<%BANNER%>

Factors Influencing the Perceived Credibility of Diet-Nutrition Information Web Sites

Permanent Link: http://ufdc.ufl.edu/UFE0041768/00001

Material Information

Title: Factors Influencing the Perceived Credibility of Diet-Nutrition Information Web Sites
Physical Description: 1 online resource (72 p.)
Language: english
Creator: Jung, Eun
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: accuracy, communication, credibility, diet, expertise, health, message, nutrition, online, perception, source, website
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, M.A.M.C.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Many people are concerned with diet and nutrition because they are strongly correlated with overall health. To ensure that online information can help educate consumers about diet and nutrition, the perception of online information credibility has been emphasized. In this context, the purpose of the present study is to investigate the factors that impact the perceived credibility of online diet and nutrition information. By employing the elaboration likelihood model (ELM) to explain how users process online diet and nutrition messages, this study suggests factors that may influence Web site credibility. In view of that theoretical background, this study examines how source expertise and message accuracy influence the perceived credibility of online health information. In this sense, the literature review discusses why people go online to obtain diet and nutrition information and assesses the importance of online information credibility. An online experiment was conducted with 374 undergraduate students. The first research question asked whether there is a relationship among source expertise, message accuracy, and the perceived credibility of Web sites providing diet and nutrition information. The result of a Pearson s correlation analysis showed that perceived Web site credibility was positively and significantly correlated with source expertise and message accuracy. Hypothesis 1 and 2 predicted interaction effects between source expertise and message accuracy. The two-way ANOVA results indicated no significant interaction effect between source expertise and message accuracy on perceived Web site credibility, and therefore, Hypotheses 1 and 2 were not supported. However, there were statistically significant main effects of source expertise and message accuracy on perceived Web site credibility. H3 and H4 assumed that issue involvement will influence the impact of source expertise and message accuracy on perceived Website credibility. An ANCOVA was conducted, using involvement as the co-variate and the result revealed that issue involvement had a significant main effect on perceived Web site credibility. Based on these results, a two-way ANOVA was conducted to examine the relationship between involvement and source expertise, and involvement and message accuracy, separately. As a result, there was no significant interaction effect of issue involvement and source expertise, but there was a significant interaction effect between issue involvement and message accuracy. This analysis showed that message accuracy had a greater impact on perceived Web site credibility for highly involved participants (M= 4.17) than for low involvement participants. Therefore, hypothesis 3 was rejected but hypothesis 4 was supported. From these findings, this study will contribute to our understanding of the factors that affect the ways people process online diet and nutrition information. The results of this study could also help practitioners, such as health professionals and educators, understand the importance of sources of information and related user behavior (Kenkel, 1990; Nayga, 2000; Pauly & Satterthwaite, 1981).
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.
Statement of Responsibility: by Eun Jung.
Thesis: Thesis (M.A.M.C.)--University of Florida, 2010.
Local: Adviser: Walsh-Childers, Kim B.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041768:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041768/00001

Material Information

Title: Factors Influencing the Perceived Credibility of Diet-Nutrition Information Web Sites
Physical Description: 1 online resource (72 p.)
Language: english
Creator: Jung, Eun
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: accuracy, communication, credibility, diet, expertise, health, message, nutrition, online, perception, source, website
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, M.A.M.C.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Many people are concerned with diet and nutrition because they are strongly correlated with overall health. To ensure that online information can help educate consumers about diet and nutrition, the perception of online information credibility has been emphasized. In this context, the purpose of the present study is to investigate the factors that impact the perceived credibility of online diet and nutrition information. By employing the elaboration likelihood model (ELM) to explain how users process online diet and nutrition messages, this study suggests factors that may influence Web site credibility. In view of that theoretical background, this study examines how source expertise and message accuracy influence the perceived credibility of online health information. In this sense, the literature review discusses why people go online to obtain diet and nutrition information and assesses the importance of online information credibility. An online experiment was conducted with 374 undergraduate students. The first research question asked whether there is a relationship among source expertise, message accuracy, and the perceived credibility of Web sites providing diet and nutrition information. The result of a Pearson s correlation analysis showed that perceived Web site credibility was positively and significantly correlated with source expertise and message accuracy. Hypothesis 1 and 2 predicted interaction effects between source expertise and message accuracy. The two-way ANOVA results indicated no significant interaction effect between source expertise and message accuracy on perceived Web site credibility, and therefore, Hypotheses 1 and 2 were not supported. However, there were statistically significant main effects of source expertise and message accuracy on perceived Web site credibility. H3 and H4 assumed that issue involvement will influence the impact of source expertise and message accuracy on perceived Website credibility. An ANCOVA was conducted, using involvement as the co-variate and the result revealed that issue involvement had a significant main effect on perceived Web site credibility. Based on these results, a two-way ANOVA was conducted to examine the relationship between involvement and source expertise, and involvement and message accuracy, separately. As a result, there was no significant interaction effect of issue involvement and source expertise, but there was a significant interaction effect between issue involvement and message accuracy. This analysis showed that message accuracy had a greater impact on perceived Web site credibility for highly involved participants (M= 4.17) than for low involvement participants. Therefore, hypothesis 3 was rejected but hypothesis 4 was supported. From these findings, this study will contribute to our understanding of the factors that affect the ways people process online diet and nutrition information. The results of this study could also help practitioners, such as health professionals and educators, understand the importance of sources of information and related user behavior (Kenkel, 1990; Nayga, 2000; Pauly & Satterthwaite, 1981).
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.
Statement of Responsibility: by Eun Jung.
Thesis: Thesis (M.A.M.C.)--University of Florida, 2010.
Local: Adviser: Walsh-Childers, Kim B.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041768:00001


This item has the following downloads:


Full Text

PAGE 1

1 FACTORS INFLUENCING THE PERCEIVE D CREDIBILITY OF DIET -NUTRITION INFORMATION WEB SITES By E UN HWA JUNG A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR T HE DEGREE OF MASTER OF ARTS IN MASS COMMUNICATION UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Eun Hwa Jung

PAGE 3

3 To my supportive and beloved family

PAGE 4

4 ACKNOWLEDGMENTS I wish to thank all the people who have supported me in completing my Master s degree. First of all, I am deeply grateful to my advisor and committee chair, Dr. Kim Walsh -Childers for her invaluable advice, generous support, and encouragement. Her guidance always broaden ed my idea s and led me the right w ay. I also want to thank my committee members, Dr. Sylvia Chan -Olmsted and Dr. Cory Armstrong for their time and attentive support. Their invaluable suggestions helped develop my thesis. Moreover I would like to thank with the warmest heart, Dr. ChoonRy ul Ryu for his encouragement He helped me think in depth about communica tion phenomena when I was in my undergraduate and graduate years in South Korea. I am also thankful for Dr. Michael H. McBride who is a professor at the Texas State University for his support and advice since we met at Kookmin University in 2008. Additionally, I would like to thank my great friends in Gainesville for their encouragement and affection. I spent a wonderful time during my master s years, thanks to them. My special thanks also go to my good friends in South Korea. They always helped me when things were hard for me. Finally I want to express my deepest gratitude to my parents, Byeong -Sam Jeong and Bu Deok Kim for their unwavering love and strong belief. Also, I would like to thank my older brother, DongHyun for his affection and encouragement. T heir sacrifice, love, and caring have allowed me to achieve all the things today.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 4 TABLE OF CONTENTS ..................................................................................................................... 5 LIST OF TABLES ................................................................................................................................ 7 LIST OF FIGURES .............................................................................................................................. 8 ABSTRACT .......................................................................................................................................... 9 CHAPTER 1 INTRODUCTION ....................................................................................................................... 11 2 LITERATURE REVIEW ........................................................................................................... 16 The Importance of Diet and Nutrition Information ................................................................... 16 Online Diet -Nutrition Communication ...................................................................................... 17 The Ris k of Diet and Nutrition Misinformation........................................................................ 20 Assessment of Online Information Credibility: The Effects of Source Expertise and Message Accuracy ................................................................................................................... 21 Elaboration Likelihood Model of Persuasion ............................................................................ 25 The Role of Involvement ............................................................................................................ 27 3 RESEARCH METHODS ........................................................................................................... 30 Sample Selection ......................................................................................................................... 30 Online Experimental Procedure ................................................................................................. 30 Web Page Stimuli ........................................................................................................................ 31 Online Experimental Instrument ................................................................................................ 32 Independent Variables ......................................................................................................... 32 Dependent Var iable ............................................................................................................. 33 Control Variables ................................................................................................................. 34 Pilot Test Measures: Pre M anipulation Check .......................................................................... 36 Data Analysis ............................................................................................................................... 38 4 RESULTS .................................................................................................................................... 39 Profile of the Sample ................................................................................................................... 39 Media Usage and Diet Nutrition Information ........................................................................... 40 Post -M anipulation Check ........................................................................................................... 41 Re liability Test ............................................................................................................................ 42 Research Question 1 .................................................................................................................... 43

PAGE 6

6 Hypotheses Testing ..................................................................................................................... 44 Hypotheses 1 and 2 .............................................................................................................. 44 Hypotheses 3 and 4 .............................................................................................................. 46 Additional Findings: T he Role of Internet Experience and Demographic variables as Covariates ................................................................................................................................ 49 5 DISCUSSION .............................................................................................................................. 50 Findings for the Research Question and Hypotheses ................................................................ 50 The Correlation between Independe nt Variables and the Dependent Variable ............... 50 Interaction Effect between Source Expertise and Message Accuracy on Web site Credibility ......................................................................................................................... 51 The Relationship among Source Expertise, Message Accuracy and Issue Involvement ...................................................................................................................... 51 The Possibility of Other Covariates: Internet Experience and Demographic Variables ........................................................................................................................... 53 Theoretical and Practical Implications ....................................................................................... 53 Limitations and Future Research ................................................................................................ 55 APPENDIX A WEB PAGES STIMULI ............................................................................................................. 58 B Q UESTIONNAIRE ..................................................................................................................... 60 LIST OF REFERENCES ................................................................................................................... 65 BIOGRAPHICAL SKETCH ............................................................................................................. 72

PAGE 7

7 LIST OF TABLES Table page 3 1 Web page versions ................................................................................................................. 32 3 2 S ummary of Variables ........................................................................................................... 35 3 3 T test for Web site manipulation check: Web site sponsor expertise ................................. 37 3 4 T test for Web site manipul ation check: Message accuracy ............................................... 38 4 1 Demographic description ....................................................................................................... 39 4 2 Use of the Internet and other source for diet and nutrition information ............................. 41 4 3 T test for Web site manipulation check: Web site sponsor expertise, message accuracy, and issue involvement ........................................................................................... 42 4 4 Means and reliability check for each variable ...................................................................... 42 4 5 Correlations among source expertise, message accuracy, and Web site credibility .......... 43 4 6 Two -way ANOVA for Web site credibility ......................................................................... 45 4 7 Web site credibility ratings for groups with different levels of source expertise and message accuracy ................................................................................................................... 45 4 8 Correlations between issue involvement and Web site credibility ..................................... 46 4 9 ANCOVA for the effect of issue involvement, source expertise and message ac curacy on perceived Web site credibility .......................................................................... 46 4 10 Two -way ANOVA for the effects of Issue involvement and source expertise on perceived Web site credibility ............................................................................................... 47 4 11 Two -way ANOVA predicting perceived Web site credibility: Message expertise vs. Issue involvement ................................................................................................................... 48 4 12 Means for Web site credibility of the groups with di fferent levels of issue involvement and message accuracy ...................................................................................... 49

PAGE 8

8 LIST OF FIGURES Figure page 4 1 Interaction effect (Source expertise X Message accuracy) on Web site credibility ........... 45 4 2 Interaction effect (Issue involvement X Source expertise) on Web site credibility .......... 47 4 3 Interact ion effect (Issue involvement X Message accuracy) on Web site credibility ........ 48

PAGE 9

9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts in Mass Communication FACTORS INFLUENCING THE PERCEIVE D CREDIBILITY OF DIET -NUTRITION INFORMATION WEB SITES By Eun Hwa Jung May 2010 Chair: Kim B. Walsh -Childers Major: Mass C o mmunication M any people are concerned with di et and nutrition because they are strongly c orrelated with overall health T o ensure that online information can help educate consumers about diet and nutrition, the perception of online information credibility has been emphasized In this context, the pur pose of the present study is to investigate the factors that impact the perceived credibility of online diet and nutrition information By employing the ela boration likelihood model (ELM) to explain how users process online diet and nutrition messages, thi s study suggests factors that may influence Web site credibility. In view of that theoretical background, this study examines how source expertise and message accuracy influence the perceived credibility of online health information. In this sense, the lit erature review discusses why people go online to obtain diet and nutrition information and assesses the importance of online information credibility. An online experiment was conducted with 374 undergraduate students. The fi r st research question asked whe ther there is a relationship among source expertise, message accuracy, and the perceived credibility of Web sites providing diet and nutrition information The result of a Pearson s correlation analysis showed that perceived Web site credibility was positi vely and significant ly correlat ed w ith source expertise and message accuracy

PAGE 10

10 Hypothesis 1 and 2 predicted interaction effect s between source expertise and message accuracy. T he two -way ANOVA results indicated no significant interaction effect between sou rce expertise and message ac curacy on perceived Web site credibility and t herefore, H ypothes e s 1 and 2 were not supported However, there were statistically significant main effect s of source expertise and message accuracy on perceived Web site credibilit y H3 and H4 assumed that issue involvement will influence the impact of source expertise and message accuracy on perceived Website credibility. A n ANCOVA was conducted using involvement as the covariate and the result revealed that issue involvement ha d a significant main effect on perceived Web site credibility Based on these results, a two -way ANOVA was conducted to examine the relation ship between i nvolvement and source expertise and involvement and message accuracy, separately. As a result, there was no significant interaction effect of issue involvement and source expertise but there was a significant interaction effect between issue involvement and message accuracy T his analysis show ed that message accuracy had a greater impact on perceived Web site credibility for highly involved participants (M= 4.17) than for low involvement participants Therefore, hypothesis 3 was rejected but hypothesis 4 was supported. From these findings, this study will contribute to our understanding of the factors tha t affect the ways people process online diet and nutrition information. The results of this study could also help practitioners, such as health professionals and educators, understand the importance of sources of information and related user behavior (Kenk el, 1990; Nayga, 2000; Pauly & Satterthwaite, 1981).

PAGE 11

11 CHAPTER 1 INTRODUCTION With the advent of new technology, people are now paying more attention to science based health information. In particular, many people are concerned about eating related issues, including diet and nutrition information, because many diseases are caused primarily by poor eating habits (Glanz, Basil, Maibach, & Goldberg, 1998). Most of all, obesity has become a serious health problem, especially in Western countries (Madden & Chambe rlain, 2004). People who are obese people with an abnormally high and unhealthy proportion of body fat are at increased risk of developing a number of adverse health conditions (National Cancer Institute, 2004). Obesity is related to many life threatening diseases such as diabetes mellitus, cardiovascular disease, hypertension, and certain cancers and is associated with increased risk of disability and a modestly elevated risk of all -cause mortality (Centers for Disease Control, 2007, p.1). To examine this issue, the Centers for Disease Control (CDC) investigated obesity trends among U.S. adults from 1985 to 2008 and reported that the rate of obesity has increased dramatically d uring the past 20 years. In 2008, 20% of the population was found to be obes e in all states except one, Colorado. Moreover, the rate of obesity in six states (i.e. Alabama, Mississippi, Oklahoma, South Carolina, Tennessee, and West Virginia) was equal to or greater than 30% of the whole population. Based on these results, the CDC views obesity as an epidemic (Harrison & Marske, 2005). In addition, according to Reuters, approximately $147 billion a year, nearly 10 percent of all medical costs in the United States, is spent on the treatment of obesity -related diseases. In other word s, the CDC reported that the increase in U.S. obesity rates (up from 37 percent in 1998) has led to an 89 percent increase in spending on treatments for obesityrelated diseases

PAGE 12

12 such as diabetes, heart disease and arthritis (CDC, 2006). Notably, obesity re sults in a loss of self -esteem as well as diseases such as cancer, diabetes, stroke, and so on (Madden & Chamberlain, 2004). Approximately 60% of adults in the U.S. are overweight, which increases the risk of developing diabetes and certain forms of cancer (Mokdad, Serdula, Dietz, Bowman, Marks, & Koplan, 1999; Must, Spadano, Coakley, Field, Colditz, & Dietz, 1999; Neuhauser & Kreps, 2003). Based on this, obesity is clearly a threat to peoples physiological well -being and a drain on financial resources. Th us, in order to curtail the negative effects of obesity, people need to become more aware of the lifestyle and nutritional factors that contribute to obesity and obesity related diseases, and they need to be equipped with the right tools and information to understand their health status accurately. Although the cause of obesity is associated with a wide range of complex factors, the National Cancer Institute (NCI) has demonstrated that overconsumption of highcalorie foods plays a primary role in the obesit y epidemic. Therefore, to avoid weight gain and lower the risk of developing obesityrelated diseases such as diabetes and cancers, people need to learn ways of controlling their dietary habits or caloric intake (NCI, 2004). With regard to the issue of die tary habits, Glanz, Basil, Maibach, and Goldberg (1998) have suggested that one reason many Americans are overweight is a lack of understanding of what constitutes a healthy diet. Thus, people need access to appropriate diet and nutrition information, whic h could help stem the obesity epidemic ( Ayoob, Duyff, & Quagliani, 2002) In the past, people often have obtained diet and nutrition information from mass media, such as television, radio, magazines, and so on. Today, people tend to seek such information o nline because the Internet provides quick, easy, and on-demand access (Liszka, Steyer, & Hueston, 2006). However, people are sometimes confused by the information they find online

PAGE 13

13 simply because of the overwhelming volume of data the Internet provides (Har per, 1988). This characteristic causes people to wonder whether the information they find online is indeed correct (Abbott, 1997; Sankofa & JohnsonTaylor, 2007). Despite this problem, the Pew Internet & American Life Project in 2008 found that more than 6 0% of Internet users not only seek health information online but also tend to be influenced by the online health information they read. Likewise, the Internet has the potential to influence health knowledge by providing audiences with useful information, w hich may play an essential role in determining users treatment of their illness or condition (Evers, Prochaska, Prochaska, Driskell, Cummins, & Velicer, 2009). Although studies have shown that people tend to prefer to seek counsel from their own doctors r egarding health problems, the Internet, nonetheless, now functions as the first channel of information consulted (Eysenbach, 2008, p.125). For example, in a study conducted by Nicola, Jonathan, Klein, Noyce, Sesselberg, and Cantrill (2005), the findings showed that consumers are used to employing the Internet to find healthrelated information, ranging from information about relatively minor health problems such as plantar warts to serious diseases such as cancer. In addition, Health Information National Trends Survey (HINTS) have investigated Internet use in seeking health information and found that the number of people who use the Internet to seek health information has been increasing. Specifically, people who l ooked for health or medical information for yourself via Internet in 2005 (58.39%) outnumbered those in 2003 (50.55%). This suggests that the convenience of accessing online information supersedes visiting a doctor in terms of satisfying peoples desire to obtain as much information as possible to assist with understanding their health problems. On this note, many health professionals also have suggested that the Internet helps users gather health information

PAGE 14

14 by providing consumers with useful knowledge about healthrelated issues (Neuhauser & Kreps, 2003). The ability to search for information online has led to a rapid increase in both information seekers and providers. As a result, the Internet could play an essential role in helping to educate people about diet and nutrition. Moreover, understan ding how users perceive this information is crucial (Eastin, 2001; Pew Internet & American Life Project, 2000). In this context, studies about online health information have become particularly important because research indicates that individuals indeed u se what they learn online to make health related decisions for themselves and their loved ones (Kreps & Kunimoto, 1994). However, to ensure that online information can help educate consumers about diet and nutrition, the perception of online information cr edibility has been emphasized (Eastin, 2001; Eysenbach, 2008; Freeman & Spyridakis, 2004). Previous studies have suggested that credibility is a significant influence on users perceptions of healthrelated messages and their behavioral responses to those messages (Dutta -Bergman, 2004; Eastin, 2001; Harris, Sillence, & Briggs, 2009). Although many studies have examined the factors that contribute to the perceived credibility of online diet and nutrition information, the relationship of credibility with mess age features and source cues has not been investigated comprehensively (Hong, 2006). To fill this gap in online health communication research, this study will investigate the impact of two key factors that may influence how people determine the credibility of online health information. In order to explain the factors that impact the perceived credibility of online diet and nutrition information, the elaboration likelihood model (ELM) is employed. The ELM focuses on message processing and holds that people may process information using either a central or peripheral route (Flanagin & Metzger, 2000). In view of that theoretical background, this study

PAGE 15

15 examines how source expertise and message accuracy influence the perceived credibility of online health inform ation. To put this issue into perspective, we must first learn why understanding diet and nutrition is so important to the improvement of public health. Thus, the literature review discusses why people go online to obtain diet and nutrition information an d assesses the importance of online information credibility. Further, by employing the ELM to explain how users process online diet and nutrition messages, this study suggests factors that may influence Web site credibility. Therefore, this study will cont ribute to our understanding of the factors that affect the ways people process online diet and nutrition information. The results of this study could also help practitioners, such as health professionals and educators, understand the importance of sources of information and related user behavior (Kenkel, 1990; Nayga, 2000; Pauly & Satterthwaite, 1981). In this context, this document proceeds as follows: Chapter 2 describes the importance of diet and nutrition information. In addition, this chapter explains the advantages and disadvantages of using the Internet as a health related information resource and discusses how various factors influence perceptions of online information credibility. Drawing on previous studies, it also provides the rationale for using the E laboration L ikelihood M odel (ELM) as the underpinning for the study. Lastly, it states the research questions and hypotheses that inform this study. The methods used in this research project are discussed in Chapter 3. Chapter 4 provides the results of the study, and Chapter 5 consists of a discussion of the findings, their implications for diet and nutrition communicators, and suggestions for further research, as well as discussing limitations of the study.

PAGE 16

16 CHAPTER 2 LITERATURE REVIEW The Importanc e of Diet and Nutrition Information Diet is defined as an individuals regular pattern of consumption of food and drink for nourishment purposes (Griggs, 1988; Newman, 1986; Qandt & Ritenbaugh, 1986). M any people are concerned with diet and nutrition because they are strongly correlated with overall health (Lindeman, Keskivaara, & Roschier, 2000). The American Dietetic Association (ADA) reported that approximately three in five consumers (67%) consider diet and nutrition to be very important ( Nutrition and You : Trends survey, 2008). In addition, Madden and Chamberlain (2004) note that food and its consumption are involved not only in alleviating hunger but in causing dietary diseases as well. P oor dietary habits not only cause diseases, such as obesity and d iabetes, but they contribute significantly to the cost of health care (Leichter, 1999). In particular, because obesity can cause life -threatening diseases, it is regarded as a major problem, typically associated with poor dietary habits (Charlton, Brewitt, & Bourne, 2004; Harrison & Marske, 2005; Madden & Chamberlain, 2004). Thus, people should understand that diet and nutrition are important in helping to control body weight and that maintaining a healthy diet is crucial. This concern about dietary health explains the need for people to have access to accurate information related to diet and nutrition. In order to obtain diet and nutrition information, people have consulted many sources, including individual counseling provided by health professionals, mas s media, health education materials, and others (Charlton, Brewitt, & Bourne, 2004). Of these sources, people tend to prefer face -to -face communication with health professionals, including dietitians, doctors, and nurses (Freeland Graves & Nitzke, 2002). A lthough interpersonal communication is considered a reliable source from which to obtain information related to health, there are some limitations

PAGE 17

17 regarding delivering health information to people through interpersonal contact. First, to receive informatio n this way, consumers must take the time to meet with a health professional and must pay for the medical service they receive, and many consumers are unwilling to seek health professionals advice on diet. For example, although Americans spend tens of bill ions of dollars annually on weight loss products, fewer than half of obese patients have been treated by qualified professionals (Ayoob, Duyff, & Quagliani, 2002; Colditz, 1999; Galuska, Will, Serdula, & Ford, 1999) Instead, people tend to use public information provided by the media. Therefore, media play an essential role in delivering the information necessary to help achieve the common goal of public health (Neuhauser & K re ps, 2003). Many studies have revealed that media conduits play a key role in del ivering information about diet and nutrition, especially because people have such easy access to that information. For example, the American Dietetic Association reported that the media are the primary providers of nutrition information (Sankofa & JohnsonTaylor, 2007). Ayoob, Duyff and Qualgliani (2002) emphasized the role of media in disseminating information about nutrition, food, and food safety issues. Moreover, they suggested that diet and nutrition messages provided by health professionals through th e media are not only more persuasive for audiences but that the information media provide is also easier to use (Ayoob, Duyff, & Qualgliani, 2002; Charlton, Brewitt, & Bourne, 2004). This holds true outside the United States as well; Holgado, Martinez Goza lez, Irala Extevez, Gibney, Kearney, and Martinez (1999) confirmed that mass media in the United Kingdom have been the main actors in delivering messages related to healthy eating and nutrition to the public. Online Diet -Nutrition Communication Nowadays, people who have health concerns can search for rel evant information via the media, especially the Internet. Before the Internet emerged, several studies were conducted on

PAGE 18

18 individuals use of health information obtained through traditional mass media, such as TV, radio, and magazines. In particular, the effect of messages from broadcast media, namely television, has been investigated for many years (Marcus, Owen, Forsyth, Cavil, & Fridinger, 1998). For example, Dietz and Gortmaker (1985) found that televisio n viewing is strongly related with childrens risk of obesity, and Hancox, Milne, and Poulton (2004) also demonstrated that television consumption has an influence on obesity, for several reasons T elevision viewing influences food purchases and therefore is correlated with consumption of foods advertised on television (Dussere, 1976). In particular, calorically dense foods such as chocolate, candy, and cookies advertised on television contribute to the obesity epidemic (Clancy-Hepburn, Hickey, Nevill, 1974). Also, an increase in the amount of time spent watching television is connected to a reduction in the amount of time spent in energy-expending activities because people expend little energy in watching television (Galst & White, 1976). Therefore, it seem s likely that the increased amount of time spent watching television has a significant relationship with obesity. In addition to the research focused on the effects of television watching, studies have shown that consumers understanding of diet and nutrit ion information and their food choices tend to be based on how food is described on TV (Harrison & Marske, 2005). For example, Dixon, Scully, Wakefield, White, and Crawford (2007) examined the effect of TV advertisements for foods on children s dietary kno wledge, attitudes and intentions and found that heavier TV use and more frequent viewing of TV ads were related to more posit ive attitudes toward foods that were presented on TV. Consistent with this argument Morley Wakefield, Dunlop, and Hill (2009) conducted an experiment using commercial television programs delivering obesity and can cer information. T heir study showed that the mass media campaign increased awareness of the link between weight gain and cancer.

PAGE 19

19 Although traditional mass media conduits have been consumers primary sources for information for a long time, the Internet now provides an important additional source of information (Marcus, Owen, Forsyth, Cavil, & Fridinger, 1998). In particular, online media features (i.e., mass customization, interactivity, and convenience) have boosted online health communication in the digital era (Neuhauser & Kreps, 2003). In other words, people actively seek information related to diet and nutrition, and their desire for that information leads them to sear ch online. Specifically, the Harris Interactive Poll reported that 67% of adults, representing some 154 million people in the United States, have looked for health information on the Internet since the beginning of 2009 (Harris poll, 2009). Also, the numbe r of Web sites that provide health information has increased and now includes individually produced web pages such as blogs and personal homepages. Further, diet related institutions and organizations, including the ADA, the CDC, and the FDA, have Web site s as well (Hong, 2005; Neuhauser & Kreps, 2003). For these reasons, the role of the Internet in delivering diet and nutrition information has become an important topic of study. Some studies have focused on the factors that influence users online informa tion seeking behavior relative to diet and nutrition. For example, Mitchell and Boustani (1993) suggested that females pay more attention to nutrition -related risks than do males. Indeed, Nayga (1997) found that, compared to females, males tend not to perc eive nutrition as being important when they buy food. In addition, Szwajcer, Hiddink, Koelen, and Woerkum (2005) investigated how pregnant women seek nutrition information. The data used in the study were collected through in -depth interviews, and the rese archers found that the Internet is an important source among first time pregnant women because of its anonymous and up-to -date information sources. These studies imply that the effectiveness of diet and nutrition communication via the media is

PAGE 20

20 determined b y how well the media communicate with people and how aptly people choose scientific information (Ayoob, Duyff, & Quagliani, 2002; Goldberg, 2000). Although previous studies have suggested factors that have an impact on users perceptions and behavior in t erms of diet and nutrition communication, they usually have focused on demographic variables of the audience, such as gender and health status. Thus, further studies are needed for a comprehensive understanding of the role of online diet and nutrition info rmation (Charlton, Brewitt1, & Bourne, 2004). Regarding this online information, Abbott (1997) evaluated public understanding of diet and nutrition information and underscored the importance of online users recognition of misinformation among the flood of information available online. This is because m isinformation can harm receivers. Especially when the information relates to health problems, such misinformation can even pose a serious threat to consumers health (Ayoob, Duyff, & Quagliani, 2002). In this context, media credibility is considered an important determinant in health communication. Although TV has been the most popular and credible form of media until now, the Internet has increasingly become a crucial source of information ( Nutrition and You: trend survey, 2008). Thus, organizations must deliver health information conscientiously through the Internet because it has the potential to influence the health of audiences (Evers, Prochaska, Prochaska, Driskell, Cummins, & Velicer, 2003). The Risk of Diet and Nutrition Misinformation Holgado et al. (1999) suggested that when people seek diet and nutrition information, they should determine the source and assess its reliability. From the providers perspective, messages should be created carefully to ad dress consumers needs for effective diet and nutrition communication (Holgado, Martinez -Gonzalez, Irala -Extevez, Gibney, Kearney, & Martinez,

PAGE 21

21 1999). Most of all, messages provided by the media must be considered carefully because the content they provide may have a significant impact on changing the behavior of the public. Despite the concern for delivering diet and health information, the American Dietetic Association (ADA) reported that one in five consumers said they are sometimes confused by the dieta ry advice they received ( Ayoob, Duyff, & Quagliani, 2002; Nutrition and You: Trends Survey, 2000). In particular, people have been suspicious about online health information because of the largely unregulated online environment (Abbott, 1997). Further, the copious amounts of online health information foster confusion among people because of the difficulty involved in determining the veracity of the information they find ( Ayoob, Duyff, & Quagliani, 2002) Misinformation can be provided to people quickly and easily because the online environment enables people to share their opinions publicly. For example, inaccurate advice provided by unqualified individuals can be exchanged in chat rooms, especially since most chat rooms allow users to remain anonymous (Flan agin & Metzger, 2003; Neuhauser & Kreps, 2003). Also, the exchange of e -mail frequently delivers to the public false information about health issues or products (Baker, Wagner, Singer, & Bundorf, 2003) This characteristic of the Internet causes users to b ecome confused and to change their perceptions and behavior, which can certainly be harmful but could also be good if it discourages people from acting on misinformation ( Ayoob, Duyff, & Quagliani, 2002; Benigeri & Pluye, 2003; Neuhauser & Kreps, 2003). A ssessment of Online Information Credibility: The Effects of Source Expertise and Message Accuracy Although advances in technology help people acquire tremendous amounts of information easily, this flood of new information may allow misinformation to inform peoples perspectives ( Harris, 1995; Lindeman, Keskivaara, & Roschier, 2000) Misinformation can be

PAGE 22

22 harmful to receivers, especially in the case of health related issues. In fact, it can even threaten peoples health ( Nutrition and You: trend survey, 2002).Therefore, the ongoing debates over media credibility have attracted substantial attention (Gaziano & McGrath, 1986; Sundar & Nass, 2001). Until now, most of these debates have focused on the medium of television. For decades, television has been the mos t popular and, for many people, the most credible media channel ( Nutrition and You: trend survey, 2008). However, with an increasing number of people turning to online sources for health information, the Internet has become a powerful influence on health m atters (Evers, Prochaska, Prochaska, Driskell, Cummins, & Velicer, 2003). Although the Internet makes useful information readily available to users, it nonetheless has inherent potential hazards that people may encounter when seeking health information onl ine. Sillence, Briggs, Fishwick, and Hattis (2007) suggested that these hazards stem from the unregulated nature of the Internet. Thus, many studies into online information have focused on Web site credibility (e.g., Fogg, Marshall, Laraki, Osipovich, Varn ma, & Fang, 2001; Walther, Wang, & Loh, 2004). Media credibility refers to the notion that which channel is used to deliver messages to audiences has an impact on audiences perception of the credibility of those messages (Gunther, 1992). In the early re search into media credibility, Hovland and Weiss (1951) found that the trustworthiness of a source positively affected both the acceptance of messages and the changes the messages made in the mind of the receiver. Focusing on media channels, Ganziano and M cGrath proposed a credibility scale for newspaper and television content (as cited in Kiousis, 2006). In addition, Hawkins (1999) suggested some criteria for determining online media credibility. The criteria include the currency of the sites information, the purpose or audience for the site, apparent bias or objectivity of the information, and questions about the author or publisher of the information, including whether the authors name appears on the site, whether

PAGE 23

23 the author is well known and well quali fied, and whether the publisher represents a reputable organizatio n Although different concepts are available to explain media credibility, Walther (2004) mentioned that credibility is considered multidimensional, and that its concepts overlap considerabl y. Centering on the concept of media credibility, several studies have focused on the factors that affect the credibility of online information. Of these factors, two major criteria seem to be the most important to determine the perception of credibility: source and content effects (Austin & Dong, 1995; Kiousis, 2006). Regarding source effects on media credibility, a Pew Internet Survey reported that most people (86%) who use online health information are worried about unreliable source s online. Therefor e, online health information seekers tend to evaluate the sources of online information more so than other Internet users. Eysenbach (2008) stated that users are influenced by surface credibility such as Web site design (p.126) that made sites appear ed p rofessional For example, markers such as the picture of the site owner (p.126) have an impact on users perceptions of Web site credibility (Eysenbach, 2008). In this context, Rieh and Belkin (1998) found that source credibility operates on two levels i n users evaluation of online information: institutional and individual. Institutional credibility is related to the domain of the Web sites, while individual credibility is related to the names of the authors or creators of the Web pages. Specifically, Ri eh and Belkin (1998) found that at the institutional level, respondents generally perceived .edu and .gov sites as providing better qualified information than .com sites. Also, Treise et. al (2003) measured Web site expertise through domain credibility an d found that science Web sites with a .gov domain are perceived to be more credible than Web sites with a .com domain. At the individual level, author affiliations, such as the persons occupation or the institution for which he or she worked, have a signi ficant impact on respondents perceptions of information

PAGE 24

24 credibility. Similarly, Flanagin and Metzger (2003) identified the credibility of the Web site sponsor as source credibility, and Web site sponsor credibility was measured using the five elements of sponsor believability, integrity, positive reputation, success, and trustworthiness. Consistent with these conceptualizations of source credibility, studies related to online health information have sought to examine the relationship between source experti se and Web site credibility. Eastin (2001) proposed that knowledge of content and source expertise have a strong influence on users perception of the credibility of health Web sites. Treise, et al. (2003) and Walther (2004) agreed that source expertise is a primary factor in perception of Web site credibility. Another important factor that influences the assessment of online information is message credibility, also sometimes called content effects (Kunst, Groot, Latthe, Latthe, & Khan, 2002). Message cred ibility refers to the degree to which the message alone is believable, regardless of the source (Flanagin & Metzger, 2000; Meyer, 1988). In terms of message credibility, Meyer (1988) suggested five dimensions that influence message credibility in news channels: fairness, bias, depth, accuracy, and trustworthiness. Flanagin and Metzger (2000) also used five items to measure message credibility: believability, accuracy, trustworthiness, bias, and completeness. In a more recent study, Bucy (2003) proposed an i ndex of message credibility that included the elements of believability, fairness, accuracy, informativeness, and depth. Based on these arguments, accuracy is a common factor in message credibility. Kunst, Groot, Latthe, Latthe, and Khan (2002) emphasized the importance of message accuracy for Web site credibility. Specifically, they measured the relationship between message accuracy and perceived credibility for five health topics and confirmed that content accuracy was strongly related to the credibility of a Web site. In addition, Dutta Bergman (2004) observed that message accuracy is a key factor

PAGE 25

25 in assessing the credibility of online health information, in particularly f o r informative messages. Therefore, this studys attempt to measure the predictors of perceived Web site credibility takes into consideration the important role source expertise and accuracy seem to play in determining users perception of Web site credibility. However, few previous studies have attempted to investigate whether the relat ionship between message accuracy and source expertise influences perceived credibility. There appear to have been no previous studies exploring the relationship between message accuracy and source expertise. Harris, Sillence, and Briggs (2009) suggested th at future research needs to explore what happens when sites containing credible design cues present incorrect information but noncredible sites present correct information (35). To supplement these previous studies, the present research seeks to elucida te how source expertise and message accuracy affects users perceptions of the credibility of Web sites providing nutrition information. In doing so, this study proposes the following research question: RQ1 : How, if at all, are perceived source expertise and message accuracy related to consumer s evaluations of the credibility of Web sites providing diet and nutrition information? Elaboration Likelihood Model of Persuasion To explain how people use online health information, the elaboration likelihood mod el (ELM) is employed. The ELM stems from the persuasion approach, which explains how people process messages presented in media (Petty, Cacioppo, & Schumann, 1983). Petty and Cacioppo (1986) proposed two routes involved in this process: the central route, which reflects a careful and thoughtful assessment of arguments; and the peripheral route, in which responses are based on assessments of the cognitive, affective, or behavioral cues in the message. Specifically, of these two different information processi ng mechanisms, the central route involves a deeper processing of the message provided by media. This means that the receiver who processes via the

PAGE 26

26 central route pays more attention to the arguments that appear in the messages and tends to understand the me ssage more thoroughly. Alternatively, the peripheral route involves attention to source cues provided by the message. In other words, a peripheral route decision indicates a simple inference about the argument without complex cognitive processing. For exam ple, the peripheral route typically focuses on factors such as the attractiveness of source, source credibility, and the number of sources endorsing a position (Dutta -Bergman, 2004, p.255). Central route processing relies on the quality of the message, w hile peripheral route processing takes into consideration shortcut cues such as design features of the sites themselves (e.g., layout, graphics, fonts, colors, etc.) (Flanagin & Metzger, 2007; Miniard, Bhatla, Lord, Dickson, & Unnava, 1991). In the context of the dual processing model, previous studies related to online media have focused on surfing versus searching. Mulphy (1998) explained that surfing involves the peripheral route, whereas searching involves the central route. That is, information seekers generally use the central route, which focuses on goal -directed action, while surfing occurs in an unplanned environment that is serendipitous in nature (Carmel, Crawford, & Chen, 1992; Dutta Bergman, 2004; Marchionini, 1987; Marchionini & Shneiderman, 1988). In a recent study, Wise and Kim (2008) investigated how content acquired through searching and surfing exerts an influence on the cognitive processing of online media content. They found that searching is more effective than surfing at encoding conten t, which means that individuals who take the searching approach are more likely to retain the information they find. Therefore, different ways of acquiring information searching and surfing may have a crucial influence on determining cognitive outcome s (Wise & Kim, 2008).

PAGE 27

27 Although the ELM provides a useful perspective to explain how message and source cues work in information processing for users, few studies using the ELM relate directly to online diet and nutrition information. To supplement the prev ious studies, this study uses an experiment to learn how people use online message and source cues in their processing of information. Overall, the literature review suggests which factors source expertise and message accuracy consumers should take int o account in evaluating the credibility of online health information. The ELM offers an effective perspective for explaining how message and source cues work in information processing. In this sense, most previous studies on source cues have been considere d representative of the peripheral route, while message cues have been seen as indicative of the central route (Dutta Bergman, 2004; Petty & Cacioppo, 1986). Here, the reasoning is that evaluating message cues, like message accuracy, require greater cognit ive effort than recognizing source cues, such as Web site sponsorship. Based on these arguments, this study examines the following hypothesis. H1 : Source expertise will have greater impact on Web site credibility when message accuracy is low rather than hi gh. H2 : Message accuracy will have greater impact on Web site credibility when source expertise is low rather than high. That is, it seems logical that when people are exposed a Web site providing a low level of message accuracy, they will perceive the inf ormation attributed to a highly expert source as more credible than information from a source low in expertise. The Role of Involvement Involvement in communication messages has been identified using various conceptual and operational definitions ( Slater, 1997; Stephenson & Palmgreen, 2001). Regarding the issue,

PAGE 28

28 Johnson and Eagly (1989) proposed two dimensions: value relevant and outcome relevant involvement. Value relevant involvement is related to enduring values, such as freedom, while outcome relevant involvement is related to goals that are desired currently, such as winning a game (Johnson & Eagly, 1989; Petty & Cacioppo, 1990). Petty and Cacioppo (1990) suggested that involvement refers to issue involvement, defined as the extent to which the attitu dinal issue under consideration is of personal importance (Petty & Cacioppo, 1979, p. 1915). For this study, involvement is conceptualized as issue involvement, as suggested by Petty and Cacioppo. Involvement has been treated as an important factor in pe rsuasion. Prior studies that have employed the ELM have shown that issue involvement is a significant moderator that affects users perceptions and attitude change. In particular, studies in social psychology have observed that there are different influenc es on persuasion under high and low involvement conditions. Online health communication research has demonstrated that issue involvement is strongly related to users perceptions and behavior regarding their health. This is explained by the interactivity users have with the Internet (Neuhauser & Kreps, 2003). This interactivity allows people to become engaged with issues that appear on Web sites (Rice & Katz, 2000). According to the ELM, it is reasonable to predict that health -oriented messages produce the strongest impact on users perceptions when users are highly involved in the message being communicated (Street & Piziak, 2001). Consistent with this approach, the ELM can be useful in predicting how involvement moderates the influence of perceived Web si te credibility. In applying the ELM to this study, central cues, such as message accuracy, are expected to have a greater effect on credibility judgments under high involvement than under low involvement conditions. On the other hand, peripheral cues, such as the expertise of the source, should have a

PAGE 29

29 greater impact on persuasion under low as opposed to high involvement conditions. Based on this argument, this study also examines the following research hypothesis: H 3 : T he relationship between source experti se and web site credibility will be stronger under the low -involvement compared the high -involvement condition H 4 : T he relationship between message accuracy and credibility will be stronger under high -involvement compared the low involvement condition Th at is, when people are highly involve d they will use the central route (i.e. message accuracy) rather than the peripheral route (i.e. source expertise). On the other hand, when people are not much involved they will use the peripheral route (i.e. source expertise) rather than the central route (i.e. message accuracy). In addition, different information processing activities through the central and peripheral routes will lead to different perceptions of Web site credibility.

PAGE 30

30 CHAPTER 3 RESEARCH METHODS Thi s study was designed to investigate the impact of source and message variations on assessments of Web site credibility. The present study has a between-participants experimental design with two independent variables: source expertise and message accuracy. In addition, involvement was measured as a co-varia te Sample Selection Participants for the present study were recruited from large classes in the College of Journalism and Communications at the University of Florida, and they were compensated for their time with extra credit. Undergraduate students of any age were allowed to participate in the analysis P articipants were randomly assigned to one of the four study conditions. The rationale for focusing on college students is that the research demonstrates that the primary users of online information are the younger and better educated segment s of the population. According to a 2008 survey conducted by the Pew Research Center, approximately one third of college students younger than age 30 go online for the ir information. In addition, according to the National College Health Risk Behavior Survey, college students pay a great deal of attention to diet and nutrition information because they tend to be particularly concerned with weight loss and body shape (Low ry, Galuska, Fulton, Wechsler, Kann, & Collins, 2000). Therefore, using college students is reasonable for th e present study. Online Experimental Procedure To collect the data, an online experiment was conducted using the Qualtrics online survey system T he online experiment used four versions of the questionnaire to create four experimental conditions: low source expertise inaccurate message; low source expertise accurate message ; high source expertise inaccurate message; and high source expertise

PAGE 31

31 accurate message. Each participant was randomly assigned to complete o ne version of the questionnaire When participants logged in to the survey site, they first saw the informed consent document. Students who agree d to participate were asked to follow a link to sample Web site which presented a Web page stimulus. After reading the Web page, participants rated the level of source expertise and the Web pages message accuracy, and then they answered questions about overall Web site credibility. In the last page of the questionnaire, they were asked about demographic information, including gender age, education, current marital status, ethnic background height, and weight. Web Page Stimuli Four different Web page designs were created. The Web pages contain ed health messages related to diet and nutrition and varied in terms of source expertise and message accuracy (see Appendix A). The content of the Web pages was determined based on recent trends regarding consumers knowledge about healthy foods. This desi gn model takes its lead from the ADAs 2008 survey, which inquired into the consumption of specific foods and found that more than half (56%) of the respondents ha d recently increased their intake of whole grain foods in response to health information. Spe cifically, most respondents (94%) said that they believe whole -grain bread is healthier than white bread (Nutrition and You, 2008, p. 8). In this context, previous nutrition studies have revealed that whole grain foods provide significant health benefit s in comparison to refined grain foods (Jacobs, Pereira, & Slavin 2000) For example, McIn tosh, Noakes, Royle, and Foster (2003) found that whole grain food consumption increase d m etabolic health among overweight middle aged men, and therefore they argued that w hole grain foods seem likely not only to improve a normal diet but also to reduce the risk of cancer s such as colon cancer.

PAGE 32

32 For this study, the information about whole grain intake was taken from a press release provided by the ADA, and the message wa s manipulated. Specifically, the message about whole grain foods was used as it appeared in the press release for the accurate message and then was rewritten to create an inaccurate message. (e.g., I ts important to include whole grains in your diet vs. Whole grains are not that important in your diet.). T o manipulate source expertise, the Web page s were identified as coming from one of two Web sites: the CDC Web site and a personal blog. In addition, the same author was listed in all four versions of th e Web page. Table 3 1 shows the information about each stimulus. Table 3 1. Web page versions Versions Conditions Websites 1 High source expertise High message accuracy Centers for Disease Control ( http://www.cdc.gov ) 2 Low message accuracy 3 Low sou rce expertise High message accuracy Susan's Nutrition Weblog ( http://vegan.typepad.com ) 4 Low message accuracy Online Experimental Instrument After viewing the Web page participants complete d the questionnaire, which asked participants to rate the Web site s credibility and also include d post -manipulation check items measuring general knowledge about whole grains perceptions of source expertise and message accuracy as well as issue involvement Internet experience and demographic variables. Independent Variables Source expertise Source expertise is concerned with the perceived credibility of the sponsor (Flanagin & Metzger, 2003). Flanagin and Metzger (2003) measured Web site sponsor credibility and used a factor analysis to identify five items that influenced perceived source expertise: the extent to which the sponsor was perceived to be credible, to have high integrity, to have a positive reputation, to be successful, and to be trustworthy (Flanagin & Metzger, 2003, p. 692). These

PAGE 33

33 same five it ems w ere used to measure perceived source expertise. Each Web message was rated on a set of 7 -point bipolar rating scales. The five questions assessing source expertise include: This site sponsor is credible. This site sponsor has high integrity. This site sponsor s reputation is positive. This site sponsor is successful This site sponsor is trustworthy. Message accuracy Message accuracy is one of the characteristics contributing to message credibility. Previous studies (e.g., Bucy, 2003; Flanagin & Metzg er, 2000; Meyer, 1998) measure d message accuracy using a single item and this study used the same item Specifically, participants rated message accuracy using a 7 -point Likert scale ranging from 1 ( Totally inaccurate ) to 7 ( Totally accurate ). Dependent V ariable Web site credibility Credibility has been identified and measured in many ways (Eastin, 2006). To assess Web site credibility, this study used 1 2 items to measure four dimensions that are commonly recognized in studies about credibility (Gaziano & McGrath, 1986; Hong, 2005; Hovland & Weiss, 1951). That is, Web site credibility includes fairness, depth, goodwill, and trust. Specifically, questions assessing fairness include: This site provides information that is neutral. This site provides informa tion that is balanced. This site is biased in the information it provides. (reverse -coded) This site is slanted in the information it provides. (reverse -coded.) This site is even handed in presenting information. Questions about depth are: This site provi des in -depth information. This site is comprehensive.

PAGE 34

34 This site offers everything you need to know on the topic. To evaluate goodwill, questions are: This site has my interests at heart. This site is concerned about its visitors. Questions about trust incl ude: This site is ethical. This site is trustworthy. All items were rated on 7 -point bipolar rating scales from Strongly disagree to Strongly agree Control Variables Issue involvement According to the definition of involvement suggested by Petty and Cacioppo three items measure involvement. These items include concern, relevance, and importance of the issue of whole -grain food intake. Thus, involvement was measured using three statements: I am concerned about including whole grain foods in my diet. Information about whole grain foods is very relevant to me. Understanding the value of whole grain foods is important to me. Each statement w as rated on a 7 point rating scales, with 1 meaning Strongly disagree and 7 meaning strongly agree The means o f involvement were used to divide the subjects into two groups: high involvement and low involvement. Internet experience and demographic variables Flanagin and Metzger (2000) have suggested that Internet experience and demographic characteristics have cri tical influences on perceived Web site credibility. They measured Internet experience using five items: Intemet/WWW use, experience, expertise, familiarity, and access (Flanagin & Metzger, 2000, p.522). Specifically, respondents were asked to assess the frequency

PAGE 35

35 of their Internet usage, their experience using the Internet, their level of expertise, their level of familiarity with the variety and the amount of information available on the Internet/WWW, and their level of Internet access in terms of the degree of ease with which they are able to access the Internet. Thus, the questions about Internet experience include: I frequently use the Internet. I have a great deal of experience using the Internet. I am expert in using the Internet. I am very famili ar with the Internet. It is easy for me to access the Internet. Each statement w as rated on a 7 point rating scale, with 1 meaning Strongly disagree and 7 meaning strongly agree. Also, to investigate whether demographic characteristics had an impact o n perceived Web site credibility, respondents were asked about their gender, age, current level of education, ma rital status, and racial/ethnic background. Table 3 2 shows the summary of independent variables, the dependent variable, and control variables used in the present study. Table 3 2. Summary of Variables Variable Operational definition Items Sources Source expertise Perceived Web site sponsor credibility (whether the Web site sponsor is a qualified expert or not) This site sponsor is credible. F lanagin & Metzger (2003) This site sponsor has high integrity. This site sponsors reputation is positive. This site sponsor is successful. This site sponsor is trustworthy. Message Accuracy Whether or not the message on a Web site is accu rate This message is accurate. Bucy (2003) Flanagin & Metzger (2000) Meyer (1998)

PAGE 36

36 Table 3 2. Continued Variable Operational definition Items Sources Web site credibility The extent to which the Web site is perceived to demonstrate fairness, depth, goodw ill, and trustworthiness. This site provides information that is neutral Gaziano & McGrath (1986) Hong (2005) Hovland & Weiss (1951) This site provides information that is balanced. This site is biased in the information it provides. (reverse cod ed) This site is slanted in the information it provides. (reverse coded) This site is even handed in presenting information. This site provides in depth information. This site is comprehensive. This site offers everything you need to kn ow on the topic. This site has my interests at heart. This site is concerned about its visitors. This site is ethical. This site is trustworthy. Issue involvement The degree to which the subject of whole grain foods is of concern, relev ance, and importance to participants. I am concerned about including whole grain foods in my diet Petty & Cacioppo (1979, 1986, 1990) Information about whole grain foods is very relevant to me. Understanding the value of whole grain foods is import ant to me. Internet experience The degree to which participants are comfortable with using the Internet I frequently use the Internet. Flanagin & Metzger (2000 ) I have a great deal of experience using the Internet. I am expert in using the Intern et. I am very familiar with the Internet. It is easy for me to access the Internet. Pilot Test Measures: Pre -M anipulation Check Before the main study, a pilot test was performed of the Web site stimuli constructed by the researcher. The purpose of the pilot test was to evaluate the manipulations of message accuracy and source expertise in four versions of a Web page providing the same basic

PAGE 37

37 information. Thirty -three participants for the pilot test w ere recruited from two undergraduate classes at the College of Journalism and Communication at the University of Florida ; they were different from participants in the main study. Pilot test p articipants w ere presented with each Web page and asked to rate the source expertise and message accuracy of the pages. The questionnaire pages was developed using Qualtrics software Of the 33 questionnaires collected, 5 provided incomplete responses and were d ropped from subsequent analysis and 2 8 responses were analyzed Source expertise items addressed the exten t to which participants recognized the expertise of the Web sites source and included of 5 items (Chronbach s alpha=0.97) and determined that respondents do, in fact, perceive pages coming from the Centers for Disease Control as having higher source exper tise. In addition, message accuracy was rated using a 7 point Likert scale ranging from 1 ( Totally inaccurate ) to 7 (Totally accurate ). D ata collected in the pilot test w ere examined using T test s to check the statistical significance of ratings of source expertise and message accuracy The analysis confirmed that there were significant differences in perceived sponsor expertise between the CDC Web site and the personal blog. In terms of Web site sponsor expertise, the level of source expertise differed si gnificantly between the two Web sites (t = 4.69, p < 0.001) and the respondents perceived the CDC Web site (M = 5.28, SD = 1.13) is more expertise than personal blog (M = 3.48, SD = 0.89) as shown in Table 3 3 Table 3 3 T test for Web site manipulation check: Web site sponsor expertise N Mean Std. Dev. Std. Error Mean t test for Equality of Means t df CDC Web site 16 5.28 1.13 0.28 4.69** 25.92 Personal blog (Susan s nutrition weblog) 12 3.48 0.89 0.26 **p < .001

PAGE 38

38 In addition, the level of me ssage accuracy also differed significantly between the two versions of the whole grain message (t = 3.77, p < 0.001). Table 3 4 shows that the original (accurate) message (M = 4.94, SD = 1.18) was perceived as more accurate than the re -written (inaccurate) message (M = 3.17, SD = 1.27). Table 3 4 T test for Web site manipulation check: Message accuracy N Mean Std. Dev. Std. Error Mean t test for Equality of Means t df Accurate message 16 4.94 1.18 0.3 3.77** 22.88 Inaccurate message 12 3.17 1.27 0.37 **p < .001 The pilot test findings confirmed that the four versions of the Web message could be used in the main study. Data Analysis All collected data were analyzed using SPSS the Statistical Package for the Social Sciences program. To measure the sample profile and reliability of each variable, descriptive statistics and Cronbach s alpha reliability coefficients were used. For the first research question, Pearson s correlation analysis was conducted to determine the relation ship between the ind ependent variables (source expertise and message accuracy) and the dependent variable (Web site credibility) Two -way a nalysis of variance (ANOVA) w as used to determine whether there were statistically significant interaction effects between source experti se and message accuracy on Web site credibility to test H ypothes e s 1 and 2. Lastly, the relationship between the independent variables and the dependent variable w as tested in an analysis of covariance (ANCOVA) with issue involvement as a covariate to test H ypothes e s 3 and 4.

PAGE 39

39 CHAPTER 4 RESULTS This chapter discusses the characteristics of study respondents their ratings of the source expertise and message accuracy of the web pages. In addition, the chapter includes discussion of the result s of and the hypo thes is test s conducted using Pearson s correlation analysis, two -way ANOVA, and ANCOVA. The dependent variable for all analyses was Web site credibility. Profile of the Sample Four hundred students completed questionnaires. However, 26 incomplete responses were dropped from subsequent analyses, for a total valid sample size of 374. As shown in Table 4 1, f emale participants ( 68. 2 % n=255) outnumbered male participants ( 31.8% n=119). T he ages of participants ranged from 18 to 2 9 with a mean age of 19.86 ye ars old and a median age of 20 years old (29.9%, n=112). Of all participants 62 were freshmen (16.6%), 123 were sophomore s (32.9%), 133 were junior s (35.6%), and 56 were senior s (15%). The majority of participants were single (96%, n=359) In terms of eth nic background, more than half of the participants (59.6%, n=223) w ere nonHispanic whites and 45 were Hispanic (12%), 24 were African -American (6.4%), 18 were Asian (4.8%) and 64 were other ethnic groups (17.1%). Table 4 1. Demographic description Varia ble Frequency Percent Gender Male 119 31.8 Female 255 68.2 Age 18 21 337 90.1 22 25 35 9.3 26 30 2 0.6 Education Level Freshmen 62 16.6 Sophomore 123 32.9 Junior 133 35.6 Senior 56 15.0

PAGE 40

40 Table 4 1. Continued Variable Fr equency Percent Marital status Single 359 96 Married 4 1.1 Other 11 2.9 Ethnic background Non Hispanic 223 59.6 Hispanic 45 12.0 African American 24 6.4 Asian 18 4.8 Other 64 17.1 N=374 Media Usage and Diet -Nutrition Information To explain how subjects use the media, especially the Internet, the online experimental instrument included questions about participants media use for diet and nutrition information. First, regarding the questions about whether subjects ha d ever visit ed the CDC Web site and /or the personal blog used as Web site stimuli in the present study, 31 reported that they had visited the CDC Web site (8.29%) but 171 had not visited the CDC Web site (45.72%) None of the subjects reported having visit ed the perso nal blog (46.00%, n=172). Participants also were asked about what kind of media they usually use in order to get diet and nutrition information. T he majority of subjects reported us ing the Internet (39.3%, n=147) to get diet and nutrition information, fol lowed by other people such as parents and friend s (20.3%, n=76), health professionals (19%, n=71), newspaper/magazines (14.2%, n=53), television/radio (5.6%, n=21), and other sources (1.6%, n=6). This finding indicates that the Internet is the medium most frequent ly used to get diet and nutrition information. Third, in terms of frequency of Internet usage for diet and nutrition information, most subjects had used the Internet once or twice for diet and nutrition information (44.1%, n=165), followed by once a

PAGE 41

41 month (23%, n=86), once a week (7.5%, 28), and daily (2.1%, n=8), while 87 subjects had never sought diet and nutrition information online (23.3%). Table 42 summarizes these results. Table 4 2. Use of the Internet and other source for diet and nutrition information Variables and Items Frequency Percent Visit CDC Web site Yes 31 8.29 No 171 45.72 Visit Personal blog (Susan s Weblog) Yes 0 0 No 172 46.00 Media type Television/Radio 21 5.6 Newspaper/Magazines 53 14.2 Internet 147 39.3 Health p rofessionals 71 19.0 Other people 76 20.3 Others 6 1.6 Frequency of Internet usage Never 87 23.3 Once or twice 165 44.1 Once a month 86 23.0 Once a week 28 7.5 Daily 8 2.1 N=374 Post -M anipulation Check As predicted by the pilot test manipula tion check in the results of the main study showed that there were significant differences in respondents perceptions of source expertise (t = 6.99, p < 0.001) and message accuracy (t = 15.41, p < 0.001) depending on which version of the Web page they s aw. As shown in Table 4 3, subjects perceived that the CDC Web site as higher in source expertise (M = 4.94, SD = 0.98) than the personal blog (M = 4.27, SD = 0.85) and perceived the original ( accurate ) message as more accurate (M = 5.38, SD = 0.77) than t he re -written ( inaccurate ) message (M = 3.73, SD = 1.25). Additionally, there were significant differences between subjects who were highly involved (M = 6.09, SD = 0.58) and those who were lowly involved (M = 3.96, SD = 0.98) in whole grain issue (t = 24.61, p < 0.001).

PAGE 42

42 Table 4 3. T test for Web site manipulation check: Web site sponsor expertise, message accuracy, and issue involvement N Mean Std. Dev. Std. Error Mean t test for Equality of Means t df Source expertise CDC Web site 202 4 .94 0.98 0.07 6.99** 372 Personal blog (Susan s nutrition weblog) 172 4.27 0.85 0.06 Message accuracy Accurate message 190 5.38 0.77 0.06 15.41** 372 Inaccurate message 184 3.73 1.25 0.09 Issue involvement High involvement 165 6.09 0.58 0.45 24.61** 372 Low involvement 209 3.96 0.98 0.68 **p < .001 Re liability Test Generally, Cronbach s alpha is used to ensure internal reliability which indicates internal consistency between items that measure the same variable ( Spathis & Ananiadis, 2005). Nunnally (1978) suggested that research instruments could be considered acceptable if they produced Cronbach s alpha reliability coefficients of 0.70 or higher. As shown in Table 4 4, all Cronbach s alpha values were above 0.70 and therefo re all scales were used in the present study. Table 4 4 Means and reliability check for each variable Variables and Items Mean Cronbachs alpha Source expertise This site sponsor is credible. 4.65 0.95 This site sponsor has high integrity. 4.57 Th is site sponsors reputation is positive. 4.70 This site sponsor is successful. 4.64 This site sponsor is trustworthy. 4.62 Website Credibility This site provides information that is neutral 3.59 0.89 This site provides information that is balan ced. 3.73 This site is biased in the information it provides. (reverse coded) 3.66 This site is slanted in the information it provides. (reverse coded) 3.41 This site is even handed in presenting information. 3.99 This site provides in depth in formation. 3.74 This site is comprehensive. 4.71

PAGE 43

43 Table 4 4. Continued Variables and Items Mean Cronbachs alpha This site offers everything you need to know on the topic. 3.08 This site has my interests at heart. 4.75 This site is concerned about its visitors. 4.86 This site is ethical. 4.89 This site is trustworthy. 4.50 Issue involvement I am concerned about including whole grain foods in my diet 4.82 0.84 Information about whole grain foods is very relevant to me. 4.84 Under standing the value of whole grain foods is important to me. 5.06 Internet experience I frequently use the Internet. 6.78 0.88 I have a great deal of experience using the Internet. 6.65 I am expert in using the Internet. 5.81 I am very familiar w ith the Internet. 6.53 It is easy for me to access the Internet. 6.75 N=374 Research Question 1 RQ1 : How, if at all, are perceived source expertise and message accuracy related to consumer s evaluations of the credibility of Web sites providing diet and nutrition information? The fi r st research question asked whether there is a relationship among source expertise, message accuracy, and the credibility of Web sites providing diet and nutrition information. To answer this question, a Pearson s correlat ion analysis was conducted. As shown in Table 4 5, Web site credibility was positively and significant ly correlat ed with source expertise (r = 0.53, p < 0.01) and message accuracy (r = 0.73, p < 0.01). Table 4 5 Correlations among source expertise, messag e accuracy, and Web site credibility Web site credibility S ource expertise Message accuracy Web site credibility 1.00 Source expertise 0.53** 1.00 Message accuracy 0.73** 0.54** 1.00 **p <0.01 (2-tailed)

PAGE 44

44 Hypotheses Testing Hypotheses 1 and 2 H1 : Sou rce expertise will have greater impact on Web site credibility when message accuracy is low rather than high. H2 : Message accuracy will have greater impact on Web site credibility when source expertise is low rather than high. Hypothesis 1 and 2 predicted interaction effect s between source expertise and message accuracy. In other words, subjects exposed to a low source expertise Web site (i.e., personal blog) w ere predicted rate Web site credibility as higher when message accuracy was high rather than low. On the other hand, participants exposed to a Web site containing a low accuracy message were expected to perceive the Web site as more credible when source expertise was high rather t han low. To test these hypotheses, a two -way ANOVA was conducted to dete rmine if there was an interaction effect between source expertise and message accuracy. As shown in Table 4 6, the ANOVA results indicated no significant interaction effect between source expertise and message accuracy on Web site credibility [F (1, 370) = 0.06, p = 0.82]. However, there were statistically significant main effect s of source expertise [F (1, 370) = 3.73, p < 0.05] and message accuracy [F (1, 370) = 163.61, p < 0.001] on Web site credibility. Specifically, Table 4 7 shows that subjects exposed to a high source expertise Web site (M = 4.16, SD = 0.91) rated the Web site as more credib le than those exposed to a low expertise Web site (M = 3.97, SD = 0.92). Similar ly, subjects exposed to a n accurate message (M = 4.58, SD = 0.70) rated the Web s ite s credibility as higher, compared to those exposed to a n inaccurate message (M = 3.55, SD = 0.84). Therefore, H ypothes e s 1 and 2 were not supported ; source expertise and message accuracy do not produce a statistically significant interaction effect (se e F igure 4 1).

PAGE 45

45 Table 4-6. Two-way ANOVA for Web site credibility Type III Sum of Squares df Mean Square F Sig. Source expertise 2.19 1 2.19 3.73 0.04 Message accuracy 96.05 1 96.05 163.61 0.00 Source expertise Message accuracy 0.03 1 0.03 0.06 0.82 Error 217.22 370 0.59 Total 6528.00 374 R2 = 0.32 Figure 4-1. Interaction e ffect (Source expertise X Message accuracy) on Web site credibility Furthermore, a comparison of the means for a ll four groups showed that perceived Website credibility was highest among those exposed to the accurate messages, regardless of the level of source expertise (See Table 4-7). Table 4-7. Web site credibility ratings for groups with different levels of source expertise and message accuracy Group Source expertise Message accuracy M SD N 1 High High 4.66 0.63 106 2 Low 3.62 0.87 96 3 Low High 4.48 0.76 84 4 Low 3.48 0.79 88 Overall High 4.58 0.70 184 Low 3.55 0.84 190 Total 4.08 0.92 374

PAGE 46

46 Hypotheses 3 and 4 H 3 : T he relationship between source expertise and web site credibility will be stronger under the low -involvement compared the high -involvement condition H 4 : T he relationship between message accuracy and credibility will be stronger under high -involvement compared the low involvement condition. H3 an d H4 assumed that issue involvement will influence the impact of source expertise and message accuracy on perceived Website credibility. As the first step to test these hypotheses, Pearson s correlation analysis was conducted to examine the relation ship be tween issue involvement and Web site credibility. Table 4 8 shows that issue involvement ( r = 0.14, p < 0.01) was strongly correlated to Web site credibility. Table 4 8 Correlations between issue involvement and Web site credibility Issue involvement We b site credibility Issue involvement 1.00 Web site credibility 0.14** 1.00 **p <0.01 (2-tailed) For the next step, an ANCOVA was conducted using involvement as the co -variate As shown in Table 4 9, issue involvement ha d a significant main effect on We b site credibility [F (1, 369) = 5.64, p < 0.05]. This result revealed that issue involvement is also an important variable relate d to Web site credibility. Table 4 9 ANCOVA for the effect of issue involvement, source expertise and message accuracy on pe rceived Web site credibility Mean Type III Sum of Squares df Mean Square F Sig. Source expertise 2.33 1 2.19 3.73 0.04 H igh 4.16 Low 3.97 Message accuracy 93.05 1 96.05 163.61 0.00 High 4.57 Low 3.55 Co Variate Issue involve ment 3.27 1 3.27 5.64 0.02 Error 213.95 369 0.59

PAGE 47

47 Table 4 9. Continued Mean Type III Sum of Squares df Mean Square F Sig. Total 6528.00 374 R2 = 0.33 Based on these results, a two -way ANOVA was conducted to examine the relation ship between in volvement and source expertise. Table 4 10 shows the results of that analysis which revealed a significant main effect for source expertise [F (1, 370) = 3.99, p < 0.05] but not for involvement In addition, there was no significant interaction effect of i ssue involvement and source expertise, as shown in Figure 4 2. Therefore, H ypothesis 3 was rejected. Table 4 10. Two -way ANOVA for the effects of Issue involvement and source expertise on perceived Web site credibility Mean Type III Sum of Squares df Me an Square F Sig. Issue involvement 3.00 1 3.00 3.57 0.06 High 4.17 Low 3.99 Source expertise 3.35 1 3.35 3.99 0.04 High 4.16* Low 3.97* Issue involvement Source expertise 0.02 1 0.02 0.02 0.88 Error 311.19 370 Total 6528.00 374 R2 = 0.02, p < 0.05 Fig ure 4 2 Interaction effect (Issue involvement X Source expertise) on Web site credibility

PAGE 48

48 To test H ypothesis 4, a two -way ANOVA was performed to test for an interaction effect between issue involvement and messa ge accuracy. As shown in Table 4 11, there was no significant main effect for issue involvement but message accuracy [F (1, 370) = 170.47, p < 0.05] did produce a significant main effect on perceived Web site credibility ; as would be predicted, the accurat e message produced higher Website credibility ratings (M= 4.58) than the inaccurate message (M= 3.55). In addition, there was a significant interaction effect between issue involvement and message accuracy [ F (1, 370) = 6.67, p< 0.05], as shown in Figure 4 3. Table 4 11. Two -way ANOVA predicting perceived Web site credibility: Message expertise vs. Issue involvement Mean Type III Sum of Squares df Mean Square F Sig. Issue involvement 0.42 1 0.42 0.72 0.40 High 4.17 Low 3.99 Message accura cy 99.08 1 99.08 170.47 0.00 High 4.58* Low 3.55* Issue involvement Message accuracy 3.88 1 3.88 6.67 0.01 Error 215.04 370 0.58 Total 6528.00 374 R2 = 0.32, p < 0.05 Fig ure 4 3 Interaction effect (Issue involvement X Messag e accuracy) on Web site credibility

PAGE 49

49 Table 4 12 shows the impact of the message accuracy -byinvolvement interaction on Web site credibility ratings. This analysis show ed that message accuracy had a greater impact on perceived Web site credibility for highly involved participants (M= 4.17) than for low involvement participants (M= 4.00), as predicted in H ypothesis 4. This finding is consistent with the predictions of the ELM because message accuracy is related to central route processing, as discussed earlier Therefore, H ypothesis 4 was supported. Table 4 12. Means for Web site credibility of the groups with different levels of issue involvement and message accuracy High level of message accuracy Low level of Message accuracy High involvement 4.72 3.61 Lo w involvement 4.44 3.47 Additional Findings: T he Role of Inter net Experience and Demographic V ariables as Covariates Previous studies have mentioned that Internet experience and demographic variables play important roles in online users perceptions of w ebsite credibility (Flanagin & Metzger, 2000). In the present study, the relationship among independent variables (i.e., source expertise and message accuracy) and the dependent variable (i.e., Web site credibility) was examined in an ANCOVA with Internet experience and demographic variables (i.e., gender, age, education level, ma rital status, and ethnic background) as covariates. The result showed that none of the covariates w as significant after the main effects of source expertise and message accuracy ha d been accounted for. Therefore, Internet experience and demographic variables are not likely to have a significant relation ship with users perceptions of the credibility of a Web site providing diet and nutrition information.

PAGE 50

50 CHAPTER 5 DISCUSSION The pr imary purpose of the present study was to understand key influences on how people perceive the credibility of Web sites that provide diet and nutrition information. In particular, the study examined two major variables that are related to perceived Web sit e credibility: message accuracy and source expertise. The study was grounded in the Elaboration Likelihood Model, which predicts effects on credibility of message processing via a central versus a peripheral route. T his chapter discusses the findings of the study in three subsections: interpretation of the findings regarding the research question and hypotheses, discussion of the theoretical and practical implications, and limitations and suggestions for further research. Findings for the Research Question and Hypotheses The Correlation between Independent Variables and the Dependent Variable The first research question asked what relationships exist among source expertise, message accuracy, and Web site credibility. Pearson s correlation analysis confirmed that source expertise and message accuracy were strongly related with Web site credibility. This finding is consistent with those of previous studies that have identified two major criteria, source and content effects, as critical variables related to the perception of Web site credibility (e.g., Austin and Dong, 1995; Kiousis, 2006). Similarly, Rieh and Belkin (1998) investigated the factors that determine the credibility of online information using in -depth interviews with 14 scholars and concluded that authority of Web sites (i.e., identified institutional sites vs. individual sites) and accuracy of content were used to evaluate online information. In this context, the present study is meaningful to confirm that source expertise and message accuracy also play essential roles in evaluating the credibility of Web sites providing diet and nutrition information.

PAGE 51

51 Interaction Effect between Source Expertise and Message Accuracy on Web site Credibility Hypothesis 1 and 2 predicted that source expertise and messa ge accuracy would interact in their effects on Web site credibility In other words, the researcher predicted that source expertise w ould increase Web site credibility when message accuracy was low rather than high and that message accuracy w ould increase Web site credibility when source expertise was low rather than high. However, the result of ANOVA revealed that there was no significant interaction effect on Web site credibility. Therefore, H ypothes e s 1 and 2 were not supported. However, there were si gnificant main effects of source expertise and message accuracy on Web site credibility confirming that when source expertise and message accuracy are both high, perceived Web site credibility will be highest. Therefore, source expertise and message accu racy seem to affect separately Web site credibility separately. These findings are consistent with some prior research showing that source expertise has a significant influence on Web site credibility (e.g., Eastin, 2006; Flanagin & Metzger, 2003; Treise e t al., 2003; Walther, 2004). The results related to message accuracy also support a previous study by Kunst et al. (2002) who argued that message accuracy is an important variable in evaluating Web site credibility. Furthermore, comparing the effect size s between source expertise and message accuracy show that message accuracy has a greater effect on Web site credibility than source expertise. The Relationship among Source Expertise, Message Accuracy and Issue Involvement Hypothes e s 3 and 4 predicted that source expertise and message accuracy would have different effects on Web site credibility depending on the users level of issue involvement. Specifically, H3 predicted that message accuracy w ould have greater effects on Web site credibility when subject s are highly involve d in the issue (i.e., whole grain intake) and H4 predicted that source expertise w ould have greater impact on perceived Web site credibility

PAGE 52

52 when involvement was low. Before the test of these hypotheses, Pearson s correlation and ANC OVA analyses were conducted to examine whether the issue involvement was a covariate of Web site credibility. The results showed that issue involvement was strongly cor related with perceived Web site credibility, and there was a significant main effect on Web site credibility. Therefore, issue involvement was predicted to interact significantly with source expertise and message accuracy in influencing perceived Web site credibility and two ANOVA tests were performed. The ANOVA test for Hypothesis 3 was co nducted to investigate the interaction effect of source expertise and issue involvement on Web site credibility. The result s indicated that there was no significant interaction effect between source expertise and issue involvement T he main effect of sourc e expertise was significant but issue involvement had no significant effect on perceptions of Web site credibility after controlling for source expertise effects Therefore, H ypothesis 3 was not supported. The results imply that the combination of source expertise and issue involvement do es not have a direct effect on W e b site credibility. The ANOVA test for Hypothesis 4 was performed to examine the effects of the interaction between message accuracy and issue involvement on perceived Web site credibility. The results showed a significant interaction effect M essage accuracy had a greater impact on perceived Web site credibility for those who were highly involved with the issue, compared to those with low issue involvement Perceived credibility of the high accuracy web sites was higher for those who were highly involved in the whole grain intake but accuracy had less impact for participants who were less involved in the issue. Therefore, H ypothesis 4 was supported. This finding supports the ELM predicting central route cues will be more important for those who are highly involved (Petty & Cacioppo, 1979).

PAGE 53

53 The Possibility of Other Covariates: Internet Experience and Demographic Variables ANCOVA test s also were conducted to determine if Internet experience o r any demographic variables function ed as covariates. The result revealed that none of the demographic variables tested including Internet experience, gender, age, educational level, marital status, or ethnic background, had significant effects on perceiv ed Web site credibility. Flanagin and Metzger (2003) actually had proposed that I nternet experience and gender differences affect Web site credibility perceptions, but the present study did not support this argument. This may have occurred because the pres ent study used Web sites providing diet and nutrition information, not overall online information. In other words, one possible explanation is that Internet experience and gender make no significant difference in influencing perceived credibility for onlin e diet and nutrition information. To summarize the results, the present study suggests that 1) source expertise and message accuracy are positively related to Web site credibility 2) source expertise and message accuracy do not produce significant inter action effects, and 3) issue involvement is positively related to Web site credibility. In particular, there is a significant interaction effect between message accuracy and issue involvement, and therefore message accuracy has a greater impact on Web site credibility for those who are highly involved in the issue, compared to those who are less involved. Theoretical and Practical Implications The findings resulting from this online experiment have important implication s for online health information. From a theoretical perspective, the present study attempt ed to investigate the factors influencing the perceived credibility of online diet and nutrition information. Although interest in diet and nutrition has increased, previous studies regarding online heal th communication typically have explored either overall health issues or have focused

PAGE 54

54 on certain disease such as cancer or diabetes. Therefore, the study helps us understand how people perceive online diet and nutrition information and which factors influe nce users perceptions of the credibility of online nutrition information. Specifically, some of these findings support the results of previous online credibility studies. In particular, prior studies have suggested that both source effects and message e ffects have a positive influence on Web site credibility, and this study particularly suggests that message accuracy is a more important influence on Web site credibility than source expertise. This may indicate that most people felt qualified to judge the accuracy of the message about whole grain intake. In particular, high involvement in diet and nutrition information seems to raise peoples concerns about message accuracy. Therefore, message accuracy is likely to be a priority consideration in determining the credibility of Web sites providing diet and nutrition information. The present study also offers some essential practical implications. First, the findings confirmed that the Internet is an important source where people learn diet and nutrition infor mation. Consistent with most previous studies, the present study investigated how subjects use online diet and nutrition information and found they usually use the Web sites to fill their desire to get diet and nutrition information. This suggests that die tary professionals should work to ensure that people can easily find appropriate online sources of accurate diet and nutrition information Second, despite the importance of online information, there remain several restrictions of using online information That is, online health information -seekers often hav e trouble finding appropriate and accurate information because the online world is a n unregulated space (Eysenbach, 2008). S ome past reports have suggested that the risk of online information -seekers

PAGE 55

55 en countering diet and nutrition mis information is serious (e.g., ADA reports). Holgado and colleagues (1999) have suggested that information providers should create messages carefully in reduce the likelihood of information seekers being confused by what the y learn online. The findings of the present study revealed that message accuracy seems to strongly influence Web site credibility. Thus, online health information providers should be particularly conscious of ensuring high message accuracy when they provid e diet and nutrition information through Web sites. In addition, highly credible Web sites encourage people to perceive the information they provide as being of high quality and to make decisions based on that information (Bates, Romina, Ahmed, & Hopson, 2006). In this context, findings related to the factors influencing perceptions of online diet and nutrition information will help online health information providers learn to more effectively manage their m essages. Furthermore, well -made Web sites will be more effective in educat ing people about how to prevent certain disease such as obesity and diabetes caused by poor dietary habits U nderstanding how people perceive the credibility of Web sites providing diet and nutrition will assist both information use rs and information providers in communicat ing more effectively. Limitations and Future Research This study has significant potential contribution s, but the study also had several limitations, some of which could be addressed through further research. Firs t, there is a limitation in terms of generalizability. Although college students typically use online information the subjects recruited in this study were not representative of the entire audience who search es online for diet and nutrition information T hus, the results of this study must be replicated in other non -student populations. In addition, the present study focused specific ally on one element of diet and nutrition -whole grain intake -so the findings do not necessarily mean

PAGE 56

56 that the credibili ty of all online diet and nutrition information will be affected in the same ways by message and expertise cues. F uture stud ies should examine other diet and nutrition topics to reconfirm the effect s of source expertise and message accuracy. Second, since the study was conducted under experimental condition, the findings may reflect artificial and /or short term effect s In other words, t he subjects could be distracted by their circumstances because they were not controlled by the researcher in the present stud y Therefore, the future studies need to be repetitively measured by a long-term observation to harden the relationship between variables, Third, although this study employed some variables related to message and source cues, future research needs to consider other factors that influence how people perceive the credibility of online diet and nutrition information. For instance, k nowledge of the content likely is one of the significant factors in determining credibility for web sites related to that con tent (Eagly & Chaiken, 1993; Eastin, 2001). In other words, there is a possibility that different level s of knowledge mediate users evaluation s of persuasive messages delivered by Web sites. In addition, Houston and Allison (2002) have suggested that hea lth status and physical condition can influence online health information usage. Therefore, future research should examine a wide range of independent variables that seem to impact the perception s or behavior of online health information seekers. Finally, the present study explored the relationships among message accuracy, source expertise and perceived Web site credibility using ANOVA and ANCOVA tests. Additional research is uncover causal relation ships among these variables by using other statistical too ls such as multiple regression analysis and structural equation model.

PAGE 57

57 Despite some drawbacks of the present study, the study represents a step toward understanding how people perceive online diet and nutrition information and offers a valuable contribut i on to the development of online health communication practices

PAGE 58

58 APPENDIX A WEB PAGES STIMULI Version 1. Version 2. Whole Grains = Whole Health Lona Sandon, an assistant professor of clinical nutrition at UT Southwestern Medical Center and a spokesperson for the American Dietetic Association, says its important to keep whole grains in your diet. Research shows that whole grains are good for your heart, lower risk of diabetes and stroke, and may help prevent certain cancers, she says. They also help in managing weight. Ms. Sandon says whole grains are chock full of good for you nutrients including fiber, folate and niacin, vital B vitamins, and magnesium. The phytochemicals found in whole grains have been shown to have health promoting and disease prevention benefits, she adds. Ms. Sandon recommends that adults aim for three servings, or 48 grams, of whole grains a day. Be sure to look for the words Made with whole grain and 100 percent whole grain on packages. Whole Grains = Whole Health? Not Necessarily Lona Sandon, an assistant professor of clinical nutrition at UT Southwestern Medical Center and a spokesperson for the American Dietetic Association, says whole grains are not that important in your diet. Research shows that whole grains dont have much of a health impact on your heart, risk of diabetes and stroke, or likelihood of developing cancer, she says. They also do not help much in managing weight. Ms. Sandon says whole grain foods are not significantly different from nonwhole grain foods in term s of their nutrient levels, including fiber, folate and niacin, vital B vitamins, or magnesium. Researchers do not know yet whether the phytochemicals found in whole grains have health promoting and disease prevention benefits, she adds. Ms. Sandon said it isnt that important for adults to have whole grains every day. In choosing foods, people are making good choices so long as they see the phrase 100 percent wheat on the package. In addition, brown bread is always healthier than white bread.

PAGE 59

59 Version 3. Version 4 Whole Grains = Whole Health Lona Sandon, assistant professor of clinical nutrition at UT Southwestern Medical Center and spokesperson for the American Dietetic Association, says its important to keep whole grains in your diet. Research shows that whole grains are good for your heart, lower risk of diabetes and stroke, and may help prevent certain cancers, she says. They also help in managing weight. Ms. Sandon says whole grains are chock full of good for you nutrients including fiber, folate and niacin, vital B vitami ns, and magnesium. The phytochemicals found in whole grains have been shown to have health promoting and disease prevention benefits, she adds. Ms. Sandon recommends that adults aim for three servings, or 48 grams, of whole grains a day. Be sure to look for the words Made with whole grain and 100 percent whole grain on packages. Whole Grains = Whole Health? Not Necessarily Lona Sandon, an assistant professor of clinical nutrition at UT Southwestern Medical Center and a spokesperso n for the American Dietetic Association, says whole grains are not that important in your diet. Research shows that whole grains dont have much of a health impact on your heart, risk of diabetes and stroke, or likelihood of developing cancer, she says They also do not help much in managing weight. Ms. Sandon says whole grain foods are not significantly different from nonwhole grain foods in terms of their nutrient levels, including fiber, folate and niacin, vital B vitamins, or magnesium. Resear chers do not know yet whether the phytochemicals found in whole grains have health promoting and disease prevention benefits, she adds. Ms. Sandon said it isnt that important for adults to have whole grains every day. In choosing foods, people are makin g good choices so long as they see the phrase 100 percent wheat on the package. In addition, brown bread is always healthier than white bread.

PAGE 60

60 APPENDIX B QUESTIONNAIRE T he following questions ask your knowledge about whole grain s. Please check True or False for each statement. 1. W hol e grains are better for your health than refined grains. True False I do not know. 2. Eating whole grains does not increase fiber intake. True False I do not know 3. The benefits of consuming whole grains include a reduced risk of stroke, type 2 diabetes, and heart disease. Additionally, whole grain intake helps to control weight gain and prevent obesity. True False I do not know. 4. Nutrition provided by whole grains has no effect on health. True False I do not know. The following questions ask about your thought s about whole grain food. The following questions ask about your Internet experience. Strongly disagree Strongly agree 1 I am concerned about including whole grain foods in my diet 1 2 3 4 5 6 7 2 Information about whole grain food s is very relevant to me. 1 2 3 4 5 6 7 3 Understanding the value of whole grain foods is important to me 1 2 3 4 5 6 7 Strongly disagree Strongly agree 1 I frequently use the Internet. 1 2 3 4 5 6 7 2 I have a great deal of experience using the Internet. 1 2 3 4 5 6 7 3 I am expert in using the Internet. 1 2 3 4 5 6 7 4 I am very familiar with the Internet. 1 2 3 4 5 6 7 5 It is e asy for me to access the Internet. 1 2 3 4 5 6 7

PAGE 61

61 Have you ever visited the CDC website ( http://www.cdc.gov/ ) before today? Yes No I do not know. Have you ever visited the Susan's Nutrition Weblog (http://veg an.typepad.com/) before today? Yes No I do not know. T he following W eb page provides information about whole grain s and health. Please read the article and answer the following questions (Please do not use the back button.) After seeing the Web site h ow would you rate the Web site ? Strongly disagree Strongl y agree 1 This site provides information that is neutral 1 2 3 4 5 6 7 2 This site is even handed in presenting information. 1 2 3 4 5 6 7 3 This site is comprehensive. 1 2 3 4 5 6 7 4 This site is concerned about its visitors. 1 2 3 4 5 6 7 5 This site offers everything you need to know on the topic. 1 2 3 4 5 6 7 6 This site provides information that is balanced. 1 2 3 4 5 6 7 7 This site is slanted in the information i t provides. 1 2 3 4 5 6 7 8 This site is ethical. 1 2 3 4 5 6 7 9 This site has my interests at heart. 1 2 3 4 5 6 7 10 This site provides in depth information. 1 2 3 4 5 6 7 11 This site is biased in the information it provides. 1 2 3 4 5 6 7 1 2 This site is trustworthy. 1 2 3 4 5 6 7 Insert each Web page stimulus

PAGE 62

62 After seeing the Web site, h ow would you rate the Web site sponsor ? After reading the message on the Web site, how would you rate the accuracy of message ? The following q uestions ask about your use of diet and nutrition information. 1. What kind of media do you usually use in order to get diet and nutrition information? _______ Television/Radio _______ Newspaper s /Magazine s _______ Internet _______ Health professionals (doctors, nurses, nutritionists, etc.) _______ Other people (parents, friends, etc.) _______ Others __________ 2. How often have you use d diet and nutrition Web sites? _______ N ever _______ Once or twice _______ Once a month _______ Once a week _______ Daily 3. When you search for diet and nutrition information on the Internet, what are your considerations in determining the credibility of the Web site? (openended) ___________________________________________________ The following questions ask about your health concern Strongly disagree Strongly agree 1 This site sponsor is credible. 1 2 3 4 5 6 7 2 This site sponsor has high integrity. 1 2 3 4 5 6 7 3 This site sponsors reputation is positive. 1 2 3 4 5 6 7 4 This site sponsor is successful. 1 2 3 4 5 6 7 5 This site sponsor is trustworthy. 1 2 3 4 5 6 7 Total ly in accurate Totally accurate 1 2 3 4 5 6 7 Strongly disagree Strongly agree 1 I am concerned about following a low fat diet. 1 2 3 4 5 6 7 2 I am interested in weight loss. 1 2 3 4 5 6 7 3 I tend to check the nutrition information about what I eat. 1 2 3 4 5 6 7 4 I am concerned about following a low carbohydrate diet. 1 2 3 4 5 6 7 5. I try to follow a low calorie diet. 1 2 3 4 5 6 7

PAGE 63

63 The following questions include some basic bi ographical data about you. 1. Gender _______ Male _______ Female 2. How old are you? _________ years old 3. What is your current level of education? _______ Freshmen _______ Sophomore _______ Junior _______ Senior _______ Graduate student 4. What is your current marital status? _______ Single _______ Married _______ Other 5 What is your ethn ic background? _______ NonHispanic _______ Hispanic _______ African American _______ Asian _______ Other 6 Please write your height and weight. Height _____feet _____inches Weight _________ pounds

PAGE 64

64 Notice! This message contained in the web site was manipulated for this study. The actual information about whole grain food is as follows. The information below was taken from an American Dietetic Association ( ADA) report ______________________________________________________________________________ Truth about Whole Grains It reduce s the risk of heart disease diabetes and stroke by decrea sing cholesterol levels, blood pressure, and blood coagulation. It help s prevent certain cancers It provides good-for-you nutrients including fiber, folate and niacin, vital B vitamins, and magnesium Health professionals recommend that adults aim for thr ee servings, or 48 grams, of whole grains a day It helps in weight control. In other words, people who consume more whole grains consistently weigh less than those who consumed less whole grain products ______________________________________________________________________________ For more information, please refer to the ADA Web site, http://www.eatright.org/default.aspx Thank you for your participation!

PAGE 65

65 LIST OF REFERENCES American Dietetic Association.(2002). Nutrition and you: Trends 2000. Journal of the American Dietetic Association, 100 (6), 626. American Dietetic Associ ation.(2008). Nutrition and You: Trends 2008. ADA Food and Nutrition Conference and Expo. Chicago. 26 Oct. Abbott, R. (1997). Food and nutrition information: a study of sources, uses, and understanding. British Food Journal 99(2), 43 49. Austin, E.W., & D ong, Q. (1994). Source v. content effects on judgments of news believability. Journalism Quarterly 71, 973 983. Ayoob, K., Duyff, R. L., & Quagliani, D. (2002). Position of the American dietetic association: Food and nutrition misinformation. Journal of t he American Dietetic Association, 102 (2), 260 266. Baker, L., Wagner, T. H., Singer, S. & Bundorf, M. K. (2003). Use of the Internet and E -mail for Health Care Information: Results From a National Survey. Journal of the American Medical Association 289(18), 2400 2406. Benigeri, M. & Pluye, P. (2003). Shortcomings of health information on the Internet. Health Promotion International 18(4), 381 386. Bucy, E. P. (2003). Media credibility reconsidered: Synergy effects between on air and online news. Journalis m & Mass Communication Quarterly, 80(2), 247 264. Carmel, E., Crawford, S., & Chen, H. (1992). Browsing in hypertext: A cognitive study. IEEE Transactions on System s, Man and Cybernelics, 22, 865 884. Centers for Disease Control. (2007). Obesity among adu lts in the United States: No statistically significant change since 20032004. Retrieved December 4, 2007, from: http://www.cdc.gov/nchs/data/databriefs/db01.pdf Centers for Disease Control and Prevention. (2009). Trends by State 1985 2008. ONLINE, http://www.cdc.gov/obesity/data/trends.html#State Charlton, K.E., Brewitt, P. & Bourne, L. T. (2004). Sources and credibility of nutrition information among black urban South African women, with a focus on messages related to obesity. Public Health Nutrition 7, 801 811. Clancy -Hepburn, K., Hickey A. A., & Nevill, G. (1974). Childrens behavior responses to TV food advertisements. Journal of Nutrition Education, 6, 93 96. Colditz, G. A.(1999). Economic costs of obesity and inactivity. Medicine & Science in Sports & Exercise 31(11), S663 S667.

PAGE 66

66 Dietz Jr., W. H., & Gortmaker, S. L. (1985). Do we fatten our children at the television set? obesity and television viewing in children and adolescents. Pediatrics, 75 (5), 807. Dixon, H. G., Scully, M. L., Wakefield, M. A., White, V. M., Crawford, D. A. (2007). The effects of television advertisements for junk food versus nutritious food on c hildren s food attitudes and preferences. Social Science & Medicine, 65, 1311 1323. Dussere, S. A. (1976). The Effects of Television Advertising on Childrens Eating Habits, thesis. Amherst, University of Massachusetts. Dutta Bergman, M. J. (2004). The imp act of completeness and web use motivation on the credibility of e -health information. Journal of Communication. 253 269. Eagly, A., & Chaiken, S. (1993). The psychology of attitudes Orlando, FL: Harcourt Brace Jovanovich College Publishers. 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). Evers, K. E., Prochaska, J. M., Prochaska, J. O., Driskell, M., Cummins, C. O., & Ve licer, W. F. (2003). Strengths and weaknesses of health behavior change programs on the internet. Journal of Health Psychology, 8(1), 63. Eysenbach, G. (2008). Credibility of health information and digital media: New perspectives and implications for yout h. Ca mbridge, MA; The MIT Press, 123 154. Flanagin, A. J., & Metzger, M. J. (2000). Perceptions of Internet credibility. Journalism and Mass Communication Quarterly 77(3): 515 40. Flanagin, A. J., & Metzger, M. J. (2003). The perceived credibility of perso nal Web page information as influenced by the sex of the source. Computers in Human Behavior 19(6), 683 701. Flanagin, A. J., & Metzger, M. J. (2007). The role of site features, user attributes, and information verification behaviors on the perceived credibility of web -based information. New media & Society 9(2), 319 342. Fogg, B., Marshall, J., Laraki, O., Osipovich, A., Varma, C., Fang, N., Paul, J., Rangnekar, A., Shon, J., Swani, P., Treinen, M. (2001). What makes websites credible? A report on a larg e quantitative study, Proceedings of CHI'01 conference on Human factors in Computing Systems, Vol. 3 No. 1, 61 68. Freeland Graves, J. & Nitzke, S. (2002). Position of the American Dietetic Association: Total diet approach to communicating food and nutriti on information. Journal of the American Dietetic Association 102(1), 100 108. Freeman, K. S. & Spyridakis, J. H. (2004). An examination of factors that affect the credibility of online health information. Technical communication, 51(2), 239 263.

PAGE 67

67 Galst, J. P. & White, M. A. (1976). The unhealthy persuader: The reinforcing value of television and childrens purchase influencing attempts at the supermarket. Child Development, 47(4) 1089 1096. Galuska, D. A., Will, J. C., Serdula, M. K., & Ford, E. S. (1999). Are health care professionals advising obese patients to lose weight? JAMA: Journal of the American Medical Association, 282(16), 1576. Gaziano, C., & McGrath, K. (1986). Measuring the concept of credibility. Journalism Quarterly 63(3), 451 462. Glanz, K., Basil, M., Mailbach, E., Goldberg, J. and Snyder, D. (1998). Why Americans eat what they do: Taste, nutrition, cost, convenience, and weight control concerns as influences on food consumption. Journal of the American Dietetic Association, 98, 1118 1126. Goldberg, J. P. (2000). Nutrition communication in the 21st Century: What are the challenges and how can we meet them?. Nutrition, 16(7/8), 644 646. Griggs, B.(1988). The Food Factor: An Account of the Nutrition Revolution Penguin Books. Gunther, A. C. (1992). Biased press or biased public? Attitudes toward media coverage of social groups. Public Opinion quarterly 56, 147 167. Hancox, R. J., Milne, B. J., & Poulton, R. (2004). Association between child and adolescent television viewing and adult health: A longitudinal birth cohort study. Lancet, 364(9430), 257 262. Harris, L. M. (1995). Differences that make a difference. In L. M. Harris (Ed.), Health and the new media: Technologies transforming personal and public health (pp. 163 183). Hillsdale, NJ: Erlbaum. Harris, P. R., Sillence, E., & Briggs, P. (2009). The effect of credibilityrelated design cues on responses to a web -based message about the breast cancer risks from alcohol: randomized controlled trial. Journal of Medical Internet Research, 11(3) Harrison, K., & Marske, A. (2005). Nutritional content of foods advertised during the television programs children watch most. American Journal of Public Health, 95(9), 1568 1574. Hawkins, D. T. (1999). What is credible information? ONLINE 86 89. Availa ble from http//www.onlineinc.com/onlinemag. Health Information National Trends Survey (HINTS). (20032007). URL: hints.cancer.gov; HINTS database, National Cancer Institute, DCCPS, Behavioral Research program, Health Communication and Informatics Research Branch. Holgado, Martinez Gozalez, Irala Extevez, Gibney, Kearney, & Martinez. (1999). Sources of information about diet and health in a Mediterranean country. The European Journal of Public Health 10(3), 185 191.

PAGE 68

68 Hong, T. (2005). The influence of structural and message features on Web site credibility. Journal of the American Society for Information Sci ence and Technology, 57(1), 114 127. Hovland, C., & Weiss, W. (1951). The influence of source credibility on communication effectiveness. Public Opinion Quarterly 15, 635 650. Jacobs D R Pereira M Slavin M J Marquart L. (2000). Defining the impact of wholegrain intake on chronic disease. Cereal Foods World, 45, 51 53. Johnson, B.T., & Eagly, A.H. (1989). Effects of involvement on persuasion: A m eta analysis. Psychological Bulletin 106, 290 314. Kenkel, D.(1990). Consumer Health Information and the Demand for Medical Care. Review of Economics and Statistics, 72(3), 587 95. Kiousis, S. (2006). Exploring the impact of modality on perceptions of c redibility for online news stories. Journalism Studies 7 (2), 348 359. Kreps, G. L., & Kunimoto, E. N. (1994). Effective communication in multicultural health care settings Thousand Oaks, CA: Sage. Kunst, H., Groot, D., Latthe, P., Latthe, M., & Khan, K. (2002). Accuracy of information of apparently credible websites: Survey of five common health topics. British Medical Journal 7337, 581 582. Leichter, S. B. (1999). The business of diabetes education before and after new medicare regulations. Clinical Dia betes, 17 (3), 127. Lindeman, M., Keskivaara, P., & Roschier, M. (2000). Assessment of magical beliefs about food and health. Journal of Health Psychology, 5(2), 195. Liszka, H. A., Steyer, T. E., & Hueston, W. J. (2006). Virtual medical care: How are our patients using online health information? Journal of Community Health, 31(5), 368 378. Lowry, R., Galuska, D. A., Fulton, J. E., Wechsler, H., Kann, L., & Collins, J. L. (2000). Physical activity, food choice, and weight management goals and practices am ong U.S. college students. American Journal of Preventive Medicine, 18(1), 18 27. Madden, H., & Chamberlain, K. (2004). Nutritional health messages in womens magazines: A conflicted space for women readers. Journal of Health Psychology 9(4), 583 597. Mar chionini, G. (1987). An invitation to browse: Designing full text systems for novice users. Canadian Journal of Information Science 12(3). 69 79. Marchionini, G., & Shneiderman, B. (1988). Finding facts vs. browsing knowledge in hypertext systems. IEEE Co mputer 21(3), 70 79.

PAGE 69

69 Marcus, B. H., Owen, N., Forsyth, L. H., Cavill, N. A. & Fridinger, F. (1998). Physical activity interventions using mass media, print media, and information technology. American Journal of Preventive Medicine, 15(4), 362 378 McIntosh, G. H., Noakes, M., Royle, P. J., & Foster, P. R. (2003). Whole grain rye and wheat foods and markers of bowel health in overweight middle aged men. The American Journal of Clinical Nutrition 77, 967 974. Meyer, P. (1988). Defining and measuring credibil ity of newspapers: Developing an index. Journalism Quarterly 65, 567574. Miniard, P. W., Bhatla, S., Lord, K. R., Dickson, P. R., & Unnava, H. R. (1991). Picture -based persuasion processes and the moderating role of involvement. Journal of Consumer Rese arch 18. 92107. Mitchell, V W. & Boustani, P. (1993). The effect of demographic variables on measuring perceived risk. in Levy, M. and Grewal, D. (Eds), Academy of Marketing Science Conference, Developments in Marketing Science Vol. XVI, Miami, May, 663 669. Mokdad, A. H., Serdula, M. K., Dietz, W. H., Bowman, B. A., Marks, J. S., & Koplan, J. P. (1999). The spread of the obesity epidemic in the United States, 19911998. The Journal of the American Medical Association, 282(16), 15191522. Morley, B., Wak efield, M., Dunlop, S., & Hill, D. (2009). Impact of a mass media campaign linking abdominal obesity and cancer: a natural exposure evaluation. Health Education Research 24(6), 1069 1079. Murphy, J. (1998). What makes people click? An analysis of Web page navigation. Unpublished doctorial dissertation, Florida State University. Must, A., Spadano, J., Coakley, E. H., Field, A. E., Colditz, G., & Dietz, W. H. (1999). The disease burden associated with overweight and obesity. JAMA: Journal of the American Med ical Association, 282(16), 1523. National Cancer Institute. (2004). Obesity and cancer: questions and answers. Retrieved March 16, 2004, from: http://www.cancer.gov/cancertopics/factsh eet/Risk/obesity Nayga, R.M. (1997). Impact of sociodemographic factors on perceived importance of nutrition in food shopping. Journal of Consumer Affairs, 31 (1), 1 9. Nayga, R. M., Jr (2000). Nutrition knowledge, gender, and food label use. Journal of Consumer Affairs 34(1), 97 112. Neuhauser, L., & Kreps, G. L. (2003). Rethinking communication in the E -health era. Journal of Health Psychology, 8(1), 7 23. Newman, J.(1986). Melting Pot: An Annotated Bibliography and Guide to Food and Nutrition Informa tion for Ethnic Groups in America Garland Publishing.

PAGE 70

70 Nicola, J., Gray, J. D., Klein, P. R., Noyce, T. S., Sesselberg, & Judith, A. C. (2005). Health information -seeking behavior in adolescence: The place of the Internet. Social Science and Medicine 60, 1467 1478. Pauly, M. V. & Satterthwaite. M. A.(1981). The pricing of primary care physicians services: A test of role of consumer information. The Bell Journal of Economics 12(2), 488 506. Petty, R.E., & Cacioppo, J.T. (1979). Issue involvement can increa se or decrease persuasion by enhancing message relevant cognitive responses. Journal of Personality and Social Psychology, 37, 1915 1926. Petty, R. E. & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. In: Berkowitz L, editor. Advanc es in Experimental Social Psychology. Ne w York: Academic Press. 19. 123 205. Petty, R. E. & Cacioppo, J. T. (1990). Involvement and Persuasion: Tradition Versus Integration. Psychological Bulletin 107(3), 367 374. Petty, R. E., Cacioppo, J. T., & Schuman n, D. (1983). Central and peripheral routes to advertising effectiveness: The moderating role of involvement. Journa l of Consumer Research, 10, 135 146. Pew Internet & American Life Project. (2000). The online health care revolution: How the Web helps Amer icans take better care of themselves. Retrieved June 5, 2001, from the World Wide Web: http://www.pewinternet.org Rice, R. & Katz, J. E. (2000). The Internet and Health Communication: Experiences and Expectations. T housand Oaks, CA: Sage. Rieh, S. Y., & Belkin, N. J., (1998). Understanding judgment of information quality and cognitive authority in the WWW. Proceedings of the 61st Annual Meeting of the American Society for Information Science 35, 279 289. Reuters. (2 009). Obesity costs U.S. health system $147 billion: study. From the World WideWeb: http://www.reuters.com/article/healthNews/idUSTRE56Q36020090728?pageNumber=1& virtualBrandChannel=0 Sankofa, J. & JohnsonTaylor, W. L. (2007). News coverage of diet related health disparities experienced by black Americans: A steady diet of misinformation. Journal of Nutrition Education and Behavior 39(2), S41 S4 4. Sillence, E., Briggs, P., Harris, P., & Fishwick, L. (2007). Health websites that people can trust the case of hypertension. Interacting with Computers, 19(1), 32 42. Slater, M. (1997). Persuasion processes across receiver goals and message genres. Communication Theory, 7 125 148

PAGE 71

71 Stephenson, M.T., Palmgreen, P., Hoyle, R.H., Donohew, L., Lorch, E.P., & Colon, S.E. (1999). Short -term effects of an anti -marijuana media campaign targeting high sensation seeking adolescents. Journal of Applied Communication Research 27, 175 195. Street, R. L., Jr, & Piziak, V. K. (2001). Improving diabetes care with telecomputing technology. In R. E. Rice & J. E. Katz (Eds.), The Internet and health communication (pp. 287 327). Thousand Oaks, CA: Sage. Sundar, S. S., & Nass, C. (2001). Conceptualizing sources in online news. Journal of Communication, 51(1), 52 72. Szwajcer, E. M., Hiddink, G. J., Koelen, M. A., & van Woerkum, C. M. J. (2005). Nutrition related information -seeking behaviours before and throughout the cou rse of pregnancy: Consequences for nutrition communication. European Journal of Clinical Nutrition, 59, S57 S65. The Harris poll. (2009). Internet Provides Public with Health Care Information that They Value and Trust and Which Often Stimulates Discussion with Their Doctors. ONLINE, www.harrisinteractive.com Treise, D., Walsh Childers, K., Weigold, M., & Friedman, M. (2003). Cultivating the science Internet audience: Impact of brand and domain on source credi bility for science information. Science Communication, 24(3), 309 332. Quandt, S. A. and Ritenbaugh, C. (eds) (1986) Training manual in nutritional anthropology 20th, American Anthropological Association Washington, DC. Walther, J. B., Wang, Z. & Loh, T (2004). The effect of toplevel domains and advertisements on health web site credibility. Journal of Medical Internet Research. 6(3). Wise, K., & Kim, H. (2008). Searching versus surfing: How different ways of acquiring content online affect cognitive processing. CyberPsychology & Behavior, 11(3), 360 362.

PAGE 72

72 BIOGRAPHICAL SKETCH Eun Hwa Jung earned a Bachelor of A rts with a double major in c ommunication (advanced) and advertising, and a minor in public administration from Kookmin University, Seoul, Sout h Korea After graduation in 2006, she began her graduate studies in communication and completed a master s degree at the Graduate School of Communication from Kookmin University. In August of 2008, s he entered the master s program in the College of Journa lism and Communications at the University of Florida and received her Masters of Arts in the spring of 2010. S he will pursue her doctora l studies in mass communication, concentrating on new media effect and health communication.