regression is that the independent variables collectively are linearly related to the dependent variable and also that each independent variable is linearly related to the dependent var iable (Agresti & Finlay, 2009 ; Cohen et al., 2002 ) To test for a linear relationship studentized residuals were plotted against the ( unstandardized ) predicted variables to see if the plots formed horizontal bands which would suggest a linear relationship according to Lund and Lund (2013) in their guide on multiple regression. To aid in understanding the procedure above the following is a brief definition of studentized residuals. In statistics, a studentized residual is the quotient resulting from the div ision of a residual by an estimate of its standard deviation (Cook & Weisberg, 1982) Typically the standard deviations of residuals in a sample vary greatly from one

PAGE 87

87 data point to another even when the errors all have the same standard deviation, particul arly in regression analysis; thus it does not make sense to compare residuals at different data points without first studentizing (Cook & Weisberg, 1982) It is a form of a t statistic, with the estimate o f error varying between points. Thus, studentizing is an important technique in the detection of outliers (Cook & Weisberg, 1982) Studentizing is named in honor of William Sealey Gosset, who wrote under the pseudonym Student, and dividing by an estimate of scale is called studentizing, which is in simila rity with standardizing and normalizing (Cook & Weisberg, 1982). Partial regression plots were also plotted to check for a linear relationship among the dependent variable and independent variables. As suggested by Lund and Lund (2013) and Cohen et al. (20 02 ) after checking for a linear relationship, homoscedasticity was then tested by taking another look at the plots of the studentized residuals and unstandardized predicted va lues ensuring that the data point s were equally spread a m ong the variables. Multi collinearity occurs when two or more independent variables are highly correlated with each other. This leads to problems with understanding which variable contributes to the variance explained and technical issues in calculating a multiple regression model (Morrow Howell, 1994) Thus, checking for multicollinearity is suggested by Cohen et al. (2002). There are two stages to identifying multicollinearity: inspection of correlation coefficients and Tolerance/VIF values The correlations table was checked to see that none of the independent variables had correlations greater than 0.7. In the coefficients table, the tolerance was checked to see that all variables

PAGE 88

88 were greater than 0.1. Both stages were performed for this research question according to Lund and Lund (2013), as explained above. From here the data wa s checked for outliers through casewise diagnostics as explained by Lund and Lund (2013) where cases larger than 3 standard deviations are prompted. Studentized deleted residuals were also checked f or cases with more than 3 SDs or less than 3 SDs. Outliers were removed. Following this analysis, leverage points were checked, where values less than 0.2 were considered safe to the regression (Lund &Lund, 2013). Influential points were also checked by c onsidering ; values over 1 would require investigation, however influential points over 1 were not encountered. In order to be able to run inferential statistics (i.e., determine statistical significance), the errors in prediction the residuals need to be normally distributed (Cohen et al., 2002 ; Lund & Lund, 2013 ) According to Lund and Lund (2013) a commonly used method for checking normality of residuals is the use of a histogram with a superimposed normal curve and the use of a P P plot. This method was utilized for this analysis where the histogram and the P P Plot were produced by the use of the software, IBM SPSS Statistics (version 22). The histogram was checked to make sure it followed the normal distributio n line and double checked by observing the produced P P Plot to ensure that the observations of the residuals follow the diagonal line as closely as possible. These methods were pursued in a related study performed by Walsh, Hyde, Hamilton, and White (20 12) OHI Once all

PAGE 89

89 assumptions were checked and met, for this study, the interpreting and reporting of the findings pertaining to the multiple regression commenced. RQ5: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social support, age, and education predict self efficacy to seek out health care services among CMLF ? To address the rese arch question all aspects of the data analysis were kept constant to what was done for the previous research question. The dependent variable was changed to self efficacy to seek out health care services. The variable is the sum of the items assessing sel f efficacy to seek health care services. As explained earlier in Chapter 3, the variable self efficacy to seek health care services is composed of 5 units, each on a sliding scale of 0 100, where the addition of scores range from 0 through 500 and scores o f 0 333 signify the variable was kept on a continuous scale, although interpretation of scores c an take place as stated above, the higher the score the more confiden ce a respondent expresses. The independent variables were dummy coded as stated for the previous research question. The data was analyzed using several bootstrapping options (1000, 3000, 5000 and no bootstrapping). The best option for this particular multiple regression was a simulated sample size of 1,000 w ithout the use of a transformed dependent variable, since the data was not skewed. From there t he assumptions of lin earity, indepen dence of errors, homoscedasticity, unusual points and normality of residuals were checked for and met as suggested by Agresti and Finlay (2009), Cohen et al. (2002), Morrow Howell (1994) and Lund and Lund (2013) and explained above for the previous researc h question. Once all assumptions for running a multiple regression were met the

PAGE 90

90 reporting and interpretation of results commenced. As previously stated, t hese methods were also pursued in a related study by Walsh et al. (2012) where predictive modeling ex OHI was attempted. P lease refer to Table 3 3 for a summarized description of data analysis plan to answer each of the 5 research questions. Apriori Analysis To determine the number of participants required to reach adequate power for data anaylses, an apriori analysis was conducted. Statistical significance is the likelihood that a finding or a result is caused by something other than just chance. Usually, this is set at l ess than 5% probability (p< 0.05), meaning that the result may be produced by chance no more than 5% of the time ( Bartlett, Kotrlik, & Higgins 2001). Thus, for the quantitative portion of the study, the researcher set the alpha level a priori at .05. The congregate population of enrolled Latino college students in the investigated Florida schools is 74,753 (Forbes.com, 2012). The data type for the quantitative portion of the study is continuous. According to Bartlett, Kotrlik, and Higgins (2001) both data type (categorical or continuous) and population size should be considered prior to data (1977) sample size formula for continuous data, n_o=((t)^2x(s)^2)/(d)^2, to dete rmine the sample size needed for this study. Where t value for selected alpha level is .025 in each tail = 1.96, the alpha level of .05 indicates the level of risk the researcher is willing to take that true margin of error may exceed the acceptable margin of error. Where s estimates standard deviation in the population=1.167. For example, the variables level of influence on attributes of OHI (source, message, and content) are all based on a 7

PAGE 91

91 point Likert scale, we can calculate the sample size needed base d on the following. The coefficient s will be the estimate of variance for a 7 point scale calculated by using 7 (inclusive range of scale) divided by 6 (number of standard deviations that include almost all (approximately 98%) of the possible values in th e range). Where d=acceptable margin of error for mean being estimated=.21. Calculated by multiplying number of points on primary scale x acceptable margin of error; points on primary scale=7; acceptable margin of error=.03 (error researcher is willing to a ccept). Thus, we get, n_o = ((t)^2 x (s)^2) / (d)^2 = ((1.96)^2 x (1.167)^2) / (7x.03)^2 = 118.63. At an alpha level ( ) of 0.05 and margin of error of 0.03 for continuous data for a population of over 10,000 indicate s that a convenience sample of 119 Lati no male fraternity college students is needed for the study. Summary Chapter 3 provided reasons decision to pursue a study at the selected locations. Chapter 3 also gave the reader a detailed description of the participants for the current study. R ecruitment strategies steps taken to ensure confidentiality and instrumentation were discussed in Chapter 3 subjectivity statement was als o given and d ata collection procedures for focus group sessions, pre testing, an d survey implemen tation where also discussed Chapter 3 concluded with an explanation of the data analysis steps for the research at hand. Chapter 4 will discuss the results of the data collected from the focus groups, the think aloud protocols, and the su rvey.

PAGE 92

92 Table 3 1 List of college campuses investigated in Florida location, student population and percentage of Latino students School Location Enrolled Latino Student Population (%) University of Florida Gainesville, FL 50,116 6,855 (1 3.7 ) Universit y of Central Florida Orlando, FL 50,968 7,264 (14 .2 ) Florida International University Miami, FL 44,616 29,446 (66) University of South Florida Tampa, FL 30,963 6,881 (14.9) Nova Southeastern University Fort Lauderdale, FL 28,457 9,391(33) Embry Riddle Aeronautical University Daytona Beach, FL 4,597 460 (10) University of Miami Miami, FL 10,237 2,731 (26.7) Florida State University Tallahassee, FL 40,838 6,125 (15) Florida Atlantic University Boca Raton, FL 26,857 5 102 (19) Barry University Miami Sh ores, FL 4,698 498 (10.6) The source for the above information came from Forbes.com (2012). Table 3 2 List of Lambda Theta Phi Chapters in Florida and their related s ector affiliation. Chapter School Location Sector Number of current active members Phi University of Florida Gainesville, FL I 12 Alpha Alpha University of Central Florida Orlando, FL I 43 Alpha Kappa Florida International University Miami, FL II 6 3 Alpha Psi University of South Florida Tampa, FL I 72 Beta Gamma Nova Southeastern Uni versity Fort Lauderdale, FL II 16 Beta Delta Embry Riddle Aeronautical University Daytona Beach, FL I 15 Beta Theta University of Miami Miami, FL II 17 Beta Iota Florida State University Tallahassee, FL I 35 Beta Xi Florida Atlantic University Boca Rat on, FL II 1 9 Gamma Xi Barry University Miami Shores, FL II 1 3

PAGE 93

93 Table 3 3 Summary of data analyses used in answering each research question. Research Question Statistical Analysis RQ 1: How often do CMLF use of the Internet to locate health information ? Frequency and percentage tables were developed for each item under the construct. The analysis mirrors those performed by Rice (2006) and Fox and Duggan (2013) where descriptive statistics were used to assess questions on health information seeking. RQ 2: What types of health information do CMLF look for online? Focus groups were performed and transcript data were coded and analyzed following methods done by Casey (1998). The results from focus group data led to the addition of answer choices to the questionnaire items assessing the types of health information searched by participants. Think aloud protocol interviews were performed to pretest and refine the questionnaire prior to survey implementation. The questionnaire items assessing types of health information searched were surveyed and results were calculated using IBM SPSS Statistics (version 22). The results were reported using frequency and percentage tables. These methods are in line with those steps taken by Easaw (2010) in her analysis of wom for OHI RQ 3: What are factors associated with OHI seeking among CMLF ? Focus groups were performed and transcript data were coded and analyzed following methods done by Casey (1998). The results from focus group data led to the addition of answer choices to the questionnaire items assessing factors that influence OHI seeking among CMLF Think aloud protocol interviews were performed to pretest and refine the questionnaire prior to survey implementation. The questionnaire items assessing fac tors that influence OHI seeking were surveyed and results were calculated using IBM SPSS Statistics (version 22). The results were reported using frequency and percentage tables. These methods are in line with those steps taken by Easaw (2010) in her analy for OHI These methods are also in line with the works done by Pea Purcel l (2008) and Livingston et al. (2008) in which factors that influence OHI seeking were previously researched among other population groups.

PAGE 94

94 Table 3 3. Cont inued Research Question Statistical Analysis RQ 4: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social support, age, and education predict self efficacy to engage in health behavior among CMLF ? Internal co nsistency reliability measures were calculated based on collected survey data following methods explained by Nunnally and Bernstein (1994) and were facilitated by the use of IBM SPSS Statistics (version 22). To seek statistically significant predictors amo ng the independent variables in terms of the dependent variable, multiple regression was pursued, similar to the statistical analyses performed in a related study by Walsh et al. (2012). The variables used in the multiple regression were dummy coded as exp lained by Hardy (1993). The data was bootstrapped and the dependent variable was transformed following suggestions by Lund and Lund (2013) and statistics consultant, Dr. Hui Bian. From there the assumptions of linearity, independence of errors, homoscedast icity, unusual points and normality of residuals were checked for and met as suggested by Agresti and Finlay (2009), Cohen et al. (2002), Morrow Howell (1994) Once assumptions were met, interpretations commenced. All statistical analyses were done using I BM SPSS Statistics (version 22). RQ 5: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social support, age, and education predict self efficacy to seek out health care services among CMLF ? To seek statistic ally significant predictors among the independent variables in terms of the dependent variable, multiple regression was pursued, similar to the statistical analyses performed in a related study by Walsh et al. (2012). The variables used in the multiple reg ression were dummy coded as explained by Hardy (1993). The data was bootstrapped following suggestions by Lund and Lund (2013) and statistics consultant, Dr. Hui Bian. From there the assumptions of linearity, independence of errors, homoscedasticity, unusu al points and normality of residuals were checked for and met as suggested by Agresti and Finlay (2009), Cohen et al. (2002), and Morrow Howell (1994) Once assumptions were met, interpretations commenced. All statistical analyses were done using IBM SPSS Statistics (version 22).

PAGE 95

95 CHAPTER 4 FINDINGS Organization of Findings The findings are organized by first giving a summary of participant characteristics for focus groups, think aloud protocol interviews and for survey respondents. Findings pertainin g to think aloud protocol interviews and internal consistency reliability testing are then presented, followed by findings pertaining to each research question. Participant Characteristics Participants for all phases of this investigation were recruited b etween July 2013 and October 2013. As discussed in Chapter 3, part icipants for the various methods in this study were recruited through secure emails containing informed consent forms and either links to scheduling pages or a direct link to the online ques tionnaire. The primary investigator made initial contact through email to invite individuals to participate and to explain the general purpose of the study. The selected men were between the ages of 18 26. All the men in the study had searched the Internet for health information sometime in the preceding 12 months. Pseudonyms were used in transcribing and during write up of results to protect participant identities. Other identifying information within quoted material has been removed or has been replaced w ith a bracketed phrase identifying what was replaced. Pa rticipant Characteristics for Focus Groups The primary investigator conducted four focus groups. The target focus group size ranged from eight to twelve participants. Focus groups were over sampled by 10 to 20

PAGE 96

96 percent to ensure sufficient participation and attendance (Doyle, 2009). A total of 41 men participated in the four groups. Demographic characteristics of focus group participants are described in detail in Table 4 1. All participants were male and enrolled in college. The first focus group session w as held at the Institute of Hispanic Latino Cultures at the University of Florida The location is affectionately named La Casita, and This location presented the ideal setting to conduct a focus group session due to its secured for hosting the first focus group. Focus group I included only participants from the University of Florida. The first focus group had 11 participants attend on July 22, 2013. Focus groups 2 room GC314 located at 11200 SW 8 th Street, Modesto A. Maidique Campus, Miami, Florida, 33199, on August 5, 2013, August 12, 2013 and August 19, 2013. Room GC314 offered the perfect meeting for the focus groups due to its privacy and convenient location for FIU students. The ove rwhelming majority of focus group participants (93%) considered themselves to be of Latino or Hispanic heritage. Focus group II had 9 attendees, focus group III had 10 attendees and focus group IV had 11 attendees. Participant Characteristics for Think a loud Protocols To guide questionnaire item modifications and to pretest the questionnaire, think aloud protocols were conducted by the primary investigator. A total of 8 think aloud protocols were conducted on August 31, 2013. Demographic characteristics o f think

PAGE 97

97 aloud protocol participants are described in detail in Table 4 2. All participants were male and enrolled in college. The think aloud protocols were conducted in the Richter Library, Rm. 125, located at 1300 Memorial Drive, Coral Gables, FL 33146, which offered both convenience and privacy for participants. The mean age for participants of think aloud protocols was 20.50 years (SD 1.31). All participants identif ied as being of Latino, Hispanic, or Spanish origin. Two participants identif ied with mor e than one Latino group origin. Of the eight participant s, all selected White as their race and two of the eight also selected Black or African American as their race. For academic year participants selected Sophomore ( n =3), Junior ( n =2), and Senior ( n =3) In terms of selected category of major/program of study, the selection was similar among participants ; however, three participants selected physical/biological sciences, making this selection the most selected major In terms of relationship status, thre e participants were single and not in a relationship, two participants were single and dating casually, two were single, dating one person exclusively, and one participant refused to answer. In terms of living situation, the majority of respondents ( n= 5) w ere living off campus with roommates, while three participants were living on campus with roommates. Participant Characteristics for Survey Respondents Table 4 survey respondents was 20.6 5 years (SD 1.43). Only one respondent preferred not to disclose their gender, while all other respondents ( n= 126) selected male as their gender. The colleges/universities with the greatest number of respondents were the University of South Florida ( n= 46), Florida State University ( n =23), and the University of Central Florida ( n =21). Out of 127 participants, only a small number ( n =12) considered themselves not to be of Latino, Hispanic, or Spanish origin. The category of Latino

PAGE 98

98 origin with the greatest numb n= 37). For this item, participants could select more than one of Latino, Hispanic, or Spanish origin, if they n =110) more frequently than other optio ns. Also, for this item, participants were allowed to select was the next most frequented selection ( n= 31). As for academic classification, the majority of respondents ( n = 51) were sophomores, followed by juniors ( n =38) seniors ( n= 21), and freshmen ( n= 13 ) among others. In terms of category of major, the most selected category was Engineering ( n= 34), followed by Health Science/Nursing ( n= 23), Business ( n= 22), and Physica l/Biological Sciences ( n= 20), among others. As for relationship status, a great majority of respondents selected either single, dating one person exclusively ( n= 48), single, not in a relationship ( n= 39) or single, dating casually ( n= 35). Finally, for livin g situation, survey n= off n= campus with n= 13), among others. Think aloud Protocol Interviews T hink aloud protocol interviews along with qualitative data from the focus groups allowed for refining the questionnaire prior to implementation. Table 4 4 depicts categories, associated codes, and their definition of nuances that arose during think aloud p rotocol interviews. The following are selected statements pertaining to the coded themes.

PAGE 99

99 Errors in the Questionnaire Errors with wording or spelling in the questionnaire were identified by think aloud protocol participants. Participant Ethan identified an error in the questionnaire wording when stating: number, but you can drag and select a number. I think it should it say The error Ethan highlighted helped to improve the wording of the particular question. Participant Ben helped to reduce errors in the questionnaire wording by bringing the following to the attention of the researcher: Please indicate fo r each of the statements the extent to which they apply to your situation. The way you feel now. Please circle the appropriate answer I can click and choose one Additionally, slight errors in grammar and wo rding errors in spelling were also errors, along with other minor errors were addressed prior to survey implementation. Misunderstanding or Misrepresentation of It ems Items that could be misunderstood were also highlighted during think aloud protocol interviews and addressed prior to survey implementation. Participant Ethan stated: I know if I know where I can find the information about my local health department so that I can contact them. to inquire if participan ts feel confident that they know where to find information on how to contact their local health department.

PAGE 100

100 the difference between graduate and professional? I think that address this issue, examples of graduate degrees and professional degrees were added to the answer choices to provide direction to survey respondents during survey implementation. Several participants indicated difficulty with unders tanding an item inquiring about income. Participant Henry stated: Participant Joe also shared similar misunderstandings by stating: the school semester to the end of the school year, or is it from when I filed ta xes to the following year that I have to file taxes. Due to the many different interpretations of the question by respondents and since a majority of think aloud protocol participants shared similar misunderstandings of the question, the item was removed d ue to the misinterpretation that may take place if results of the particular item were analyzed. Socioeconomic status of respondents was of high interest to investigate but realizing that students sometimes take out loans, work to pay for school, borrow mo ney from parents or have a combination of economic circumstances to finance school; accurately accessing socioeconomic status from a single item in this population would present a challenge. As such future research focusing solely on the socioeconomic stat us of CMLF would be of much interest to pursue.

PAGE 101

101 Suggestions for Improvements Several suggestions were given by participants. Suggestions were considered and changes were applied to improve the overall understanding, wording, and format of the questionnair e. Participant Sabastian stated: sort of weird. Why would I look up birth control or for condoms? Maybe you should think about another word or something for this one. In depth consideration was taken at the end of interviews to address how the subject of birth control and safe sex practices could be reworded to be better understood by male respondents taking the questionnaire. Thus, to address the issue, the items were combined The realization that participants could identify with more than one race was a mishap that was quickly corrected when a participant suggested the ability to select multiple races. Participant Sabastian also later stated: ity than one race. Maybe that should be added in there. The ability for respondents to select multipl e race and ethnicities was made possible by changing settings in Qualtrics to allow for multiple selections of response options. demographic information related to housing and m arital status. The following

PAGE 102

102 describe your living situation? I live off campus with roommates, but I think there should ion was added to the questionnaire. Thus, pretesting by means of think aloud protocol interviews allowed for the refinement and overall improvement of the questionnaire prior to survey implementation. Internal Consistency Reliability Analysis The questio nnaire was employed to measure different underlying constructs One construct, measured using DSSI consisted of 11 questions. The scale had a high level of internal consistency, as determi ned by a Cronbach's alpha of 0.794 The construct, alpha of 0.690. As previously stated, Nunnally and Bernstein (1994) suggest that a eliability coefficient of .70 and greater is an acceptable reliability statistic for a scale that is newly developed. Efficacy to Engage in Health behavior questions and had a high level of internal consistency with a strong level of internal consistency, again, as determined 0.736. Table 4 examined.

PAGE 103

103 Research Questions The main study findings are divided into 5 sections, each of which corresponds to a research question. RQ1: How often do CML F use of the Internet to locate health information? A total of 127 respondents, whose characteristics were described earlier, answered the item on the extent of online use. According to the data, depicted in Table 4 6, the majority of respondents (44.1%; n = 56) go online several times a month to search for health information. A lesser percentage, 25.2% ( n= 32), go online every few months to search for health information. Table 4 7 depicts the findings of the device most used when accessing the Internet. For the item assessing device most used when accessing the Internet, approximately 41% ( n= 52) selected that they use a laptop. Tablet computers were the second most used device for accessing the Internet, with 22.8% ( n= 29) selecting this option. Cell phones we re the third most used device for accessing the Internet, with 21.3% ( n= 27) selecting this option. Table 4 8 depicts the place respondents mostly find themselves while online. A majority of respondents (46.5%; n= 59) indicated browsing online while at home while 29.1% ( n =37) browse online while at school. Browsing online at work (22%; n=28) was the third most frequently reported response. RQ2: What types of health information do CMLF look for online? Focus groups and survey results guided the findings of this research question. Over 184 distinct codes arose during focus group analyses. Object 4 1, below, depicts the definition of these distinct codes. Object 4 1 Definition of codes (.pdf file 436 KB)

PAGE 104

104 C odes arising in focus group analysis and their associated quotations are depicted in Object 4 2. Over 98 pages of codes with associated quotations arose from the focus group data. Object 4 2 Codes wi th Associated Quotations (.pdf file 349 KB) Object 4 3, depicts the themes and associated codes that arose from focus group data analysis. Object 4 3 Themes with Associated Codes (.pdf file 57.3 KB) For focus groups, the main health information items that study participants their own symptoms, diagnoses, physical appearance (weight concerns and exercise, diet, nu trition), natural living (organic diet, natural products), and treatments for conditions or diseases. Additional types of health information searched included monetary issues (e.g., insurance coverage management), immigration (exams needed for obtaining a green card), home remedies, mental health issues, sexually transmitted diseases, family planning (pregnancy and birth control), and curiosity triggered by interpersonal or mass media content. Informational needs of others Health information was sought on line by Latino men in college over concern of family and friends. Such instances included searching for health information online for usually search online for my fam OHI concerning family and a significant other: jungles for zip lining.

PAGE 105

105 Health information was also sought for family members who request assistance from their children to search for such information. According to participant Carlos, et helps Another Latino male, participant George, sought information for his mo ther concerning a specific condition: This is a little embarrassing but earlier today I got a text from my mom. She asked me to help her look online for any info I find on uterine fibroids. The doctor told her [that] hers were the size of cantaloupes. She wants me to Online searches were also prompted by Latino male students in concern for siblings as is demonstrated by Peter: Most of the time my brother is always getting sick. He has asthma and g ets bronchitis almost all the time. Last Christmas he had pneumonia for 3 weeks. My mom too she has high cholesterol and gets headaches daily. Since my boss gave me a Samsung Galaxy tablet for work use for doing homework and I look online t o see if there is anything that can help my mom and brother when they get sick. Symptoms Seeking OHI was also prompted by a majority of CMLF with concern for different types of physical symptoms linked to suspected illnesses or conditions. Participant Kirk had specific concerns for a potential heart attack and searched the Internet for related symptoms: symptoms. I see palpitations and chest pains are both symptoms and warning sign s. So I freak out whenever I start to feel any type of pain in my chest.

PAGE 106

106 Another example for searching OHI related to a specific type of symptom involved searching for symptoms associated with using a prescription drug administered for individuals who suff er from hypothyroidism, as expressed by participant Alvin: I know if I look for symptoms of Synthroid symptom if I start reading the random blogs that I find online. I guess if you farfetched. Other searches included seeking information for symptoms related to, for example, a sore throat. An example is participant Aaron: Additional online searches included searching for symptoms specific to pain and discomfort. Participant Adam remarked: When I spoke to my co worker she said she had the same thing several months before she told me I should look to see if I have carpal tunnel syndrome. I researched the net to find out if I had some of the symptoms. I thought I had carpel tunnel for sure. Diagnoses Latino col lege men also sought health information online concerning specific diagnoses or used the Internet as a medium for self diagnoses. Participant Diego Six years ago my mom got di agnosed with pancreatic cancer. She was the school principal at ( local elementary school ]) Everyone loved my mommy. place really. As mentioned previously, OHI served as a medium for self diagnosis. Participant online searches resulted in, for example, self diagnosis of hernia and mouth sores. Participant Marvin made a self diagnosis of conjunctivitis:

PAGE 107

107 This Saturday I looked online to see if I had pink eye. I mean it was all red and bloodshot. I was wondering if eye drops might help. So I G oogled pink eye Physical appearance and weight concerns Other searches online concerning weight loss also involved seeking gyms via the Internet and searching for websites with diet programs. An example of this is shown in Participant Wayne who was anxious to lose weight and remarked: ; I even went to nutrisystem.com. I just want to know which program might be best for me. I got membership to LA Fitness but I feel exercise alone is not helping me lose weight. Other concerns which prompted online searches concerning physical appearances are exemplified by participant Kirk who uses the Int ernet to search for body building Natural living For the focus group studies, some participants also expr essed searching for OHI for organic foods and using social media as a tool to advertise the benefits of organic foods. Participant Diego remarked: Me and some friends organized a rally to educate others about organic foods and being vegan. The farmers mark et on (major street in the city) wanted for us [to] tell others how being vegan helps with being fit and healthy. We used both Twitter and Facebook to advertise for it. We also uploaded a short video telling others of the free food and giveaways. I hope th Another participant, Jeff, had a change in lifestyle when his mother was diagnosed with cancer and expressed concern related specifically to food preservatives. Jeff remarked:

PAGE 108

108 I just wanted to know everything and anything rela ted to pancreatic cancer, especially about treatments. The biggest thing that I ran into was that changing your diet can really help you avoid a bunch of cancer causing things that you find in some foods. BHT, BHA and nitrates all of these can cause cance r. A bunch of major food companies put these preservatives in A majority of the participants who commented on leading a healthy lifestyle through consuming not only organic foods but also using organic hygie ne products commented on using the Internet to research such produce and products. The overall Treatments for condi tions or diseases Searches also included seeking information for specific diseases and conditions for the participant themselves or for others. Participant Brad commented on searching for information related to pancreatic cancer. He searched online for inf ormation Another participant, Humbert, commented: One thing I wish got better is my se izures. I want to get off of my I feel it makes me way too drow sy in the mornings. So I try to take it mostly at night even though the thing do that. Lately atments to my seizures. I found some things on this web page called We bMD The site has a bunch of links to everything. Participants also shared similar thoughts as it related to searching online for treatments and discussing such treatments with their healthcare providers. Participant Dan exemplified this remarking: I think that you can balance going online for health information with going to the doctors. I know some doctors that encourage you to help in your

PAGE 109

109 treatment by looking online for health information that may seem beneficial to you. Some work with you to get to exa ctly what the problem might be. Monetary issues Online health information searches also included concern for monetary issues. The specific type of issues searched for online vastly included health insurance. One representative example of those who mentione d health insurance as a type of monetary issue was participant Sam, where he remarked policies due to healthcare provider request s. For example, Participant Peter remarked: Overall, participant now we have no insurance and hardly an health insurance associated with monetary concerns. Immigration Another type of health information sought online included concerns of immigration status. This is exemplified by participant Carlos, where he remarke rather use the I Participant Carlos went on to remark Last time I check ed for health on the I nternet was yesterday. I was seeing how to get a healt h check and what I Home remedies Health information concerning health remedies was also sought on the I nternet. Specific types of home remedy searches online were related to searching for injuries or discomfort. Participant Eric

PAGE 110

110 Stan, I l ike home remedies, [they] may be for Curiosity and clarification through social media Another prevailing theme that certain diagnoses or for gaining greater understanding of health information in general, from social media sources. P articipant Mike used Twitter and Facebook as a means to elicit responses: I tweeted a picture of my nosebleed on Twitter and Facebook the other day and got a couple comments from some of my high school friends on how they had the same thing happen to them. One participant also commented on other specific types of social media sources, which may be used as a tool to provide health information and clarification for health based issues. Participant Kenny remarked: As a sociology major tha happening because of Facebook, Tw itter, Skype and those. Can you imagine if all the brightest people got together by these m eans for helping others? I definitely think the I nternet is helping all of us with our health. Thus, the findings from the focus group s clarification of health issues through social media. Quantitative findings of types of OHI sought Table 4 9 depicts the frequency and percentage of types of health information sough t by survey respondents. Only valid percent is reported for the following types of health information sought by participants. For health information sought for others, 55.1% ( n= o nline. For concern over symptoms, 50.4% ( n= 64) of respondents sought this type of

PAGE 111

111 health information online In regards to a recent diagnosis, 73.0% ( n= to searching for this type of health information online For exercise, 62.2% ( n= 79) of online Concerning searching for health information online related to nutrition, 64.6% ( n= 82) of ealth information sought online, 30.7% ( n= For a certain medical treatment or procedure, as a type of health information sought, 77.2% ( n= 98) of respondents did not search for OHI related to medical treatments or procedures As for a specific disease or medical problem as a type of health information sought, 70.6% ( n= A few respondents, 17.3% ( n= online for informatio n on immigration health concerns. For prescription drugs as a type of health information sought, 33.1% ( n= the Internet for this type of health information. Pertaining to over the counter drugs, 30.4% ( n= 38) o the Internet for this type of health information. As for home remedies as a type of health information sought, 44.9% ( n= type of health information sought, 92.1% ( n= of health information. Approximately 33% (n=42) of participants indicated searching online for information on mental health issues. As for family planning as a type of OHI sought, 89.0% ( n= sexually transmitted diseases, as a type of health information sought online, 93.7%

PAGE 112

112 ( n= insuranc e options and concerns, as a type of OHI sought, 27.6% ( n= 35) of respondents OHI As for exploring online posts, tweets or videos on health, 48.0% ( n= this type of health information. Finally, in regards to concerns on weight or physical appearance, as a type of OHI sought, 33.1% ( n= RQ3: What are factors associated with OHI seeking among CMLF ? As detailed in Chapter 3, focus gr oup data analysis and survey results lead to the findings for RQ3. For focus groups, the main factors mentioned that influence OHI seeking among study participants were family and loved ones, monetary issues (e.g., insurance coverage management), fear of illnesses, and physical appearance. Family and loved ones Family members and significant others were found to be a factor which emerged from the focus group discussions that influences OHI seeking among CMLF One p diagnosis of breast cancer. He remarked: My mom was also diagnosed with breast cancer. She had taken a routine invasive, stage 0 breast cancer in both breasts. I remember my mom crying, and my dad I look online to see what the chances are that my mother might still get si ck. I hope God always keeps watch over her. Participants also expressed concern over potential genetic predispositions based on their family history. Participant Jose, commented:

PAGE 113

113 The biggest thing that scares me is if I get a stroke like what happened to m y gramps. My family in Guatemala said that he died from a stroke in his kid. I think I might have been four or five years old when he passed. So even though I look for things to lower m y blood pressure I also watch Yoga and fitness videos on YouTube. I have a smart TV so I can just search for things in my living room and do any of the exercises I see right then and there. A further concern which emphasize the importance of family which p rompts Latino men to search online is exemplified by participant Steve. He remarked: online to see if I can help. My brother just had another baby girl this summer. Back in early Jun broke in the morning o that night. My little niece ended up getting a meconium infection because the doctors took too long to decide to induce her with Pitocin Quite frankly, searched them online. Significant others were also a factor that prompted participants to seek health information. One participant, Adam, expressed concerns about the c ravings of his pregnant girlfriend and about her general health and the health of the unborn child. turns into a witch bugging. I also want to make sure the baby is healthy, so I search to see if my girl should be taking vitamins or avoiding some foods. Monetary issues Concerns regarding monetary expenses appeared to be another factor, which prompted p articipants to search for health information online. Participants searched online for less expensive alternatives to prescription drugs, health insurance (as earlier mentioned), and in some instances, monetary issues led participants to neglect care or

PAGE 114

114 to self diagnose. The result of self diagnosis was exemplified by participant Byron, who claimed: Remember for me to look up what probably happened to my finger than to go to some [doctor office for them to tell me the same thing I found online. I save time and money that way. Monetary expenses related to family was also expressed as a factor since some participants stated health insurance was unaffordable to them due to taking up finan cial responsibilities for their family. Participant Sam remarked: Yeah that would be easier if I had some money. My mom got laid off back in 2011 works right now her paying the bills. Her unemployment check got renewed for a few more months threatening to take that away from her for months now. So for now we have no insurance and hardly any money. For some participants, monetary issues were int imately linked with immigration status. Participant Peter commented: My mom thinks that onions, honey and vinegar cures everything. I want to buy insurance part time paycheck until I get my papers. An additional example of concerns for the cost of health insurance as deemed that I have to pay Fear of condition, disease or disorder Participants were also prompted to search for health information online due to personal fears of a particular condition or diagnosis for themselves and others. Specifical ly, participants feared the severity of their health issues, a fear of cancer and

PAGE 115

115 other illnesses, as well as a fear of surgery. Participant Marcus expressed concern for The thing is that my younger sister is now 16. She was 13 back then so it line, at the Others expressed fears of speaking to a doctor as participant Carlos, commented: Google, I think upcoming surgical procedures prompted participant Ray to search online. I searched for what a bone spur is. I found surgery videos on YouTube. They were nasty. I also found some links on Google that I glanced at. Finding these things make me nervous about my upcoming surgery. Physical appearance Another factor that prompted Latino college men to search health information online involved body image and weight iss ues. Online searches included diet, bodybuilding, lifestyle changes, and overall physical fitness. Participant Wayne reported that his main purpose for searching OHI issues and his weight loss: I think she [mother ] also has heart problems die young ; online these days I try looking for weight loss in search for ways to get a lap band.

PAGE 116

116 One participant exclaimed concern about his significant other who was dealing with body image issues. Participant Simon states: g irlfriend] is going to therapy for her eating disorder with a specialist. I looked to see how long the usual treatment lasts. This Quantitative findings of factors relating to OH I seeking among participants. Table 4 10 describes the frequency and percentage of factors that influence OHI (n=73) selected very influential The second most selected c 33.1% ( n= 42) of participants selecting this option. Only 7.1% ( n= 9) of participants n= 53) of n= n= 16) selected n= as an answer choice while 1.6% ( n= non n= 20) of participants n= 28.3% ( n= % ( n= 22) of participants n= 4) of participants selected 26.0% ( n= n= 44) selected n= accordingly 11.8% ( n=

PAGE 117

117 ( n= n= n= 62) of participants selected n= n= n= ( n= n= 2) selected 11.1% ( n= 14) of re n= 26) selected n= n= 22) of n= 15) selected and 14.3% ( n= ( n= physical appearance as a factor influencing OHI seeking. Table 4 11 depicts the frequencies and percentages of resourc es visited by majority 86.6% ( n= respondents 69.3% ( n= source was used relatively evenly with 57.5% ( n= n= 54) en it comes to n= 114), n= is resource for

PAGE 118

118 n= 84) of n= n= 84) of respondents answe seeking this resource for health information while 33.3% ( n= 42) of respondents n= 49) to using this resource for health information Finally, for 40.9% ( n= n= 75) of respondents Table 4 12 port rays the frequencies and percentages of the resources said to be line information by 37.8% ( n= 48) o the second most frequently selected option with 22.8% ( n= 29) of respondents choosing next most chosen with 11.8% ( n = 15) and 9.4% ( n= 12), respectively choosing these options as a resource most frequented to acquire health information. Among other (4.7%; n= 6) (3.9; n =5) line (3.1%; n =4) (1 .6%; n =2 ) were selected by respondents. RQ4: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, social support, age, and education predict self efficacy to engage in health behavior among CMLF ? A multiple regressi on with the bootstrap function of 3,000 simulated sample was ru n to predict self efficacy to engage in health behavior from source, message content, frequency of OHI seeking, DSSI (level of social support), age, and education (year in

PAGE 119

119 college) The assump tions of linearity, independence of errors, homoscedasticity, unusual points and normality of residuals were met. Specifically, the Durbin Watson statistics was 1.910 indicating near normal distribution. Both the variance inflation ) and tolerance ) were compliant with the recommended cut off points of less than 10 and greater than 0.10 respectively (Pallant, 2010) which suggested that the model would not be influenced by collinearity. All of the variables were simult aneously entered into the regression model which accounted for 43.2% of the variance in self efficacy to seek out health care services ( R 2 = .432, R 2 adj = .246, F (31, 95) = 2.328, p < .001). Table 4 13 presents a summary of the regression coefficients gen erated by this analysis. The statistically significant predictors of self efficacy to engage in health behavior were Frequency of online use for health information seeking (Several Times a Week), Source (Somewhat Non Influential), Content (Somewhat Non Inf luential), and Age (18 20). independen t variables suggested that all four statistically significant predictors made relatively large contributions to the regression model. With this, it must be mentioned that when multicollinear ity (or collinearity) is present between the predictors in multiple regression, both beta weights and structure coefficients must be interpreted (Tong, 2006). Since, multicollinearity was checked for previously, and found not to be an issue in this study, interpretation of results through beta weights was sufficient. The model showed that as age (18 20) (b = .387 ) increased by one unit the total self efficacy to engage in health behavior score decreased by .387 points. The selection OHI seeking (b = .401) decreased the total self efficacy to engage in health behavior score by .401 points holding all other factors

PAGE 120

120 in the model constant. related t o the source of OHI (b = .772) decreased the total self efficacy to engage in health behavior score by .772 points holding all other factors in the model constant. of OHI (b = .796) increased the total self efficacy to engage in health behavior score by .796 points holding all other factors in the model constant. Frequency of online use for health information seeking (Less Often, Every Few Months, Several Times a Mo nth, and Everyday), Source (Very Non Influential, Non Influential, Neither Influential nor Non Influential, Somewhat Influential, Influential, and Very Influential), Message (Very Non Influential, Non Influential, Somewhat Non Influential, Neither Influent ial nor Non Influential, Somewhat Influential, Influential and Very Influential), Content (Very Non Influential, Non Influential, Neither Influential nor Non Influential, Somewhat Influential, Influential and Very Influential), DSSI (Low, Moderate, High), age (21 23), and year in school (Freshman, Sophomore, Junior, Senior, Graduate, Professional) were not statistically significant predictors in the model. RQ5: To what extent do attributes of OHI (source, message, and content); frequency of OHI seeking, soc ial support, age, and education predict self efficacy to seek out health care services among CMLF ? A multiple regression with the bootstrap function of 1,000 simulated sample was ra n to predict self efficacy to seek out health care services from source, me ssage content, frequency of OHI seeking, DSSI (level of social support), age and education (year in college) The assumptions of linearity, independence of errors, homoscedasticity, unusual points and normality of residuals were met. Specifically, the Dur bin Watson statistics was 1.707 indicating near normal distribution. Both the variance ) and tolerance ) were compliant

PAGE 121

121 with the recommended cut off points of less than 10 and greater than 0.10 respectively (Pallant, 2010) which suggested that the model would not be influenced by collinearity. All variables were simultaneously entered into the regression model which accounted for 57.2% of the variance in self efficacy to seek out health care services ( R 2 = .5 72, R 2 adj = .108, F (31, 95) = 1.491, p < .05). Table 4 14 presents a summary of the regression coefficients generated by this analysis. The statistically significant predictors of self efficacy to seek out health care services were Mess age (Neither Influe ntial nor Non Influential) and Age (21 23) t variables suggested that both statistically significant predictors made relatively large contributions to the regression model. The model showed that as age (21 23) (b = .243 ) increased by one unit the total self efficacy to seek out health care services score decreased by .243 points. The (b = .312) decreased the total self efficacy to seek out health ca re services score by .312 points holding all other factors in the model constant. Frequency of online use for health information seeking (Less Often, Every Few Months, Several Times a Month, Several Times a Week, and Everyday), Source (Very Non Influentia l, Non Influential, Somewhat Non Influential, Neither Influential nor Non Influential, Somewhat Influential, Influential, and Very Influential), Message (Very Non Influential, Non Influential, Somewhat Non Influential, Somewhat Influential, Influential and Very Influential), Content (Very Non Influential, Non Influential, Somewhat Non Influential, Neither Influential nor Non Influential, Somewhat Influential, Influential and Very Influential), DSSI (Low, Moderate,

PAGE 122

122 High), age (18 20), and year in school (Fre shman, Sophomore, Junior, Senior, Graduate, Professional) were not statistically significant predictors in the model. Summary Chapter 4 presents demographic characteristics of focus group, think aloud protocol participants and survey respondents. The study participants, aged 18 26 years, were primarily White of Latino origin. Most participants were single and living off campus with roommates. Think aloud protocol findings were described, as were internal consistency reliability analysis findings. The findin gs of the posed research questions were also discussed. The present study found that most participants expressed going online several times a month to search for health information. Most participants also expressed using a laptop to access OHI Focus grou p analysis informed the selection of survey items which was used to explore the several types of health information that are searched by CMLF Types of health information searched include informational needs for others, information on symptoms, diagnoses a nd physical appearance. Information on natural living, nutrition and diet were also types of health information searched, as were treatments for conditions and monetary health concerns such as health insurance. Factors that affect OHI searching, expressed by respondents, were family and loved ones, monetary issues, fear of health condition, and physical appearance. Most respondents expressed that their own health status most influences them to search online for health information. A great majority of respon dents use online search engines as a resource for seeking OHI Most respondents again choose online search engine as their preferred resource for health information, followed by family and health professionals.

PAGE 123

123 Data collected during this study revealed, t hrough multivariable analysis, that the variables Frequency of online use for health information seeking (Several Times a Week), Source (Somewhat Non Influential), Content (Somewhat Non Influential), and Age (18 20) were statistically significant predictor s for self efficacy to engage in health behavior Findings also revealed that the variables Mess age (Neither Influential nor Non Influential) and Age (21 23) were statistically significant predictors for self efficacy to seek out health care services. The implications for these findings will be discussed in Chapter 5 as will the limitations for this study and the directions for future research. Table 4 1 Focus group demographic characteristics. Characteristics n=41 Age (mean SD) 20.88 1.42 Educati on level (Year in College) 1 st 0 2 nd 15 3 rd 11 4 th 13 5 th 1 Graduate School 1 Identify as Latino/Hispanic /Spanish Origin 38 *Those not self identifying as Latino were still included in analysis.

PAGE 124

124 Table 4 2 Think alou Characteristics n=8 Age (mean SD) 20.50 1.31 Gender Male 8 Female 0 Prefer not to disclose 0 Identify as Latino/Hispanic/Spanish Origin No, not of Hispanic, Latino, or Spanish Origin 0 Yes, Mexican, Mexican American, Chicano 2 Yes, Puerto Rican 2 Yes, Cuban 2 Yes, another Hispanic, Latino or Spanish origin 3 Race* White 8 Black or African American 2 American Indian or Alaska Na tive 0 Asian Indian 0 Japanese 0 Native Hawaiian 0 Chinese 0 Korean 0 Guamanian or Chamorro 0 Filipino 0 Vietnamese 0 Samoan 0 Other Asian 0 Other Pacific Islander 0 Academic classification Freshman 0 Sophomore 3 Junior 2 Senior 3 Category of Major/Program of Study Humanities 1 Computer Science 1 Health Science/Nursing 1 Engineering 1 Physical/Biological Sciences 3 Business 1 Other 0

PAGE 125

125 Table 4 2. C ontinued Characteristics n=8 Relationship Status Single, not in a relationship 3 Single, dating casually 2 Single, dating one person exclusively 2 Living with my partner 0 Married/Committed 0 Divorced 0 Other 0 Refuse to answer 1 Living situation Live on campus (campus housing) with roommates 3 Live off campus with roommates 5 Live off campus with parents/relatives/family 0 Other, please specify 0 Ref use to answer 0 Table 4 3 Characteristics n= 12 7 Age (mean SD) 20.65 1.43 Gender Male 126 Female 0 Prefer not to disclose 1 College/University Attended Barry University 5 Embry Riddle Aeronautical University 7 Florida Atlantic University 14 Florida State University 23 NOVA Southeastern University 11 University of Central Florida 21 University of South Florida 46 Identify as Latino/Hispani c/Spanish Origin No, not of Hispanic, Latino, or Spanish Origin 12 ** Yes, Mexican, Mexican American, Chicano 27 Yes, Puerto Rican 35 Yes, Cuban 37 Yes, another Hispanic, Latino or Spanish origin 3 5 Living situation** Live on campus (campus housing) with roommates 13 Live off campus with roommates 66 Live off campus with parents/relatives/family 35 Other, please specify 12 Refuse to answer 0

PAGE 126

126 Table 4 3. Continued Characteristics n=127 Race* White 110 Black or African American 31 American Indian or Alaska Native 5 Asian Indian 3 Japanese 1 Native Hawaiian 0 Chinese 4 Korean 0 Guamanian or Chamorro 0 Filipino 0 Vietnamese 1 Samoan 0 Other Asian 0 Other Pacific Islander 0 Academic classification** Freshman 13 Sophomore 51 Junior 38 Senior 21 Graduate (e.g. Masters or Ph.D) 2 Professional (e.g. J.D. or M.D.) 1 Other 0 Ca tegory of Major/Program of Study Humanities 10 Computer Science 8 Health Science/Nursing 23 Engineering 34 Physical/Biological Sciences 20 Business 22 Other 8 Relationship Status Single, not in a relations hip 39 Single, dating casually 35 Single, dating one person exclusively 48 Living with my partner 2 Married/Committed 1 Divorced 0 Other 1 Refuse to answer 1 *Respondents selected more than one answer choice; **So me respondents did not select an answer choice ; ***Not excluded from analysis.

PAGE 127

127 Table 4 4. Categories, codes and associated definitions of think aloud protocol analysis. Category Code Definitions Errors in Questionnaire Error in Wording Used to signal gr ammatical errors, or errors in directions. Spelling Error Used to signal highlighted misspellings. Misunderstandin g or misinterpretation of items Misunderstanding Used to signal instances were respondents were confused on interpretation of question. Clarification Needed Used to depict were examples would clarify the meaning of certain questions and words. Suggestions for improvements Item Removal Used to signal where suggestions for item removal were given. Format Related Used to signal where rest ructuring of items or formatting of items on Qualtrics was needed. Wording Related Used to signal where different words for questions might be necessary. Table 4 5. Internal Consistency Reliability Analysis Scale Alpha d on Standardized Items N of Items Duke Social Support Index ( DSSI ) .794 .796 11 Self Efficacy to Seek Health Care .636 .690 5 Self Efficacy to Engage in Health behavior .841 .860 7 Level of Influence associated with Attributes of OHI (Source, Message, Content) .736 .738 3 Table 4 6 Extent of online use for health information seeking. Frequency Percent Everyday 4 3.1 Several Times a Week 17 13.4 Several Times a Month 56 44.1 Every Few Months 32 25.2 Less Often 18 14.2 Total 127 100.0

PAGE 128

128 Table 4 7 Device most used when accessing the internet. Frequency Percent Cell phone 27 21.3 Desktop 17 13.4 Laptop 52 40.9 Tablet computer 29 22.8 Smart TV 1 .8 Gaming console (e.g. XBOX 360, PLAYSTATION 3, Wii, etc.) 1 .8 Total 127 100.0 Table 4 8 Place respondent mostly find themselves while searching online. Frequency Percent At Home 59 46.5 At Work 28 22.0 At School 37 29.1 Somewhere else, if so, please specify: 3 2.4 Total 127 100.0

PAGE 129

129 Table 4 9 Frequency and percentage of types of onlin e health sought by participants. Type Frequency Percent Health information for others (e.g. family, friends, or significant other) Yes 70 55.1 No 53 41.7 Unsure 4 3.1 Concern over symptom(s) (e.g. pain, bruising, discomfort) Y es 64 50.4 No 60 47.2 Unsure 3 2.4 A recent diagnosis Yes 31 24.4 No 92 72.4 Unsure 3 2.4 Exercise Yes 48 37.8 No 79 62.2 Unsure 0 0.0 Nutrition Yes 44 34.6 No 82 64.6 Unsure 1 .8 Natural living (e.g. organic food and products) Yes 39 30.7 No 82 64.6 Unsure 6 4.7 A certain medical treatment or procedure Yes 25 19.7 No 98 77.2 Unsure 4 3.1 A spec ific disease or medical problem Yes 33 26.0 No 89 70.1 Unsure 4 3.1 Information on immigration health concerns Yes 22 17.3 No 103 81.1 Unsure 2 1.6 Prescription drugs Yes 42 33.1 No 84 66.1 Over the counter drugs Yes 38 29.9 No 84 6 6.1 Unsure 3 2.4 Home remedies (e.g. splints, sprains) Yes 57 44.9 No 68 53.5 Substance abuse (alcohol or drugs) Yes 7 5.5 No 116 91.3 Unsure 3 2.4

PAGE 130

130 Table 4 9. Continued Type Frequency Percent Mental health issue s (i.e. stress, anxiety, depression) Yes 42 33.1 No 83 65.4 Unsure 2 1.6 Family planning (e.g. condoms, birth control, safe sexual practices or pregnancy of significant other) Yes 14 11.0 No 113 89.0 Unsure 0 0.0 S exually transmitted diseases* Yes 7 5.5 No 118 92.9 Unsure 1 .8 Health insurance options and concerns Yes 35 27.6 No 90 70.9 Unsure 2 1.6 Exploring online posts, tweets, or videos on health Yes 61 48.0 No 63 49.6 Unsure 3 2.4 Concerns on weight or physical appearance Yes 42 33.1 No 83 65.4 Table 4 10 Factors that influence OHI seeking. Factor Frequency ( n ) Percent My own health status Very Influential 73 57.5 Inf luential 42 33.1 Somewhat Influential 9 7.1 Neither Influential nor Non Influential 2 1.6 Somewhat Non Influential 0 0.0 Non Influential 1 .8 Very Non Influential 0 0.0 The health status of a family member Very Influe ntial 53 41.7 Influential 48 37.8 Somewhat Influential 16 12.6 Neither Influential nor Non Influential 7 5.5 Somewhat Non Influential 2 1.6 Non Influential 0 0.0 Very Non Influential 0 0.0 The health status of a friend Very Influential 20 15.7 Influential 45 35.4 Somewhat Influential 36 28.3 Neither Influential nor Non Influential 22 17.3 Somewhat Non Influential 4 3.1 Non Influential 0 0.0 Very Non Influential 0 0.0

PAGE 131

131 Table 4 10. Continued Factor Frequency ( n ) Percent The health status of a significant other (e.g. girlfriend) Very Influential 33 26.0 Influential 44 34.6 Somewhat Influential 28 22.0 Neither Influential nor Non Influential 15 11.8 Somewhat Non Influential 4 3.1 Non Influential 3 2.4 Very Non Influential 0 0.0 Monetary issues Very Influential 62 48.8 Influential 27 21.3 Somewhat Influential 17 13.4 Neither Influential nor Non Influential 15 11. 8 Somewhat Non Influential 3 2.4 Non Influential 2 1.6 Very Non Influential 0 0.0 Concerns over weight or physical appearance Very Influential 14 11.0 Influential 26 20.5 Somewhat Influential 17 13.4 Neither Inf luential nor Non Influential 22 17.3 Somewhat Non Influential 15 11.8 Non Influential 18 14.2 Very Non Influential 14 11.0

PAGE 132

132 Table 4 11 Resources visited for health information. Resource Frequency Percent Health care provider (Nurse or Doctor) Yes 110 86.6 No 17 13.4 Pharmacist Yes 39 30.7 No 88 69.3 Friends/Peers Yes 73 57.5 No 54 42.5 On line Search Engine (e.g. Google, Yahoo, Bing) Yes 114 89.8 No 13 10.2 On line video site (e.g. YouTube, Vimeo) Yes 89 70.1 No 38 29.9 Mobile Health App(s) Yes 43 33.9 No 84 66.1 Family Member(s) Yes 84 66.1 No 42 33.1 On line forum (e.g. blog for FAQ) Yes 49 38.6 No 77 60.6 Social N etworking Site (e.g. Facebook, Twitter, Pinterest) Yes 52 40.9 No 75 59.1 Table 4 1 2 Resource most frequently visited by participants. Resource Frequency Percent Health care provider (nurse or doctor) 15 11.8 Pharmacist 2 1.6 Friends/Pe ers 6 4.7 On line Search Engine (e.g. Google, Yahoo, Bing) 48 37.8 Mobile health App(s) 6 4.7 Family member(s) 29 22.8 On line forum (e.g. blog for FAQ) 4 3.1 Social Networking Site (e.g. Facebook, Twitter, Pinterest) 5 3.9 On line video site (e.g. YouTube, Vimeo) 12 9.4

PAGE 133

133 Table 4 13 Multiple linear regression predicting self efficacy to engage in health behavior Model B SE B t Sig. Constant 2.321 .465 4.986 .000 Extent Less Often .178 .247 .139 .720 .473 Extent Every Few Months .046 229 .045 .201 .841 Extent Several Times a Month .174 .219 .193 .799 .427 Extent Several Times a Week .542 .242 .401 2.236 .028 Source (Very Non Influential) .029 .515 .006 .057 .955 Source (Non Influential) .283 .420 .056 .674 .502 Sour ce (Somewhat Non Influential) 1.982 .430 .772 4.610 001* Source (Neither Influential nor Non Influential) .042 .205 .023 .205 .838 Source (Somewhat Influential) .070 .140 .054 .501 .617 Source (Influential) .070 .095 .073 .734 .465 Message (Very Non Influential) .018 .413 .004 .045 .964 Message (Non Influential) .658 .438 .130 1.505 .136 Message (Somewhat Non Influential) .204 .291 .080 .702 .485 Message (Neither Influential nor Non Influential) .148 .225 .085 .658 .512 Message (Somewhat I nfluential) .025 .141 .019 .175 .862 Message (Influential) .033 .102 .035 .325 .746 Content (Very Non Influential) .156 .452 .031 .345 .731 Content (Non Influential) .040 .320 .011 .125 .901 Content (Somewhat Non Influential) 2.350 .518 .796 4.534 001* Content (Neither Influential nor Non Influential) .206 .171 .134 1.201 .233 Content (Somewhat Influential) .054 .118 .050 .457 .649 Content (Influential) .076 .111 .082 .683 .496 DSSI Low .195 .267 .076 .730 .467 DSSI Moderate .046 083 .051 .557 .579 Age (18 20) .993 .274 .387 3.629 001* Age (21 23) .008 .116 .007 .073 .942 Year in School (Freshman) .360 .437 .243 .823 .413 Year in School (Sophomore) .207 .407 .227 .510 .611 Year in School (Junior) .106 .412 .109 .258 .797 Year in School (Senior) .293 .416 .242 .703 .484 Year in School (Graduate) .121 .509 .034 .238 .812 *Indicates statistical significance.

PAGE 134

134 Table 4 14 Multiple linear regression predicting self efficacy to seek health services. Model B SE B t Sig. Constant 434.297 74.288 5.846 .000 Extent Less Often 22.785 39.492 .121 .577 .565 Extent Every Few Months 49.132 36.586 .328 1.343 .182 Extent Several Times a Month 10.073 34.876 .076 .289 .773 Extent Several Times a Week 13.209 38 .687 .067 .341 .734 Source (Very Non Influential) 84.439 82.160 .113 1.028 .307 Source (Non Influential) 88.995 67.104 .120 1.326 .188 Source (Somewhat Non Influential) 10.909 68.615 .029 .159 .874 Source (Neither Influential nor Non Influential) 20.464 32.709 .076 .626 .533 Source (Somewhat Influential) 15.610 22.287 .083 .700 .485 Source (Influential) 17.391 15.132 .124 1.149 .253 Message (Very Non Influential) 71.774 65.903 .096 1.089 .279 Message (Non Influential) 46.783 69.833 .063 670 .505 Message (Somewhat Non Influential) 26.470 46.475 .070 .570 .570 Message (Neither Influential nor Non Influential) 79.986 35.881 .312 2.229 .028 Message (Somewhat Influential) 12.149 22.434 .063 .542 .589 Message (Influential) 9.759 16.2 12 .071 .602 .549 Content (Very Non Influential) 25.811 72.190 .035 .358 .721 Content (Non Influential) 42.870 51.111 .081 .839 .404 Content (Somewhat Non Influential) 63.245 82.752 .146 .764 .447 Content (Neither Influential nor Non Influen tial) 7.739 27.357 .034 .283 .778 Content (Somewhat Influential) 12.246 18.820 .077 .651 .517 Content (Influential) 4.144 17.762 .030 .233 .816 DSSI Low 9.388 42.604 .025 .220 .826 DSSI Moderate 14.456 13.300 .108 1.087 .280 Age (18 20) 4.448 43.673 .012 .102 .919 Age (21 23) 41.592 18.505 .243 2.248 .027 Year in School (Freshman) 17.720 69.729 .082 .254 .800 Year in School (Sophomore) 3.314 64.951 .025 .051 .959 Year in School (Junior) 38.327 65.696 .269 .583 .561 Year in Sch ool (Senior) 16.074 66.437 .091 .242 .809 Year in School (Graduate) 36.359 81.229 .069 .448 .655 R = .572, R 2 = .327, R 2 adj = .108.

PAGE 135

135 CHAPTER 5 DISCUSSION AND CONCL USION Introduction Latinos are set to become the largest minority in the United Stat es and increased Latino student enrollment in colleges and universities nationwide warrants a study that investigates health information seeking on the Internet by CMLF T he purpose of the present study was to explore health information seeking among CML F Most health studies, to date, have underrepresented or misrepresented samples of Latino participants (Betancourt, Green, Carillo, & Maina, 2004), and the literature lacks studies examining dynamics among CMLF that predict or explain use of the Internet for locating health information. In an effort to bridge the gap which exists in the literature, this mixed methods pilot study sampled college men from a Latino fraternity to investigate the interrelationships between frequency of OHI seeking types of hea lth information sought on the Internet level of social support, education, age, self efficacy to seek health care services and self efficacy to engage in health behavior Chapter 5 discusses study findings, along with associated study limitations and r ecommendations for future research pertaining to health information seeking on the Internet among Latino populations Discussion of Findings Demographics As expected due to inclusion criteria, t he overwhelming majority of focus group participants (93% ; n = 38 ) considered themselves of Latino or Hispanic heritage. For the think aloud protocols, all eight participants stated being of Latino or Hispanic origin. As

PAGE 136

136 for survey respondents, the majority of respondents (90.5% ; n =115 ) selected being of Latino or Hi spanic heritage. The majority of respondents (40% ; n =51 ) were college sophomores studying mostly engineering (27 % ; n =34 ), health science (18% ; n =23 ), business (17 % ; n =22 ), and physical or biological science (16 % ; n =20 ). These findings are significant in t hat they give credence to information reported by Fry and Lopez (2012) in which it is stated Engineering and Mathematics (STEM) fields than ever before. Pretesting and Inter nal Consistency Reliability Analysis The use of think aloud protocol interviews allowed for the correction of overlooked nuances and errors present in the questionnaire before it was administered The methods described by Van Someren et al. (1994) and DeMa io et al. (1998) mentioned earlier in Chapter 3 proved to be invaluable in both encouraging participants to be descriptive and in helping to highlight issues that might have gone overlooked. Internal consistency reliability testing further supported previo us literature on the DSSI such as the study undertaken by Wardian et al. (2012) where Chronbach alpha scores of .75 as internally consistent. Previous literature did not report Chronbach indicate that the scales used were internally consistent. The internal consistency reliability analysis results of this study help to give a foundation for future studies utilizing the same scales.

PAGE 137

137 Frequency of Internet Use to Locate Health Information T he frequency in which the majority of Latino m ales belonging to a fraternity (44.1%) search the Internet for health information is several times per month. Approximately 25% of respondents go online every few months to search for health information. These findings differ from those of Fox and Rainie ( 2002) in a nationally representative sample, where they report (25%) of respondents seeking OHI several times a month and (58%) of respondents seeking OHI every few months. This finding is significant in that it suggests that OHI seekers are now seeking fo r information more frequently. However, such comparison between Fox and Rainie (2002) and the current between both studies. The current study was college student specific and differs with Fox and Rainie (2002) in that aspect, as such an updated look at whether variables such as access to computers and lifestyles requiring online access, play a role in the increased frequency of searching the Internet for OHI The majority of Latino men in a fraternity selected laptops (40.9%) as the device used most to access the Internet. Also, most respondents (46.5%) selected the home as the place they most likely find themselves when searching the Internet for health information. Inter estingly, this finding is identical to findings reported in a larger national devices to access the Internet (Duggan & Smith, 2013). From this finding, it is suggested that m obile phones have yet to take over as the dominant device for assessing the Internet among CMLF However, future research should delve further into this occurrence to give further credence to this finding.

PAGE 138

138 Types of Searched OHI F indings reported by Escoffe ry et al. (2005) and Fox (2006) indicated that users seek several types of OHI own symptoms, diagnoses, physical appearance in terms of weight and physique, natural living and treatments for c onditions or diseases. Contrary to previous studies performed to examine OHI seeking among college students (Easaw, 2010), this college student population did not seek on topics such as substance abuse (alcohol or drugs), condoms, or safe sexual practices. Although, such topics may not be sought online, it remains to be seen whether these topics are of priority for CMLF outside of the Internet realm; investigating this phenomenon in future research would definitely be a unique avenue to pursue. T he lack of conversation on safe sexual practices participants i s cause for concern, especially due to the concern over their girlfriends unplanned pregnanc ies This suggests that there is an apparent need of mor e health education for this population to assist in improving safe sex practices. conducted investigating the phenomenon of unintended/unplanned pregnancy in the Latino college populat ion and available research suggests that the phenomenon may be associated with acculturation ( Rafaelli, Zamboanga, & Carlo, 2005 ), thus investigating this phenomenon warrants an updated look. Factors Influencing OHI Seeking The main factors mentioned that influence OHI seeking among study participants visits, insurance coverage management), fear of illnesses, and physical appearance. A majority of the sample also reported usin g at least three or more different sources to

PAGE 139

139 seek health information, including the Internet, parents or family members, a doctor or nurse, and friends. From these sources, online search engines and family members were selected as the most frequent ly cite d source of health information. This finding shows that the Internet is not the only source for seeking health information by CMLF This finding is consistent with previous research in which LaJoie and Ridner (2009) found that college students seek health information from parents as well as from the Internet. Self Efficacy to Engage in Health behavior Multiple linear regression results identified four statistically significant predictors for self efficacy to engage in health behavior The statistically sign ificant predictors of self efficacy to engage in health behavior were Frequency of online use for health information seeking (Several Times a Week), Source (Somewhat Non Influential), Content (Somewhat Non Influential), and Age (18 20). Previous research such as that performed by Zuckerman (2009) suggests that college students search for health information regardless of the number of hours spent quency of OHI seeking (b = .401) decreased the total self efficacy to engage in health behavior score by .401 points, holding all other factors in the model constant. This suggests that the possibility exists that more time spent online searching for heal th information may have a negative impact on self efficacy to engage in health behavior among Latino men who belong to a fraternity Previous research conducted by Eastin (2001) stated: The findings did suggest that source expertise (and/or age and/or gen der) and knowledge of content (and/or order effects) affect perception of message credibility. This suggests that when people evaluate online health

PAGE 140

140 information, the heuristic cues they attend to vary, depending on the subject matter. (Discussion section, para. 1). The findings of this study are similar to Eastin (2001) findings in that they suggest that a in that these variables may play a role in predicting self efficacy to eng age in health behavior of OHI (b = .772) decreased the total self efficacy to engage in health behavior score by .772 points, holding all other factors in the model constant. This suggests that when a n online efficacy to engage in health behavior may be adversely impacted. to the content of OHI (b = .796) increased the total self efficacy to engage in health behavior score by .796 points, holding all other factors in the model constant. This suggests that impact self efficacy to engage in health behavior From this, one can gather that in order to have a positive impact on self efficacy to engage in health behavior content or the manner (e.g. videos, pictures, format) in which the messa ge is portrayed, may be crucial to positively influence CMLF to take action in managing their health. Prior research by Atkinson, Sape rstein and Pleis (2009) suggests that age may be related to OHI seeking in increased likelihood of taking action in term s of purchasing vitamins or medicine. Similar findings were found in the present study where the model showed that as age (18 20) (b = .387) increased by one unit, total self efficacy to engage in health behavior score decreased by .387 points. Thus, self efficacy to engage in health behavior may be adversely impacted for those in the age group of 18

PAGE 141

141 efficacy to engage in health behavior is adversely impacted is warranted. sent age as a statistically significant predictor for self efficacy to engage in health behavior for those in the age groups of 21 23 and 24 26. Thus, another look into this incidence in future studies will help to clarify whether age is a strong predictor for self efficacy to engage in health behavior for those ages 21 26. Self Efficacy to Seek Out Health Care Services Multiple linear regression analyses identified two statistically significant predictors for self efficacy to seek out health care services The statistically significant predictors of self efficacy to seek out health care services were message (Neither Influential nor Non Influential), and age (21 23). The model showed that as age (21 23) (b = .243) increased by one unit, the total self eff icacy to seek out health care services score decreased by .243 points. This result is similar to the study performed by Dunlop, Manheim, Song and Chang (200 2) where age was also suggested to share an inverse relationship with health care seeking and utiliz ation in minority men .312) decreased the total self efficacy to seek out health care services score by .312 points, holding all other factors in the model constant. This sugges ts that for those who stated feelings of neutrality towards health messages obtained from online sources their self efficacy to seek out health care services scores were negatively impacted. To date, published literature investigating this variable is not available. However, research undertaken by de Boer, Wijker, and de Haes (1997) in their systematic literature review of predictors t o health care utilization give credence to the fact that the variable social support was not found to be a predictor of sel f efficacy to seek out health care services

PAGE 142

142 in this study, as was also the case in de not found to be a statistically significant predictor for those selecting anything other than In another look at these results will do much in clarifying the issue. Implications The findings from this study highlight some of the opportunities and challenges that those researching Latino health, c ollege health, and OHI seeking face moving forward. Theoretical Implications Efficacy Theory to inform the research and to develop research questions (Bandura, 2006). Self efficacy theory was the basis for several questions u sed in the developed questionnaire. As noted earlier, a key principle of the Self efficacy theory is that individuals are more likely to engage and put forth more effort and persistence in activities that they have higher feelings of efficacy for and less likely to engage in those activities for which they have less feelings of efficacy (Van der Bijl & Shortridge Baggett, 2002). efficacy to engage in health behavior may be adversely impacted. This finding suggests Efficacy Theory was correct in describing the relationship of higher feelings of efficacy and engagement versus lower feelings of efficacy and less e xpected engagement. may predict self efficacy to engage in health behavior to lend further knowledg e and credence to the findings of the current study.

PAGE 143

143 The study at hand self efficacy to engage in health behavi or This finding warrants a second look as to self efficacy does not in fact impact engagement among CMLF Due to the fact that the study at hand was a pilot study, it is suggested that future research takes a more in depth view Efficacy Theory and its relationship with OHI seeking among CMLF Implications for Health Educators Although some research projects do not lend themselves to practical applications, the resu lts of this study suggest an important implication for health educators This health information technology to improve overall health for minority college students The u tilization of health information technology to improve health outcomes is a national health priority as is the overarching goal to achieve health equity, eliminate disparities, and improve the health of all groups ( U.S. Department of Health and Human Servi ces, n.d. ). Health educators need to focus their efforts on education with regard to improving knowledge about the benefits of seeking professional medical attention promptly rather than self medicating and self diagnosing which the results of this study i ndicate are consequences of OHI seeking dealt by CMLF in trying to reduce costs and time falsely believed to be associated with health care utilization. Recommendations for Future Research Nearly one seventh of the planet is on some form of social media (B ernstein, 2013). Due to social media, distance is no longer an issue, and ideas, suggestions and care for health matters can be shared half a world away in a matter of seconds.

PAGE 144

144 Although the study at hand recognized that social media was an important factor on how CMLF search for OHI it was not the central focus of the investigation. It is suggested that future research endeavors take a closer look on whether rates of social media usage for OHI seeking have increased and whether health information sought th rough social media influences Latino males to take action more or less than traditional avenues. Collins and Lewis (2013) have begun on the path of investigating the role of social media as it relates to directing Internet users to credible health informat ion. Clearly, investigating the relationship between social media, OHI seeking and related health outcomes is an avenue to pursue. Other areas to also consider in future research studies relate to several findings from this study During focus group sessi ons, an overarching theme that arose was that of weight concern and body image issues that young men in college face. Historically, weight concerns and body image concerns have been investigated in various populations of women (Grabe & Hyde, 2006; Varnes, Stellefson, Janelle, Dorman, concern seems to not only be a topic among women but seems to be a topic that should be further investigated specifically within a Latino college m ale population. Although, studies such as those by Cohane and Pope (2001) and McCabe and Ricciardelli (2004) have looked into investigating the topic among males differences among fraternity and non fraternity college men in weight and body image concerns is definitely of interest. As suggested by focus group and survey results, another issue for concern and a great direction for future research is the apparent lack of attention participants gave to searching for OHI regarding safe sex practic es and sexually

PAGE 145

145 transmitted diseases. It would be of much interest for future research to empirically look at reasons concerning this phenomenon among college men in fraternities and whether health education and health promotion programs aimed at teaching safe sex practices can help to turn around the low interest in the topics among Latino men in college To date, a study looking specifically at these aspects has not been published. t seem to significantly impact OHI seeking behaviors analyzed. As briefly stated earlier, de Boer, Wijker, and de Haes (1997) reported that social support was not a statistically significant predictor of self efficacy to seek out health care services. This occurrence is of special concern and interest and warrants further investigation. T here is yet to be a study investigating social support as it relates to predicting health behaviors among CMLF Thus, further research investigating social support as a pre dictor to health behavior among CMLF would certainly add to the current knowledge base. Limitations of the Study Throughout the course of this study, several limitations were encountered and fully experimental approach to investigate the research questions I n the absence of random sampling, results from the sample cannot be generalized to larger populations. Second, the OHI seeking behaviors over time. In addition, the majority of study participants were of Latino background ; the specificity of participants limits the generalizability of study results to the general public. Also, the qualitative sessions was another limitation in this study. In this study, self report could have biased the data

PAGE 146

146 when seeking OHI although steps were taken to Taki ng into consideration the limitations stated and those which might have been overlooked, it is suggested that this Conclusion Research on Latino health has been historically lacking and published information is often based on misrepresented or underrepresented samples of Latino participants despite Latinos being the largest minority in the United States. Research on college Lati nos in fraternities. Despite the fact that a large percentage of Americans go online to seek health information, literature pertaining to OHI seeking among CMLF has been non existent, until now. Colege men in Latino fraternities have been turning to the i n ternet for health information, and this study attempted to understand the dynamics related to their OHI searches to bridge the existing knowledge gap Despite limitations, the present study is the first to provide a detail ed description of OHI seeking amon g CMLF This study represents a systematic effort to examine self efficacy to engage in health behavior and self efficacy to seek out health care services as they relate to OHI seeking among CMLF The present investigation revealed that CMLF are mostly sea rching for OHI for persons other than themselves. College men in Latino fraternities are not searching for information on safe sex practices, or sexually transmitted diseases which is a cause for concern. The findings of this study suggest that the most ut ilized resources for health information are family members. Study results also suggest that monetary apprehensions in this population have led to an increase in self diagnoses and self treatment. The findings of this empirical endeavor also highlight the s ignificance of

PAGE 147

147 illuminating the role of social support and other demographic factors as covariates explaining and/or predicting self efficacy to engage in health behavior and self efficacy to seek out health care services Social media has become a leading communication platform and will continue to attract users across all segments; thus, it is important to n OHI seeking among college students in general F urther research is still needed to fully understand the influence o f OHI seeking on health behaviors and decision making among CMLF

PAGE 148

148 APPENDIX A R ECRUITMENT EMAIL FOR UNIVERISTY OF FL ORIDA AND FLORIDA INTERNATIONAL UNIVER SITY STUDENTS (FOCUS GROUP)

PAGE 149

149 APPENDIX B INFORMED CONSENT FOR M FOR FOCUS GROUP PA RTICIPANTS Protocol Title: Health Information Seeking on the Internet by Latino Fraternity College Men Please read this consent document carefully before you decide to participate in this study. Purpose of the research study: To bridge the knowledge gap as to the types of online health information sought by Latino males in college who are members of a fraternal organization. The study is meant to also give insight on how or whether self efficacy to seek out health care services and self efficacy to engage in health behavior may or not relate to online heath information seeking. Along with the above information, the study aims to determine whether or not social support among Latino men in college who are members of a fraternal organization influences self efficacy to seek out health care services and self efficacy to engage in health behavior Note: This study involves research for dissertation purposes. What you will be asked to do in the study: You will be asked to volunteer to participate in a group discussion (focus group ), with other participants who share similar interests. You will be asked a series of questions by the focus group moderator (Principal Investigator). Please feel free to answer and share your thoughts on the subject matter. You do not have to answer any q uestion that you do not wish to answer Each participant, and yourself are asked Time required: 1.5 2 hours. Risks and Benefits: We anticipate little to no risk involved with par ticipation in this study. We do not anticipate that you will benefit directly by participating in this experiment. Alternative Advantageous Procedures: It is anticipated that little to no risk as well as no direct benefit will be experienced as a result of participating in this research study. Hence, no immediate alternative advantageous procedures are closely apparent or related to this research study. Compensation: You will be given a $5.00 Amazon.com gift card as compensation for participating in this r esearch. Confidentiality: Your identity will be kept confidential to the extent provided by law. Recordings will be transcribed within 24 48 hours after the group meeting, identifying information will be removed and the recordings erased at

PAGE 150

150 that time. Whil e we treat the discussion as confidential and ask all members participating to do the same, Reporting: The collected data are for dissertation research purposes. Thus, a final report of analyzed data will be publi c domain and available through the U.S. Library of Congress and online in ProQuest Digital Dissertations. Voluntary participation: Your participation in this study is completely voluntary. There is no penalty for not participating. Right to withdraw from the study: You have the right to withdraw from the study at any time without consequence. Whom to contact if you have questions about the study: Enmanuel Chavarria, MS, CHES, Graduate Student, Principal Investigator, Department of Health Education and Beha vior, FLG 23, P.O. Box 118210, Gainesville, FL 32611 8210, phone: (352) 294 1807. Elizabeth Chaney, PhD, MCHES, College of Health and Human Performance, FLG 12, P.O. Box 118210, Gainesville, FL 32611 8210, phone: (352) 294 1813. Whom to contact about your rights as a research participant in the study: IRB02 Office, Box 112250, University of Florida, Gainesville, FL 32611 2250; phone: (352) 392 0433. Agreement: I have read the procedure described above. I voluntarily agree to participate in the procedure and I have received a copy of this description. Participant: ___________________________________________ Date: _________________ Principal Investigator: ___________________________________ Date: _________________

PAGE 151

151 APPENDIX C FOCUS GROUP MODERATO I. W elcome Enmanuel Chavarria, Research Assistant, University of Florida, Department of Health Education and Behavior a. Self Introduction and Thank you b. Reminder about refreshments c. Hand out informed consent form d. identiality II. Overview and Introduction of Meeting a. Speak about what prompted the research b. Remind participants about their rights, and the audio recording that will take place c. Remember there are never any wrong answers III. (Ice Breaker) Introduction of Particip ants Please introduce yourselves and tells us a little bit about yourself IV. Ask 1 st broad question. a. Please describe the most recent experience where you looked online for health information Please make sure that you address the following issues: What we re you looking for? How did you find it? What was the outcome? What was challenging? b. Begin transition with narrower topic. I hear several types of health information being discussed (mention a few examples). If you find yourself online looking for health i nformation, what types of health information do you seek? Why? V. Ask 2 nd broad question. a. What circumstances or situations have typically influenced or affected your experiences of going online to search for health information? (In other words, what prompts you to search for health information online? What are factors that influence your search?) Prompts: Symptoms New medication Conversation with friends Something you read in the newspaper or saw on TV.

PAGE 152

152 b. Transition again to a narrower topic. When do you search the Internet for health information versus going to other sources of health information? VI. (For focus groups 2 4) Add broad questions that arise from earlier focus group sessions. 1. Do you feel the general public knows how to critique the health information and sources they find online? (Added for Focus Groups II, III, & IV). 2. Do you feel searching online for health information helps? Or creates more of a problem? (Added for Focus Groups II, III, & IV). 3. What are the problems/disadvantages wi th finding health information on the Internet? What are the benefits/advantages with finding health information on the Internet? (Added for Focus Groups III & IV). 4. How do you believe social media affects you when searching for health information online? ( Added for Focus Group IV). Thank you for taking the time to partic ipate in this important process. I will now distribute to you a small token of our appreciation.

PAGE 153

153 APPENDIX D RECRUITMENT EMAIL FO R UNIVERSITY OF MIAMI STUDENTS (THINK ALOUD PROTOCOL INTE RVIEW)

PAGE 154

154 APPENDIX E INFORMED CONSENT FOR THINK ALOUD PROTOCOL PARTICIPANTS Protocol Title: Health Information Seeking on the Internet by Latino Fraternity College Men Please read this consent document carefully before you decide to participate in this st udy. Purpose of the research study: To bridge the knowledge gap as to the types of online health information sought by Latino males in college who are members of a fraternal organization. The study is meant to also give insight on how or whether self effi cacy to seek out health care services and self efficacy to engage in health behavior may or not relate to online heath information seeking. Along with the above information, the study aims to determine whether or not social support among Latino men in coll ege who are members of a fraternal organization influences self efficacy to seek out health care services and self efficacy to engage in health behavior Note: This study involves research for dissertation purposes. What you will be asked to do in the stu dy: asked to the best of your ability. You will be asked a series of questions by the interviewer (Principal Investigator). Please feel free to answer and share your thoughts on the question and work in front of you Please feel free to answer as detailed as possible. Please also remember, you do not have to answer any question that you do not wish to answer Your participation will be audio recorded and the computer screen you will be working on will be video captured. Time required: 45 minutes 1 hour. Risks and Benefits: We anticipate little to no risk involved with participation in this study. We do not anticipate that you will benefit directly by participating in this experiment. Alternative Advantageous Procedures: It is anticipated that little to no risk as well as no direct benefit will be experienced as a result of participating in this research study. Hence, no immediate alternative advantageo us procedures are closely apparent or related to this research study. Compensation: You will be given a $5.00 Amazon.com gift card as compensation for participating in this research.

PAGE 155

155 Confidentiality: Your identity will be kept confidential to the extent provided by law. Recordings will be transcribed within 24 48 hours after the interview identifying information will be removed and the recordings erased at that time. Your name will not be used in any report. Reporting: The collected data are for disser tation research purposes. Thus, a final report of analyzed data will be public domain and available through the U.S. Library of Congress and online in ProQuest Digital Dissertations. Voluntary participation: Your participation in this study is completely voluntary. There is no penalty for not participating. Right to withdraw from the study: You have the right to withdraw from the study at any time without consequence. Whom to contact if you have questions about the study: Enmanuel Chavarria, MS, CHES, Grad uate Student, Principal Investigator, Department of Health Education and Behavior, FLG 23, P.O. Box 118210, Gainesville, FL 32611 8210, phone: (352) 294 1807. Elizabeth Chaney, PhD, MCHES, College of Health and Human Performance, FLG 12, P.O. Box 118210, G ainesville, FL 32611 8210, phone: (352) 294 1813. Whom to contact about your rights as a research participant in the study: IRB02 Office, Box 112250, University of Florida, Gainesville, FL 32611 2250; phone: (352) 392 0433. Agreement: I have read the proce dure described above. I voluntarily agree to participate in the procedure and I have received a copy of this description. Participant: ___________________________________________ Date: _________________ Principal Investigator: ____________________________ _______ Date: _________________

PAGE 156

156 APPENDIX F INTERVIEWER QUESTION GUIDE FOR THINK ALOUD PROTOCOLS I. Introduce Self Enmanuel Chavarria, Research Assistant, University of Florida, Department of Health Education and Behavior II. Give Out Informed Consent III. Describe goal of task. a. I am collecting information for a study investigating whether Latino male college students use the internet for looking up health information. b. Your participation will help to give insight on how others may view the questionnaire in front of you. IV. Briefly explain the Think Aloud procedure to participant. a. I will be observing you as you complete the questionnaire on your screen. During this time I will be asking you to talk aloud about your thoughts on each question and your decision making proc esses. b. questions on the questionnaire. c. The most important thing is to remember to keep talking. V. Do practice Think Aloud task to familiarize participant with the procedure. a. To g et used to the talk to imagine we were at your front door. From there take me around your house or apartment, and count the windows in your home, starting from the front door and walk your way around your home. As you do this explain to me what you are seeing and doing. b. Make sure participant is explaining their process in sufficient detail as they attempt to describe their home and number of windows. VI. Reassure the participant a. I am testing the instructions, and not you, an y difficulties you encounter are not your fault, it is my fault. b. You can stop this task any time you want if you become uncomfortable. c. You can ask questions at any point but I will be unable to answer them. d. You will have to determine whether to continue to the next question yourself. VII. Begin Task a. Do you have any other questions about what to do? b. If not, feel free to begin. Just click on the link and begin. c. If necessary, prompt the participant to continue talking. VIII. Example of prompts if needed: a. What are you t hinking now? b. Why did you do that? c. IX. Debrief participant a. When the subject believes they are done, thank them. b. Ask if they have any additional feedback. c. Thank again and give out gift card. X. Prepare for next participant a. Reset all equipment and materials for the next participant. b. To reduce redundancy in data, correct errors in questionnaire before beginning with the next participant.

PAGE 157

1 57 APPENDIX G RECRUITMENT EMAIL FOR STUDENTS OF BARR Y UNIVERSITY, EMBRY RIDDLE AERONAUTICAL UNIV ERSITY, FLORIDA ATLA NTIC UNIVERSITY, FLO RIDA STATE UNIVERSITY, NOVA SOU THEASTERN UNIVERSITY UNIVERSITY OF CENT RAL FLORIDA, AND THE UNIVERSITY OF SOUTH FLORIDA

PAGE 158

158 APPENDIX H INFORMED CONSENT FOR M FOR SURVEY PARTICIPA NTS Protocol Title: Health Information Seeking on the Internet by Latino Fraternity College Men Please read this consent document carefully before you decide to participate in this study. Purpose of the research study: To bridge the knowledge gap as to the types of online health information sought by Latino males in college who are members of a fraternal organization. The study is meant to also give insight on how or whether self efficacy to seek out health care services and self efficacy to engage in health behavior may or not relate to onli ne heath information seeking. Along with the above information, the study aims to determine whether or not social support among Latino men in college who are members of a fraternal organization influences self efficacy to seek out health care services and self efficacy to engage in health behavior Note: This study involves research for dissertation purposes. What you will be asked to do in the study: You will be asked to volunteer to complete a survey by answering the attached questionnaire to the best of your ability. The questionnaire asks a series of questions concerning online health information seeking utilization of health care resources, and social support. Again, please answer each question to the best of your ability. Remember, there is no right o r wrong answer to each item. Please feel free to answer as detailed as possible. Please also remember, you do not have to answer any question that you do not wish to answer Time required: 45 minutes 1 hour. Risks and Benefits: We anticipate little to no risk involved with participation in this study. We do not anticipate that you will benefit directly by participating in this experiment. Alternative Advantageous Procedures: It is anticipated that little to no risk as well as no direct benefit will be e xperienced as a result of participating in this research study. Hence, no immediate alternative advantageous procedures are closely apparent or related to this research study. Compensation: You will be given a $2 .00 Amazon.com gift card as compensation fo r participating in this research.

PAGE 159

159 Confidentiality: Your identity will be kept confidential to the extent provided by law. Data collected through questionnaires are expected to have no identifiers that could distinguish participants individually. The only demographics anticipated to be obtained in this study are gender, race, age range, and school. Data collected will be coded accordingly to prevent direct association with given responses Data by means of filled questionnaires will be kept in a locked fil e in my faculty supervisor's office. When the study is completed and the data have been analyzed, the questionnaires will be destroyed. Your name will not be used in any report. If you elect to have a $2 Amazon gift card emailed to you upon completion, you r email address and IP address will not be maintained after the gift card has been emailed to you. Your email address and IP address will be deleted once your email is sent out. Reporting: The collected data are for dissertation research purposes. Thus, a final report of analyzed data will be public domain and available through the U.S. Library of Congress and online in ProQuest Digital Dissertations. Voluntary participation: Your participation in this study is completely voluntary. There is no penalty fo r not participating. Right to withdraw from the study: You have the right to withdraw from the study at any time without consequence. Whom to contact if you have questions about the study: Enmanuel Chavarria, MS, CHES, Graduate Student, Principal Investiga tor, Department of Health Education and Behavior, FLG 23, P.O. Box 118210, Gainesville, FL 32611 8210, phone: (352) 294 1807. Elizabeth Chaney, PhD, MCHES, College of Health and Human Performance, FLG 12, P.O. Box 118210, Gainesville, FL 32611 8210, phone: (352) 294 1813. Whom to contact about your rights as a research participant in the study: IRB02 Office, Box 112250, University of Florida, Gainesville, FL 32611 2250; phone: (352) 392 0433. Agreement: I have read the procedure described above. I voluntari ly agree to participate in the procedure and I have received a copy of this description. Clicking this button means that you agree to participate in this study as described. I ACCEPT

PAGE 160

160 APPENDIX I AD HOC QUESTIONNAIRE

PAGE 161

161

PAGE 162

162

PAGE 163

163

PAGE 164

164

PAGE 165

165

PAGE 166

166

PAGE 167

167

PAGE 168

168

PAGE 169

169

PAGE 170

170

PAGE 171

171 LIST OF REFERENCES Agresti, A., & Finlay, B.F. (2009). Statistical methods for the social sciences (4 th Ed.). San Francisco, CA: Dellen Publishing Co. Ai, A.L., No l, L.T., Appel, H. B., Huang B., & Hefley, W.E. ( 2013 ). Overall health and health care utilization among Latino American men in the United States. American ( 1 ), 6 17 Aitken, L.M., Marshall, A., Elliott, R., & McKinley, S. ( 2011 and observation as data collection methods in the study of decision making regarding sedation in intensive care patients. International Journal of Nursing Studies, 48 (3), 318 325. Anderson, M., & Fienberg, S.E. (2000). Race and ethnicity and the controversy over the U.S. Census. Current Sociology, 48 (3), 87 110. Aragones, N. (n.d.) Top 25 Hispanic colleges. Retrieved from http://www.sunyrockland.edu/current students/student services/multicultural student support services/multicultural student support services/files/top25hi.pdf aws ire of prospective juror in New York. CNN Retrieved from http://www.cnn.com/2014/01/11/us/new york juror form negro/ Atkinson N L & Gold R S. (2002). The promise and challenge of eHealth interventio ns. American Journal of Health Behavior, 26 (6) 494 503. Atkinson, N.L., Saperstein, S.L., & Pleis, J. (2009). Using the internet for health related activities: findings from a national probability sample. Journal of Medical Internet Research, 11 (1), e4. Bandura, A. (1977 a ). Self efficacy: Toward a unifying theory of behavioral change. Physchological Review, 84 (2), 191 215. Bandura, A. (1977b). Social learning theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1986). Social foundations of thought an d action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A., & Adams, N.E. (1977). Analysis of self efficacy the ory of behavioral change. Cognitive Therapy and Research, 1 (4), 287 310. Bandura, A. (2006). Guide for constructing s elf efficacy scales. In Pajares F. & Urdan T. (Eds.). Self efficacy beliefs of adolescents (Vol. 5., pp. 307 337). Greenwich, CT: Information Age Publishing.

PAGE 172

172 Baker, L, Wagner, T.H., Singer, S., Bundorf, M.K. (2003). Use of the Internet and e mail for he alth care information: results from a national survey. Journal of the American Medical Association, 289 (18), 2400 2406. Barkley, T.W., Jr., & Burns, J.L. (2000). Factor analysis of the condom use self efficacy scale among multicultural college students. He alth Education Research, 15 (4), 485 489. Bartlett, J.E., Kotrlik, J.W., Higgins, C.C. (2001). Organizational research: Determining appropriate sample size in survey research. Information Technology, Learning, and Performance Journal, 19 (1), 43 50. Bass, S .B., Ruzek, S.B., Gordon, T.F., Fleisher, L., McKeown Conn, N., & Moore, D. (2006). Relationship of Internet health information use with patient behavior and self efficacy: Experiences of newly diagnosed cancer patients who contact the National Cancer Inst Journal of Health Communications, 11 (2), 219 236. Batten, L., & Dutton, J. (2011). Young tertiary stude nts and help seeking for health advice. Nursing Praxis In New Zealand, 27 (3), 31 42. Bellatin, A.M., Eddy, B., Perez, M., & Wolf, E. (2008). Closing the mental health gap: Eliminating disparities in treatment for Latinos. Retrieved from http://srdc.msstate.edu/opportunities/samhsa_full_report.pdf Bernstein, J. (2013, November 6). Social media in 2013: by the numbers Ret rieved from http://socialmediatoday.com/jonathan bernstein/1894441/social media stats facts 2013 Bernstein, A.B., Hing, E., Moss, A.J., Allen, K.F., Siller, A.B., Tiggle, R.B. (2003). Health care in America: T rends in utilization Hyattsville, Maryland : Na tional Center for Health Statistics. Betancourt, J.R., Carillo, J.E., Green, A.R. & Maina, A. (2004). Barriers to health promotion and disease prevention in the Latino population. Clinical Cornerstone, 6 (3), 16 29. Biswas, R., Martin, C. M., Sturmberg, J. Shanker, R. Umakanth, S., Shanker, S., & Kasturi, A. S. (2008). User driven health care -answering multidimensional information needs in individual patients utilizing po st EBM approaches: a conceptual model. Journal of Evaluation in Clinical Practice, 14 (5), 742 749. doi: 10.1111/j.1365 2753.2008.00998.x Borsari, B. E., & Carey, K. B. (1999). Understanding fraternity drinking: Five recurring themes in the literature, 1980 1998. Journal of American College Health, 48 (1), 30 37.

PAGE 173

173 Boulter, T., Moran, S., & Cole, G. (2011). Investigation of knowledge and perception of Tuberculosis among Utah County Hispanics. Retrieved from http://www.lung.org/associations/states/utah/assets/investigation of knowledge.pdf Buhi, E. R., Daley, E. M., Fuhrmann, H. J., & Sm ith, S. A. (2009). An observational study of how young people search for online sexual health information. Journal of American College Health, 58 (2), 101 111. Retrieved October 31, 2012, from http://search.proquest.com/docview/61826909?accountid=10920 Burt on, T. M. (2005). Health information on the inte rnet: Who seeks it and how does it affect the utilization of physician services? (University of California, Los Angeles). ProQuest Dissertations and Theses. Retrieved November 1, 2012, from http://search.proq uest.com/docview/305003678?accountid=10920. (305003678). Bush N E Bowen J Wooldridge J Ludwig A Meischke H & Robbins R. (2004). What do we mean by internet a cce ss? A framework for health r esearchers. Preventing Chronic Disease, 1 (4), 1 17. C asey, M.A. (1998). Analysis honoring the stories. In R.A. Krueger (Ed.), Analyzing and reporting focus group results (pp. 80 85) Thousand Oaks, CA: Sage Publications, Inc. Cashin, J. R., Presley, C. A., & Meilman, P. W. (1998). Alcohol use in the Greek system: Follow the leader? Journal of Studies on Alcohol, 59, 63 70. Centers for Disease Control and Prevention [CDC]. (2011, January 14). CDC health disparities and inequalities report United States, 2011. MMWR. Morbidity and Mortality Weekly Reports R etrieved December 10, 2012, from http://www.cdc.gov/mmwr/pdf/other/su6001.pdf Chaney, B.H., Barry, A.E., Chaney, J.D., Stellefson, M.L. & Webb, M. (2012). Using screen video capture software to aide and inform cognitive interviewing. Quality & Quantity D OI 10.1007/s11135 012 9669 4. Chisolm, D. J. (2010). Does online health information seeking act lik e a health behavior?: a test of the behavioral model. Telemedicine Journal and e Health,16 (2), 154 160. doi: 10.1089/tmj.2009.0102 Cline, R.J.W. & Haynes, K. M. (2001). Consumer health information seeking on the internet: the state of the art. Health Education Research, 16 (6), 671 692. Cochran, W.G. (1977). Sampling techniques (3rd e d. ). New York, NY: Wiley. Cohane, G.H., & Pope Jr., H.G. (2001). Body image in boys: a review of the literature. International Journal of Eating Disorders, 29 (4), 373 379.

PAGE 174

174 Cohen, J., Cohen, P., West, S.G., Aiken, L.S. (2002). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). New York, NY: Routle dge Cohen, S., Mermelstein, R., Kamarck, T., & Hoberman, H. (1985). Measuring the functional components of social support. In I. G. Sarason & B. R. Sarason (Eds.), Social support: Theory, research, and application (pp. 73 94) The Hague, Holland: Martinus Nijhoff. Collins, S.E., & Lewis, D.M. (2013). Social media made easy: guiding patients to credible online health information and engagement resources. Clinical Diabetes, 31 (3), 137. Cook, R.D., & Weisberg, S. (1982). Residuals and influence in regression London, UK: Chapman & Hall. Cotten, S.R. & Gupta, S.S. (2004). Characteristics of online and offline health information seekers and factors that discriminate between them. Social Science and Medicine, 59 (9), 1795 1806. Creswell, J.W. (2014). Research desi gn: Qualitative, quantitative and mixed methods approaches (4th ed.). Thousand Oaks, CA: SAGE Publications, Inc. Crozier, S. (2011, November 21). Insights gained in Hispanic survey: oral health information, access, insurance sought. Retrieved from https:// www.ada.org/news/6579.aspx De Boer, A.G.E.M., Wijker, W., & de Haes, H.C.J.M. (1997). Predictors of health care utilization in the chronically ill: a review of the literatur e. Health Policy, 42 (2), 101 115. De la Rosa, M. (1989). Health care needs of Hispa nic Americans and the responsiveness of the health care system. Health and Social Work, 14 (2), 104 113. Delorme, D. E., Huh, J., & R eid, L. N. (2010). Evaluation, use, and usefulness of prescription drug information sources among Anglo and Hispanic America ns. Journal of Health Communication, 15 (1), 18 38. doi: 10.1080/10810730903460526 DeMaio, T.J., Rothgeb, B., & Hess, J. (1998). Improving survey quality through pretesting. Working papers in survey methodology (No. 98/03). Washington, D.C.: U.S. Bureau of the Census. Dillman, D.A., Smyth, J.D., Christian, L.M. (2008). Internet, mail, and mixed mode surveys: the tailored design method. (3rd ed.). Hoboken, NJ:Wiley.

PAGE 175

175 Dobransky, K., & Hargittai, E. (2012). Inquiring minds acquiring wellness: Uses of online and offline sources for health info rmation. Health Communication, 27 (4), 331 343. doi: 10.1080/10410236.2011.585451 Doyle, J. (2009). Using focus groups as a research method in intellectual disability research: A practical guide Galway, Ireland: National Fed eration of Voluntary Bodies. Duggan, M., & Smith, A. (2013). Cell internet use 2013 Washington, DC: Pew Internet & American Life Project. Retrieved December 3, 2013, from http://www.pewinternet.org/~/media//Files/Reports/2013/PIP_CellInternetUse2013. pdf D unlop D.D. Manheim, L.M., Song J., & Chang R.W. (2002) Gender and ethnic/racial disparities in health care utilization among older a dults Journal of Gerontology: SOCIAL SCIENCES, 57B (3), s221 S233. Easaw, S. (2010). Health information version 2.0: Fe male students in cyberspace. (M.S., San Jose State University). Retrieved October 29, 2012, from http://proquest.umi.com/pqdweb?did=2180056891&Fmt=7&clientId=20179&RQT= 3 09&VName=PQD. (1482537). Eastin, M.S. (2001) Credibility assessments of online health information: the effects of source expertise and knowledge of content. Journal of Computer Mediated Communication, 6 (4), 0. doi: 10.1111/j.1083 6101.2001.tb00126.x Ericsson, K., & Simon, H. (1993). Protocol analysis: Verbal reports as data (2nd ed.). Cambr idge, MA: MIT Press. Escoffery, C., Miner, K. R., Adame, D. D., Butler, S., McCormic k, L., & Mendell, E. (2005). Internet use for health information among college students. Journal of American College Health, 53 ( 4 ), 183 183. Retrieved October 29, 2012, fro m http://search.proquest.com/docview/62127121?accountid=10920 Eysenbach, G. (2008). Medicine 2.0: social, networking, collaboration, participation, apomediation, and openness. Journal of Medical Internet Research, 10 (3), e22. doi: 10.2196/jmir.1030 Fama, J Retrieved from http://abcnews.go.com/Politics/OTUS/us census bureau drops negro surveys/story?id=18591761 Fernz, E. F. (1982). The use of foc us groups for idea generation: the effects of group size, acquaintances, and moderator on response quantity and quality. Journal of Marketing Research, 19 (1), 1 13.

PAGE 176

176 Fogel, J., Fajiram, S., & Morgan, P. D. (2010). Sexual hea lth information seeking on the I n ternet: Comparisons between White and African American college students. Journal of the Association of Black Nursing Faculty in Higher Education, 21 (4), 79 84. http://www.forbes. com/top colleges/list/ Foubert, J. D. (2000). The longitudinal effects of a rape prevention program on fraternity Journal of American College Health, 48 (4), 158 163. Fox, S. (2006). Online health search 200 6. Washington, DC: Pew Internet & American Life Project. Retrieved November 3, 2012, from http://www.pewinternet.org/~/media/Files/Reports/2006/PIP_Online_Health_2006. pdf.pdf Fox, S. (2011). The social life of health information, 2011. Washington, DC: Pew Internet & American Life Project. Retrieved November 4, 2012, from http://pewinternet.org/~/media/Files/Reports/2011/PIP_Social_Life_of_Health_Info .pdf Fox, S. & Duggan, M. (2013). Health Online 2013. Washington, DC: Pew Internet & American Life Project. Retrieved from http://www.pewinternet.org/~/media/Files/Reports/2013/Pew%20Internet%20Healt h%20Online%20report.pdf Fox, S., & Rainie, L. (2002). Vital decisions Washington, DC: Pew Internet & American Life Project. Retrieved from http://www.pewinternet.or g/~/media//Files/Reports/2002/PIP_Vital_Decisions_May 2002.pdf.pdf Fraternal Organization. (n.d.). Investopedia. Retrieved November 18, 2012, from http://www.investopedia.com/terms/f/fraternal organization.asp#axzz2CXBIfJYt Freeman, E., Barker, C., & Pistra ng, N. (2008). Outcome of an online mutual support group for college students with psychological problems. CyberPsychology & Behavior, 11 (5), 591 593. doi: 10.1089/cpb.2007.0133 Fry, R. & Lopez, M.H. (2012). Hispanic student enrollments reach new highs in 2011: now the largest minority group on four year college campuses. Washington, DC: Pew Hispanic Center, Pew Research Center. Retrieved November 4, 2012, from http://www.pewhispanic.org/files/2012/08/Hispanic Student Enrollments Reach New Highs in 2011_FIN AL.pdf

PAGE 177

177 Gamboa, S. (2012). Hispanic college enrollment reaches two million, Latinos are largest minority on campus. Retrieved from http://www.huffingtonpost.com/2012/08/20/hispanics college enrollment largest minority_n_1813655.html#slide=1322872 Garfield, R. (2010). enhance couples therapy. Family Process, 49 (1), 109 122. Garrard, J. (2010). Health science literature review made easy: the matrix method Sudbury, MA: Jones & Bartlett Learning. Ghaddar S. F., Valerio, M. A., Garcia, C. M., & Hansen, L. (2012). Adolescent health l iteracy: The importance of credible sources for online health information. Journal of School Health, 82 ( 1), 28 36. doi: 10.1111/j.1746 1561.2011.00664.x Glanz, K., Rimer, B.K & Lewis, F.M. (2002). Health behavior and health e ducation. Theory, Research and Practice. San Fransisco CA : Wiley & Sons. Goldman, N., Kimbro, R.T., Turra, C.M. & Pebley, A.R. (2006). Socioeconomic gradients in health for White and Mexican origin popula tions. American Journal of Public Health, 96 (12), 2186 2193. Grabe, S., & Hyde, J. S. (2006). Ethnicity and body dissatisfaction among women in the United States: A meta analysis. Psychological Bulletin, 132 622 640. Gray, N., Klein, J., Noyce, P., Sessel berg, T. & Cantrill, J. (2005). Health information seeking behaviour in adolescence: the place of the I nternet. Social Science & Medicine, 60 ( 7), 1467 1478. Grudens Schuck, N., Allen, B.L., & Larson, K. (2004). Focus group fundamentals Methodology Breif PM 1989b. Ames, IA: Iowa State University Extension. Guardia, J.R. & Evans, N.J. (2008). Factors influencing the ethnic identity development of Latino fraternity members at a Hispanic serving institution. Journal of College Student Development, 49 (3), 163 Gurman, T., & Borzekowski, D. L. G. (2004). Condom use among Latino college students. Journal of American College Health 52 (4),169 178. doi: 10.3200/JACH.52.4.169 178 Hanik B. & Stellefson M.L. (2011). E health literacy com petenci es among undergraduate health education students: a preliminary s tudy. International Electronic Journal of Health Education, 14 46 58. Retrieved October 27, 2012, from http://search.proquest.com/docview/964181303?accountid=10920 Har bour, R. & Miller, J. (2001). A new system for grading recommendations in evidence based guidelines. BMJ, 323 (August 11), 334 336.

PAGE 178

178 Hardy, M.A. (1993). Regression with dummy variables. Series: Quantitative applications in the social sciences. Thousand Oaks CA: Sage Publications, Inc. Harris, K.J., Ahluwalia, J.S., Catley, D., Okuyemi, K.S., Mayo, M.S., Resnicow, K. (2003). Successful recruitment of minorities into clinical trials: The Kick It at Swope project. Nicotine & Tob acco Res earch 5 (4), 575584. Ha sebrink, U, & Domeyer, H. (2012). Media repertoires as patterns of behaviour and as meaningful practices: A multimethod approach to media use in converging media environments. Journal of Audience and Reception Studies, 9(2), 757 779. Hassani, S.N. (2006). Locating digital divides at home, work, and everywhere else. Poetics, 34 ( 4 5 ), 250 272. Health Behavior. (n.d.). Medical Dictionary The Free Dictonary. Retrieved November 17, 2012 from http://medical dictionary.thefreedictionary.com/heal th+behavior Health Care Cost Institute. (2012). Health care cost and utilization report: 2011. Washington, DC. Helem L. (2004, Nov 01). Greeks go L atin or vice versa. Newsweek, 144, 51 51. Hesse, B., Nelson, D., Kreps, G, Croyle, R.T., Arora, N.J., Rimer B.K., Viswanath, K. (2005). Trust and sources of health information. Archives of Internal Medicine, 165 (22), 2618 2624. Hesse Biber, S.N., & Leavy, P. (Eds.). (2003). Approaches to qualitative research: A reader on theory and practice. New York, NY: Oxfo rd University Press. Hesse Biber, S.N., & Leavy, P. (Eds.). (2008). Handbook of emergent methods. New York, NY: The Guildford Press. Hochbaum, G.M. (1958). Public participation in medical screening programs: A socio psychological study (Public Health Serv ice Publication No. 572). Washington, D.C.: Government Printing Office. Hispanic Association of Colleges and Universities (HACU) (n.d.).List of HACU members. Retrieved from http://www.hacu.net/hacu/HACU's_Members.asp Hu, Y. (2007). Healt h information on the I nternet: I nfluence of online sources on credibility and behavioral intentions. (Pennsylvan ia State University). ProQuest Dissertat ions and Theses. Retrieved November 3, 2012, from http://search.proquest.com/docview/304822532?acc ountid=1 0920. (304822532).

PAGE 179

179 Hurtado, S., & Carter, D. F. (1997). Effects of college transition and perceptions of the campus racial climate on Latino colleg e students' sense of belonging. Sociology of Education, 70 (4), 324 345. Ingelmo, J.J.V. (2012). The p erception of belonging: Latino undergraduate students participation in the social and academic life at a predominately White private university. Retreived from http://digitool.library.colostate.edu///exlibris/dtl/d3_1/apache_media/L2V4bGlicmlzL 2R0bC9kM18xL 2FwYWNoZV9tZWRpYS8xNzAxMTU=.pdf Jarama Alvan, S. L., Belgrave, F. Z., & Zea, M. C. (1996). Stress, social support, and c ollege adjustment among La tino students. Cultural Diversity and Mental Health, 2 (3), 193 203. doi: 10.1037/1099 9809.2.3.193 Jones, W. J (1989). Personality and epistemology: Cognitive social learning theory as a philosophy of science. Zygon, 24 (1), 23 38. Kerwin, J., Willis, G. (2011). Cognitive interview pretesting of a health information survey. In L. Finney Rutten B. Hesse, R. Moser, & G. Kreps (E ds.). Building the evidence base in cancer communication: Health Information National Trends Survey (HINTS) ( pp. 47 60 ) New York, NY: Hampton Press Inc. Kim, H., Park, S., & Bozeman, I. (2011). Online health information search and evaluation : Observations and semi structured interviews with college students and maternal health experts. Health Information & Libraries Journal, 28 (3), 188 199. doi: 10.1111/j.1471 1842.2011.00948.x Kim, J. & Park, H.A. (2012). Development of a health information technology Journal of Medical Internet Research, 14 (5), e133. d oi: 10.2196/jmir.2143 Koenig, H.G., Westlund, R.E., George, L.K., Hughers, D.C., Blazer, D.G., Hybels, C. (1993). Abbreviating the D uke Social Support Index for use in chronically ill elderly individuals. Psychosomatics, 34 (1), 61 69. Kontos, E.Z., Bennett, G.G., Viswanath, K. (2013). Barriers and facilitators to home computer and internet use among urban novice computer users of low socioeconomic p osition Journal of Medical Internet Research, 9 (4), e31. Kontos, E.Z., Emmons, K.M., Puleo, E., & Viswanath, K. (2010). Communication inequalities and public health implications of adult social networking site u se in the United States Jour nal of Health Communication: International Perspectives 15 (supp 3) 216 235. Kontos, E.Z., Emmons, K.M., Puleo, E., & Viswanath, K. (2012). Contribution of communication inequalities to disparities in human papillomavirus vaccine awareness and k nowledge. American Journal of Public Health 102 (10), 1911 1920

PAGE 180

180 Krueger, R.A. (1998). Developing questions for focus groups. In D.L. Morgan & R.A. Krueger (Eds.). The focus group kit (Vol. 3). Thousand Oaks, CA: Sage. Krueger, R. A., & Casey, M. A. (2000). Focus gro ups: a practical guide for applied research (3rd ed.). Thousand Oaks: Sage Publications, Inc. Kuh, G. D., & Arnold, J. C. (1993). Liquid bonding: A cultural analysis of the role of alcohol in fraternity pledgeship. Journal of College Student Development, 3 4, 327 334. LaCaille, R. A., & Kuvaas, N. J. (2011). Coping styles and self regulation predict complementary and alternative medicine and herbal supplement use among college students. Psychology, Health & Medicine, 16 (3), 323 332. doi:0.1080/13548506.2010. 543909 LaJoie, A.S., & Ridner, S.L. (2009). Health information and health risk behaviors in a sample of college students. Journal of the Kentucky Medical Association 107 (2), 58 63. Larimer, M. E., Anderson, B. K., Baer, J. S., & Marlatt, G. A. (2000). An individual in context: Predictors of alcohol use and drinking problems among Greek and residence hall students. Journal of Substance Abuse, 11 (1), 53 68. Larimer, M. E., Irvine, D. L., Kilmer, J. R., & Marlatt, G. A. (1997). College drinking and the Greek system: Examining the role of perceived norms for high risk behavior. Journal of College Student Development, 38 (6), 587 598. LaVange, L.M., Kalsbeek, W., Sorlie, P.D., Avil s Santa, L.M., Kaplan, R.C., Barnhart, J., Elder, J.P. (2010). Sample design and cohort selection in the Hispanic health study/study of Latinos. Annals of Epidemiology, 20 (8), 642 649. Leaffer, T., & G onda, B. (2000). The I nternet: a n underutilized tool in patient education. Computers in Nursing, 18 (1), 47 52. PMID: 10673816 Lee, S. Y. Hwang, H., Hawkins, R., & Pingree, S. (2008). Interplay of negative emotion and health self efficacy on the use of health information and its outcomes. Communication Research, 35 (3), 358 381. Leech, N. L., & Onwuegbuzie, A. J. (2007). An array of qualita tive analysis tools: A call for data analysis triangulation. School Psychology Quarterly, 22 557 584. doi:10.1037/1045 3830/22/4/557 Lie, D., Lee Rey, E., Gomez, A., Bereknyei, S., & Braddock, C. H. (2011). Does cultural competency training of health pro fessiona ls improve patient outcomes? A systematic review and proposed algorithm for f uture research. Journal of General Internal Medicine, 26 (3), 317 325. doi:10.1007/s11606 010 1529 0

PAGE 181

181 Lilley, S. (2012, August 20). Report: Latinos are the largest minority group on college campuses. Retrieved from http://nbclatino.com/2012/08/20/latinos are now the largest minority group on college campuses/ Lincoln, Y.S. & Guba E. G. ( 1985 ). Naturalistic inquiry. Thousand Oaks, CA: Sage publications. Lindlof, T.R. & Taylor, B.C. (2010). Qualitative communication research methods (3rd ed.). Thousand Oaks, CA: SAGE Publications, Inc. Liu, M.C. (2011, July). Investing in higher education for Latinos: Trends in Latino college access and success. Retrieved from http://www. ncsl.org/documents/educ/trendsinlatinosuccess.pdf Livingston, G., Minushkin, S., & Cohn, D. (2008). Hispanics and health care in the United States: Access, information and knowledge. (Web, Government Report). Washington, DC: Robert Wood Johnson Foundation and Pew Hispanic Center. Lo, C. C., & Globetti, G. (1995). The facilitating and enhancing roles Greek associations play in college drinking. The International Journal of the Addictions, 30 (10), 1311 1322. tions can no longer ignore Hispanic marketing like Mitt Romney did. Retrieved from http://www.forbes.com/sites/glennllopis/2012/11/12/americas corporations can no longer ignore hispanic marketing like mitt romney did/ Lu, H. (2005). Factors affecting inten tions to seek information about STDs and HIV/AIDS on the I nternet among Taiwanese college students. (University of Kentucky). ProQuest Dissertations and Theses. Retrieved November 16, 2012, from http://search.proquest.com/docview/304 996065?a ccountid=10920. (304996065). Lund, A., & Lund, M. (2013 ). Multiple regression in SPSS. Retrieved from https://statistics.laerd.com/premium/mr/multiple regression in spss 4.php Lust, K., Ehlinger, E.P., Golden, D., Sanem, J., Bakke, B., & Bartkus, A. (2007) University of Minnesota systemwide student health report Minneapolis, MN: Boyton Health Service, University of Minnesota. Retrieved December 10, 2012, from http://www.bhs.umn.edu/survey s/survey results/Systemwide_Report_07.pdf Luszczynska, A., & Schwarzer, R. (2005). Social cognitive theory. In M. Conner & P. Norman (Eds.), Predicting health behaviour (2nd ed. rev., pp. 127 169). Buckingham, England: Open University Press.

PAGE 182

182 Madden M. & Fox, S. (2006). Finding answers online in sickness and in health. Washington, DC: Pew Internet and American Life Project Report. Retrieved November 2, 2012 from http://www.pewinternet.org/Reports/2006/Finding Answers Online in Sickness and in Health.aspx Marshall, C. & Rossman, G.B. (1999). Designing qualitative research (3rd ed.). London: Sage Publications Inc. Martin, J. & Fogel, S. (2006). Projecting the U.S. population to 2050: four immigration scenarios. Washington, DC: FAIR Horizon Press. Retrieved November 7, 2012, from http://www.fairus.org/site/DocServer/pop_projections.pdf Martinez, M., & Ariosto, D. (2011, March 24). Hispanic population exceeds 50 million, firmly nation's No. 2 group. Re trieved from http://www.cnn.com/2011/US/03/24/census.hispanics/ McAlister, A.L., Fernandez Esquer, M.E., Ramirez, A.G., Trevino, F., Gallion, K.J., Villareal, R., Pulley, L.V., Hu, S., Torre s, I., Zhang, Q. (1995). Community level cncer control in a Texas barrio: Part II baseline and preliminary outcome findings. Journal of the National Cancer Institute Monographs, 18 123 126. McCabe, M.P., & Ricciardelli, L.A. (2004). Body image dissatisfaction among males across the lifespan: a review of past literatur e Journal of Psychosomatic Research, 56 (6), 675 685. Meade, C.D., Calvo, A., Rivera, M.A., & Baer, R.D. Focus groups in the design of prostate ca ncer screening information for Hispanic farmworkers and African American men. Oncology Nursing Forum, 30 (6), 96 7 975. Menendez, R. (n.d.) Latinos and college access: ensuring young Latinos can achieve the American dream. Latino Leadership Link: A Democratic Update on the Hispanic Community Retrieved November 13, 2012 from http://www.menendez.senate.gov/pdf/HigherE dLLL072407eng.pdf Miller, L.M.S. & Bell, R.A. (2011). Online health information seeking: the influence of age, information trustworthiness, and search challenges. Journal of Aging and Health, 24 (3), 525 541. doi: 10.1177/0898264311428167 Miranda M.L (199 9). Greek life...with a little sazon. The World & I, 14 (9), 191. Mischel, W., & Shoda, Y. (1995). A cognitive affective system theory of personality: Reconceptualizing situations, dispositions, dynamics, and invariance in personality structure. Psychologic al Review, 102 ( 2 ), 246 268.

PAGE 183

183 Morahan Martin, J., & Anderson, C. D. (2000). Information and misinformation online: recommendations for facilitating accurate mental health info rmation retrieval and evaluation. CyberPsychology & Behavior, 3 (5), 731 746. doi: 10.1089/10949310050191737 Morgan, D. L. (1997). Focus groups as qualitative research (2nd ed.). Thous and Oaks, CA: Sage Publications, Inc. Morgan, D.L. (1998). Computerized analysis. In R.A. Krueger (Ed.), Analyzing and r eporting focus group results (pp. 89 93). Thousand Oaks, CA: Sage Publications, Inc. Morgan, D.L., & Krueger, R.A. (1998). The focus group kit (Vol 1 6). Thousand Oaks, CA: Sage Publications, Inc. Morrow Howell, N. (1994). The m word: multicollinearity in multiple regression. Social Work Research 18 (4), 247 251. Mubarak, A. R., Rohde, A., & Pakulski, P. (2009). The social benefits of online chat rooms for university students: an explorative study. Journal of Higher Education Policy & Management, 31 (2), 16 1 174. doi: 10.1080/13600800802559310 National Association of Latino Fraternal Organizations, Inc. [NALFO]. (n.d.). List of institutions housing member organizations. Retrieved from http://www.nalfo.org National Center for Health Statistics. (2012). Heal th, United States, 2011: with special feature on socioeconomic status and health. Hyattsville, MD. National Museum of the American Latino Commission [NMAL]. (2011). To illuminate the American story for all : A report to the President and Congress of the Un ited States. Washington, D.C.: National Museum of the American Latino Commission Nguyen, J.D., Carson, M.L., Parris, K.M., Place, P. (2003). A comparison pilot study of public health field nursing home visitation program interventions for pregnant Hispani c adolescents. Public Health Nursing, 20 (5), 412 418. Nielsen. (2013, May 1). Retrieved from http://www.nielsen.com/content/corporate/us/en/newswire/2013/latino populations are growing fastest where we arent looking.html Nunnally, J.C., & Bernstein, I.H. (1994). Psychometric theory (3rd ed.). New York, N.Y.: McGraw Hill.

PAGE 184

184 Nustad, J., Adams, T., & Moore, M. (2008). Health information sources accessed by college females: d ifferences between bod y i mage distorted and non body image distorted. Health Marketing Quarterly, 25 (3), 241 253. doi: 10.1080/07359680802081837 Oates, C. (2002). The use of focus groups in social research. In Burton, D. (eds.). Research tra ining for social sciences. London, UK: Sage Publications. http://stateimpact.npr.org/florida/2011/08/25/hispanic college enrollment surges Oh, H.J., & Lee, B. (2012). The e ffect of computer mediated s ocial support in online c ommunities on patient empowerment and doctor patient c ommunication. Health Communication 27 (1), 30 41. Onwuegbuzie, A.J., Dickinson, W.B., Leech, N.L., & Zoran, A.G. (2009). Toward more rigor in focus group research: A new framework for collecting and analyzing focus group data. International Journal of Qualitative Methods, 8 (3), 1 21. Ormrod, J.E. (2006). Education psychology: Developing learners (5th ed.). Upper Saddle River, NJ: Pearson/Merrill Pren tice Hall. Owens, D.K., Lohr, K.N., Atkins, D., Treadwell, J. R., Reston, J.T., Bass, E.B., Chang, S., & Helfand, M. (2004). Grading quali ty of evidence and strength of recommendations. BMJ, 328 (7454), 1490 1494. Owsley, C., McGwin, G., Scilley, K., Girkin, C.A., Phillips, J.M., & Searcey, K. (2006). Perceived barriers to care and attitudes about vision and eye care: focus groups with older African Americans and eye care providers. Investigative Ophthalmology & Visual Science, 47 (7), 2797 2802. zcan, N., & Buzlu, S. (2007). Internet use and its relation with the psychosocial situation for a sample of university student s. Cyberpsychology & Behavior, 10 (6), 767 772. doi:10.1089/cpb.2007.9953 Pagliari, C., Sloan, D., Gregor, P., Sullivan, F., Detmer, D., Kahan J.P., Oortwijn, W., MacGillivray, S. (2005). What is eHealth (4): a scoping exercise to map the field. Journal of Medical Internet Research, 7 (1), e9. doi: 10.2196/jmir.7.1.e9 Pallant, J. (2010). SPSS survival manual. A step by step guide to data analysi s using SPSS. New York, NY: McGraw Hill. Pap p a, E. & Niakas, D. (2006). Assessment of health care needs and utilization in a mixed public private system: the case of the Athens area. BMC Health Services Research, 6 146. d oi: 10.1186/1472 6963 6 146

PAGE 185

185 Pe a P urcell N. (2008). Hispanics' use of I nternet heal th information: an exploratory study. Journal of the Medical Library Association, 96 (2), 101 107. doi: 10.3163/1536 5050.96.2.101 Percheski, C., & Hargittai, E. (2011). Health informat ion seeking in the digital age. Journal of American College Health, 59 (5), 379 386. doi: 10.1080/07448481.2010.513406 Pleis, J.R. & Lethbridge ejku, M. (2007). Summary health statistics for U.S. adults: national health interview survey, 2006. National Center for Health Statistics. Vital and Health Statistics, 10 (235). Ponce, C. & Comer, B. (2003). Is acculturation in Hispanic health research a flawed concept? East Lansing, MI: The Julian Samora Research Instit ute, Michigan State University. Powers, J., Goodger, B., Byles, J.E. (2004). Assessment of the abbreviated Duke Social Support Index in a cohort older Australian women. Australian Journal on Ageing, 23 (2) 71 76. Qualtrics. (n.d.). Security statement Retrieve d from http://qualtrics.com/security statement/ Raffaelli, M., Zamboanga, B.L., & Carlo, G. (2005). Acculturation status and sexuality among female Cuban American college students. Journal of American College Health, 54 (1), 7 13. Rice, R.E. (2006). Influen ces, usage, and outcomes of Internet health information searching: Multivariate results from the Pew surveys International Journal of Medical Informatics, 75 (1), 8 28. Rideout, V. (2001). Generation Rx.com: how young people use the I nternet for health inf ormation. Menlo Park, CA.: publication 3202. Ross, M. W., Rosser, B. R. S., & Stanton, J. (2004). Beliefs about cybersex and I nternet media ted sex of Latino men who have I nternet sex with men: relationships with sexual practices in cyberse x and in real lif e. AIDS Care, 16 (8), 1002 1011. doi: 10.1080/09540120412331292444 Sabo, D. (2000). Men's health studies: o rigins and trends. Journal of American College Health, 49 (3), 133 142. doi:10.1080/07448480009596295 Salabarr a Pe a, Y., Trout, P.T., Gill, J.K., Mor isky, D.E., Muralles, A.A., Ebin, V.J. related behaviors. Ethnicity & Disease, 11 (4), 661 675.

PAGE 186

186 Salazar, G., Byers, S. R., Salas, R. A., & Yang, R. K. (2009). Cultural adapt ation to a university campus: the case of Latino students. Journal of Hispanic Higher Education, 8 (2), 115 129. doi: 10.1177/1538192709331975 Scott Sheldon, L.A.J., Carey, K.B. & Carey, M.P. (2008). Health behavior and college students: does Greek affiliat ion matter? Journal of Behavioral Medicine, 31 (1), 61 70. Selsky, C., Luta, G., Noone, A.M., Huerta, E.E., Mandelblatt, J.S. (2013). Internet access and online cancer information seeking among Latino immigrants from safety net clinics. Journal of Health C ommunication, 18 (1), 58 70. Shanahan, M. (2008). Transforming information search and evaluation practices of undergraduate students. International Journal of Medical Informatics, 77 (8), 518 526. doi: 10.1016/j.ijmedinf.2007.10.004 Shanahan, M. (2009). Lear ning centred approac h for developing the electronic information search processes of students. Medi cal Teacher, 31 (11), 994 1000. doi: 10.3109/01421590802572726 Smith, A. (2011). 35% of American adults own a smartphone: one quarter of smartphone owners use their phone for most of their online browsing Washington, DC: Pew Internet & American Life Project. Retrieved November 5, 2012, from http://pewinternet.org/~/media//Files/Reports/2011/PIP_Smartphones.pdf Smith McLallen, A., Fishbein M., & Hornik, R.C. ( 2011) Psychosocial determinants of cancer related i nform ation seeking among cancer p atients. Journal of Health Communication 16 (2) 212 225. Sorlie, P.D., Avils Santa, L.M., Wassetheil Smoller, S., Kaplan, R.C., Daviglus, M.L., Giachello, A.L., Heiss, G. (2010). Design and implementation of the Hispanic community health study/study of Latinos. Annals of Epidemiology, 20 (8), 629 641. Spraggins, A. (2009). Problematic use of online social networking sites for college students: Prevalence, predictors, and as sociation with well being. ( University of Florida). Pro Quest Dissertations and Theses. Retrieved November 16, 2012, from http://search.proquest.com/docview/761142588?accountid=10920. (761142588). Standford Patient Education Research Center. (n.d.) Health care utilization. Retrieved from http://patienteducation.stanford.edu/research/utilization.html Statistics Canada (n.d.). Canadian community health survey (CCHS): Social support. Retrieved from http://www.st atcan.gc.ca/concepts/health sante/pdf/support soutien eng.pdf Tichenor, P.J., Donohue, G.A., & Olien, C.N. (1980). Community conflict and the press. Newbury Park, CA: Sage Publications.

PAGE 187

187 Tong, F. (2006). A comparison between the use of beta weights and stru cture coefficients in interpreting regression results Austin, TX: Southwest Educational Research Association. Trainor, A.A. & Graue, E. ( 2012). Reviewing qualitative research in the social sciences. New York, NY: Routledge. U.S. Census Bureau. (2008, Aug ust 14). An older and more diverse nation by midcentury. Retrieved November 2, 2012 from https://www.census.gov/newsroom/releases/archives/population/cb08 123.html U.S. Census Bureau. (2009, January). The 2010 Census Questionnai re: Informational Copy. Ret rieved from http://2010.census.gov/2010census/pdf/2010_Questionnaire_Info_Copy.pdf U.S. Department of Health and Human Services. Office of Disease Prevention and Health Promotion. Healthy People 2020. Washington, DC. Available at http://www.healthypeople.g ov/2020/topicsobjectives2020/objectiveslist.aspx?topicI d=18. Accessed November 11, 2013. Van der Bijl, J. J., & Shortridge Baggett, L. M. (2002). The theory and measurement of the self efficacy construct. In E. A. Lentz & L. M. Shortridge Baggett (Eds.), Se lf efficacy in nursing: Research and measurement perspectives (pp. 9 28). New York: Springer. Retrieved from http://books.google.com/books?id=J6ujWyh_4_gC Van Someren, M.W., Barnard, Y.F., & Sandberg, J.A.C. (1994). The think aloud method: A practical guid e to modelling cognitive processes. London, UK: Academic Press. Varnes, J.R., Stellefson, M.L., Janelle, C.M., Dorman, S.M., Dodd, V., & Miller, M.D. (2013). A systematic review of studies comparing body image concerns among female college athletes and no n athletes, 1997 2012. Body Image, 10 (4), 421 432. Vega, W.A., Rodrguez M.A. & Gruskin, E. (2009). Health disparities in the Latino population. Epidemiologic Reviews, 31 (1), 99 112. Viswanath, K., Ramanadhan, S., & Kontos, E.Z. (2007). Mass media and po pulation health: A macrosocial view. In Macrosocial determinants of population health. Galea, S. (ed.). New York, NY: Springer, pp. 275 294. Viswanath, K., & Demers, D. (1999). Mass media from macrosocial perspective. In Demers, D., & Viswanath, K., (Eds. ). Mass media, social control and social change: A macrosocial perspective. Ames, I A: Iowa State University Press. Walsh, A.M., Hyde, M.K., Hamilton, K., & White, K.M. (2012). Predictive modelling: ation to increase their BMC Medical Informatics and Decision Making, 12 (1), 144 154.

PAGE 188

188 Wang, W. (2010). An investigation of online health support groups: e ffects of narrative exposure and social su pport on th e experience of sympathy, self disclosure and cognitive changes. (Pennsylvania State University). ProQuest Dissertat ions and Theses. Retrieved November 13, 2012 from http://search.proquest.com/docview/817791316?accountid=10920. (817791316). Wardian, J., Robbins, D., Wolfersteig, W., Johnson, T., & Dustman, P. (2012). Validation of the DSSI 10 to measure social support in a general population. Research on Social Work Practice, 23 (1), 100 106. Warren, J.R., Kvasny, L., Hecht, M.L., Burgess, D., Ahluwalia, J.S. & Okuyemi, K.S. (2010). Barriers, control and identity in health information seeking among African American women. Journal of Health Disparities Research and Practice, 3 (3), 68 90. Weaver, J.B., Mays, D., Weaver, S.S., Hopkins, G.L., Ero lu, D. & Bernhardt, J.M. (2010). Health information seeking behaviors, health indicators, and health risks. American Journal of Public Health, 100 (8), 1520 1525. Wei, M., Valdez, R.A., Mitchell, B.D., Haffner, S.M., Stern, M.P. & Hazuda, H.P. (1996). Migr ation status, socioeconomic status, and mortality rates in Mexican Americans and non Hispanic Whites: the San Antonio heart study. Annals of Epidemiology, 6 (4), 307 313. Workman, T.A. (2001). Finding the meanings of college drinking: an analysis of fratern ity drinking stories. Health Communication, 13 (4), 427 447. Wright, K.B. & Bell, S.B. (2003). Health related support groups on the I nternet: linking empirical findings to social support and computer mediated communication theory. Journal of Health Psycholo gy, 8 (1), 39 54. Huffington Post. Retrieved from http://www.huffingtonpost.com/2013/02/25/us census surveys will no longer use negro_n_2759306.html Young, M.M. (200 1). Hispanic health information outreach:Recommendations for NLM strategy and tactics. Retrieved from http://nnlm.gov/evaluation/tools/hispanicoutreach.pdf Younger, M.S. (1979). A handbook for linear regression. North Scituate, MA: Duxbury Press. Zuckerman M. (2009). Online health information seeking behaviors among college undergraduates. (University of Maryland, Baltimore County). ProQu est Dissertat ions and Theses. Retrieved November 6, 2012, from http://search.proquest.com/docv iew/305071036?accountid=10 920. (305071036).

PAGE 189

189 BIOGRAPHICAL SKETCH Enmanuel Antonio Chavarria was born in Weslaco, Texas. The youngest of three children, he grew up mostly in Miami, Florida, where he graduated from Coral Reef Senior High in 2005. In 2009, he graduated from the Unive rsity of Florida in Gainesville, Florida, receiving a Bachelor of Science degree with a major in Health Education and Behavior and a minor in leadership. In 2010, he graduated again from the University of Florida, receiving a Master of Science degree with a concentration in health education and human behavior. During his time at the University of Florida, Enmanuel became a Certified Health Education Specialist (CHES). In the fall of 2010, Enmanuel gained admittance to the University of Florida, to pursue a doctorate degree in health and human performance. During his time at the University of Florida, Enmanuel was committed to teaching undergraduate and fellow graduate students, participating in research that led to several published manuscripts and presentat ions at national conventions, as well as maintaining active membership in several professional and fraternal organizations. Enmanuel and his wife Nikita, now reside in Rockville Maryland where he contin ues his line of research as a postdoctoral fellow at G e o r g e M a s o n U n i v e r s i t y


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E6IMHREG1_3CMK9B INGEST_TIME 2014-10-03T21:42:29Z PACKAGE UFE0046247_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES