|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 FRAMING FOOD HEALTH RISKS IN A WEB 2.0 WORLD: EFFECTS OF MESSAGE FRAMING AND WEB 2.0 INTERACTIVITY ON ATTITUDE CHANGE AND RECALL OF INFORMATION By KAREN J. CANNON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIV ERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Karen J. Cannon
3 To my grandparents, Stan and Marian Olson and Jim and Kay Cannon
4 ACKNOWLEDGMENTS The jo urney to completing a doctoral degree, while seemingly an individual endeavor, is not one that I traveled alone. I am indebted to a number of dear friends, family and faculty members who have helped me along the way. I would first like to thank the members of my doctoral committee for their unending support and expertise, advice and counsel. Debbie Treise, my external committee member, provided invaluable insight and perspective on my study from its inception. Debbies seminar course during my first semester at UF had an immeasurably positive impact on not only my choice of topic for this dissertation, but also helped me see that I could indeed be successful as a graduate student and future researcher. Her personal support and friendship has helped me throug h some of the most difficult and enjoyable times of my life, and I will forever treasure her friendship and mentorship. To my methods man, Brian Myers, thank you for always finding time in your busy schedule to meet with me and discuss a new and wacky i dea I had about how to carry out my study. Your input and expertise were essential to my understanding the process and I am thankful that you were willing to step out of your comfort zone and be a part of a communication study! I greatly appreciate your support and participation; you have indeed helped me in my quest to become a researcher. I wish to say a special thank you to Tracy Irani, my agricultural communications expert and faculty supervisor for much of my teaching assignment while in the department Tracy, your help has been tremendous and thank you seems an inadequate phrase for the feelings of gratefulness that I have for the time, dedication, mentorship, and expertise that you have shared with me. It has been an honor and a pleasure to work with you and have you as one of my cherished committee members.
5 Thank you for your unending faith in my abilities, especially when I could not see them. I look forward to my future role as a university professor so that I can continue working with someone that I so admire. To my committee chair, adviser, mentor and true friend, Nicole Stedman, I thank you for opening your heart, your house, your family, your office door, and your bank of experiences and knowledge to me. I could not have made this journey without you and your support and advice have helped me become a better student, learner, teacher, researcher and person. Your undying faith in me kept me going when the end of the tunnel seemed so very far away. I am so lucky to have you, and your darling husband Jeremy, and daughters Jordan and Sarah, in my life. I look forward to a future where we continue to work together on projects and where I get to see my Florida family frequently. As anyone who has been a graduate student knows, in addition to faculty and family, graduate student colleagues become friends, and I am indebted to a number of them for their support and advice during my time at Florida. Ann De Lay was, and continues to be, a dear friend, who was there at the beginning of my journey as a Flori da Gator and has always words of support ready when needed, even from across the country. Andrew Thoron has not been just an officemate, but has been a friend and colleague who was always available for a laugh and good research discussion. I am thrilled he will soon be a Florida faculty member and would like to remind him that payback for grad office shenanigans never dies. Angie Lindsey, Katie Abrams, Rochelle Strickland, Kate Shoulders and Alexa Lamm have been invaluable officemates and research partners who made my life in Rolfs Hall enjoyable and full of laughter. Jason
6 Davison, Ryan Conklin, Viviana Giraud, and Laura Kubitz, as well as Melissa Mazurkiewicz, Mary Rodriguez, Micah Scanga, Lauren Hrnrick, Allison Britton Wagner and Andrea Andrews have cont ributed many wonderful hours of fun during my experience in the Department, and I thank them for their friendship. My deep appreciation goes to Jodi Modica, Holly OFerrell, and Rachel Harris for their understanding, assistance, and knowledge, as well as t heir friendship and unending smiles; AEC could not be the department it is without you. Glenn Israel has become a mentor and friend during my time here at Florida, and I will always remember his kind words and helpful advice. Dr. O, Greg Gifford, Ricky Tel g, and Christy Chiarelli have all contributed valuable time, support, mentorship and friendship to my time as a graduate student and I thank them all. I have no words to express my feelings of gratitude to the members of Team Festrogen for their love, support and understanding during this intense time in my life. You are my nearest and dearest despite our miles apart. To Jen and James, thank you for always being willing to have a visitor from up above on the map. Thank you to Pat and Roger Winston who alw ays opened their arms and their home to a Gator despite being diehard Bulldogs; I will be forever grateful for your love and kindness. To my dear friend and adopted brother Jason Greer, I am thankful everyday for your smiling face and your ability to make an outstanding latte. And finally, thank you to my family, without whom this journey, and my success in life, would not be possible. Cannons, Conways, Deipenbrocks, Estrellas, Mori Pranges, Olsons, and Zwifelhofers, you have shown me love and support tha t has carried me through this adventure thank you. To my dearest sister Adrienne who has endured
7 this crosscountry separation with remarkable pluck, I am so glad to have always found you on the other end of the phone when I needed. I am so lucky to have you as my sister! To my parents Kris and Jim, I am unable to find the words to express what your love, understanding, and support mean to me. My only wish is to continue to make you proud to be my parents. Thank you for being such remarkable people, for t eaching me how to be a lifelong learner, and for always believing in me. I do not yet know what the next step in this adventure of life may be, but I am beyond blessed to have the above group of individuals with me in my heart as I take on the next projec t. As always, GO GATORS!
8 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES .......................................................................................................... 12 LIST OF FIGURES ........................................................................................................ 14 LIST OF DEFINITIONS ................................................................................................. 15 ABSTRACT ................................................................................................................... 17 CHAPTER 1 INTRODUCTION .................................................................................................... 19 Food Concerns and Media Coverage ..................................................................... 21 Organic Food on the Rise................................................................................. 25 Mass Media Embrace Organic ......................................................................... 30 Role of Mass Media in Contemporary Society ........................................................ 31 News Consumption ................................................................................................. 33 Agriculture and the Media ....................................................................................... 34 Todays Media Landscape ...................................................................................... 36 News and New Media ...................................................................................... 37 News and Millennials ........................................................................................ 39 New Media .............................................................................................................. 40 Web 2.0 and Interactivity .................................................................................. 43 Blogs ................................................................................................................ 45 Problem Statement ................................................................................................. 47 Purpose and Objectives .......................................................................................... 48 2 RELEVANT LITERATURE ...................................................................................... 50 Overview of Literature ............................................................................................. 50 Website Interactivity ................................................................................................ 50 Framing Theory ...................................................................................................... 55 Media Frames .................................................................................................. 58 Audience Frames ............................................................................................. 59 Frames and New Media ................................................................................... 62 Framing Science .............................................................................................. 64 Media Framing of Agriculture and Food ........................................................... 66 Framing Organics ............................................................................................. 69 Framing Risk .................................................................................................... 70 Risk Communication ............................................................................................... 72 Social Amplification of Risk .............................................................................. 74 Food Risk ......................................................................................................... 74
9 Persuasion .............................................................................................................. 76 Elaboration Likelihood Model .................................................................................. 77 Routes to Persuasion and Information Processing ........................................... 78 Issue Involvement and Argument Quality ......................................................... 80 Need for Cognition ........................................................................................... 81 Fear Appeals .................................................................................................... 83 Fear and Persuasion ........................................................................................ 86 Fear Appeals and Health Messages ................................................................ 87 Attitude .................................................................................................................... 88 Attitude Formation ............................................................................................ 89 Role of Schema ................................................................................................ 90 Attitude Change ................................................................................................ 91 Attitudes and Information Processing ............................................................... 92 Priming and Attitudes ....................................................................................... 94 Information Recall ................................................................................................... 96 Recall of Online v. Print Messages ......................................................................... 98 Conceptual Model ................................................................................................. 101 Chapter Summary ................................................................................................. 102 3 METHODOLOGY ................................................................................................. 104 Overview ............................................................................................................... 104 Hypotheses and Research Objective .................................................................... 105 Research Design .................................................................................................. 105 Subjects ................................................................................................................ 109 Message Stimuli ................................................................................................... 111 Message Development ................................................................................... 111 Message Testing ............................................................................................ 113 Independent Variables .......................................................................................... 114 Message Frame ............................................................................................. 114 Website Interactivity ....................................................................................... 115 Dependent Variables ............................................................................................ 116 Attitude Change .............................................................................................. 116 Information Recall .......................................................................................... 117 Attribute Variables ................................................................................................ 118 Need for Cognition ......................................................................................... 118 Issue Involvement .......................................................................................... 118 Instrum entation ..................................................................................................... 119 Pilot Study ...................................................................................................... 121 Scale Reliabilities ........................................................................................... 121 Instrument Content ......................................................................................... 123 Procedure ............................................................................................................. 126 Data Analysis ........................................................................................................ 127 Final Instrument Scale Reliabilities ................................................................. 127 Hypothesis Testing ......................................................................................... 128
10 4 RESULTS ............................................................................................................. 130 Overview ............................................................................................................... 130 Descriptive Analysis .............................................................................................. 130 Demographics ................................................................................................ 131 Individual Att ribute Variables .......................................................................... 137 Issue involvement .................................................................................... 137 Need for cognition .................................................................................... 137 Prior knowledge ....................................................................................... 138 Prior attitude ............................................................................................. 140 Descriptive Analysis of Variables of Interest ......................................................... 141 Attitude Change .............................................................................................. 141 Information Recall .......................................................................................... 142 Manipulation Checks ............................................................................................. 142 Hypothesis Testing and Research Objective ........................................................ 145 5 CONCLUSION ...................................................................................................... 151 Overview ............................................................................................................... 151 Key Findings ......................................................................................................... 152 Conclusions and Implications ............................................................................... 154 Theoretical Implications .................................................................................. 154 Practical Implications ...................................................................................... 158 Recommendations ................................................................................................ 159 For Future Research ...................................................................................... 159 For Practice .................................................................................................... 161 Limitations ............................................................................................................. 162 Conclusion ............................................................................................................ 164 APPENDIX A VERBAL RECRUITMENT SLIDE ......................................................................... 165 B FIRST EMAIL CONTACT ..................................................................................... 166 C SECOND EMAIL CONTACT REMINDER .......................................................... 167 D FINAL EMAIL CONTACT THANK YOU ............................................................. 168 E INSTRUMENT ...................................................................................................... 169 F MESSAGE TREATMENTS ................................................................................... 181 G HIGH INTERACTIVE WEBSITE TREATMENT EXAMPLE .................................. 185 H LOW INTERACTIVE WEBSITE TREATMENT EXAMPLE ................................... 186
11 LIST OF REFERENCES ............................................................................................. 187 BIOGRAPHICAL SKETCH .......................................................................................... 210
12 LIST OF TABLES Table page 3 1 The PII Scale used to measure issue involvement. .......................................... 122 3 2 Attitude scale. ................................................................................................... 123 3 3 Efficient assessment of need for cognition. ...................................................... 124 4 1 Personal Involvement Inventory (PII) scale inter item consistency statistics. ... 138 4 2 Means table for need for cognition score. ......................................................... 138 4 3 Need for cognition scale inter item consistency statistics. ................................ 139 4 4 Means table for prior knowledge score. ............................................................ 139 4 5 Means table for prior attitude score. ................................................................. 140 4 6 Prior attitude tow ard organic beef. .................................................................... 140 4 7 Means table for attitude change by treatment group. ....................................... 141 4 8 Means table for information recall score by treatment group. ........................... 142 4 9 Openended message frame manipulation check. ........................................... 1 43 4 10 Close ended message frame manipulation check. ........................................... 144 4 11 Website interactivity manipulation check. ......................................................... 144 4 12 Means table for attitude change by treatment group. ....................................... 145 4 13 Means table for information recall score by treatment group. ........................... 146 4 14 One way betweensubjects ANOVA for attitude change differences bet ween subjects receiving risk framed messages and benefit framed messages. ........ 146 4 15 One way betweensubjects ANOVA for information recall differences between subjects receiving risk framed mess ages and benefit framed messages. ........................................................................................................ 147 4 16 One way betweensubjects ANOVA for attitude change differences between subjects receiving messages on high interactive websites and low interactive w ebsites. .......................................................................................................... 147
13 4 17 One way betweensubjects ANOVA for information recall differences between subjects receiving messages on high interactive websites and low interactive websites. ......................................................................................... 148 4 18 Independent samples t test for information recall differences between subjects receiving risk framed messages on a high interactive website and subjects receiving benefit framed messages on low interactive websites. ....... 149 4 19 Independent samples t test for attitude change differences between subjects receiving risk framed messages on a high interactive website and subjects receiving benefit framed messages on low interactive websites. ...................... 149 4 20 Backwards multiple linear regression analysis for variables predicting attitude change. ............................................................................................................. 150 4 21 Backwards multiple linear regression analysis for variables predicting information recall. ............................................................................................. 150
14 LIST OF FIGURES Figure page 2 1 Conceptual model. ............................................................................................ 102 4 1 Subjects age distribution. ................................................................................. 132 4 2 Subjects college affiliation. .............................................................................. 133 4 3 Subjects year in college. .................................................................................. 133 4 4 Subjects reported internet use. ........................................................................ 134 4 5 Subjects reported internet tasks. ..................................................................... 135
15 LIST OF DEFINITIONS Blog An abbreviation of the term weblog, a blog is a site maintained by an individual or organization with frequent/regular written entries appearing in reverse chronologic al order, usually containing news or commentary on a specific subject. A key characteristic of blogs is their extensive use of hyperlinks to other blogs and online content (Carlson, 2007). Conventional/ Beef products produced using any standards other tha n USDAs traditional beef National Organic Program. Elaboration A dual process communication model of persuasion developed by Likelihood Model R. E. Petty and J. T. Cacioppo, which illustrates that there are two routes to persuasion, one central wher e elaboration and cognition are high, the other peripheral where elaboration and cognition are lower, which are located at each end of a continuum. (Petty and Cacioppo, 1984). Framing Framing refers to the way events and issues are organized and m ade sense of, especially by media, media professionals, and their audiences (Reese, 2003, p. 7). Frames are organizing principles that are socially shared and persistent over time, that work symbolically to meaningfully structure the social world (Reese, 2003, p. 11). Interactivity The presence of f unctions that allow users to interact with the content on a website through the use of technology. Interactive functions help users to make choices and exert control over the communication process performing activities such as clinking on hyperlinks, entering comments on a website, and so forth (Chung & Yoo, 2008, p. 378). Mass media A ll media produced by an individual or organization that is intended for distribution to mass audiences, including traditional media sources such as print newspapers and magazines, as well as broadcast media such as radio, television, and Internet media (including blogs, podcasts, videos and message boards). Millennials The generation of young people, born in or after 1982, who ar e now in college and are considered the generation following Gen X, according to Howe and Strauss (2000). Need for Cognition Refers to an individuals tendency to engage in and enjoy effortful cognitive endeavors. Research on need for cognition suggest s that this characteristic is predictive of the manner in which people deal
16 with tasks and social information (Cacioppo, Petty, & Kao, 1984, p. 306). News Schudson (2003) defined news as something people make; things and events that are publicly notable (within a framework of shared understanding that judges it to be both public and notable) (p. 6). News habit Habitual or repeated use or consumption of news information. The Pew Research Center uses the term news habit to discuss the way in which, how f requently, and what kinds of news people use on a regular basis. News source/outlet Any professional news organization or journalist that has as its business the production of news information for consumption by the public. Online news News information delivered via the Internet. Organic beef Any beef produced within specifications outlined by the USDA in the National Organic Program Standards. These standards and regulations include the statement that animals raised on an organic operation must be fed organic feed and given access to the outdoors. They are given no antibiotics or growth hormones (U.S. Department of Agriculture, 2008). Recall Recall is defined as the recollecting of an earlier event and its surrounding circumstances" ( Booth, 1970, p. 604) In this study, recall of information is used as a proxy measurement for knowledge/learning. Whether an individual learns or gains knowledge from a news story is influenced by a variety of factors pertaining to the social and behavioral situation of h ow the user receives the story and the environment in which he finds himself (DeFleur, Davenport, Cronin, & DeFleur, 1992). Traditional website A website with low levels of interactivity/interactive functions. Dynamic website A website with high levels o f interactivity/interactive functions. Web 2.0 Characterized by interactivity and collaboration, Web 2.0 is a set of technologies, which are intended to encourage participation and socialness among individuals on the World Wide Web. Commonly used Web 2.0 technologies include podcasts, wikis, blogs, videos, photo sharing services, as well as social networking sites and services (Madden & Fox, 2006).
17 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FRAMING FOOD HEALTH RISKS IN A WEB 2.0 WORLD: EFFECTS OF MESSAGE FRAMING AND WEB 2.0 INTERACTIVITY ON ATTITUDE CHANGE AND RECALL OF INFORMATION By Karen J. Cannon May 2011 Chair: Nicole Lamee Perez Stedman Major: Agricultural Education and Communication News media coverage of issues and events affects public attitudes and perceptions and stories and messages appearing in news media are framed, often highlighting conflict to discuss issues of the day. In the past two decades, news stories highlighting problems related to food safety have increased due to well publicized foodrelated crises such as E. coli in spinach and beef, salmonella in peanut butter, and bovine spongiform encephalopathy (mad cow) in beef. Consumers have consequently reported increasing concerns over the safety of food and have moved toward organic products to alleviate such fears. At the same time, the manner in which consumers get news has changed; today s consumer most often turns to the Internet for news information. However, previous research indicates that messages delivered online may be processed differently than those read on the page, especially when they include elements of fear. This study utiliz ed an experimental design with a convenience sample of 410 university students to examine how risk/benefit message framing and website interactivity impacted individuals information recall and change in attitude after exposure to an online news story.
18 The results of the study indicated that significant differences existed in attitude change between individuals receiving risk framed message treatments and those receiving benefit framed treatments, with those receiving risk framed messages having a greater overall change in attitude. No significant differences among treatment groups were present in this study, possibly due to the high prior level of knowledge about the subject on the part of the participants. This study attempted to determine if risk/benefit message framing and website interactivity impacted subjects ability to recall information after exposure to a news story and affect a change in attitude. The results suggest that the level of website interactivity employed in this study was not sufficient to impact subjects information recall or attitude change, however, the frame employed did affect attitude change.
19 CHAPTER 1 INTRODUCTION First it was hamburger, then municipal water, apple ciderand now, spinach. Is nothing safe? Maki, 2006. In Se ptember 2006, a largescale outbreak of the bacteria E. coli O157:H7 was found in fresh spinach from Californias Central Valley region. During the course of the outbreak, more than 200 individuals in 26 states were hospitalized due to illness and three died of their infections. The Centers for Disease Control and Prevention (CDC) and the Food and Drug Administration (FDA) worked in collaboration with Natural Selection Foods, the company found to have distributed substantial portions of the affected product, to recall all bagged and loose fresh spinach. The sale of spinach nationwide, a $107 million industry, ground to a halt and is estimated to have cost California growers as much as $3,500 an acre. Almost overnight organic spinach went from an ultimat e health food to the ultimate health risk (Curtin & Gaither, 2007, p. 1). During the month of January 2009, an outbreak of salmonella in peanut products distributed by Peanut Corporation of America (PCA) sickened hundreds of people and prompted extensive recalls of peanut products ( DeNoon, 2009). More than 3,900 different consumer items were recalled due to the contamination, many of which were produced at the companys plant in Blakely, Georgia, which had previously failed a federal inspection by the Food and Drug Administration (Maugh & Engel, 2009; Wittenberger & Dohlman, 2010). The outbreaks size and scale raised the attention of legislators on Capitol Hill who then called for stricter food safety standards and increased federal inspections (Harris, 2009).
20 Media coverage of these two foodsafety related events, E. coli in spinach and salmonella in peanut products, received considerable media attention. Food scares such as these and others involving salmonella, listeria and bovine spongiform encephalopathy (BSE) have resulted in the public s increased awareness and concern about food safety issues (Shaw, 2004). The extent of coverage these events received supports Tucker, Whaley and Sharps (2006) assertion that food safety is a growing concern worldwide. Indeed, it underlies the importance of studying the relationship among these events, how they are portrayed in the mass media and how consumers perceive messages regarding these topics The influence of mass media should not be underestimated because perceptions and attitudes go handin hand. Many of the attitudes expressed through the media directly affect consumers perceptions of agriculture (Haygood, Hagins, Akers, & Keith, 2002, p 3). For decades, the predominant concern regarding food in most of the world was whether enough was produced and available to feed the population (Sobal & Maurer, 1995). In much of the developing world this is still true. However, today in the United States as in other developed countries, consumer concerns about q uantity have been replaced by concerns about quality (Sobal & Maurer, 1995). As our rapidly modernizing society continues to move further from its agrarian roots and towards an increasingly urbanized landscape, public knowledge about agricultural producti on practices diminish (Doerfert, 2003) and fears about food safety risks increase (Anderson, 2000). Fewer than 1% of the American workforce is engaged in production agriculture occupations (Hurt, 2002). This lack of first hand agricultural knowledge contr ibutes to attitudes and perceptions that are primarily based on media accounts (Ten Eyck, 2000;
21 Whaley & Tucker, 2004). For most people, there is a fear that, since food production is increasingly centralized, they have lost the ability to have control ov er the safety of their food (Anderson, 2000). With this come increased fears of food safety incidents. Indeed, Gregory (2000) noted the decline in public confidence [of food safety] has been associated with an increase in the frequency of food scares ( p. 251). It also seems that as agricultural production becomes more technically oriented, food safety concerns related to these technologies (biotechnology and genetically modified foods, for example) increase as well (Gregory, 2000). Food Concerns and Media Coverage Concerns about risks associated with eating foods produced in a certain manner are on the rise, now that the focus on food has shifted from producing enough to producing food of high quality that is safe to eat. Over the last decades, several notable incidents of agricultural products prompting food scares have arisen, including pesticide residues, bacterial contamination, and concerns about biotechnology and genetically engineered foods (Tucker, Whaley, & Sharp, 2006). Yiridoe, Bonti Ankomah, and Martin (2005) noted: Food scares have spanned several years, includingproblems of mercury in fish, botulism in tinned (canned) salmon and hormone residues in veal and beef in the 1970s; salmonella in the 1980s, bovine spongiform encephalopathy (BSE) and E. coli in the 1990s; and food and mouth disease in the 2000s (p. 198). The controversy over the use of the chemical Alar (used by producers to keep apples from ripening prematurely, both on the tree and during transport) on apples came to a head in 1989 when a major story regarding chemical residues and the threat to children aired on CBSs 60 Minutes news magazine program (Herrman, Warland, & Sterngold, 1997). As Alar is a carcinogen, consumers were outraged to learn about the
22 extent of the threat. Mon ths of negative publicity followed, keeping the issue on the radar of consumers, despite the fact that many apple producers and food transport companies had voluntarily agreed to ban the use of the chemical on their products. Meat products, specifically beef, are prominent on the radar of concern for consumers due in part to a few significant beef related food safety scares. In 1993, the Health Department in Washington State was contacted by Seattles Childrens Hospital and told that a high number of chil dren were being treated for infections of the bacteria E. coli O157:H7 (Ulmer & Sellnow, 2000). Two days after the report from the Seattle area hospital, the president of the fast food restaurant Jack in the Box was alerted that there was a connection bet ween hamburgers sold by the company and this highly unusual number of illnesses, considered by that time an outbreak of E. coli (Ulmer & Sellnow, 2000). One month later, three children under the age of three had died from their illnesses; 400 people in al l were found to have been infected due to the outbreak in three separate states: Washington, Idaho, and Nevada (Adelson, 1993). The contaminated burgers served at Jack in the Box restaurants were traced back to a lot of 500,000 contaminated hamburger patti es. As a result, Powell (2000) noted: E. coli O157:H7 became the focus of Congressional debates on regulatory reform, tragic tales from bereaved parents, and the subject of investigative journalism. More importantly, in the wake of the [outbreak] stories about microbiological food safety began appearing more frequently and more prominently in American media (p. 1). A Canadian scientist first identified E coli O157:H7, a variant of normal human gut bacterium, in 1982 as a cause of disease in humans Scienti sts determined that ruminant animals (cattle, sheep, deer) can serve as reservoirs for the bacteria, and it is often found in their guts and on their hides. When humans eat food contaminated with E. coli O157:H7, the bacteria attach to the surface of the i ntestines and cause the
23 release of a toxin that is then absorbed through the intestinal wall, which can cause hemolytic uremic syndrome (HUS). HUS is a renal disease causing severe anemia and loss of liver function. This disease is the leading cause of ki dney failure in children and also poses a risk to adults with suppressed immune systems (Leiss and Powell, 2004). Media coverage of the outbreak helped Americans become familiar with terms like s almonella, shigella, campylobacter listeria and of cours e, E. coli The incident in Washington state sparked a surge in the public awareness of bacterial risks in food and catalyzed political reform of the entire meat inspection system in the United States (Leiss & Powell, 2004, p. 78). However, concerns reg arding food safety and meat, specifically beef, have not been limited to bacterial contamination issues. The first outbreak of Bovine Spongiform Encephalopathy (BSE, more commonly known as mad cow disease) occurred in Britain in 1986, but by 1992, the dise ase had been classified as an epidemic, with more than 1,000 cases reported (Jin, Skripnitchenko, & Koo, 2004). BSE, a degenerative neurological disease found in cattle, is caused by misfolded proteins (called prions), which build up in the central nervous system and eventually kill the animal (BSEinfo.org). The impact of the disease on the British cattle industry was significant. The arrival of bovine spongiform encephalopathy (BSE) in the UK during the late 1980s was a turning point in the relationship between the consumer and the food industry (Anderson, 2000, p. 260) British consumers, increasingly distrustful of their government and its ability to regulate food safety, blamed intensive farming techniques for the tragedy (Leiss & Powell, 2004). Br itish media seized on the crisis and sensationalized the story, citing an epidemic of unknown size and drew analogies to devastating human diseases such
24 as AIDS, while emphasizing the damage to the British beef industry (Brookes, 1999, p. 252). By all ac counts, the massive media coverage of the outbreak had brutal effects on the British beef industry and contributed to doubts about whether or not the British government could protect its citizens from this unfamiliar and devastating disease (Brookes, 1999; Leiss & Powell, 2004). Media reports in the U.S. (and in Canada, which also has a large cattle industry) during 1996 extensively commented on this link between the disease and intensive farming, but also emphasized that the risk of BSE appearing in the United States was small (Leiss & Powell, 2004). In May 2003, Canada reported its first case of BSE (Jin, Skripnitchenko, & Koo, 2004). Canada, much like the United States, has a farm economy that is heavily impacted by its beef industry. The majority o f Canadian beef is exported to other countries, however, and according to the Canadian Ministry of Agriculture, Food, and Rural Development, total beef sales to the United States from Canada during 2002 was approximately $1.3 billion (Ministry of Agriculture, Food, and Rural Development, 2003, as cited in Ruth, Eubanks, & Telg, 2005). Mere months after Canadas BSE outbreak, the disease appeared in a cattle herd in Washington State (BSEinfo.org) changing the landscape of the United States beef industry for ever (Ashlock, Cartmell, & Keleman, 2006). Several new governmental regulations were put in place following the BSE outbreak in the United States including an expansion of the disease surveillance program, a ban on downer cattle cattle that are too i njured, weak, or sick to stand and walk (Stull, Payne, Berry, & Reynolds, 2007, p. 227), regulation of certain technologies used to process beef, and the development of
25 a national animal identification program (Economic Research Service, 2006). Sometimes referred to as the cow that stole Christmas (Ashlock, Cartmell & Keleman, 2006; Economic Research Service, 2006), the arrival of BSE in the United States was characterized in media coverage of the incident as an industry crisis, an economic calamity, and as yet another health risk to be concerned about due to the link between BSE and a human variant of the disease, Cretuzfeldt Jakob disease (Ashlock, Cartmell & Keleman, 2006). In September 2006, the large scale outbreak of the bacteria E. coli O157:H7 fou nd in fresh spinach from Californias Central Valley region illustrated that food safety concerns are not limited to meat products. The media had a field day reporting on the possible cause(s) and extent of this food safety scare, frequently capitalizing on the paradox that spinach and leafy greens more generally, once considered healthy foods, were now the source of serious bacterial contamination, illness, and worse (DeLind & Howard, 2008, p. 302). Concern about food safety is one of several reasons consumers have begun to turn to organic food. Organic Food on the Rise Interest in organically produced food is increasing globally, and researchers have determined that this interest is a response to concerns about agricultural practices, animal welfare, f ood safety, human health and the environment (Hwang, Roe, & Teisl, 2005; Onyango, Hallman, & Bellows, 2007; Yiridoe, Bonti Ankomah, & Martin, 2005). The International Federation of Organic Agriculture Movements (IFOAM) reported that as of the close of 2008 more than 35 million hectares of land are certified to organic production standards, a rise in 3 million from the 2007 year, and the global organic product market was worth more than $50 billion in 2008 (IFOAM, 2010). In the U.S.
26 since the late 1990sor ganic production has more than doubled, but the consumer market has grown even faster. Organic food sales have more than quintupled, increasing from $3.6 billion in 1997 to $21.1 billion in 2008 (Greene, Dimitri, Lin, McBride, Oberholtzer, & Smith, 2009, p. iii). The USDAs Economic Research Service reported that milk and produce are the top two sales categories of organic food (Greene, et al., 2009), and fresh fruits and vegetables also dominate the market (Yiridoe, et al., 2005). While organic food cos ts more to produce, evidence suggests consumers are willing to pay the price premiums. In addition to organic food products, organic inputs such as soybeans are increasing in popularity. USDA statistics show that organic soybeans were a more profitable cr op than conventionally raised soybeans during 2006 (Greene, et al., 2009). In a significant statement of the importance of organic food to the consumer market, Congress included money in the 2008 Farm Bill, the first such provision, to provide support to f armers converting to organic production methods (Greene et al., 2009). While there is no declarative date associated with the organic and natural food movement or rise to prominence, according to Belasco (1993) organic foods have been around since the natural foods social movements of the 1960s (p. 285). Gregory (2000) noted the food safety scares of the 1980s helped to direct consumers attention towards organic and health foods. Demand for organic foods increased during the late 1990s (p. 253). As consumers react to food safety issues and show an interest in the environmental effects of pesticides and genetically modified foods, organic foods have become more and more attractive to consumers as safe and healthy alternatives
27 (Hughner, McDonagh, Prothero, Schultz, & Stanton, 2007). In Britain, concern about health is the primary reason consumers choose to purchase organic foods (Harper & Makatouni, 2002). In the United States, consumers choose organic (in order of decreasing importance) because of its p erceived safety, freshness, general health benefits, nutritional value, environmental impact, and flavor (Bourn & Prescott, 2002). According to the Organic Trade Association (2010), fruit and vegetables account for almost 40% of the organic market, followed by dairy at 15%, beverages at 13%, breads and grains at 11%, and meat, poultry and fish at 2%. Chryssochoidis (2000) indicated that considerable confusion existed about the exact meaning of the term organic for consumers. This confusion led the U.S. De partment of Agriculture to develop the National Organic Program (NOP) in October 2002 (California Institute for Rural Studies, 2005), which has helped the organic market to grow rapidly (Onyango, Hallman, & Bellows, 2007). The NOP requires that all produc ts labeled and marketed as certified organic are inspected by the USDA. Crops certified as organic cannot be grown with conventional pesticides or fertilizers, and organic livestock must be given organic feed, access to the outdoors, and cannot be fed anti biotics or growth hormones. In addition, the NOP regulations prohibit the use of genetic engineering, ionizing radiation, and sewage sludge in organic production and handling (U.S. Department of Agriculture, 2008, para. 6). In a review of studies profil ing the characteristics of organic consumers, Hugher et al., (2007) noted that consumers tend to purchase organic products in order to avoid chemicals used in conventional food production. Similarly, Bourn & Prescott (2002)
28 determined pesticide residues an d other conventional inputs were the primary reasons that consumers chose the perceived healthier alternative organic products. Consumers also often believe that organic food is more nutritious than conventional food (Jolly, 1991; Hill & Lynchehaun, 2002), however evidence in support of this belief is mixed. Studies investigating the nutritional superiority of organic foods have yielded contrasting conclusions (Yiridoe, Bonti Ankomah, & Martin, 2005). Dangour, Dodhia, Hayter, Allen, Lock, and Uauy (2009) reviewed studies conducted on the nutritive value of organic versus conventional foods and determined that there was no conclusive evidence of a difference in nutrient quality, a conclusion which echoed Williamss (2002) findings. Overall, most studies report that consumers purchase organic foods because of a perception that such products are safer, healthier, and more environmentally friendly than conventionally produced alternatives (Yiridoe, et al, 2005, p. 198). While concern about health is the pri mary motivator, consumers are also concerned about the environmental impact on intensive farming, and about the welfare of intensively produced farm animals (Harper & Makatouni, 2002, p. 288). The organic market in the United States has grown rapidly, in part as a result of consumer perceptions that organic foods alleviate food safety concerns (Onyango, Hallman, & Bellows, 2007). Recent scares associated with E. c oli, salmonella, and BSE incidence are likely to have further intensified consumers already substantial interest in organic foods ( Onyango, Hallman, Bellows, 2007, p. 400) Bernard, Zhang, and Gifford (2006) noted that the increased availability and variety of organic food products, driven by consumer demand, is a trend that has had a
29 substant ial impact on the U.S. food system in recent years. According to the U.S. Department of Agricultures Economic Research Service ( Greene, et al., 2009) organic production has risen to meet the demands of consumers, more than doubling since 1997 when sales were $3.6 billion, to over $21 billion in 2008. While the top two organic food product categories are organic produce and milk, during 2006 organic soybeans were more profitable than conventional soybeans (p. iii) illustrating that while organic food c osts more to produce, it also commands premium prices for producers. Reflecting consumers desires, one of the fastest growing segments of the organic food industry is organic meat and poultry, which is increasingly offered at university dining facilitie s, airport restaurants, hotels and at vacation destinations (Storck, 2008). A joint survey between the American Meat Institute and the Food Marketing Institute (2008) indicated that in the previous three months, 19% of consumers had purchased organic or natural meat. The rise in popularity of organic and natural foods can be seen in a number of consumer market metrics In the marketing of food, the rise can be seen in the expansion of Whole Foods Market and other chains like it. Since its initial public offering in 1992, the stock price of the company has risen almost 25 fold (McGinn, 2005) and in February 2007 the company purchased the Wild Oats supermarket chain which focuses on the same market and declared that it intended to compete against larger retail stores such as Wal m art and Safeway (Martin, 2008) to provide organic and natural foods. Wal lm art itself became a part of the organic food market and launched a USDA certified organic line of products in 2006 (Burros, 2006).
30 Mass Media Embrace Or ganic Marian Burros, a prominent food journalist for the New York Times wrote that 2006 was the year that Americans got in touch with their food, and its varied political and social connections came into focusThe Nation devoted an entire issueto food. Michael Pollans [book] Omnivores Dilemma, a book that took a look at industrial agriculturebecame a bestseller. Fast Food Nation [the book by Eric Schlosser] was made into a film by the director Richard Linklater (Burros, 2006, para. 6). Burros declar ed 2006 the year that organic food went mainstream, noting that even Wal Mart began carrying a line of USDA certified organic products in its stores. Several popular magazines including Newsweek and Glamour ran features encouraging consumers to be certain to purchase food items produced organically, advocating that milk and other dairy products, beef, and produce grown conventionally contain the highest levels of contaminates (McGinn, 2008; Green, 2009). In a content analysis of organic coverage in major newspapers in the United States, the United Kingdom, and Australia, Lockie (2006) determined that organic food and agriculture were portrayed in the news media as newsworthy in their own right, fashionable, high quality, a safe and natural alternative to c onventionally grown foods, the solution to environmental problems caused by conventional farming, and the focus of scientific controversy over the risks and benefits of organic food and farming practices. In media outlets, organic foods were constructed often in terms of what they were not genetically engineered or the products of industrialized agriculture. However, while the link between food scares, genetic engineering, and organic food was strong, it was certainly not the only or even dominant, fr aming of organic food ( Lockie, 2006, p. 319)
31 Role of Mass Media in Contemporary Society Just as food and eating are universal, few would argue that mass media influence almost every aspect of contemporary society, especially in the areas of politics, cul ture, daily social life and economics (McQuail, 2005). Mass communications, coupled with technology, are the two elements that set modern society apart from previous societies (Dexter, 1964). Before the arrival of mass media, communication happened primarily in a faceto face manner (Dexter, 1964). However, communication, by definition, is a social activity, and cannot exist outside of some manner of social interaction (Mendelsohn, 1964). Schudson (2003) defines communication as the social coordination o f individuals and groups through shared symbols and meanings (p. 11) and states that news dominates how members of society construct reality and obtain a sense of what is real and important. Mass media also affect and impact the values and attitudes we hold as a society. In return, the values and attitudes of society, and individual users, impact mass communication (Mendelson, 1964). When a story appears in the media, its presence is conferred a kind of public legitimacy, providing a certification of its importance (Reese, 2003). Indeed, Price, Huang, and Tewksbury (1997) explored individuals beliefs about the power of news coverage to persuade and about audience vulnerability to the potential persuasive impact of media. Durfee (2006) noted that these res earchers determined news information produced an impact similar to that of persuasive communication (p. 462). Durfee also commented that research in the mediaeffects tradition shows that news coverage is the most uncritically accepted type of media message (p. 462). Journalists and those who construct the news often referred to as gatekeepers, are responsible for regulating much of what the public learns about the world
32 (Schudson, 2003). This often results in members of the media having an ability to construct news according to their own perceptions of events (Schudson, 2003). In Making News Tuchman (1978) explored the process of news construction, observing that it is inherently a social process. This process provides a kind of window to the worl d through which Americans learn about themselves, their country, leaders and events, other countries, and their leaders and events. By seeking to disseminate information that people want, need and should know, news organizations both circulate and shape knowledge (Tuchman, 1978, p. 2). This shaping of knowledge happens through the use of what Gitlin (1980) called information packages, or media frames, which allow journalists to communicate their meaning to audiences and readers more effectively. Frames, t hen, are considered principles of selection, emphasis and presentation composted of little tacit theories about what exists, what happens, and what matters (Schudson, 2003, p. 35). Price, Tewksbury, and Powers (1995) noted that framing of events and news in mass media can thus systematically affects how recipients of the news come to understand these events (p. 4). Schudson (2003) referred to framing as as central a concept there is in the study of news (p. 35). He pointed out that news almost inherently emphasizes conflict (mainly stemming from the journalistic tradition), which holds that there are two sides to any story. In constructing news, complex social processes are simplified to emphasize melodrama and tell stories of battles between good and evil. In short, news coverage is often triggered by things going badly (Schudson, 2003). Shoemaker and Reese (1996) pointed out: Media content may be based on what happens in the physical world, but it singles out and highlights certain elements over other s; and the medias
33 own structural logic is imposed on those elements. Reality is necessarily manipulated when events and people are relocated into news or prime time stories. The media can impose their own logic on assembled materials in a number of ways, including emphasizing certain behaviors and people and stereotyping (p. 37). It may be that often, what the news highlights is conflict, negativity and elements of risk and fear. Altheid (1997) has written about fear being pervasive in American society. H e noted that our societal sense of fear has been produced through the interaction of commercial media, entertainment formats and programming, and the rise of the problem frame (p. 648) According to Altheid, the problem frame focuses the discussion of events and creates boundaries for what will be discussed and how, as well as what will not be discussed. This problem frame has been promoted by the news media, especially in the post 9/11 era, and the terms crime, victim, and fear are joined with news reports with specific connections to terrorism (Altheid, 2006, p. 416). Specific fears related to food are reflected in mass media, and according to Anderson (2000) concerns about food safety are rarely out of the media spotlight (p. 254). Verbeke (2005) noted food quality and safety issues have received intensive media coverage in recent years (p. 347). Shaw (2004) noted that food scares such as salmonella and listeria outbreaks, and the incidence of BSE, have cause the public to be more aware of and concerned with food safety. Altheid (1997) noted shared knowledge about the social world in a mass mediated society tends to be about bad news (p. 658). Indeed it seems that the problem frame is indeed employed in relation to agricultural news, specif ically in relation to food safety fears. News Consumption Across the country, news consumption is declining. The Pew Research Center for People and the Press (2008) reported in its Biennial News Consumption Survey that the
34 overall percentage of individuals using traditional news sources (newspapers, magazines, radio, and television) decreased from 90% in 1994 to 73% today. Less than half of the American population reads a newspaper regularly, and that percentage of individuals fell from 40% to 34% between 2006 and 2008. Recently, the web surpassed television news as the most commonly used medium. While the Pew Research Center (Kohut, Doherty, Dimock, & Keeter, 2008) reported that television was most commonly used medium for obtaining news, a more recent st udy indicated that the Internet is now the most essential medium when individuals were given a choice among television, radio, newspapers and the web (Edison, 2010). The Pew Center determined that the amount of time people tend to spend with the news on a given day is just more than one hour, but despite this online news use has increased notably (Kohut et al., 2008). A full 37% of those surveyed by Pew reported getting news online each week, about the same percentage of people who regularly watch cable n ews and more than the percentage of people who watch at least one of the nightly network news broadcasts. Daily use of online news sources has increased, but those who go online spend an average of only nine minutes with the news on a given day (Kohut et al., 2008). Agriculture and the Media One of the oldest uses of mass communication began as an effort to provide information about improving growing and production practices to farmers and agriculturalists in the country following the American Revolution (B oone, Meisenbach, & Tucker, 2000). Agricultural communications, a term often interchanged with agricultural journalism, focused on helping farmers improve the quantity of food they produce through agricultural journals, pamphlets, and bulletins.
35 Agricultu ral societies and organizations initially filled the role of distributing information, creating monthly publications to reach their members (Boone, Meisenbach, & Tucker, 2000; Seevers, Graham, Gamon, & Conklin, 1997). Seaman Knapp, dubbed by some as the f ather of modern day cooperative extension, worked diligently to provide farmers with information to help them become more efficient through moveable schools and demonstrations. Communication about agriculture at that time was primarily about improving tec hniques to provide enough food for the growing population. In 1862, the federal government passed the Morrill Act, designed to provide land for the establishment of colleges to educate people in the agricultural and mechanic arts. Seen as an educational revolution, the idea was to provide education to the everyman. But soon, instructors realized there was a lack of information from which to teach. The establishment of the experiment stations followed with passage of the Hatch Act in 1887 (Seevers et al., 1997). In 1914, the Smith Lever Act was implemente d to coordinate the efforts of experiment stations, agricultural colleges and a new organization called cooperative extension, which was designed to carry information developed at the experiment stations to those not able to attend the land grant colleges (Seevers et al. 1997; Campbell, 1995; Rasmussen, 1989). Over the years, a threepronged approach to carrying agricultural information and research to those who need it was developed teaching, research and outreach. This tri partite mission is carried out today through the cooperation of local (county), state, and federal organizations ( Boone, Meisenbach, & Tucker, 2000).
36 Some decades later communicating agriculture has evolved into a diverse field responsible for developing and disseminating news and marketing information related to food, agricultural, and environmental systems (Tucker, Whaley, & Cano, 2003). In the early days of agricultural communications, documents were limited to books, comme rcial farm papers, magazines and journals but with the expansion of communications also came to embrace radio, television, film, the Internet and other mediated technologies (Zumalt, 2004a). T he field has since expanded from a focus on production of food and fiber and information for those who produce it to a complex enterprise, encompassing research, production, processing, marketing, consumption, nutrition, and health (Zumalt, 2004b). Today, food appears in the news and in popular literature, and is v iewed as a complex interplay of cultural, economic, social, political, and technological forces (Kniazeva & Venkatesh, 2007, p. 420). Food scares and crises such as those mentioned above (outbreaks of E. coli and BSE) have caused the public to have an i ncreased awareness and level of concern about food safety (Shaw, 2004). Powell and Leiss (2004) pointed out that as the distance between consumers and food producers increases, the most likely sources of information on food safety for consumers are mass me dia. The manner in which food safety issues are portrayed in the media impact how members of the public perceive these issues (Ruth, Eubanks, & Telg, 2005). Perception is often reality, and statements by the media play a large role in shaping public perception (Ashlock, Cartmell, & Keleman, 2006, p. 44). Todays Media Landscape In society today, consumers have an abundance of information at their fingertips due primarily to the ubiquity of the Internet. The Internet is a dominant force in our lives,
37 as i s the ability to access it almost anywhere we go. People can shop online, socialize online, do business online, and get their news online. However, as noted above, even with the popularity of the Internet and digital communication technology which makes news more accessible than ever before, fewer people are consuming news, and the manner in which they prefer to receive their news is changing the way news media do business (Kohut et al., 2008). News and New Media According to the Pew Research Center for P eople and the Press (Kohut, et al., 2008), 29% of Americans report having gotten news online yesterday, an increase from 23% in 2006. Edison Research (2010) found that the Internet is now considered the most essential mass medium for consumers. However, f ew people use the Internet as their sole outlet for news and the average online news user spends approximately 35 minutes with online news on a given day (Kohut, et al., 2008). Where do online users go to get their news? A wide variety of websites, portal s and interfaces is the answer. Web portals Yahoo! and MSN are among the most popular, as well as television news organization websites (CNN, MSNBC, NBC, Fox News). Several traditional print publication companies are accessed via their online version, including The New York Times the Wall Street Journal USAToday and The Washington Post (Kohut, et al., 2008). In addition, almost 50% of online news users receive emails, automatic updates or view posts from social networking sites to get news information (Smith, 2010). In a study of users online news seeking preferences, Curtain, Dougall, and Mersey (2007) found that salience of issues was a significant factor motivating online
38 news seeking behavior. They found that web portals tended to attract readers because this platform allows users to follow l inks to mainstream stories while also actively seeking information on particular topics and looking beyond the top layers of a site to find news of interest (p. 33). When given the choice, online news users wanted hard news on topics they considered important, particularly those in the national, political and world arenas (Curtain, Dougall, & Mersey, 2007). According to Pew, 73% of online news users say they encounter news when online for a purpose other than seeking news. Online news habits differ according to age group, with 84% of those in the 1824 year old category grazing the Internet and encountering news this way. Pew found four clear segments of users in its 2008 survey: Integrators (23% of the public), Net Newsers (13%), Traditionalists (46%) and the Disengaged (14%) (Kohut, et al., 2008). Integrators are individuals who use television as their primary news source, but do get news online during a typical day. They also spend the greatest amount of time with the news on a typical day when compared to the other groups. They have greater interest in political news and sports than other groups. Net Newsers tend to read political blogs more regularly than the other groups, use web news at high levels during a typical day, frequently view news online, and are heavy technology users with strong interest in technology news. Traditionalists rely heavily on TV news at all times of the day (morning, day, evening and at night), typically have a computer but do not get news online, have little interest in s cience and technology news, and remain the largest segment of the overall news audience ( Kohut, et al. 2008, p. 2) Disengaged news users is the smallest group of Americans surveyed, and as their name indicates they
39 have a very low level of news consump tion, extremely low interest in and knowledge of current events. Just more than half (55%) of this group gets news on a typical day (Kohut, et al., 2008). News and Millennials According to the Project for Excellence in Journalism (2009), the growth in onl ine news consumption spans generations, but is particularly driven by the younger generation. Media tycoon Rupert Murdoch described a revolution in the way young people access news; they want their news on demandand want control over their media (Murd och, 2005). Termed the Millennial Generation (Howe & Strauss, 2000), 1824 year olds today primarily get their news from the Internet. Termed Net Newsers by the Pew Center, the category of individuals who most frequently point to the web as their mai n source of news are Millennials. More than 75% of this age group identifies themselves as interested in news throughout the day rather than at regular times; Pew has termed them News Grazers (Kohut, et al., 2008). Millennials are collegeaged individuals who have been raised in a society where personal computers, televisions, and other communication technologies have been ever present (Howe & Strauss, 2007). This group uses an array of media technologies and lives in an age where information is literal ly at their fingertips (Carlson, 2007). Despite their technological prowess and their eagerness and ability to quickly adopt innovations (and use online sources to get news), Millennials are less interested in and less likely to read newspapers than their parents generation (Kohut, et al., 2008). While historical research indicates that traditionally younger individuals are less interested in the news than their older counterparts, those in the 1824 year old
40 category today are even less likely than prev ious generations to enjoy news (Kohut, et al., 2008). The largest category of disengaged news consumers, that is, individuals who express an overall lack of interest in the news, are in the 1824 year old category, and the majority of individuals under the age of 25 report getting the news from time to time rather than on a regular basis (Kohut, et al., 2008). In a study of news use among college agricultural majors, this low level of news use was confirmed (Cannon & Lamm, 2010). Gone are the days when the majority of the population turns to the daily broadsheet for the bulk of their news, regardless of age. Fewer than half of Americans still read a newspaper on a regular basis (Kohut, et al., 2008). However, when younger individuals do consume news, they often seek online sources to satisfy their needs. Diddi and LaRose (2006) found that college students predominantly used the campus newspaper, the Internet, and latenight comedians as their news sources. Web portals and news aggregator services are popular for younger online news consumers. Sites such as Yahoo! and MSN are cited frequently, as are the sites of broadcast news organizations like CNN, MSNBC and NBC, and Fox News. Increasingly, social networking sites are used by the younger generation t o get news information (Kohut, et al., 2008). Although they do not consume news regularly, when they do, the Millennial generation primarily uses the Internet to satisfy their desire for news, and are savvy at using a variety of tools and features on Internet sites to meet their needs. New Media With the rise in popularity of the Internet came a new conception of the term new media; media outlets other than the traditional print newspaper, television, and radio
41 that are able deliver information and news to audiences faster than ever before. The term new media is not new itself; Rice and Williams (1984) called new media only relatively new and noted that at one time or another, all communication media were considered new. At the time of their writing, Rice and Williams called the following new media: personal computers, videotext and teletext, interactive cable, videodiscs, electronic mail and computer conferencing, communication satellites, office information systems (p. 16). Our understanding of the conceptual definition of new media is varied. Using one perspective, one might say that the concept of the media is established and understood, while the concept of the new media is less understood (Lister, Dovey, Giddings, Grant, & Kelly, 2003). Today the term new media pertains more to cultural objects and perspectives than to specific technologies. New media are the cultural objects which use digital computer technology for distribution and exhibition, such as the Internet, web sites, computer mul timedia technologies, digital television, and the like (WardripFruin & Monfort, 2003). Again, however, it is evident that what is considered new media technology today will soon no longer be new and will have been replaced by technologies not yet inv ented. Baran and Davis (2003) commented that new media allow us possibly even compel us to extend our senses in new ways (p. 373). These technologies help new media become situated in the collective mind as part of a larger landscape of change, inclu ding social, cultural, and technological change; in short, new media have contributed to a new technoculture (Lister, et al., 2003). This new culture has created vast new possibilities for media experiences, many of which center around the Internet
42 and the web. The Pew Internet and the American Life Project noted that the information ecosystem has changed radically in the Internet era, which has impacted the way people relate to each other, and the way they relate to information. The Internet and mobile t echnologies are at the center of the story of how peoples relationship to news is changing. In todays new multi platform media environment, news is becoming portable, personalized and participatory (Purcell, Rainie, Mitchell, Rosensteil, & Olmstead, 201 0, para. 3). The Pew Center ( Rainie & Horrigan, 2005) reported that the Internet has the effect of bridging gaps among people and functioning as a bonding technology where people believe they are more accessible to friends and colleagues, are able to save time and improve interactions with large entities (such as the government and large corporations) and execute work and personal tasks more efficiently. In essence, the Internet is seen as a technology that improves lives. The Internet has indeed changed t he way people communicate, and the way they access news information. The Pew Center reported a growth in online news use, 37% in 2008, up from 31% in 2006, and more individuals regularly turn to online news sources than watch nightly network broadcast new s shows (37% in 2008, up from 29% in 2006). The use of online newspapers has increased, specifically among the 25 to 34 year old age group. Online news users turn to news feeds, social networking sites, and receive news updates by cell phone, as well as use customizable webpages such as iGoogle Respondents indicated primarily using news portals (such as Yahoo! and MSN), television news organization websites (such as CNN, MSNBC, Fox News), print newspaper organizations (such as The New York Times, The W all Street Journal,
43 USAToday and The Washington Post), as well as search engines and news ranking websites (such as reddit.com and dig.com) to access news information ( Kohut, et al. 2008). Research shows that online audiences increasingly use features tha t allow them the ability to share information and control the news information they are exposed to (Purcell, Rainie, Mitchell, Rosensteil, & Olmstead, 2010). Investigations into the phenomenon of Web 2.0 document the collaborative capabilities of the Int ernet and the sense of community that it can inspire (Birdsall, 2007; Madden & Fox, 2006; OReilly, 2005; Ullrich, Borau, Luo, Tan, Shen & Shen, 2008). Indeed, some of the tools that allow filtering of news and information also promote collaboration and c ommunity. Web 2.0 and Interactivity The term Web 2.0 is attributed to Dale Dogherty and was made popular by OReilly Media (Madden & Fox, 2006), which held the first Web 2.0 conference in 2004 (OReilly, 2005). Researchers note that there is often confus ion when attempting to define Web 2.0 as a concept. OReilly (2005, para. 5) remarked that Web 2.0 doesnt have a hard boundary, but rather a gravitational core which can be envisioned as a set of principles and practices applied to common threads and tendencies observed across many different technologies. Birdsall (2007) called Web 2.0 a state of mind, an attitude, a new business model, the next generation of webbased software and services, a set of development principles, a revolution (para. 1). Madden and Fox (2006) provided a set of defining characteristics of Web 2.0 and noted that these technologies utilize collective intelligence, provide network enabled interactive services, and give users control over their own data. Ankolekar, Krotzsch, Tran, and Vrandecic (2007) noted that the term was originally intended to refer to the state of the art in web
44 engineering (p. 826). One thing is clear, Web 2.0 is a set of technologies, not a new edition of the Internet itself. As a concept and set o f technologies then, Web 2.0 is characterized by collaboration and the participatory role given to users of the web in its development and use (Birdsall, 2007, p. 5). Madden and Fox (2006) noted that a definitive Web 2.0 characteristic is the use of collective intelligence, providing network enabled interactive services, giving users control over their own data (p. 1). Ullrich, et al., (2008) pointed out that Web 2.0 technologies by nature make the most of the structure of the web by encouraging collaboration and participation among users, and are inherently social and open. Commonly used Web 2.0 technologies include podcasts, wikis, blogs, videos, photo sharing services, as well as social networking sites and services (Madden & Fox, 2006). Characteristically, these technologies all allow users to be contributors, and share information with relative ease, illustrating a focus on interactivity (Ankolekar, Krotzsch, Tran, & Vrandecic, 2007) and a shift from a read only web to a read and write web (Ullr ich, Borau, Luo, Tan, Shen and Shen, 2008). Interactivity has been called the primary advantage of new media (Morris & Ogan, 1996) and, as noted by Kiousis (2002), has several varying definitions. In an attempt to provide a cohesive definition for the term Kiousis (2002) reviewed varied definitions in the literature and developed this cohesive explanation: Interactivity can be defined as the degree to which a communication technology can create a mediated environment in which participants can communicate ( oneto one, oneto many, and many to many), both synchronously and asynchronously, and participate in reciprocal message exchanges (thirdorder dependency). With regard to human users, it additionally refers to their ability to perceive the experience as a simulation
45 of interpersonal communication and increase their awareness of telepresence (p. 372). Birdsall (2007) emphasized the common focus on participatory use in Web 2.0, using Wikipedia as an example. One of the most well known and used wikis, Wikipedia, has been called one of the poster children for Web 2.0 technologies (Madden & Fox, 2006, p. 3). It has also been called a radical experiment in trusta profound change in the dynamics of content creation (OReilly, 2005, p. 23). In light of Kiousi ss (2002) comprehensive definition, Wikipedia can be called a truly interactive concept. A great deal of the value of Web 2.0 is the collaborative nature of the technologies, as acknowledged above. However, this value increases as individuals use, shape, and create using Web 2.0 (Ullrich et al., 2008; OReilly, 2005; Birdsall, 2007). Blogs One of the most highly touted features of the Web 2.0 era is the rise of blogging (OReilly, 2005, p. 24). Blogs (an abbreviation of the term weblog) generally cons ist of online comments posted by citizens, groups, and news professionals outside of the normal venues provided by the mainstream news organizations (Reese, Rutigliano, Hyun, & Jeong, 2007, p. 236). Most often, blogs are styled as journal entries, with t he most recent post at the top, with previous posts appearing in a reverse chronological order, and include a number of hyperlinks to other blogs and online content (Carlson, 2007). OReilly (2005) noted that hyperlinks are fundamental to the connectivity of the web, but blogs by their nature illustrate a new level of interconnectedness (Reese, et al., 2007). Reese et al. (2007) outlined fundamental differences between traditional online news sites and blogs and noted that traditional online news sites use the story as the
46 basic element and change and add to the story throughout its life, but that these changes do not accumulate the way blog posts do, nor do these story units usually embed links to information or sources on other sites. This increased connec tivity has led researchers to suggest that blogs have a role in strengthening democracy. Touri (2009) suggested that because blogs provide a platform for interaction between traditional journalists and citizen journalists (those individuals who are not considered professional journalists, having been trained in journalistic writing and news values, etc.) they allow voices of dissent to be heard and possibly impact real world events. Using framing theory to analyze the content of several blogs focusing on w ar and conflict, Touri (2009) determined that blogs have the capacity to shift the power over framing away from the usual sources in the news reporting of political conflict and turn the media system into a greater constraining factor for governments than before (p. 170). Blog content typically contains opinions and commentaries, often in real time, that afford consumers the opportunity to offer their own opinions (Nardi, Schiano, Gumbrecht & Swarts, 2004, p. 30). Blogs thrive on reader interaction (Pohlig, 2003) and provide a kind of online community, allowing voices that are not heard through mainstream press outlets to be discussed on almost equal footing by bloggers. These communities have become so important and powerful in some cases that Schultz and Sheffer (2009) noted that stories which the mainstream media originally ignored and which were covered in the blogosphere eventually appeared in the mainstream news media cycle.
47 Despite the heralding of Web 2.0 technologies and their high level of inter activity, not all evidence illustrates interactivity is a purely positive concept. Eveland and Dunwoody (2002) determined through an experiment that individuals who read online messages learned significantly less than those who read the same messages in pr int. They also learned that those who read the online messages were more likely to selectively scan the messages, leading to this decreased learning. Other researchers have discovered that site features such as ads and design elements have an impact on how credible people find information on websites, as well as the credibility of the websites themselves (Flanagin & Metzger, 2007). According to Tremayne and Dunwoody (2001), the effects of interactivity on users are the subject of debate. Some see the higher level of user activity leading to greater involvement by the user and, subsequently, more significant media effects (p. 111). Liu (2003) commented that it remains uncertain whether interactivity really facilitates persuasion (p. 207). Problem Stateme nt As the preceding literature indicates, news media coverage of issues and events affects public attitudes and perceptions; essentially, news has a societal effect. News media frame the stories and messages they produce, often highlighting conflict and us ing a problem frame to describe and discuss issues of the day. In the past two decades, news stories highlighting problems related to food safety and food risk have increased after well publicized foodrelated crises. Consumers have reported increasing concern over the safety and health of their food, and have moved toward certain food products, such as organic, to alleviate health and safety fears associated with food.
48 At the same time as consumers concern about food safety and health have risen, the w ay members of the public get their news has changed. Todays news consumer turns most often to the Internet and online sources for news information. Online news is now available to audiences, now referred to as users and allows individuals to become involv ed with a story, rather than merely observe it or read about it. Online news consumers can comment on a story appearing on a multinational news organizations website or participate in a discussion forum, send the story to friends or family using email or print functions, can learn more about the subject by clicking on interactive links within the story or links to supplemental information and podcasts about the issue, or can post the story on one of an array of social networking sites, allowing the discuss ion to go beyond the initial news organizations webpage. Many researchers view this new increased level of interactivity present on the Internet as a positive thing, even noting that it can make positive contributions to public discussion in a democratic society. Others suggest that the interactivity of websites can interfere with information people take away after reading a news story, or even the way in which they interpret a message. What researchers have yet to discover is if and how message framing and website interactivity levels impact individuals attitudes toward a topic, and the amount of information individuals recall after reading a news story. Purpose and Objectives The purpose of this study is to examine what effect message frame and website i nteractivity level have on an individuals attitude toward a topic and information recall. The objectives of this study were to:
49 1. Describe the differences in attitude change and information recall between subjects receiving risk framed messages and benefit framed messages 2. Describe the differences in attitude change and information recall of subjects receiving messages on a high interactive website and a low interactive website 3. Explain changes in attitude and information recall based on measured individual at tribute variables ( issue involvement, need for cognition and prior knowledge of the issue )
50 CHAPTER 2 RELEVANT LITERATURE Overview of Literature The previous chapter established the need for this study and outlined the research purpose which is to investi gate the effects of message frame and website interactivity on subjects attitude change and information recall. The following literature review serves to outline the pertinent theories on which this study is based and provide an overview of the concept s of message framing and website interactivity, as well as need for cognition, and issue involvement, and the manner in which they may impact how individuals attitudes change and affect the amount of information individuals recall after exposure to a mess age. Website Interactivity A unique characteristic of new media and the Internet is the concept of interactivity, which has been called the primary advantage of new media (Morris & Ogan, 1996). Accessing the Internet, users (rather than audience members) actively make choices about the information they wish to access and view, becoming involved users of information rather than more passive audience members of mass media (Chung & Yoo, 2008). Inherently then, the Internet demands a higher level of interactiv ity from its users (Ruggiero, 2000). A variety of definitions of the term interactivity are used among academic researchers. Kiousis (2002) considered interactivity both a psychological factor and a media factor, and noted that the concept of interactivity varies across different communication technologies, contexts and peoples perceptions, and noted the
51 standard for what makes one medium more interactive than another is quite ambiguous (p. 356). Nevertheless, Kiousis provided his definition of interacti vity as: the degree to which a communication technology can create a mediated environment in which participants can communicate (oneto one, one to many, and many to many), both synchronously and asynchronously, and participate in reciprocal messages exchanges With regard to human users, it additionally refers to their ability to perceive the experience as a simulation of interpersonal communication and increase their awareness of telepresence (p. 372). In using this definition, Kiousis clarified that the term communication technology can refer to any technology from a telephone to a computer system, and a mediated environment can mean anything from a telephone wire to virtual reality (Kiousis, 2002). According to Chung & Yoo (2008) a useful way to conceptualize web interactivity is to categorize it into two groups. The first way is to see interactivity as between two or more human beings, and the second is to consider interactivity as the interaction between humans and the technology or medium itself. StromerGalley (2000) commented that human interactivity on the web more closely resembles faceto face interaction that one might have off line, and is therefore more genuinely interactive. Medium interactivity, conversely, indicates interaction between users and the technology or medium and pertains to what actions or tasks the technology allows a user to perform, such as using hyperlinks. Stromer Galley (2000) views these tasks as less genuinely interactive. Under this view, activities such as instant message chats, video and audio conferencing, and other activities where a user engages another human user are considered human interactivity features. In contrast, Chung and Yoo (2008) noted that medium interactivity indicates functions where a user inter acts with the medium itself and utilizes the technology to make choices and exert control over the communication process performing activities
52 such as clinking on hyperlinks, entering comments on a website, and so forth (p. 378). Massey and Levy (1999) w rote about interactive journalism in two realms, similarly to StromerGalley. They divided interactivity into content interactivity, the degree to which journalists technologically empower consumers over content and interpersonal interactivity which they termed the extent to which news audiences can have computer mediated conversations through journalists technological largess (Massey & Levy, 1999, p. 140). Chung and Yoo (2008) further dissected the concept of interactivity and suggested evaluating it as a multidimensional concept with three categories on a continuum: medium, human/medium, and human interactivity. The progression from medium interactivity through human interactivity is believed to be exercised through the use of various interactive features which can be categorized into three progressive levels. Medium interactive features solely rely on the technology to allow users to exert control (p. 378) Several researchers have focused their exploration on medium interactivity and what users l earn or gather from interacting with a site and its features. Chung and Zhao (2004) determined that website options which allowed users to interact with other humans were used less often when compared to use of medium interactive features such as audio and video downloading of files, viewing multimedia galleries and search functions. Additionally, Chung and Yoo (2008) investigated the ways in which online audiences were using interactive website features of online newspaper sites and found that users primar ily were interested in features that allowed them to obtain information and entertainment rather than those that facilitated socializing with others.
53 Outing (1998) proposed a list of elements he believed should be included on an ideal interactive websit e: discussion forums, live chats, reporter email addresses, some kind of article feedback mechanism, an option for individuals to create personal web pages hosted by the news site, special areas where users can create hobby sites, a portion of the site where births, marriages and deaths can be recorded by users, community group pages, a place where users can comment and criticize content and issues presented on the site, user polls, and actually using online comments as reporting tools. Including these elem ents, he asserted, would help news websites become truly interactive in nature (Outing, 1998). Sicilia, Ruiz and Munuera (2005) used the elaboration likelihood model of persuasion to study how consumers processed information from two types of websites, one interactive and one noninteractive. They determined that individuals who were exposed to an interactive site processed the information more thoroughly than those who received the information via the noninteractive site. Additionally, the results of the study indicated that as interactivity on a site increased, so did the level of information processing. Ariely (2000) determined that when individuals have control over the flow of information, they are better able to integrate and remember the information. Heeter (1989) pointed out that what an individual sees when viewing a website may be different than what another individual sees viewing the same website due to the level of interactivity, and so developed a multidimensional model of interactivity. The model includes six dimensions, the first of which, complexity of choice available represents the range of content available to the user.
54 Ramirez and Burgoon (2004) wrote about the concept of modality richness, which refers to the visual, audio, verbal and contextual information sources a medium provides. They found that this modality richness was central to producing heightened experiences of interactivity (p. 439) and increased richness enhanced interaction involvement when positive information was re ceived (p. 440) rather than negative information. Tremayne (2008) investigated how interactivity of a website affected participants learning, where learning was defined as the process of incorporating information into memory by comparing the new informa tion to preexisting schemata and establishing connections between the new and the old (p. 706). Results provided some support for the idea that information presented in hypertext (the studys representation of interactivity) improves learning among readers. Sundar, Xu, and Bellur (2010) noted a scientific understanding of the psychological effects of interactivity is quite critical for a society that is becoming saturated with interactive digital media (p. 2255). The variety of definitions, both concep tual and operational, of the term interactivity mean that challenges can arise when conducting research. The researchers suggested a framework for those wishing to explore the ways in which interactive tools in modern media contribute to user engagement wi th content, and emphasized the importance of a broad theoretical framework in future research (Sundar, Xu, & Bellur, 2010). The variety of definitions and conceptualizations of interactivity illustrate that research in this field is still in its youth.
55 S ome research has indicated that the effects of interactive websites are not without downsides. Tremayne and Dunwoody (2001) noted that while interactivity may lead to greater user involvement, it might also trigger more significant media effects (p. 111) Eveland and Dunwoody (2002) explored differences between online and print delivery of messages on how much subjects learned and how they processed information. Results illustrated that subjects exposed to the online version of the messages elaborated more on the message, which was positively related to the amount learned, however subjects receiving the online messages learned significantly less when compared to those having received the print version of the messages. Framing Theory Framing research, li ke many theories of mass communication effects, has a relatively short history. However, message framing has been studied in numerous contexts, most notably within mass mediated communications. McQuail (1994) noted that the mass communication field of st udy is based entirely on the notion that the media have significant effects on audiences. The way information is structured in mass media (ultimately, how messages are framed) affects the way that information is processed in an individuals mind, and indiv iduals schemas interact with the information to ultimately derive meaning from the message (Reese, 2003). When framing messages, writers use interpretive packages with metaphors, catchphrases, visual images, moral appeals, and other symbolic devices (Gamson & Modigliani, 1989, p. 2) as central organizing ideas to help audiences (and the writers themselves) make sense of the messages (Reese, 2003). According to Reese (2003), Goffman (1974) and Bateson (1972) are credited as the founders of framing resear ch. Both of these researchers, and subsequent
56 academics since their time, have investigated framing from a social psychological perspective, concluding that individuals utilize frames to make sense of their everyday social experiences. In Frame Analysis, G offman (1974) wrote that humans actively classify, organize, and interpret life experiences in order to make sense of them and that these schemata of interpretation, which are essentially frames, help us locate, perceive, identify, and label information for better understanding (p. 21). Because of its diversity of applications, researchers have developed a number of definitions that attempt to help explain framing theory and what it means to frame a message. A prominent definition of framing comes from E ntman (1993) who wrote about framing theory as a fractured paradigm that he believed suffered from a variety of explanations and no definitive and uniform definition. Consequently, he offered what has since become one of the most commonly used definitions of framing. Entman (1993) states to frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or t reatment recommendation (p. 52, emphasis in the original). Reese (2001) expanded on Entmans (1993) definition of framing and incorporated Gamson and Modiglianis (1989) concept of frames as organizing principles and called them organizing principles that are social shared and persistent over time, that work symbolically to meaningfully structure the social world (p. 11) According to Hertog and McLeod (2001), message framing structures our understanding in a number of significant ways, including helping us to determine what content is relevant to discussions surrounding issues of social concern, defining roles
57 that groups play in relation to the issue, outlining ways in which various values are created, and outlining goals inherent in the structuring of a content area. Gamson and Modigliani (1989) explored framing from a sociological perspective, noting that individuals construct their understanding of issues by using their own previous experiences and knowledge about a subject, and combining those elements with what the media say about it to form an opinion. Entman (1993) brought up the idea that frames can be found in four different locations in the communication process. The first location frames can be present is within the communicator as message creator. Communicators crafting messages are guided by schemas or mental organizing system present in their own minds/experiences and from these make certain judgments about what information to include or exclude in a message. Frames can also be located in the text of a message, where certain words, phrases, images, sources of information and sentence structures provide indications of the way messages should be read and understood. Receiver frames, the third location, are frames that guide the thinking of mes sage receivers, which are not always in line with the frames located in the text itself, nor the intentions of the communicators frames. The fourth location of frames, according to Entman, is culture, where culture might be defined as the empirically demonstrable set of common frames exhibited in the discourse and thinking of most people in a social grouping (p. 53). These four framing locations function similarly to highlight certain message elements and construct a given argument about the subject of the message. Scheufele (1999), similarly to Entman, wrote about the diverging areas of framing research and proposed a model to guide further research in the field. He reasserts
58 Entmans concerns about the fractured nature of framing research and this potentially weakening the theory. Scheufeles model divides framing research into two main areas media frames and audience frames. The model allows for the investigation of four separate kinds of framing research: the building or creation of frames, frame setting, exploring the effects of framing on individuals or audience members, and investigating the link between media frames and audience frames Scheufele notes that frames need to be considered as structures for both presenting news and understanding the information presented. Media Frames As structures that help communicators create meaningful and understandable messages, Gamson (1996) called framing a process by which individuals or organizations use central organizing ideas to help a reader/audience m ember make sense of a complex issue. Journalists, therefore, play a key role in constructing media frames. Frames enable journalists to quickly and efficiently categorize information and package it in a way that can readily be understood by audience members (Gitlin, 1980). Dunwoody (1992) referred to frames as a kind of set of mental maps that are central to journalists work in helping to explain issues. Hackett (1984) noted, however, that journalists do not necessarily consciously apply or construct fram es, but use what they know about an issue and their assumptions to create stories/messages. Regardless of the manner in which frames are created by communicators/journalists, the way an issue is written about has consequences for how audience members perc eive that issue (Tversky & Kahneman, 1981; Price, Tewksbury & Powers (1997). Media frames are an invitation for audience members to interpret a message about an issue in a certain way, which an audience member is free to either accept or
59 reject (Gamson, C roteau, Hoynes, & Sasson, 1992; Reese, 2003). Hallahan (1999) and Bateson (1972) both commented that frames include and/or exclude certain information, which also indicates to audiences how a message should be interpreted. Rather than specifically advocati ng a point of view, such as pro or con, on an issue, frames suggest how audience members should think about an issue. Schudson (2003) called framing as central a concept as there is in the study of news and noted that the concept of framing has generall y overtaken the study of media bias when examining news content (p. 35). He used Gitlins (1980) definition which called frames principles of selection, emphasis, and presentation composed of little tacit theories about what exists, what happens, and what matters (p. 6). News professionals simplify issues and problems in order to explain them to audiences, and in doing so tend to emphasize conflict and melodrama due in part to the traditional journalistic notion that there are always two sides to a stor y or issue, even if the issue is not inherently driven by conflict. This emphasis on framing illustrates that decisions made in regard to what appears in the news has more to do with the marketplace, the nature of organizations, and the assumptions of new s professionals than with individual bias (p. 47). In discussing the framing of news, Tankard (2003) noted that the framing process has the ability to eliminate certain voices and affect the strength or weakness of arguments. Media can frame issues in w ays that favor a particular side without showing an explicit bias (Tankard, 2003, p. 96). Audience Frames On the converse side of Scheufeles (1999) model of framing research sits the concept of audience frames, also called individual frames. Entman (1993) defined
60 individual frames as mentally stored clusters of ideas that guide individuals processing of information (p. 53). These clusters of information, sometimes referred to as schemas or schematas, are constructed using previous experiences and knowl edge, popular wisdom, and media dialog to help individuals understand an issue (Pan & Kosicki, 2003). Goffman (1974) wrote that these schemata of interpretation, allow individuals to identify and categorize issues and experiences, helping them to underst and and make sense of things. Because all of these elements go into the way in which individuals interpret issues and events, the way that an issue or problem is characterized when presented in the media has an impact on how individuals respond to and interpret the issue or problem (Reese, 2003). What we know about the nature of the social world depends upon how we frame and interpret the cues we receive about that world (Edelman, 1993, p. 231). Reese (2003) commented that frames invite us to think about social phenomena in a certain way, often appealing to basic psychological biases (Reese, 2003, p. 12). However, no guarantee exists that individuals will choose to interpret the message in the invited manner (Gamson, et al., 1992; Reese, 2003). Indi viduals do not slavishly follow the framing of issues presented in the mass media (Shah, Domke, & Wackman, 2003, p. 229). Rather, people actively filter, sort, and reorganize information in personally meaningful ways in constructing an understanding of public issues (Neuman, Just, & Crigler, 1992, p. 7677). Audience frames, then, interact with message frames to create individual understanding (Reese, 2003). Shah et al. (2003) noted that when individuals are faced with information, they are believed to immediately locate schemas to help guide their
61 information processing and understanding. There is little doubt then that the way information is structured affects cognitive processing (Reese, 2003, p. 9). Several framing researcher acknowledge that frames interact with and can trigger certain schemas with audience members. Price, Tewksbury, and Powers (1995; 1997) determined that issue frames significantly impacted cognitive responses among participants. As noted previously, Gamson and Modigligani (1989) referred to frames as schemata of interpretation. Scheufele (1999) asserted that an individuals information processing function and interpretation of messages are influenced by schema. Fredin (2003) noted: The frame as a mental construct is generally r eferred to as a schema in the psychological literature. As is the case with news frames, schemas are organizations of knowledge, ideas, values, and affect, though they are sometimes treated as being composed of knowledge only. Schemas function as efficient ways to quickly interpret events in daily life[and] fill in gaps in available information (p. 270). Related to schema is the concept of priming. According to Fredin (2003) people are highly susceptible to priming in that they can be influenced to look at an event with a particular point of view in mind (p. 271). Studies in media priming generally refer to the effects of media content on an individuals subsequent behavior related to the content processed and have often focused on the priming effect of m edia violence and political priming (Roskos Ewoldsen, Roskos Ewoldsen, & Carpentier, 2009). In contrast, priming research in the psychological tradition focuses on priming as activating an individuals schema regarding a certain object or issue, and the id eas activated in an individuals neural network as a result of that priming (Eagly & Chaiken, 1993). Priming in relation to attitude will be discussed later in this chapter.
62 Frames and New Media Pan and Kosicki (2003) noted that new media blur the lines between communicators and audiences, news and entertainment, information and opinion. The Internet has created a new culture, where people are no longer subject to a few individuals or organizations dispensing information, but where individuals are now gett ing information from any number of individuals or organizations, and people are able to participate in controlling the news they are exposed to through the use of technology (Katz, 1999). The invention of hyperlinks and the graphical browser forever chang ed the way in which information is communicated and illustrates the profound influence of computers on human lives (Engel, 2003). New media, especially the Internet, expand the possibilities of what can be included on a newspaper or magazine page, and have had an impact on the rhetoric available to audiences, previously only accessible through mainstream media sources (Engel, 2003). Pavlik (2003) asserted that web technology and tools allow storytellers, such as journalists, to create much more engaging, n avigable, contextualized reports that tell the days events more accurately, fully, and dynamically than before, and that these news reports are less bound by the typical onedimensional, episodic frame storytelling used traditionally (p. 312). Pavlik ( 2003) went on to note the significance of hypermedia with respect to framing news stories, because it enables the journalist to provide links between stories and related online content that can provide additional background, detail, and most importantly, context (p. 316). Touri (2009) investigated the ability of blogs to influence dominant media forces by utilizing voices and perspectives distinct from the popular press in the context of war
63 and conflict. Touri determined that blogs were indeed able to impact the framing of conflict by mainstream media by reusing material and producing new and different frames than those of mainstream media outlets. This, Touri noted, effectively reduced the complete control over the framing power of wars and conflicts o n the part of the mainstream media, generating a more public and democratic landscape, which speaks to the power of new media (p. 180). Hypermedia such as is found on the web, can enable journalists to present stories that reach beyond the traditionally narrow frames available for use in print and broadcast media, and by doing so, allow news consumers to consider a story or issue from multiple perspectives, enabling individuals to see and understand events in a more complete context (Pavilik, 2003, p. 318). Zillmann, Chen, Knobloch, and Callison (2004) studied framing effects of Internet news magazines and manipulated the subheadings of articles seen on an overview screen, asserting that readers of such newsmagazines make choices about what to read based on the information provided in headlines and subheads which are selected by editors. Subjects in the experiment were exposed to one of the following frame conditions: conflict, disaster, emotional/agony, economic or a control frame. Interestingly, the researchers reported that subjects chose to spend more time reading articles with conflict and agony frames. The researchers noted that, given these results, the articles containing leads in the other framing conditions were deprived of a chance to recommend t hemselves through features such as layout, images, excerpted quotes, or text segments of dramatic quality. In filing through a print magazine, in contrast, these features are allowed to exert their effect and thereby foster selective reading of these artic les for reasons other than their [framed] leads (p. 76).
64 Framing Science Several researchers have studied framing theory in connection with a variety of scientific issues, from human health, to new scientific technologies, to agricultural issues. Nelkin ( 1987) proclaimed that framing research has focused on the medias portrayal of scientific issues and processes (Nelkin, 1987, p. 223). Meyerowitz and Chaiken (1987) hypothesized that stressing the negative consequences of not performing breast self exa mination would be more persuasive than emphasizing the positive consequences in a message geared toward college aged female subjects in keeping with research by Fiske (1980) which found that individuals tend to give more weight to negative information than positive information, resulting in a negativity bias. Meyerowitz and Chaiken (1987) found that subjects exposed to messages framed in loss language (i.e. negative consequences of not performing breast self examination) had more positive attitudes toward the self examination techniques than those exposed to gainframed messages. The implication of this study was, similarly to Kahneman and Tversky (1979), that framing messages as a loss maximizes risk seeking behavior (Meyerowitz & Chaiken, 1987). In anot her study related to human health, Lawrence (2004) investigated the framing of obesity in news coverage between 1985 and 2003. The study indicated that framing of obesity shifted over the time period studied. Initially, frames were focused on the individual, medical and biological origins of obesity, and then moved to a focus on personal behavior, and then to the concept of environmental causation (Lawrence, 2004). Cobb (2005) and Scheufele and Lewenstein (2005) explored framing of nanotechnology in media and its impact on public opinion. Cobb (2005) found that
65 successful frames, those that impacted public opinion, were those that emphasized the risks of nanotechnology rather than the benefits, but that framing effects were not significant enough to alter public opinion. Scheufele and Lewenstein (2005) discovered that heuristics (cognitive shortcuts) presented by mass media factor in to the way the public thinks about nanotechnology, its risks and benefits, and whether as a useful scientific concept it des erves further research funding. Another controversial subject that has been the focus of fervent media attention and numerous framing studies is biotechnology. Lundy and Irani (2004) explored news framing of biotechnology in U.S. and U.K. national daily newspapers and discovered that biotechnology was framed in a number of ways, including as a potential food supply contaminant, a risk to human health, a risk to the environment, a representation of scientific progress, and a potential solution to world hunger issues. No single frame dominated coverage in either the U.S. or the U.K. newspapers, although the issue of biotechnology itself was more prominent on the medias agenda in Great Britain, possibly, the authors note, due to corresponding legislation in Britain during the year (Lundy & Irani, 2004). Marks, Kalaitzandonakes, Wilkins and Zakharova (2007) examined media coverage of biotechnology over a 12year period in the United States and the United Kingdom. They found that two types of biotechnologies were framed consistently in different manners medical biotechnology was framed more positively, and agricultural biotechnology was framed more negatively. Crawley (2007) also investigated new framing of biotechnology, and used a sample of local newspapers from California and Missouri. Results of the content analysis indicated that the frames appearing in these sources were more diverse and
66 complex than those appearing in analyses of national publications, sometimes representing a wider range of perspect ives. In a study utilizing international elite newspapers, Priest and Ten Eyck (2003) analyzed news coverage of biotechnology in 14 European countries, Canada, and the U.S. To conduct the study, the researchers developed eight frames to help categorize articles progress, economics, ethics, Pandoras Box the idea that if the technology is used it will only bring bad/evil, runaway technology, nature/nurture, and globalization. The progress frame was used most often across all 16 countries, but differed in the way it was characterized, appearing in articles pertaining to medical issues as opposed to articles related to agriculture and food. Media Framing of Agriculture and Food Several studies have been conducted analyzing the framing of agricultural iss ues in the news media, predominantly focusing on animal diseases related to human health, food safety, and food products, specifically organic foods. Nerlich, Hamilton and Rowe (2002) investigated the use of frames and metaphors in news coverage of the 2001 outbreak of foot and mouth disease (FMD) in the United Kingdom. The researchers determined that the central frame used in the coverage analyzed was one of the nation fighting a war against a disease, with an overwhelming use of images depicting the death and destruction that accompanied the British governments disease eradication plan which included the wholesale destruction of animals infected by the disease, as well as those potentially infected. A central image found throughout the coverage was that of fighting an enemy (FMD) in a war, which linked FMD to deeply rooted cultural concepts of death and of plaguefilled medieval villages, but also to mythical and religious images of heaven and hell, of
67 sacrifice and redemption (Nerlich, Hamilton, & Rowe, 2002, p. 96). Overall, the disease itself (FMD) was depicted as the villain, the farmers facing the destruction and devastation to their herds as the victims, and the governments desire to be the hero in the situation. Also analyzing news media frami ng of the 2001 FMD outbreak in Britain, Cannon and Irani (in press) found that coverage of the outbreak utilized three primary frames, including fear, a comparison of FMD to BSE (mad cow disease), and the economic impact of the outbreak in Britain. Shih, Wijaya, and Brossard (2008) studied the news framing of three animal diseases with implications to human health avian flu, mad cow disease, and West Nile virus. The researchers examined print media coverage of these public health epidemics, and noted that in each instance, coverage was event oriented and tended to surround new outbreaks, updates on infected cases and actions taken by authorities were the staple of coverage topics, which was different from the kinds of coverage observed when environmental issues were involved (Shih, Wijaya, & Brossard, 2008). The researchers also noted that coverage of these epidemics tended to increasingly focus on conflict as the issue became more political in nature. Studies by Ruth, Eubanks, and Telg (2005) and Ashlock, Cartmell, and Kelemen (2006) focused on analysis of news frames surrounding coverage of bovine spongiform encephalopathy (BSE, or mad cow disease). Ruth, Eubanks, and Telg (2005) evaluated frames in news coverage of the outbreak of BSE in Canada in May 2003 in newspapers from both Canada and the U.S. Overall, coverage of the outbreak was predictably negative, and coverage of the issue was more frequent in the Canadian
68 publications than in the U.S. publications. The majority of the U.S. coverage utilized a frame about the risk to human health when reporting the issue, often including misreports of BSE being the cause of the human variant CreutzfeldJakob disease (vCJD). In addition, U.S. coverage often used an economic calamity frame, highlighting the ov erall impact of the outbreak on the beef industry in Canada. Overall, implications of the coverage were of the diseases devastation of the cattle industry, and a threat to individuals who consume beef. The researchers noted that the use of the health ris k frame greatly outweighed the actual risk to human health, and that there was an overall emphasis on the role of government in handling the outbreak. In an investigation of the frames used during the coverage of the BSE outbreak in a U.S. herd in December 2003, Ashlock, Cartmell, and Kelemen (2006) used the frames identified by Ruth, Eubanks, and Telg (2005) and found industry crisis to be dominant. Again, coverage of the outbreak was negative overall, this time specifically with respect toward the beef industry in the U.S. The researchers attributed some of the different focus of frames between their results and those of Ruth, Eubanks, and Telg (2005) to the fact that during this outbreak, the threat was not from imported beef from Canada, rather it was the U.S. having an outbreak of BSE of its own, a much more imminent threat. Irlbeck and Akers (2009) studied the news frames in broadcast coverage of the outbreak of salmonella bacteria in salsa during the summer of 2008. Most often, broadcast media uti lized a frame critical of the governments response to the outbreak and coverage honed in on the fact that the true source of the outbreak was undetermined until late in the period of coverage. In addition, Irlbeck (2009) determined
69 that, contrary to the findings of Ashlock et al., (2006) and Ruth, et al., (2005), agriculture was primarily framed in a positive light, with only three stories out of the total 71 considered negative toward agriculturalists. Framing Organics Meyers and Abrams ( in press) analyz ed media framing of organic foods in five national newspapers during an 18month period during which Whole Foods Market, the leading retail outlet for natural and organic foods in the United States, achieved its status as a Fortune 500 Company. The resear chers determined that four frames were utilized throughout the coverage period: an ethical frame, a health frame, a frame that focused on methods of food production, and a frame that centered on the industrialization of the organic market. Meyers and Abra ms noted that an emphasis on the ethical and moral reasons for purchasing organic food dominated, while there appeared to be little coverage of scientific evidence for assertions that organic foods are of higher quality, are safer, and more nutritious than nonorganic foods. The health frame characterized organic food as healthier, safer, and of higher quality than those produced conventionally and painted it as an overall superior choice. Lockie (2006) analyzed newspaper coverage of sustainable food and agriculture in Australia, the United Kingdom, and the United States between 1996 and 2002. Throughout the coverage, Lockie noted the media discourse tended to be reduced to a polarization between organic foods at one end of the spectrum and conventional f oods at the other. This narrative produces discourses in which organic foods are seen as more or less synonymous with safety, naturalness and nutrition, and their alternatives as artificial, threatening, and untrustworthy (p. 313). Organic foods overall were
70 portrayed as trustworthy alternatives to foodrelated problems such as genetic engineering, food scares, pesticide residues, and regulatory failures. Framing Risk In a review of research studies that focus on the framing of risk information, Kuhberger (1998) pointed out that the notion of risk is operationalized in different ways in the literature base. He points out that some researchers approach the concept of framing information about risk from the perspective of positive versus negative frames, so me from a gain versus loss perspective. Framing of risks is even further fragmented by the amount and degree of risk presented in frames; some researchers employ multiple risk events while others focus on manipulating the degree of risk or the quality of t he risk (Kuhberger, 1998). Tversky & Kahneman (1979, 1982) are credited with the development of prospect theory, which asserts that framing choices between options which are framed positively and described in terms of the benefits which will result, and choices which are considered risky, or framed negatively and presented in the context of the loss of certain benefits, generally leads people to be risk averse. That is, when exposed to message frames emphasizing a benefit over a risk (or a gain rather than a loss), individuals tend to make choices that ensure the benefit over the risk. Iyengar (1991) studied the use of news frames and determined that the use of different frames triggers different cognitive functions in the human mind, which in turn affect in dividuals risk perceptions. Also in the area of cognitive activity in relation to framing of risks, Gonzalez, Dana, Joshino, and Just (2005) conducted a study to investigate the cognitive effort involved with making decisions between risks/losses and bene fits/gains. The researchers asked subjects to respond to problems framed as gains
71 or losses and measured the amount of brain activity using a functional magnetic resonance imaging machine. Results of the study indicated that the amount of cognitive effort expended when subjects made a decision that ensured a gain (rather than a loss) was considerably lower than the amount of mental effort expended when selecting a risky gain, or a risk option. Gonzalez et al. determined that individuals tend to try to deter mine the best decisions for themselves by using the least amount of cognitive effort, echoing the concept of cognitive miserdom established by Fiske and Taylor (1984). People appear to exhibit a general tendency to be risk seeking when confronted with neg atively framed problems and risk averse when presented with positively framed problems (Gonzalez et al., 2005). While certainly not all risks are perceived the same way by all people, Fischer, Morgan, Fischoff, Nair and Lave (1991) were interested to det ermine what risks most generally of concern among members of society. They determined that health, safety and environmental risks were of the greatest concern. In the study, female subjects and those who were collegeaged were more concerned about environm ental risks, while males and older subjects were more concerned with health and safety risks (Fischer, et al., 1991). At a basic level, the risks that concern us most are those that threaten our individual health. According to Durfee (2006), the way the m edia frame a story about environmental issues has the potential to influence the audiences perception of risk (p. 459). In an investigation of the way in which frames impacted individuals risk perceptions of newspaper reporting on environmental health ri sks, Durfee contrasted social change frames, which are intended to inspire long term thinking about the impacts of potential
72 environmental risks and suggest alternative viewpoints and solutions, with status quo frames, focusing on the current way of addres sing the risk. The use of social change frames was associated with greater risk perception, illustrating that using the same set of facts and shifting the news angle or focus by selecting a different set of quotes and sources can significantly alter the m essage received and processed by the audience (Durfee, 2006, p. 487). Risk Communication The concept of risk communication focuses primarily on issues that the public perceives as potentially threatening. According to Leiss and Powell (2004) risk is the probability of harm in any given situation...determined by two factors: a) the nature of a hazard, and b) the extent of anyones exposure to that hazard (p. 33). Overall risk related to a particular issue, then, is the combination of the hazard and an individuals exposure to it (Leiss & Powell, 2004). Our constantly changing environment and the perils of disease, war, and other threats to the survival of the human race illustrate that risk is inherent to the human condition (Plough & Krimsky, 1987, p. 5). Plough and Krimsky (1987) stated the emergence of risk communication as a research theme cannot be fully appreciated or accounted for without understanding its link to a set of issues that symbolize the discord between scientific experts and the publ ic around the issue of risk (p. 5). One of the major challenges of communicating risk is that the public sees and evaluates risk in a decidedly different way than scientists and experts (Leiss & Powell, 2004; Lofstedt, 2006; Macfarlane, 2002) and differently from one another as individuals (Leiss & Powell, 2004). In addition, language can get in the way. Communication challenges arise surrounding the description and explanation
73 of risks due primarily to the different language used by scientists, which te nds to be scientific and statistical, and the public lexicon (Leiss & Powell, 2004). In theory, information and messages about risk are communicated by mass media in a way that is intended to help the public understand these issues (Goodman & Goodman, 200 6; Frewer, Miles, & Marsh, 2002; McCarthy, Brennan, DeBoer & Ritson, 2008). However, agreement on how risks should be portrayed can be challenging. Science and policy experts sometimes view mass media coverage of health and environmental risks as oversimplified, inaccurate, and sensationalized (Lundgren & McMakin, 1998, p. 237). In addition, Leiss and Powell (2004) pointed out that some risks are more feared by the public than others, and experts seem to not understand the publics desire for more certain ty pertaining to these particular threats. Researchers have discovered, however, that risk situations that are more easily understood by the public are less likely to inspire fear or dread (Frewer, 1999; Pidgeon & Beattie, 1998). Some researchers have sug gested that when evaluating risk information, the public may face a kind of information overload that can result in a lack of integration of the information (Heller, 2006). McCarthy, Brennan, De Boer, and Ritson (2008) asserted that a risk messages co ntent, tone of its delivery, the media format used, the expert sources referred to, and the accuracy of the information communicated can all play a significant role in determining whether the public attends to, understands andacts upon risk information ( p. 377). Mass media are the publics primary source of science information (Leiss & Powell, 2004; Nelkin, 1987; Treise & Weigold, 2004), risk information (Kitzinger, 1999), and about food safety (Fisher & Chen, 1996, Pisano & Woods, 2002 as cited in Tucker
74 Whaley, & Sharp, 2006). U.S. media outlets have been shown to focus their attention disproportionately on hazards and risks that are new, cataclysmic and violent (Adams, 1986; Combs & Slovic, 1979; Lundgren & McMakin, 1998; Singer & Endreny, 1993). While mass media may not determine how people think about a particular issue, over time they may be key information sources that influence peoples knowledge and perceptions about science and technology issues (Powell, Dunwoody, Griffin, & Neuwirth, 2007). Soc ial Amplification of Risk Peoples risk perceptions are greatly affected by their previous life experiences (Tucker, Whaley, & Sharp, 2004) and the experiences of others. One theory which examines risk and communication is Kasperson, et al.s (1988) social amplification of risk framework, which posits that when a lack of direct personal experience exists about a risk, information about that risk (called a risk event) is obtained by individuals through two channels the news media and informal personal netw orks. Kasperson et al. (1988) stated risk events interact with psychological, social, and cultural processes in ways that can heighten or attenuate public perceptions of risk and related behavior (p. 178179). Even risks that pose minor physical threats can evoke strong concerns among the public (Kasperson, et al., 1988). Social amplification of risk is a dynamic process that takes into account the learning and social interactions that one has from experiences with risk; Kasperson et al. (1988) noted var ious groups present competing evidence based upon their own perceptions and social agenda (p. 178). Food Risk Eating has become a risky business according to Caplan (2000, p. 34). Growing concerns about food safety and the food supply are well document ed (Green, Draper, &
75 Dowler, 2003; Hansen, Holm, Frewer, Robinson, & Sandoe, 2003; Lofstedt, 2006; Shaw, 2004; Verbeke, 2005). Green, Draper, and Dowler (2003) pointed out that food risks are now inherent with our everyday lives and unfortunately are hidden within the ordinary and even wholesome goods and services we rely upon (p. 34). Tucker, Whaley, and Sharp (2006) determined that concerns over food safety, often driven by media coverage of specific issues, resulted in elevated levels of perceived r isk among subjects. The researchers found that pesticide residues in food and contamination of drinking water were the issues where subjects perceived the highest levels of risk. In this study, the researchers expected that an increased reliance on mass m edia for food safety information would result in lower levels of perceived risk, but actually found the opposite the increased reliance on food safety information tended to result in higher levels of perceived risk. Verbeke (2005) noted that the intensiv e mass media coverage of food quality and safety issues in recent years has led consumers to alter their beliefs, attitudes, and behaviors regarding purchasing and consuming food. Some believe that this wealth of information helps consumers to make safe a nd healthy decisions regarding their food choices. However, Verbeke noted that there is potential for consumers to be inundated with too much information, or the wrong kind of information that does not meet their specific information needs. Too much inform ation or information of the wrong kind can affect the way individuals process that information and the way in which they form perceptions about a situation or issue. Smith and Petty (1996) pointed out that negatively framed information might receive different weight in [individuals] judgments than positively framed information (p. 258). These researchers also noticed that
76 negative frames might also arouse more fear than the messages that focus on gains from engaging in a behavior and thus might be more persuasive (Smith & Petty, 1996, p. 258). Frewer, Howard, Hedderly and Shepperd (1997) investigated the impact of food risk information and discovered that low source credibility reduced risk perceptions, but also that if people believe that the informat ion is directed towards a vulnerable other person, rather than the self, they are unlikely to pay attention to the risk information itself (p. 768). Persuasion Perloff (2003) called persuasion a symbolic process in which communicators try to convince oth er people to change their attitudes or behavior regarding an issue through the transmission of a message, in an atmosphere of free choice (p. 8). Dillard and Pfau (2002) noted that there are almost endless amounts of ways in which persuasion can and does impact contemporary society, from the way consumers decide to spend their money, to influencing political decisions. All persuasive communications share five distinct features: a specific goal or issue topic, a source or communicator, a message, a channel to communicate that message, and a message recipient (Shavitt & Brock, 1994, p. 220). Early communication researchers assumed that mass media had a direct effect on audience members and initial media effects theories posited that media were so powerful me re exposure to messages could produce direct effects on individuals, persuading them to hold one position or another (Baran & Davis, 2009). Subsequent mass media scholars determined, however, that the influence of media messages was not as direct as they had once been assumed, and that a number of variables factor in to whether or not
77 individuals are affected and persuaded by media messages (Petty, Priester, & Briol, 2002). Elaboration Likelihood Model Persuasion by others messages can be more or less effective, depending on various features of the message source, the message itself, the recipient, andthe cognitive and motivational context in which the message is received (Eagly & Chaiken, 1993, p. 499). An extensively used approach to understanding t he persuasiveness of media messages is the elaboration likelihood model (ELM), developed by Petty & Cacioppo (1981). The ELM was developed in an attempt to integrate the literature on persuasion by proposing that there was a limited set of core processes by which variables could affect attitudes, and that these processes required different amounts of thought (Petty & Briol, 2010, p. 224). This model, which posits that there are two main ways in which a person processes information, is a mechanism that al lows communication researchers to explore several aspects of persuasive communication. Petty and Wegner (1999) called the ELM a theory about how the classic source, message, recipient, and contextual variables have an impact on attitudes toward various objects, issues, and people (p. 42). The theory can also be helpful in understanding how internal and external variables impact some evaluative or nonevaluative judgment (p. 42). Fundamentally, this model allows researchers to determine how a message recipient evaluates and elaborates about that message. Cacioppo and Petty (1986) outlined several principles of the elaboration likelihood theory. First, people are motivated to hold what they believe are correct attitudes. Secondly, although people are so motivated, the amount and degree to which they can and are willing to hold those correct attitudes varies according to individual difference
78 factors and situational variables. The researchers explained that individuals are not able to elaborate on each and every message they encounter each day. Human beings tend to be cognitive misers who use mental shortcuts such as prior knowledge and experiences to make judgments and arrive at decisions (Taylor, 1981). Therefore, there is a continuum along which indi viduals elaborate on messages choosing to elaborate more on certain items than others. It is this idea of a continuum that sets the ELM apart from other theories of persuasion in that it accounts for arguments contained within messages and issuerelevant thinking (Petty, Briol, & Priester, 2009). Routes to Persuasion and Information Processing Petty (1977) first proposed the idea that individuals utilize one of two separate routes to process information, a central route and peripheral route. In the central route to information processing, an individual considers the quality of arguments, tending to be persuaded by high quality and well constructed arguments, rather than those that are illogical or weak (Eagly & Chaiken, 1993). Central route information processing involves effortful cognitive activity on behalf of the individual, who also utilizes prior knowledge and experience related to the information to form judgments (Petty & Cacioppo, 1986). Individuals using the peripheral route to process information are less motivated to make the effort to cognitively evaluate the information presented. These individuals are less motivated or able to process information using issuerelevant cues and are persuaded by issueirrelevant cues, which according to Pett y and Cacioppo (1986) include elements such as the attractiveness or power of the message source, or the opportunity for a reward (Petty, 1981). Those taking the peripheral route attend to cues that are sufficient to produce an initial attitude change wi thout any active thinking about the attributes of the issue or the object under consideration (emphasis in the original,
79 Petty, 1981, p. 256). Essentially, the use of the peripheral route indicates a lower level of cognitive work to process the message at hand. These two routes yield changes in attitudes based on the differing degrees of information processing (Petty and Wegener, 1999). Several studies related to agriculture and food have utilized the ELM to analyze persuasive communications. Frewer, Howard, and Shepherd (1998) explored the impact of prior attitudes, source credibility and admission of risk uncertainty by information source in a mail survey of British consumers attitudes toward genetic engineering. The team determined that increased credibility alone does not necessarily influence the psychological impact of information provision about risks other factors such as perceived hazard characteristics and informational content are likely to be important (p. 767). Frewer, Howard, Hedderly, and Shepherd (1997) adapted the ELM to investigate how the type of hazard, source of information, and the persuasive content of information impacted individuals thoughtful/effortful cognitions about risk messages related to food. The researchers determined there was an impact of source credibility on the perception of risk; risk perceptions were lower if the information came from a government source. In addition, they determined that while information from trusted sources resulted in greater perceived risks, a greater optimistic bias effect existed. Lundy (2006) utilized the ELM to explore message framing effects and cognitive processing levels of organization employees regarding a message. She discovered that study subjects produced significantly different numbers of thoughts (using the thought -
80 listing technique) based on the message frame they received, and that the thoughts listed differed depending on frame exposure. Issue Involvement and Argument Quality Two key elements in the ELM are issue involvement and argument quality. Petty, Priester, and Briol (2002) noted that ones motivation to process a message affects persuasion. Motivation, then, becomes part of the ELM in the form of issue involvement. The level of personal involvement one has regarding an issue is directly connected to how relevant that topic is to the individual (Petty & Cacioppo, 1979). When an individual sees an issue as more personally relevant (and is therefore considered to have higher issue involvement), certain types of arguments have been determined to be more persuasive. Individuals highly involved with an issue are more affected (and therefore persuaded) by logical and strong arguments than by illogical and/or weak arguments (Petty, Cacioppo, & Goldman, 1981; Petty & Cacioppo, 1979). Individuals with a low level of issue involvement, in contrast, are more affected by weaker arguments and peripheral cues in the message (Petty, Cacioppo, & Goldman, 1981). Burnkrant and Unnava (1989) discovered that changing the pronouns in a message from third person to second person (from she or he to you) is an effective way to increase the personal relevance of messages and therefore manipulate issue involvement. Another way to increase the relevance of an issue to individuals is to frame the m essage in such a way that it fits with their values or self image (Petty & Wegener, 1998). In the ELM, the concept of argument quality refers to the elements in a message that make it persuasive. Petty and Cacioppo (1981) wrote that any information in a me ssage which helps an individual evaluate the object, issue, or person under
81 consideration can be considered an element of argument quality. Argument quality, then, is important because it helps a message recipient to decide whether to process the informati on in a more elaborative way (taking the central route), or in a less considered way (the peripheral route (Petty & Cacioppo, 1996). What constitutes a strong or weak argument? Strong arguments, according to Petty and Cacioppo (1996), are those that primar ily elicit favorable thoughts, whereas weak arguments elicit negative thoughts or counter arguments. Argument quality is important in the ELM because elements of an argument affect how an individual processes the message, and ultimately whether or not he or she is persuaded by the message. However, argument quality does not act alone in an individuals decision to process the message in a certain way. Argument quality, combined with issue involvement work to impact how individuals process messages. Need fo r Cognition Cohen, Stotland, and Wolfe (1955) are credited with the first investigations into need for cognition, which they defined as an individuals need to structure relevant situations in meaningful, integrated waysa need to understand and make reas onable the experiential world (p. 291). Cacioppo and Petty (1982) explained need for cognition as the tendency to engage in and enjoy thinking (p. 116). In their study, Cacioppo and Petty (1982) established a scale to measure need for cognition to dif ferentiate between individuals who tend towards engaging in effortful analytic activity (considered high in need for cognition) and those who do not tend to or enjoy engaging in such thinking (considered low in need for cognition). Coutinho (2006) determi ned that individuals who are high in need for cognition are internally motivated to engage in tasks requiring more mental effort, while those low
82 in cognition tend to avoid such tasks and form opinions based on peripheral cues (Haugtvedt, Petty, & Cacioppo, 1992). Cacioppo, Petty, Feinstein, & Jarvis (1996) developed the concept of a continuum of need for cognition levels, where cognitive misers, those with low intrinsic motivation to engage in effortful thinking lie at one end, and cognizers, those with hi gh intrinsic motivation lie at the opposite end. Cacioppo et al., (1996) noted both individuals low in need for cognition and individuals high in need for cognition must make sense of their world, but they tend to derive meaning, adopt positions, and sol ve problems by somewhat different means (p. 198). Also, a link has been established between NFC and academic success. Elias and Loomis (2002) conducted a study about relationship between need for cognition and college students academic performance. In their study of college students, the researchers found that academic self efficacy beliefs and need for cognition were predictors of academic success, as measured by grade point average. Individuals high in need for cognition tend to spend more time analyzi ng information, such as media messages and other communications (Cacioppo, Petty, & Morris, 1983) and individuals low in need for cognition tend to be more influenced by information in context, using peripheral cues to form attitudes (Haugtvedt & Petty, 1992). Cacioppo, et al., (1996) suggested that one of the most commonly tested hypotheses is that individuals high in need for cognition should recall more of the information to which they are exposed than individuals low in need for cognition (p. 217). Following this line of thinking, if individuals are elaborating on and thinking about information, they are more likely to remember that information than if they are not expending effortful cognitive energy on the message. These researchers also asserted
83 that that attitudes of individuals high in need for cognition are more influenced by the quality of the issuerelevant arguments in a persuasive message than are the attitudes of individuals low in need for cognition (p. 229). Cacioppo, Petty, and Morris (1983) determined that individuals high in need for cognition were more likely to gather information from and think about externally provided message arguments than individuals low in need for cognition (p. 815). Cacioppo and colleagues (1996) concluded that, in large part, differences between individuals with high and low need for cognition levels are impacted by individuals past experiences. Hallahan (1997) investigated need for cognition in relation to persuasive messages presented as news and adver tisements. He found individuals high in need for cognition used media differently from low need for cognition individuals. Those high in need for cognition tended to use media for information seeking, while their counterparts with low need for cognition tended to use media primarily for entertainment and escape. Hallahan concluded, people are not uniformly attracted to media for its informational content (p. 19). Fear Appeals The discussion of the use of fear arises frequently in persuasion literature. The body of research regarding the use of fear is rather large and relatively complex; researchers have noted that the quality and focus of fear appeal studies differ widely, and that the literature itself lacks cohesion (LaTour & Rotfeld, 1997; Witte, 19 92). LaTour and Rotfeld (1997) noted that some discussions in the literature have used the terms threats, fear appeals, and fear as almost synonymous (p. 46). Witte (1992) also pointed out that researchers often use the terms fear and threat interc hangeably, when these two concepts are actually distinctly different from one another.
84 Perloff (1993) described fear appeals as communications that use fear to induce individuals to accept the message recommendations (p. 161); they are commonly acknowledged by researchers to be effective (Perloff, 1993). Witte and Allen (2000) described fear appeals as persuasive messages that arouse fear [and] motivate behavior change (p. 591). Fear itself, was defined by Witte, Meyer, & Martell (2001) as an internal emotional reaction composed of psychological and physiological dimensions that may be aroused when a serious and personally relevant threat is perceived (Perloff, 2003, p. 187). According to Nabi (2002), when a situation is both threatening to ones phy sical or psychological self, and when it is out of ones control, it provokes a sense of fear. Witte (1998) called fear a complex emotion, with both cognitive and affective elements. In a discussion of fear appeal messages, Perloff (2003) noted that such m essages contain two parts information about the actual threat and then information about a potential solution, or efficacy information; essentially, fear appeal messages present a problem one should be afraid of, and then information to alleviate or pr ovide a solution to the fearful problem. Early researchers discovered that elements of fear in messages inhibit acceptance of the message. Janis and Feshbach (1954) conducted a study with high school students to determine the influence of fearful messages about brushing ones teeth on adopting behaviors that provided a solution (brushing ones teeth to avoid dental problems). The researchers used three different messages, containing a high, medium, and low level of fear, and determined that a negative rel ationship existed between the strength of the fearful message and a subjects compliance with the
85 behavior to prevent the problem/address a solution. Thus they concluded that the greater the fear appeal contained in a message, the greater an inhibitor that message was to performing solution behavior. Whether or not fear appeals need to contain graphic language is still a subject of debate. Witte (1992) commented that fear appeals usually contain gruesome content in the form of vivid languagepersonalis tic languageor gory pictures (p. 331). However, OKeefe (1990) pointed out that the use of such language is not guaranteed to provoke fear, and it is possible to create a sense of fear without use of vivid or gory language and imagery. Altheid (1997) as serted that fear is pervasive in American society (p. 648). In studying the use of fear in television news reports, Altheid remarked that d espite clear evidence showing that Americans today have a comparative advantage in terms of diseases, accidents, n utrition, medical care, and life expectancy, they perceive themselves to be at great risk and express specific fears about this (p. 649) He proposed a conceptual model and suggested the use of a problem frame by the news media creates a system that per petuates messages of fear, and heightens societys fears about certain issues despite actual facts and figures indicating otherwise. Altheid pointed out that s hared knowledge about the social world in a mass mediated society tends to be about bad news (p. 658) Society is told what, specifically, to fear. Carried with the messages of fear are the images and targets of what and who is to be feared (p. 665). In essence, this suggests that through the persistent use of the problem frame by media, society is programmed to expect news and information to be presented as a problem to be solved.
86 Fear and Persuasion Witte (1992) called fear appeals persuasive messages designed to scare people by describing the terrible things that will happen to them if they do not do what the message recommends (p. 329). LaTour and Rotfeld (1997) conducted a metaanalysis of fear appeals studies and noted that the majority of the research concluded increases in fear are generally associated with changes in behavior, attitudes, or intentions (p. 47). Perloff (2003) noted, however, that it is possible that messages can cause fear without motivation to change ones attitude, because the message either does not mesh with the individuals current beliefs (cognitive dissonance), or because the message does not include a solution to deal with the cause of the inspiration of fear. In a discussion of the drivereduction model of fear appeal research, Eagly and Chaiken (1993) noted that the model suggests low and moderate levels of fear in a message likely enhance persuasion; however, increases in emotional tension likely reduce persuasion by stimulating defensive behaviors. Fear is most likely to enhance persuasion when [message] efficacy is high (vs. low) and when this information i mmediately follows fear arousing message content (Eagly & Chaiken, 1993, p. 444). Keller & Block (1996) studied the way in which fear arousal affects persuasion and determined that the level of fear arousal may be positively related to the likelihood t hat an individual will elaborate on a message. In low fear conditions or situations where a persons fear has not been aroused sufficiently, they may be less likely to be persuaded because they are not sufficiently motivated to elaborate on the message. Ho wever, Keller and Block noted that high levels of fear in messages have the potential to motivate individuals to only elaborate on the problems in the message and not the potential solution(s) included in the message. The researchers also suggested that, i n
87 the absence of message elaboration, persuasion is possibly a function of memory based message recall. Fear Appeals and Health Messages Messages with fear appeals are commonly used in health communication in order to prompt individuals to adopt healthy behaviors (Das, deWit, & Stoebe, 2003). Witte (1998) noted that there is a connection between the beginnings of fear appeal research and the development of health behavior models, and PrenticeDunn and Rogers (1986) postulated that fear appeal research and t heories are varieties of the health belief model (HBM). Witte (1998) state that both the health belief model and the extended parallel process model (a fear appeal theory/model) are concerned with how certain stimuli influence perceptions of susceptibilit y and severity to influence behaviors (p. 441). However, the HBM does not specifically address the way in which fear influences an individuals behavior. The HBM itself was created to help understand and explain human health behaviors, such as the failure of individuals to perform or complete behaviors like vaccination or early detection screenings, which are proven preventative measures (Eagly & Chaiken, 1993). Kelly and Edwards (1992) pointed out that effective health campaigns have a common characterist ic their creators understand how and to what degree audiences see the topic as personally relevant, and then craft messages accordingly. Roskos Ewoldsen, Yu, and Rhoades (2004) explored fear appeal messages about breast cancer and breast self exams. Thei r research determined that messages which emphasized the efficacy of performing breast self exams yielded attitudes that were more positive toward breast self exams, and that those positive attitudes predicted behavioral intentions to perform the exams. Th e researchers noted that it can be
88 particularly important [to devise] health communications so that they are personally relevant, yet not so fear inducing that they push the recipients into a defensive processing style (Roskos Ewoldsen, Yu, & Rhoades, p. 63, 2004). LaTour and Rotfeld (1997) conducted research regarding consumer attitudes toward brands and purchase intentions in relation to fear appeals, using protection motivation theory. They asserted if the threat seems real and salient, one should feel an enhancement of value towarda product and increased desire to obtain it for ones peace of mind (p. 49). Nabi (1999) wrote about the impacts of emotions such as fear on information processing, recall, and attitude change. In proposing a model of p ersuasion which suggests emotion plays a role in message processing, Nabi pointed out that several previous studies conducted regarding fear and information processing, without explicitly reaching a consensus, suggest that chronic fear and messagerelevant acute fear may lead to less careful message processing if the need for reassurance is not met or is met by peripheral cues (p. 300). Essentially, the arousal of fear can indeed impact how an individual processes a message, and thus, whether or not an i ndividual is persuaded as the result of the message. Attitude Eagly and Chaiken (1993) defined attitude as a psychological tendency that is expressed by evaluating a particular entity with some degree of favor or disfavor (p. 1). Petty and Briol (2010) stated attitudes refer to general evaluations individuals have regarding people (including yourself), places, objects and issues. Attitudes can be assessed in many ways and are accorded special status because of their presumed influence on peoples choi ces and actions (p. 217). Attitudes motivate our behavior and
89 are complex psychological and abstract constructs not directly observable (Eagly & Chaiken, 1993; Perloff, 2003). Perloff (2003) called attitudes learned predispositions that help an individual respond in a consistently favorable or unfavorable manner regarding a specific object (p. 39). Petty and Cacioppo (1996) noted that attitudes can serve important functions for peopleand can be useful in predicting behaviors (p. 29). According to Fishbein & Azjen (1975), attitude is composed of two elements: cognition and affect. These researchers noted that attitude is a combination [of] what you believe or expect of a certain object, and how you feel about (evaluate) these expectations (p. 46). Eagl y and Chaiken (1993) considered attitude as divided into three elements: cognition, affect, and behavior. The cognitive aspect of attitude pertains to the thoughts that an individual has about an attitude object. The affective aspect refers to feelings and emotions that one has about an attitude object, while the behavioral element concerns what an individual does with respect to the attitude object. Katz (1960) suggested that attitudes help individuals structure their knowledge and environment, help us e stablish a frame of reference and also serve a social role in self expression and social interaction. Attitude Formation Attitudes are presumed to vary along an evaluative continuum ranging from a strongly positive orientation to a neutral orientation to a strongly negative orientation. This continuum can be decomposed into valence (i.e., positive or negative) and extremity (degree of favorability) (Krosnick & Petty, 1995, p. 56). The way in which we form attitudes and change our attitudes are distinctl y separate functions (Crano & Prislin, 2006). Our attitudes can be formed in many ways. We can consciously embrace or reject a specific attitude object, and we can also form our attitudes subconsciously
90 (Crano & Prislin, 2006). We are also able to construct our attitudes based purely on exposure to an attitude object or stimulus (Bargh, 2001; Lee, 2001). Perloff (2003) noted that individuals are able to hold simultaneously contradictory attitudes toward an issue or attitude object. However, it is believed that strong attitudes are more predictive of behavior, while weak attitudes are not (Miller & Peterson, 2004). Krosnick and Petty (1995) declared that attitudes are strong to the extent that they are persistent, resistant to change, impact information processing and guide behavior. The stronger the attitude an individual holds, the more predictive that attitude can be said to be of the individuals behavior. Petty and Krosnick (1995) conceptualized attitudes as present along a continuum, ranging from stron g positive to neutral to strongly negative. Role of Schema Eagly and Chaiken (1993) suggested that a useful way to think about attitudes is to consider them as a type of schema. According to these researchers, schema is a broader classification of cognit ive structures (p. 18). Fiske and Linville (1980) called schemas cognitive structures of organized prior knowledge, abstracted from experience with specific instances (p. 543). Schema and their relationship to attitudes have important implications for the way in which individuals process and recall information. Eagly and Chaiken (1993) asserted a mental representation of attitude may be stored in memory and thus can be activated by the presence of the attitude object or cues related to it (1993, p. 2) Katz (1960) noted that one of the primary functions of attitudes is to organize and simplify an individuals experience; a concept echoed by Eagly and Chaiken (1993). According to Eagly & Chaiken (1993), schemas have an influence on the way individuals processes information, affecting the amount of attention given to a piece of
91 information as well as the encoding and judgment of the information (p. 19). Schemas selectively affect the organization of memory, retention and retrieval of information (Eagly & Chaiken, 1993). Attitude Change Kruglanski and Stroebe (2005) stated, persuasion refers to attitude change in response to complex verbal messages that are intended to persuade an audience (p. 342). While the ELM is a model that allows for the examinati on of how an individual processes information, peripherally or centrally, it is also a model that contributes to the literature regarding attitude change. Petty and Briol (2010) stated the ELM was developed in an attempt to integrate the literature on persuasion by proposing that there was a limited set of core processes by which variables could affect attitudes, and that these processes required different amounts of thought (p. 224). Petty and Cacioppo (1996) noted that several theoretical approaches ca n be used to explore attitude change, but that the ELM is particularly and often used when attempting to understand attitude change as the result of persuasive communications. Petty and Wegner (1999) noted that the ELM allows the examination of the proces ses underlying changes in judgments of objects, the variables that induce these processes, and the strength of judgments resulting from these processes (p. 42). Petty (1981) explained the connection between information processing and attitude change in t he ELM. Using the central route, an individual attends to the arguments contained in the message and combines this using all of the information s/he possesses about the issue to form a coherent, reasoned position. Petty noted, for the most part, under the central route, persuasion is based on a thoughtful consideration of the object or issue at hand (p. 256). In contrast, the peripheral route to persuasion is
92 not one that utilizes a great deal of thoughtful elaboration. Using the peripheral route if a me ssage is associated with a pleasant smell or an attractive source, it is accepted. If the message takes a position that is too discrepant, it is rejected regardless of the cogency of the arguments presented (p. 256). Petty pointed out that it is not that one of these routes is rational and the other is not rather: the difference between the two routes has to do with the extent to which the attitude change that results from a message is due to active thinking about either the issue or the object relevant i nformation is the most direct determinant of the direction and amount of attitude change produced (p. 256) Petty (1981) also stated that personal relevance of the issue under consideration is the primary determining factor as to whether one is persuaded by a message, whether ones attitude is changed. When the issue has high personal relevance factors central to the issue are more important; under low relevance, peripheral factors become more potent (Petty, 1981, p. 259). Chaiken (1980) studied attitude change based on different sources and message characteristics, and manipulated personal consequences, message arguments and source likeability. From the study, Chaiken determined that in conditions where there were high consequences, subjects determined t heir attitudes mainly by the quantity of issue relevant arguments present, while the likeability of the source had no effect. Chaiken determined, however that under low consequences conditions attitudes were determined primarily by source likability; the number of arguments pres ented had no significant effect (p. 259) Attitudes and Information Processing Allport (1935) declared that for individuals, attitudes determine what he will see and hear, what he will think and what he will dothey engender meaning upon the
93 world; they draw lines about and segregate an otherwise chaotic environment; they are our methods for finding our way about in an ambiguous universe (p. 806). As noted previously, media messages are often framed in a certain manner in order to emphasize a specific perspective or point of view. Frames also impact the way in which individuals interpret messages. Specific knowledge about an issue has relevance for attitude about that issue and the way an individual processes information regarding that issue. Petty and Krosnick (1995) noted that the more knowledgeable an individual is about a subject, the more difficult it is for that individual to change his or her attitude. Additionally, Petty and Krosnick determined that knowledgeable indivi duals are expert reasoners and information processors in the relevant domain. They can use their extensive store of beliefs and prior experiences to interpret and evaluate new information on the topic (p. 283). In contrast, the less knowledgeable and the uninformed have considerable difficulty attending to, comprehending, and evaluating relevant information (Petty & Krosnick, 1995, p. 283). In their writing on working knowledge, Petty and Krosnick (1995) asserted that knowledge about a subject affects t he way individuals process new information about the subject. Festingers (1957) cognitive dissonance theory asserts that individuals actively seek out information that is consonant with their current knowledge and beliefs, often to the degree that they ar e willing to discount information that does not fit their current position or understanding of the topic at hand. According to Petty and Krosnick (1995), knowledgeable individuals are able to readily recall information stored in
94 memory about an issue or object when they are faced with subsequent interactions or encounters with it. The concept of working knowledge accounts for the schema an individual holds which is called to mind when an individual is confronted with an issue or object about which he holds an attitude (Wood, 1982). Therefore, the knowledge itself is limited to this particular topic and attitude domain and does not represent the entirety of an individuals knowledge (Wood, 1982). Priming and Attitudes The concept of priming is often discuss ed in media effects literature, in close relation to the concepts of framing and agendasetting. However, it is primarily a psychological concept that is based on a memory model of information processing (Schuefele, 2000; Schuefele, 2007). Roskos Ewoldsen, RoskosEwoldsen, and Carpentier (2008) stated that priming refers to the effect of some preceding stimulus or event on how we react, broadly defined, to some subsequent stimulus (p. 97). When applied to the media, this means that what an individual has read or seen in the media at a prior date can affect judgments or behavior later when they are exposed to that issue again (Roskos Ewoldsen, Roskos Ewoldsen, & Carpentier, 2008). Salancik (1974) suggested that priming makes concepts or issues more accessi ble in ones memory. By receiving and processing information about a concept or issue, one creates mental pathways about that topic, and then those pathways influence subsequent processing of information about that topic. Scheufele (2000) noted that primi ng is the impact that agendasetting can have on the way individuals evaluate public officials by influencing the thematic areas or issues that individuals use to form
95 these evaluations (p. 297). Iyengar and Kinder (1987) noted that priming affects how people make evaluations of political issues and candidates. Petty and Krosnick (1995) referred to framing as cognitive priming, and noted that it has an effect on attitude. According to Tourangeau, Rasinkski, Bradburn, and DAndrade (1989) individuals are m ore susceptible to the effects of priming in two instances: when the issue at hand is important to them or when their attitude toward the issue is closer to one end or the other of the attitude continuum. Priming involves three basic steps: a concept must be activated in ones memory from a previous priming experience, this concept must then affect the way the individual interprets a new related issue or concept, and finally, the combination of previous information and new information is then used to form a new perspective which then affects the individuals judgment or guides behavior (Petty & Jarvis, 1996). In a review of priming literature, Roskos Ewoldsen, Roskos Ewoldsen, and Carpentier (2008) provided examples of priming studies that have shown subst antive results. Josephson (1987) found effects of media violence on the behavior of children, while Anderson (1997) conducted similar research regarding the influence of violent media on aggression in college students. Priming has also been used in studies regarding political news coverage (Iyengar, Peters, & Kinder, 1982; Krosnick & Kinder, 1990) and the influence of stereotyping in health communication research, such as in anti smoking campaigns (Roskos Ewoldsen, Roskos Ewoldsen, & Carpentier, 2008). The researchers concluded, clearly, the media act as a prime[and] influences later judgments and behavior (p. 102).
96 Despite clear evidence that priming does indeed work, evidence illustrates that the effects of priming fade with time (Roskos Ewoldsen, Ros kosEwoldsen, & Carpentier, 2008). This phenomenon has been attributed to the fact that the impact of a prime is a combination of two separate functions: the intensity of the prime, which refers to the frequency of the prime or its duration, and the recenc y of the prime, which refers to the time between when the information is first encountered and when it is next encountered. Petty, DeMarree, Briol, Horcajo, & Strathman (2008) explored a connection between need for cognition and the influence of priming. They asserted there is reason to believe that highNFC individuals would be more influenced by primes than low NFC individuals [possibly because] the prime produces greater activation of the underlying construct (Petty, DeMarree, Briol, Horcajo, & Str athman, 2008, p. 901). They suggested that priming assumes the activated concept exerts a bias on related information processing. Petty et al. (2008) discovered that although priming effects can and do occur when perceivers are not thinking a lot, the present research offers additional evidence that they can also occur among highthought individuals (p. 910). In essence, the researchers discovered that individuals who have a high need for cognition can be more affected by priming than individuals with low er levels of need for cognition. Information Recall Booth (1970) hypothesized that four variables impact a media users recall of news items. These included the frequency with which the news item appears, the location of the item on the page or in the broa dcast, the relative time or space devoted it, and the use of images accompanying the news item. Booth noted that repetition of a message
97 is more effective than a single exposure in aiding recall, therefore the number of times a news item is seen by an individual can impact whether or not, and how much of the information is recalled. Also, the amount of time and/or space devoted to a news item in its medium provides an opportunity for individuals to recall details and put the item into context, aiding recall. The location of the news item can also affect recall, with items in locations such as the outer or end pages being more accessible, and the items at the beginning of a newscast being seen as more prominent and therefore possibly more memorable. Final ly, Booth noted that the use of images accompanying news items can impact recall through expanding the message to a multiple channel presentation of the news item, therefore aiding recall. David (1998) conducted a series of experiments testing the notion t hat adding an image to a story could increase information recall and determined that indeed, a related image did impact recall, as did the concreteness of the news topic (versus an abstract topic). DeFleur, Davenport, Cronin, and DeFleur (1992) conducted experiments to determine if one medium could prompt subjects to recall information best from radio, television, newspaper, and computer. Controlling for attendance to the information, motivation to learn, multiple exposures to the information, and potential distracting factors, the researchers found subjects were able to recall more information, both aided and unaided, at significantly higher levels from newspapers and computers than the other two mediums. Additionally, they determined that there were no s ignificant differences in recall between newspapers and computer screens. The researchers concluded that these results confirmed previous studies, which indicated that newspapers are a superior medium than the others for promoting information recall.
98 Judd and Kulik (1980) hypothesized that social attitudes towards issues have schematic effects on individuals information processing. They suggested that information which fit into individuals schemas would be recalled more readily. Results from the exper iment led them to determine that information either highly consistent to the attitude one holds about the issue/subject or that is highly contradictory to a subjects attitude (that is, information on either end of a bipolar continuum of extremes) was more easily recalled than information that is considered only moderately consistent or contradictory (p. 576). They concluded that information representing more extreme positions on a pro/anti continuum are processed and recalled more easily than information, which conveys a more moderate point of view. In a similar study, Wicks and Drew (1991) suggested that information consistent with existing schemas tend to reinforce current beliefs, but that information which is inconsistent with those current beliefs woul d cause subjects to recall more information because of the salience of the inconsistencies. The researchers also explored the differences in recall of inconsistent information when using different media text, television, and radio and hypothesized that the text form would induce subjects to recall more information than the other two media. The researchers determined that subjects receiving information consistent with their attitudes recalled more information, and were also more likely to make inferences about situations than those who received inconsistent information. Subjects who received inconsistent information, however, were more likely to recall factual information. Recall of Online v. Print Messages Several studies have been conducted regarding the differences in recall between print and online sources of news/information. Bogart (2000) reported that an experiment
99 testing identical news articles in print and on a computer screen, subjects rated the print version as more understandable. Tewksbur y & Althaus (2000) conducted an experiment comparing the effects of exposure to print and online versions of the New York Times, and determined that those who read the online version of the stories chose to read fewer international, national, and political news, and spend less time with those kinds of stories when they did read them. Additionally, those who read online news were less likely to recall national and political news than business and other news topics. The researchers determined that cues present in the online versions of the news were more likely to distract the subjects than encourage them to recall information from the stories. Instead, subjects substituted their own interest as guides. In a study of residents of the Netherlands, DHaenens, Jankowski, and Huevelman (2004) explored the recall of news presented in print and online versions of two newspapers. They found that reader attention to news stories differed depending on the newspaper and the category of the news item. Despite this, t he researchers determined there was no difference between the recall levels of individuals exposed to the online news as compared to the print news. Eveland & Dunwoody (2002) used experimental methodology to explore the influence of medium (online versus print) on subjects learning and information processing. The researchers determined that subjects exposed to the online version of the messages had an increased level of elaboration on the messages when compared to those receiving the print version, and t his increased elaboration was found to be positively related to the amount of learning exhibited. Despite this, subjects receiving
100 the online messages learned significantly less when compared to those having received the print version of the messages. Ev eland and Dunwoody also determined that subjects exposed to the online messages more frequently utilized selective scanning of information than those exposed to the print messages. They concluded that this selective scanning contributed to a decreased lev el of learning among subjects exposed to the online treatment of the messages. Davis, Akers, Cepica, Doerfert, Fraze, and Lawver (2005) conducted an experiment to measure recall of an agricultural news story presented in various media formats, including pr int and electronic text of news stories, as well as audio and video news releases. Results indicated that as in previous studies, print delivery yielded greater recall, and subjects exposed to print versions of the news stories recalled significantly more information than those exposed to electronic text versions of the stories. deWaal and Schonbach (2008) used survey methodology to explore whether subjects would differ in the kinds and number of issues (social and political) recalled when exposed to print versus online versions of newspapers. The research team determined that while print newspapers were more effective than online newspapers in increasing the overall number of topics recalled by subjects, this was only the case if subjects indicated they w ere interested in and relied on newspapers for information. Fewer studies have been conducted comparing types of online message delivery than those comparing print versus online delivery of messages. Eveland, Cortese, Park and Dunwoody (2004) investigated the way in which site design and organization influenced recall of information, factual knowledge, and knowledge structure density.
101 These researchers investigated the notion that the nonlinear structure of the web could be both a strength and a weakness in terms of these three types of learning. They determined that learning overall was inhibited by the nonlinear structure compared to linear structural delivery. In addition, they concluded that web sites designed in a more linear fashion (similarly t o traditional print media) encouraged factual learning, but that nonlinear designed sites increased knowledge structure density, which has important implications for comprehension, retention of information, and the ability to engage in problem solving (E veland, et. al., 2004, p. 210). Conceptual Model Figure 2 1 represents the conceptual model developed to guide this study and explains how the concepts of message framing and website interactivity work to affect changes in attitude and information recall among subjects exposed to a message in a specific issue context. The model represents the experimental design and includes individual attribute variables, which also may have an effect on the outcome variables. This conceptual model illustrates the studys exploration of how messages in a specific issue context, in this case the contrast between conventionally raised beef and organically raised beef, can be framed as risks or benefits in a news story. Frames are unavoidable in producing news; they are often considered the angle of a story. Message frames are an invitation for readers to interpret information in a certain way, invitations which highlight certain elements of a message over others, which can also exclude certain opinions, arguments or groups involved in an issue. This framed news story can then be delivered on either a highly interactive website or a website with a low level of interactivity. The concept of website interactivity involves how many and what kinds of functions are included on a website to engage a user.
102 Websites with high levels of interactivity have a number of functions which allow users to go beyond merely reading a news story or message and actually engage with the story itself by passing it on to others, commenting on it, or participating in a discussion regarding the topic. Figure 21 combines these concepts of message framing, specifically as risks and benefits, and the concept of website interactivity and suggests that together these elements impact a subjects ability to recall information from the message and has an effect on the subjects attitude change regarding the subject of the message. Previous literature indicates that individual attribute variables such as an individuals involvement with the issue and need for cognition can also impact the outcome of the model. Figure 21. Conceptual model. Chapter Summary This chapter has provided the theoretical and conceptual framework for the study. In addition, the material presented covers the salient research relevant to the research study. This review of literature included scholarly research in the field of the uses and
103 gratifications approach to media use, new media, website interactivity, framing theory, risk communication, the elaboration likelihood model of persuasion, attitude, and recall of information.
104 CHAPTER 3 METHODOLOGY Overview As illustrated in the previous chapters, media messages have an effect on consumer perceptions. The manner in which news is framed affects how individuals think about the issue and what arguments and evidence is taken into consideration when forming an opinion. Due to intense media coverage of food safety issues over the past few decades, this topic has become a concern for consumers across the globe. Consumers increasingly see org anic foods as safer alternatives to conventional products. At the same time, individuals are faced with new ways of obtaining information, often via the Internet. Web sites provide differing levels of interactivity, which often impact the way individuals i nterpret and process messages, their attitudes about the content and the amount of information they remember. At this time, no research exists which investigates the potential impact of the message framing and site interactivity on individuals attitudes and information recall. The purpose of this study was to investigate how messages about beef, framed as either risk or benefit, when coupled with site interactivity (low vs. high) affected subjects attitudes toward the issue and their ability to recall inf ormation about the message. The studys objectives were to describe the differences in attitude change and information recall among subjects exposed to the eight treatment conditions. In addition, this study sought to explain changes in attitude and infor mation recall based on measured individual attribute variables ( issue involvement, need for cognition and prior knowledge of the issue )
105 By assessing need for cognition prior to the experiment, the researcher was able to consider the impact of subjects wi llingness to engage in effortful thinking on attitude and information recall. Determining these effects will aid professional communicators in crafting effective messages about agricultural issues, as well as contribute to the body of knowledge regarding message framing and site interactivity. Hypotheses and Research Objective Based on the literature presented, the following hypotheses were developed: H1: Subjects exposed to risk framed messages will have significantly different attitude change and informa tion recall than subjects exposed to benefit framed messages. H2: Subjects exposed to messages on a high interactive website will have significantly different attitude change and information recall than subjects exposed to messages on a low interactive website. H3: Subjects exposed to risk framed messages on a highinteractive website will have significantly less information recall than subjects receiving benefit framed messages on a low interactive website. H4: Subjects exposed to risk framed messages on a high interactive website will have significantly greater attitude change than subjects receiving benefit framed messages on a low interactive website. In addition, a single research objective was included: Objective 1: To explain changes in attitude and i nformation recall based on the individual attribute variables of issue involvement, need for cognition, and prior knowledge of the issue. Research Design The research design for this study was experimental, using a 2 (risk of conventional/organic, or benef it of conventional/organic message) x 2 (low interactive website/high interactive website) between subjects, factorial design in order to assess whether message frame delivered via a traditional (low interactivity) or Web 2.0 (high interactivity) website i nfluenced attitude and information recall of subjects.
106 Factorial designs allow for the investigation of a combined effect of two or more independent variables (Ary, Jacobs, Razavieh & Sorenson, 2006). In addition, research studies using factorial designs allow multiple hypotheses to be tested in a single experiment (Christenson, 2001). To generate a sample size sufficient to effectively test the research hypotheses, a minimum sample size of 30 individuals per treatment condition is necessary (4 treatment conditions x 30 = 120) per Shadish, Cook, and Campbell (2002). The group design was: R O1 O2 O3 O4 X1 O5 O6 R O1 O2 O3 O4 X2 O5 O6 R O1 O2 O3 O4 X3 O5 O6 R O1 O2 O3 O4 X4 O5 O6 R O1 O2 O3 O4 X5 O5 O6 R O1 O2 O3 O4 X6 O5 O6 R O1 O2 O3 O4 X7 O5 O6 R O1 O2 O3 O4 X8 O5 O6 In the above design: R = random assignment, O1 = issue involvement scale, O2 = prior knowledge measure, O3 = prior attitude scale, O4 = need for cognition scale, X1 8 experimental treatment, O5 = information recall measure, O6 = post attitude scale. The experimental treatment conditions were as follows: X1 = benefit of organic beef on high interactive site, X2 = benefit of organic beef on a low interactive site, X3 = benefit of conventional beef on a high interactive site, X4 = benefit of conventional beef on a low interactive site, X5 = risk of conventional beef on a high interactive site, X6 = risk of
107 conventional beef on a low interactive site, X7 = risk of organic beef on a high interactive site, X8 = risk of organic beef on a low interactiv e site. Threats to internal validity were considered when designing the study. According to Ary et al. (2006), these threats include history, maturation, testing instrumentation, regression, selection, attrition interaction effects, experimenter effects subject effects and diffusion. History, maturation, instrumentation, regression, selection, interaction effects, experimenter effects, subject effects and diffusion threats were addressed by the design of the study. Testing threats are described by Cam pbell and Stanley (1963) as the effects of the administration of a first test on the results of a second test. As subjects completed two instruments, it is possible that this produced a testing effect; however, it is important to note the material on only two of the preexperimental measures were repeated on the post experimental measures. Therefore, with the exception of these two instruments (the attitude measure and the prior knowledge measure) the design of the two instruments was not pretest/post test in nature and the results of the first instrument would not affect the second instrument. Potential confounding variables such as individual need for cognition and preexisting knowledge of the issue topic were considered, controlled for, and measured as i ndividual difference variables. Validity of the researcher designed instrument was assessed through a panel of experts who evaluated its face validity. The pilot testing procedures ensured construct validity, and tested the preexperimental measures, the r esearcher designed instrument, and the experimental message stimuli. Threats to external validity considered when designing the study were population validity, ecological validity, and experimenter validity (Christensen, 2001). Population
108 validity pertains to the ability to generalize from the sample used in the study to a larger population of individuals that the sample represents and is important to consider (Christensen, 2001). This study utilized a convenience sample of college students from a large s outheastern university, and therefore is not generalizable to the whole population of college students. Convenience sampling of college students is often used in psychology research, but Peterson (2001) cautioned researchers doing so with the intent of gen eralizing their study findings to nonstudent populations. He noted that in a review of psychological studies using college student convenience samples that the subjects tended slightly, but consistently, to respond in a more homogenous manner than nonst udent subjects. However, in this study, the intention of the research was not to generalize the findings to a nonstudent population; rather it was to obtain the participation of college student subjects due to their previously identified roles as new vote rs, relatively new consumers, and individuals at a time in their lives where they are beginning to form habits, attitudes and perspectives that many of them will hold for the rest of their lives on important societal issues. While generalizations cannot be made to all college students from the sample used in this study, data collected here can inform future research and provide a snapshot of the characteristics of the students in this sample. Ecological validity was addressed through the online design of th e experiment. In addition, to reduce potential Hawthorne effects, subjects were not told they were participating in an experiment; rather they were informed that the study was a survey. An additional threat to ecological validity is the potential for novel ty or disruption (Christensen, 2001). To address this potential threat, subjects were shown an image of
109 the experimental conditions (messages on and interactivity of websites) rather than having them access a live website. Experimenter validity was address ed through the online administration of the experiment. Subjects This study used a convenience sample of subjects (N = 492) from four large courses taught in the college of agriculture at a large Southeastern university. In order to provide an incentive f or participation, students were offered extra credit in the course or courses in which they were enrolled. Data were collected for individual subjects; however, course enrollment data was collected for purposes of crosschecking to ensure subjects completed the experiment once. Subjects enrolled in more than one of the courses received extra credit for each course by participating in the experiment a single time. The courses from which subjects were recruited meet curriculum requirements for university general education credits and therefore can be considered representative of the university student population, including variation in majors and student backgrounds. The courses that comprised the sample pool of subjects were selected due to availability, and to counteract any validity threats due to selection, the message stimuli and site versions (the experimental conditions) were randomly assigned to subjects using a random number generator. A total of 410 students completed the entire experimental administr ation, which included an online preexperimental questionnaire measuring need for cognition, preexisting knowledge and attitude toward the issue topic, the message stimuli, and the questionnaire instrument. The study was administered via the Internet usi ng Qualtrics, an online questionnaire administration system. Subjects were randomly assigned to
110 one of the eight experimental conditions (benefit of organic beef on a high interactive site, benefit of organic beef on a low interactive site, benefit of conventional beef on a high interactive site, benefit of conventional beef on a low interactive site, risk of conventional beef on a high interactive site, risk of conventional beef on a low interactive site, risk of organic beef on a high interactive site, and risk of organic beef on a low interactive site). College students are a large portion of the 18 24 year old age group and are considered by researchers to be an important group due to their status as relatively new voters, their role as consumers and t heir tendency to be early adopters of new technologies (Emanuel et al., 2008). Vincent and Basil (1997) noted that among college students, current events knowledge and media consumption are related. Putnam (2000) illuminated an important connection between media use and civic engagement of individuals, pointing out that lower levels of news consumption have a relationship to the low level of civic engagement among todays American youth. College aged individuals are in the throes of a period of significant changes in their lives. Henke (1985) and Rubin and Rubin (1985) noted that this is a period of socialization to news media at a time where they are unlearning the habits of childhood and beginning to define their roles as adults, and a time where they beg in forming lifelong news consumption patterns (Al Obaidi, LambWilliams, & Mordas, 2004; Barnhurst & Wartella, 1998; Diddi & LaRose, 2006; Schlagheck, 1998). Involvement in and knowledge of the world is essential for college students success in their chosen professions and media play an important role in society (Al Obaidi, LambWilliams, & Mordas, 2004).
111 Message Stimuli This experiment was designed to test message stimuli. Using a random number generator, subjects were randomly assigned to one of eight treatment conditions containing messages about the risks or benefits of consuming organic or conventional ground beef. Message Development The messages for this study were structured as news articles highlighting research conducted at a university and wer e designed to appear as if they were published on an online newspaper. The articles (see Appendix F) used a typical news story format with headline, dateline, hypothetical web address, and used the inverted pyramid style of news writing. Each message utili zed a risk or benefit frame, paired with the emphasized subject of conventional or organic ground beef. Each of the frames was then presented on a high interactive website and a low interactive website, for a total of 8 treatment conditions. Marette, Roose n, Blanchmanche, and Verger (2006) conducted a study that was used to aid in construction of the message treatments for this experiment. Marette et al., conducted an experiment to determine the effect of health information on consumers choice of two types of fish products. The researchers created messages focusing on risks and benefits of eating the two types of fish and evaluated the impact of heath information on consumers choice between a risky type of fish and type of fish considered to offer health benefits (Marette et al., 2006, Introduction, para. 3). Marette et al. then gauged subjects willingness to pay for each kind of fish after exposure to the information.
112 The choice to contrast risk and benefits of beef was prompted by literature regarding media coverage of food safety issues, as well as the increasing popularity of organic food. As noted in Chapter 2, research has documented media coverage of food consistently raises health and safety issues, and often contrasts the health and safety of organic food to conventional food. Lockie (2006) in a content analysis of media coverage regarding organic food determined that organic foods were often framed as a way to avoid risks associated with conventional foods, and that organic foods were constructed in the mass media often in terms of what they were not genetically engineered or the products of industrialized agriculture (p. 319). Williams and Hammitt (2001) and DuPuis (2000) both noted that consumers purchase of organic foods is often motivated by the desire to avoid certain elements of conventional food production such as pesticides, supplemental hormones and antibiotics, or the involvement of biotechnology, which are often perceived as risks (Gifford & Bernard, 2004). To create the messages, the Marette et al. (2006) article was used as a model, and information about actual health benefits of beef (conjugated linoleic acid, or CLA) and actual health risks related to beef ( E. coli O157:H7) were chosen to accompany the contrasting categories of or ganic and conventional beef to exemplify the frames. In addition, the USDAs National Organic Program standards (USDA, 2008) were incorporated into the messages to add a clear definition of what constituted organic beef. The benefit framed messages focus ed on the positive health impacts of consuming CLAs and both messages included the following keywords and phrases: CLAs, beneficial to human health, lowering cancer risk, long lasting beneficial
113 effects. The risk framed messages included the follow ing key words and phrases: bacteria, E. coli kidney disease, illness, and death. Common across all four messages (benefit of organic, benefit of conventional, risk of organic, risk of conventional) were the following key terms: scientific evidence, pesticides, genetically modified organisms, animal byproducts, antibiotics, growth hormones, and beef consumption. While the messages incorporated these two actual elements, the healthbenefit aspect of CLAs contained in beef, and the health risk aspect of E. coli bacteria in beef, the messages themselves were created in order to elevate the risk/benefit frames for the purposes of this study and were not factual, which made the experiment a deception study. As such, debriefing information was included at the end of the experiment so that subjects had accurate information regarding CLAs and E. coli bacteria and their connection with beef products. The messages were approved by a panel of experts and pretested with graduate students. A messa ge frame manipulation check was included in the final instrument. Message Testing The message frames used in the final study were pretested using a convenience sample of graduate students with a variety of subject area specializations, levels of awareness and experience with the message content. For the pretest, subjects were randomly assigned to one of four news stories: benefit of organic beef, benefit of conventional beef, risk of conventional beef, and risk of organic beef. To test the messages, respondents were asked to read a news story and asked to respond to a series of questions on a survey instrument. Using Dillmans (2009) tailored design methods for distributing online surveys, a total of 70 surveys were
114 distributed and 35 were returned with usable data for a response rate of 50%. Following the news story, subjects were asked to list the thoughts and feelings that came into their minds while reading. Brock (1967) developed the thought listing technique to assess cognitive elaboration. Thought listing is often used in ELM studies to determine the amount of cognitive activity after exposure to a message (Cacioppo & Petty, 1981). Subsequent to the thought listing procedures, subjects were asked to identify the articles frame, first using their ow n words and then responding to a closeended item. A total of 30 out of the 35 subjects correctly identified the article frames on the closeended item. Independent Variables The two independent variables manipulated in this study were message frame (risk of conventional beef, risk of organic beef, benefit of conventional beef, or benefit of conventional beef) and the website type (low interactive or high interactive). The individual difference variables, need for cognition and issue involvement were measured on the basis of data collected on the preexperimental portion of the instrument, which also included preexisting knowledge and attitude measures related to the issue topic, as well as the end of the post experimental instrument. Message Frame Four news stories were prepared for this study focusing on the consumption of organic or conventional ground beef, emphasizing the risks or benefits of each. This subject matter was selected after a review of literature indicated consumer concerns over the safety and health of food is rising (Tucker, Whaley, & Sharp, 2006), that consumers have concerns about the health and safety of beef (Leiss & Powell, 2004), and that consumers view organically produced foods as a safe and healthy alternative
115 to conventional products (Gregory, 2000; Harper & Makatouni, 2002; Lockie, 2006; Onyango, Hallman, & Bellows, 2007). Eating beef poses both health risks and benefits to individuals. Therefore, each news story included key pieces of information consistent across benefit ver sions and risk versions; both benefit versions contained information about the benefits of CLAs, both risk versions contained information about the risks of E. coli bacteria, and all treatment versions included information about beef consumption. However, in the benefit framed versions of the news articles, one framed organic ground beef as healthier, while the second framed conventional ground beef as healthier, and the hypothetical research highlighted pertained to the amount of CLA in the product. Contrastingly, in the risk framed versions of the articles, one version framed organic ground beef as having a higher risk of containing E. coli, and the other version framed conventional ground beef as having a higher risk of E. coli The news stories were embe dded in two site versions described below. Website Interactivity Chung and Yoo (2008) stated that an interactive website is one where a user is able to interact with the medium itself and is able to exert control over the communication process by completing activities such as clicking on hyperlinks, entering comments, and performing other interactive actions. As noted in Chapter 2, Outing (1998) compiled a list of elements that are found on the ideal interactive website. With this list and Chung and Yoos ( 2008) definition of interactivity in mind, the images of a low interactive site and high interactive site were created (Appendix X). The image of a low interactive website included the message treatment (framed as a risk or benefit of conventional or organic ground beef) as a news story on an online
116 daily metropolitan newspaper website. The article appeared on one page and included a minimal amount of hyperlinks links to other pages on the site, an opportunity to print, save, or share the article with others via email. In this low interactive version of the site, no hyperlinks were embedded in the text and no image accompanied the article. The high interactive website also included the message treatment as a news story on a daily online newspaper website. However, included on the highinteractive version of the site were several hyperlinks within the text of the article; interactive options to share the story with others via social networking sites, print, and email; an image related to the story; and a pod cast link related to the article content which also included an image. Both the high and low interactive versions of websites included the same text and information; only interactivity level and message frames differed. Dependent Variables Attitude Change Attitudes themselves are not directly measurable or visible, and therefore must be inferred by individuals responses that illustrate a state of mind or disposition that has been engaged in (Eagly & Chaiken, 1993). The researcher therefore assumed that attitudes are formed through a process where cognitive learning occurs and individuals subsequently form beliefs. Similarly to Meyers (2008), a semantic differential scale was used to measure subjects attitudes toward organic beef. In order to reflect th e subject of interest in the study, Meyers scale was modified and the final scale measure was a researcher developed scale of 6 semantic differential items; I believe eating beef is safe/unsafe, foolish/wise, healthy/unhealthy, agreeable/disagreeable, n utritious/nonnutritious, a risk to me/a benefit to me. The bipolar adjectives were accompanied by a five point response scale, using Dillmans (2000) suggestion that three of the items be
117 reverse coded to decrease the influence of response layout. Similarly to Kelman (1953) and Woloschuk, Harasym and Temple (2004) this scale was used twice in the study in order to calculate an attitude change among subjects, first in the preexperimental section of the instrument and then again in the post experimental po rtion. An attitude change score was then computed, and then using absolute values the scores were treated as dependent variables. Reliability coefficients were calculated for the initial attitude scale measure and for the post attitude measure, yielding al pha coefficients of .92 and .94 respectively. Information Recall Previous studies have measured information recall by asking subjects to list information recalled (either aided or unaided) after being presented with stimuli (DHaenens, Jankowski, & Heuvelm an, 2004; Davis, Akers, Cepica, Doerfert, Fraze, & Lawver, 2005). Aided recall was of primary interest in this study, and thus similarly to Davis (2003) a multiple choice knowledge test was devised from information contained in the message treatments to m easure aided recall. Information included in the knowledge test was taken from the USDAs published National Organic Program Standards (2008), academic research published and highlighted by the National Cattlemens Beef Association pertaining to CLAs (2007), and information available from the Center for Disease Control and Prevention (CDC) (2010) regarding E. coli O157:H7, and then incorporated into each of the four message treatment frames. The information recall measure contained four items, which were co mbined to create a index which yielded an alpha reliability coefficient of .89.
118 Attribute Variables Need for Cognition Cacioppo and Pettys (1982) need for cognition scale was designed to identify differences among individuals in their tendency to engage in and enjoy thinking (p. 116). The need for cognition scale is comprised of 34 items and has been determined to measure a single factor and be reliable. Cacioppo, Petty, and Kao (1984) published a consolidated 18item version of the NFC scale, which was determined to have one dominant factor that explained 37% of the variance with an alpha coefficient reliability score of .90 (Cacioppo, Petty, & Kao, 1984). This consolidated NFC scale was selected a priori for use in the present study to measure subjects tendency to engage in and enjoy effortful cognitive endeavors (Petty, Cacioppo, 1986, p. 48) due to its shortened length and its high reliability. The 18item scale employs statements about cognitive activity and a five item Likert response scale, where 1=Strongly Disagree and 5=Strongly Agree; 9 items on the scale are reverse coded to avoid response bias. Issue Involvement The level of involvement a person feels in relation to an issue impacts how that person processes issuerelated messages, including what parts of the message are attended to, which in turn affects whether or not and to what degree the person is persuaded by the message (Petty & Cacioppo, 1979). This study did not manipulate issue involvement; however, measuring issue involvement was considered important due to its role in the processing of persuasive messages. It is assumed in this study that members of the population (college students at a large southeastern university) have varying levels of issue involvement with the subject of the study.
119 Issue involvement was measured using Zaichkowskys (1994) Personal Involvement Inventory scale, which is a shortened version of Zaichkowskys (1985) original issue involvement scale. The PII is designed to measure emotional and cognitive involvement in relation to a specific product. The shortened version was selected for use a priori due to its brevity and its reported comparable reliability score; the 20item scale was reported to have an alpha reliability score of .90 and the shortened 10item sc ale reported to have an alpha reliability score of over .90 (Zaichkowsky, 1985; Zaichkowsky, 1994). The 10item scale contained a series of bipolar adjectives measured on a 7point scale (a 5point scale was utilized in the current study). The scale begins with the statement To me [product X] is and is followed by the bipolar adjectives: important/unimportant, boring/interesting, relevant/irrelevant*, exciting/unexciting*, means nothing to me/means a lot to me, appealing/unappealing*, fascinating/mundane *, worthless/valuable, involving/uninvolving*, not needed/needed (* indicates item is reverse coded to avoid response bias). Subjects mark the level of personal relevance along the 5item scale for each semantic differential item. Instrumentation The instr uments for this study were distributed using and online questionnaire program and were designed to be completed in a single session. Several researchers have pointed out benefits of using online or webbased research studies. Gosling and Johnson (2010) not ed that the use of Internet based research studies, specifically experiments, allows for the use of rich multimediain ability tests, questionnaires, surveys, and experiments, creating a more lifelike and ecologically valid environment than ordinarily found in pencil and paper measures (p. 6). Tuten (2010) also noted that
120 online surveys allow for studies including visual and auditory elements, and that particularly in studies investigating concepts related to the Internet using online administration meas ures is the best way to understand the results. Reips and Krantz (2010) wrote about the advantages of Internet based experiments and noted that such studies can be distributed to a large number of participants with relative ease and a low level of effort w hen compared to laboratory research, and is often more cost effective in time, space, and labor. They further noted that Internet based experiments have better generalizability and external validity in terms of findings and greater external validity throug h greater technical variance. Conversely, researchers note potential down sides to online research studies. Tuten (2010) commented about potential sources of error stemming from the use of online research, and noted that researchers must consider their tar get audiences and whether or not online research is appropriate as response rates to online surveys are often lower than with traditional administration methods. Several design choices can influence the degree of measurement errorincluding the survey for mat (scrollable vs. interactive), complexity of design, number of items per screen, detail of instructions, the presence of progress indicators, framing, and the presentation of response options (Tuten, 2010, p. 183). In mitigation of some of these challenges, Reips (2010) suggested mentioning design factors to participants, including the approximate length of time completing the study will take, and including progress indicators as motivators for shorter studies. For this study, online administration of t he experiment was considered highly appropriate due to the interest in collegeaged individuals, who are highly
121 technologically savvy as a group overall, as well as a group who encounters the Internet multiple times a day in their daily lives and who tend to be early adopters of Internet related technologies (Jones, JohnsonYale, Millermaier & SeoanePerez, 2009). Additionally, the subject matter, online delivery of messages on websites with varying levels of interactivity, made online administration of thi s experiment ideal. This study used Reipss (2000) recommendations for online research, including the personalization of written contact with subjects and the use of several reminders to participate. Pilot Study To establish the instruments reliability and validity, a pilot study was conducted with 30 undergraduate students in an agricultural communications course. Precautions were taken to ensure no students participating in the pilot study were part of the final sample; students enrolled in the courses used for pilot testing were crosschecked and it was determined that none of these subjects were enrolled in the courses used for the final sample. Prior to the pilot test, results of the message stimuli testing were incorporated into the message/instrume nt, and a panel of experts assessed the face and content validity of the instrument. Scale Reliabilities Using SPSS 17.0 for Windows, item analyses were calculated to determine the validity of each of the scales utilized. Although Traub (1994) indicated that in the social sciences, reliability coefficients of .70 or larger are acceptable, those with a reliability of .80 or higher are considered more precise and well constructed. A total of four scales were employed in the creation of the experimental ins trument.
122 Zaikowskys (1994) 10 item issue involvement scale, which was developed to measure individuals attitudes towards products, was utilized for the study. This scale had an initial Cronbachs alpha reliability of .75. After removing one item, which had a corrected item total correlation of (.60), the final 9item scale employed on the pilot had a Cronbachs alpha reliability of .80 (Table 3.1). Table 31. The PII Scale used to measure issue involvement. To me, organic beef is Important 1 2 3 4 5 Unimportant* Boring 1 2 3 4 5 Interesting Relevant 1 2 3 4 5 Irrelevant* Exciting 1 2 3 4 5 Unexciting* Means nothing to me 1 2 3 4 5 Means a lot to me Appealing 1 2 3 4 5 Unappealing* Fascinating 1 2 3 4 5 Mundane* Worthless 1 2 3 4 5 Valuable Inv olving 1 2 3 4 5 Uninvolving Not needed 1 2 3 4 5 Needed Note: *Item is reverse coded. Bold indicates item was removed in the final instrument. In this study, prior knowledge of the issue was measured using five researcher developed items. These items were compiled using factual information gathered from the National Cattlemens Beef Association and the USDAs National Organic Program standards about beef production, E. coli O157:H7, and conjugated linoleic acid in beef. Correct answers on each of the i tems were combined to create an index score of prior knowledge. Possible scores ranged from 05. The information contained in this measure was incorporated into each of the experimental treatments so it could also be used as the post experimental measure o f information recall. To measure subjects prior attitude toward the topic, a six item scale used by Wood (2006) and Meyers (2008) was employed (Table 32). The scale used bipolar adjectives on the ends of a 5point Likert scale. Wood (2006) adapted the or iginal index
123 created by Burgoon, Cohen, Miller, and Montgomery (1978) designed to measure attitude about agricultural biotechnology. The original scale had an alpha reliability of .95. The alpha reliability for the scale on the pilot instrument was .95. The same scale was used to measure subjects attitudes toward the subject after the experimental treatment. Table 32. Attitude scale. I believe eating beef is: Safe 1 2 3 4 5 Unsafe Foolish 1 2 3 4 5 Wise* Healthy 1 2 3 4 5 Unhealthy Agreeable 1 2 3 4 5 Disagreeable Nutritious 1 2 3 4 5 Non nutritious A risk to me 1 2 3 4 5 A benefit to me* Note: *Item is reverse coded. To measure need for cognition, Cacioppo, Petty, & Kaos (1984) Efficient Assessment of Need for Cognition Index was used (Table 33 ). The 18item scale yielded a .91 alpha reliability, and thus no items were removed. Preexisting attitude toward the issue/subject was measured using a six item scale that yielded an alpha reliability of .95, so all items in the scale were retained. Manipulation checks were conducted during the pilot study to determine if subjects were able to identify the low interactive/high interactive version of the website, and to determine if subjects could correctly identify the message frames. Instrument Content Th e eight versions of the experimental instrument were identical except for the randomly assigned image of the news story on a website, which constituted the various versions of the message frame and website interactivity level (Appendix G and
124 Appendix H). T hese images were embedded in the Qualtrics questionnaire instrument and appeared as images of websites that would be seen online. The instrument itself Table 33. Efficient assessment of need for cognition. 1. I would prefer complex to simple problems. 2. I like to have the responsibility of handling a situation that requires a lot of thinking. 3. Thinking is not my idea of fun.* 4. I would rather do something that requires little thought than something that is sure to challenge my thinking abiliti es.* 5. I try to anticipate and avoid situations where there is a likely chance I will have to think in depth about something.* 6. I find satisfaction in deliberating hard and for long hours. 7. I only think as hard as I have to.* 8. I prefer to th ink about small, daily projects to long term ones.* 9. I like tasks that require little thought once Ive learned them.* 10. The idea of relying on thought to make my way to the top appeals to me. 11. I really enjoy a task that involves coming up wit h new solutions to problems. 12. Learning new ways to think doesnt excite me very much.* 13. I prefer my life to be filled with puzzles that I must solve. 14. The notion of thinking abstractly is appealing to me. 15. I would prefer a task that is intellectual, difficult, and important to one that is somewhat important but does not require much thought. 16. I feel relief rather than satisfaction after completing a task that required a lot of mental effort.* 17. Its enough for me that something gets the job done; I dont care how or why it works.* 18. I feel relief rather than satisfaction after completing a task that required a lot of mental effort.* Note: *Item is reverse coded. Items measured on a 5point Likert scale where 1 = strongly dis agree and 5 = strongly agree. was displayed as a series of pages in order to minimize the subjects need to scroll up and down on a page, and contained the following: Page 1 included the informed consent information. Page 2 included a prompt to type in th e subjects unique participant identification number, which was used to avoid collecting names in conjunction with the data, but to allow subjects to receive the extra credit incentive for participating. Page 3 contained an item for subjects to indicate which course(s) s/he was enrolled in. This item ensured that subjects enrolled in more than one course received the extra credit in each course but completed the experiment only once.
125 Page 4 contained a description of the items on the screen and included 10 item issue involvement scale. Page 5 contained a four item scale measuring subjects issue knowledge. Page 6 contained a fiveitem scale measuring subjects issue knowledge. Page 7 contained a six item attitude measurement scale. Page 8 contained a threeitem attitude measurement scale. Page 9 included an 18 item need for cognition scale. Page 10 was a transition page that contained directions regarding the images on the next screen. Page 11 included the randomly assigned treatment (which included a message frame and low/high interactive website), as well as timer invisible to subjects that measured time spent on the treatment screen. Page 12 contained an openended item asking participants to use their own words to describe the main idea of the news story on the previous screen as a frame manipulation check. Page 13 contained a multiplechoice item for respondents to indicate the subject focus of the news article s/he read as a manipulation check. o If Levels of conjugated linoleic acid in organic beef was selected as the answer, the following knowledge question was displayed Beef products contain conjugated linoleic acid. True/False o If Levels of E. coli bacteria in traditional beef was select as the answer, the following knowledge question was display ed Ground beef can be contaminated with E. coli bacteria. True/False Page 14 included three subject knowledge questions. Page 15 included a multiplechoice item for subjects to indicate the main theme of the article (frame manipulation check). Page 16 contained the 6item attitude measurement scale. Page 17 contained a 3 item belief measurement scale. Page 18 asked the subject to indicate which form of the message treatment s/he saw, the low interactive or high interactive version. If the subject indicated having seen the high interactive version of the website, the following question was displayed:
126 o If the website on which the article appeared included a comments section, did you read the comments? o If the respondent answered affirmatively, the following question was displayed: o If you read the comments which accompanied the article, did the comments seem to support the theme of the article? Page 19 included two items for subjects to indicate their frequency of Internet use. Page 20 included five openended response items for subjects to list the activities/tasks they perform while on the Internet. Page 21 included a fiveitem question pertaining to time spent on particular activities while online. Page 22 included four demographic items. Page 23 contained a thank you message and a paragraph with debriefing information indicating that the research profiled in the article was only hypothetical, and included links to two separate sites with accurate information on the substances emphasized in the message treatments (CLA and E. coli bacteria). Procedure Subjects (N = 492) were verbally notified of the opportunity to participate in the study by the researcher during an inperson recruitment session/pre notification at the beginning of a regular class meeting period (Appendix A). Following this verbal prenotification, the researcher distributed emails personalized with subjects first names, per Reipss (2000) recommendation, to subjects that included directions and a unique link to the experimental instrument ( Appendix B). The online questionnaire was available for 14 days; subjects received a follow up email five days after the initial email (Appendix C) and a thank you email the day that the questionnaire closed (Appendix D). Each communication included a link to the instrument and directions for completion. Subjects were randomly assigned, using a random number generator, to one of the eight treatment conditions. The combined message and interactive website
127 experimental conditions are shown in Appendix G (high interactive version) and Appendix H (low interactive version). The message frame and site treatment were integrated and appeared to subjects as a screen image of a website. The message frame was embedded in a news story about ground beef, and the experime ntal manipulation was based on the risk or benefit frame contained in the news story. Following exposure to the treatment (subjects were allowed as much time as they needed and were able to move on to the next screen at their leisure), subjects responded t o the items measuring attitude and information recall. Data Analysis Data analysis for this study was completed using SPSS 17.0 for Windows. The researcher used Cronbachs alpha coefficients as a measure of internal consistency reliability, appropriately employed when Likert type items are utilized (Ary, et. al., 2006). Final Instrument Scale Reliabilities The instrument in this study utilized Zaichkowskys PII scale to measure issue involvement. The final scale had a range of scores from 9 to 45 (lowest to highest level of possible issue involvement). Standard deviations ranged from .99 to 1.30, and corrected item total correlations ranged from .61 to .72. The overall alpha reliability coefficient was .90, and thus no items were removed from the scale. C acioppo, Petty, and Kaos (1984) Efficient Assessment of Need for Cognition was employed, and scores ranged from 32 to 89, with standard deviations from .1 to 1.07. Corrected item total correlations ranged from .27 to .62 with an overall alpha coefficient of .86. Attitude was measured in both a preexperimental administration and a post experimental administration, using a researcher developed scale based on Wood
128 (2006) and Meyers (2008) work. The scale had standard deviations from .92 to 1.03, and an overall alpha coefficient of .92. Scores ranged from 6 to 30 (positive to negative), with an overall mean score for prior attitude at 12.32, indicating a moderately positive attitude toward the issue prior to the experimental treatment. Hypothesis Testing To address hypotheses, analysis of variance (ANOVA) tests were run through the general linear model (GLM) function in SPSS ANOVA tests allow for exploration of both main effects and interaction effects when comparing differences between groups (Garson, 2009). Calculating ANOVAs using the GLM also calculates the eta squared value, which explains variance among data, avoiding possible error that may arise in manually calculating the value. Additionally, the results were analyzed using Cohens (1992) classifica tion of effect sizes, .10, .25, and .40, representing small, medium, and large effect sizes, respectively. To address the single research objective, backward stepwise multiple regression was used to explore the relationships among the individual attribute variables, issue involvement, need for cognition, prior knowledge, and prior attitude, and dependent variables of attitude change and information recall. Previous literature indicated that in similar studies these individual attribute variables contributed to changes in the two dependent variables. In the backward stepwise method of multiple linear regression, the predictor variables are loaded into the model and the software calculates the contribution of each of the variables, removing the variable if it does not make a statistically significant contribution to the models prediction of the outcome variable (Field, 2009). The backward method is preferable to the forward stepwise method, due
129 to the possibility of suppressor effects (when a predictor has a s ignificant effect only when another variable is held constant) when using forward stepwise regression. According to Field (2009), the forward selection method is more likely than backward elimination to exclude predictors involved in suppressor effects ( p. 213).
130 CHAPTER 4 RESULTS Overview The purpose of this study was to combine message frame manipulations with website interactivity levels and determine the effects on subjects attitude change and information recall. The message frame treatments include d news stories about hypothetical research studies regarding the risks (prevalence of E. coli bacteria) and benefits (elevated levels of CLAs) of eating conventionally/organically produced ground beef. The messages were delivered on two different types of websites, one high interactive and one low interactive. Using the structure of the ELM model of persuasion, this study manipulated the independent variables of message frame and website interactivity level. Dependent variables of interest were attitude cha nge and the amount of information recalled following message exposure. This chapter provides the data analysis of the research objectives and hypotheses set out in Chapter 3, beginning with sample demographics including need for cognition and issue involvement, then presents the variables of interest including attitude change and information recall. The chapter then includes a discussion of scale reliabilities used to measure the dependent variables. Finally, an overview of the manipulation checks included in the experimental instrument is provided, as well as tests of the hypotheses and a discussion of the single research objective are presented. Descriptive Analysis The online experimental instrument, administered through Qualtrics, contained four pretrea tment instruments containing a total of 45 items, the experimental message treatment, and two post treatment measures, attitude and information recall, as well as
131 demographics containing a total of 30 items. The entire experimental instrument was delivered to a sample of 492 college students. No invalid email addresses were found, and the final sample included 492 students, with 410 subjects completing the experiment. Demographics The demographic items included on the instrument were: age, college affiliati on, year in college, and student status (full or part time). Additional items such as prior knowledge about the subject of beef, issue involvement and media use were also measured. Descriptive analysis revealed that subjects ranged in age from between 18 and 48 years, with a mean of 20 years of age. A total of 15 subjects reported they were age 18 (3.65%), while 21% reported they were 19 years ( n = 87), 30.2% reported 20 years of age ( n = 124), 24% were 21 ( n = 99), 10.9% were 22 ( n = 45), 3.9% were 23, 2.19% were 24, and 3 subjects reported their age as 25 years old. The remaining subjects indicated they were 28 years of age ( n = 2), 26 ( n = 1), 27 ( n = 1), 29 ( n = 1), 31 ( n = 1), 32 ( n = 1), 34 ( n = 1), 45 ( n = 1), 46 ( n = 1), and 48 years of age ( n = 1 ) (see Figure 41). The majority of subjects indicated they were enrolled in the College of Agricultural and Life Sciences (72%, n = 296), Warrington College of Business (9.2%, n = 38), the College of Health and Human Performance (6.3%, n = 26), the Colleg e of Liberal Arts and Sciences (5.1%, n = 21), the College of Public Health and Health Professions (3.1%, n = 13), the College of Journalism and Mass Communications (2.1%, n = 9), two subjects were enrolled in the College of Education, and one in the
132 Fi gure 41. Subjects age distribution. College of Fine Arts (see Figure 42). Subjects were primarily in their junior year of college (40.7%, n = 167), followed closely by seniors (34.3%, n = 141), then sophomores (20.0%, n = 85), while freshman (2.9%, n = 12) and graduate students (0.2%, n = 1) represented the smallest number of subjects, with two individuals not reporting their year in school. The vast majority of subjects were full time students (98.3%, n = 403), with a total of 5 classifying themselves as part time students (1.2%, n = 5), and one individual not reporting full or part time student status.
133 Figure 42. Subjects college affiliation. Figure 43. Subjects year in college.
134 Items regarding Internet use were also included in the demograp hic section of the study. On an average weekday, subjects primarily reported spending between three and four hours per day using the Internet (41.0%, n = 168), followed by between one and two hours (23.7%, n = 97), between five and six hours (20.5%, n = 84 ), more than six hours (13.0%, n = 53), and the least amount reported spending less than an hour per day on the Internet (1.7%, n = 7) (see Figure 44). F igure 44. Subjects reported internet use. The majority of subjects in the study reported that on an average weekend day they spent between three and four hours online (34.6%, n = 142), between one and two hours, (%, n = 124), between five and six hours (30.2%, n = 53), more than six hours (12.4%, n = 51), and the fewest subjects reported spending less than an entire hour online (9.5%, n = 39) (see Figure 4.4).
135 Subjects were asked to indicate the approximate amount of time they spent online performing a variety of tasks. When looking for information, the largest number of subjects reported spending betw een 30 minutes to one hour on an average day (34.6%, n = 142), while the next largest group reported spending between 15 and 30 minutes on the task (24.9%, n =102). Subjects reporting between 1 and 2 hours on searching for information came next (21.7%, n = 89), followed by those spending more than two hours on the task (10.2%, n = 42), and the fewest number of subjects reported spending less than 15 minutes searching for information when online (8.3%, n = 34) (see Figure 45). Figure 45. Subjects reported internet tasks. The majority of subjects reported spending less than 15 minutes reading news (39.8%, n = 163), followed by subjects spending between 15 and 30 minutes on the task (29.8%, n = 122), those spending between 30 minutes and an hour (19.5%, n = 80),
136 between one and two hours (9.2%, n = 38), and the fewest number of subjects indicating they spent more than two hours reading news online in an average day (1.7%, n = 7) (see Figure 45). When asked to indicate how much time they spent online brow sing for fun, the largest number of subjects indicated they spent between 30 minutes to one hour performing this task (34.4%, n = 141), followed by subjects indicating they spent between one to two hours daily on this task (26.1%, n = 107), and subjects who spent more than two hours online browsing for fun (18.8%, n = 77). The fewest number of subjects indicated spending between 15 and 30 minutes browsing for fun (14.6%, n = 60) and fewer than 15 minutes (5.9%, n = 24) online browsing for fun (see Figure 45). Subjects were asked to indicate how much time they spent online performing school related tasks. The majority of subjects indicated spending less than 15 minutes online per day completing academic tasks (39.8%, n = 163), and the second largest majority indicated spending between 15 and 30 minutes on these tasks (29.8%, n = 122), followed by those spending between 30 minutes to one hour (19.5%, n = 80). The fewest number of subjects indicated spending between one and two hours (9.2%, n = 38) and more tha n two hours (1.7%, n = 7) (see Figure 45). Finally, subjects were asked to indicate how much time each day they spent online completing work related tasks. Subjects indicated spending primarily less than 15 minutes per day on these tasks (42.2%, n = 173 ), followed distantly by subjects who spent between 30 minutes and an hour online for work tasks (17.0%, n = 70), and those who spent between one and two hours (14.9%, n = 61). Subjects indicating spending
137 between 15 and 30 minutes each day on work related tasks (14.6, n = 60) and more than two hours (11.0%, n = 45) rounded out the groups (see Figure 45). Individual Attribute Variables Prior to receiving the experimental treatment, subjects completed a series of individual attribute measures: issue involvement, need for cognition, and prior subject knowledge. Issue involvement As described in Chapter 3, Zaichkowskys (1994) Personal Issue Inventory (PII) was used to gauge subjects involvement with the issue. Following the pilot test, one item from the orig inal scale was removed to improve reliability, resulting in a final scale with nine items. Subjects indicated their responses on a 5point Likert scale with bipolar adjectives at each end. Scores on the scale ranged from 9, representing the lowest level of possible involvement with the issue, to 45 at the highest. The scale had a range of standard deviations from .99 to 1.30, demonstrating the amount of variance in the data. Corrected item total correlations on the scale ranged from .61 to .72 (Table 41). The alpha reliability coefficient for the scale overall was = .90 and results indicated that removing items would not improve the coefficient. The grand mean for the issue involvement scale was 28.34. Need for cognition Cacioppo, Petty, and Kaos (1984) Efficient Assessment of Need for Cognition index was employed, utilizing the full 18item scale since no items were removed following the pilot. Subjects indicated their responses to each item on a 5point Likert scale, and possible scores ranged from 18 r epresenting the lowest level of need for
138 Table 41. Personal Involvement Inventory (PII) scale inter item consistency statistics. Mean SD Corrected Item Total Correlation Alpha if Item Deleted Important/Unimportant* 3.10 1.30 .70 .89 Boring/Interesting 3.12 1.11 .62 .89 Relevant/Irrelevant* 3.36 1.17 .72 .89 Exciting/Unexciting* 2.67 1.06 .67 .89 Means nothing to me/Means a lot to me 2.86 1.14 .70 .89 Appealing/Unappealing* 3.34 3.34 .67 .90 Fascinating/Mundane* 2.92 2.92 .61 .90 Worthless/Valuabl e 3.53 1.06 .76 .89 Not needed/Needed 3.44 1.17 .67 .89 Note: Scores based on semantic differential scale with 1 = important and 5 = unimportant. *Item was reverse coded. Valid n = 409. cognition, to 90 representing the highest possible level. Subjects scores on the NFC scale ranged from 32 to 89 (Table 42), and had a range of standard deviations from .1 to 1.07 (Table 43). Corrected item total correlations ranged from .27 to .62, and the overall alpha coefficient for the scale was .86, indicating no potential improvement if items were deleted. Table 42. Means table for need for cognition score. Treatment Group Mean SD n Benefit of organic on a high interactive site 58.56 8.37 55 Benefit of organic on a low interactive site 62.78 8.91 54 Benefit of conventional on a high interactive site 61.86 8.68 56 Benefit of conventional on a low interactive site 58.90 11.13 50 Risk of conventional on a high interactive site 60.39 8.96 49 Risk of conventional on a low interactive site 60.71 7.98 45 Risk o f organic on a high interactive site 61.73 9.03 51 Risk of organic on a low interactive site 61.46 9.39 50 Total 60.81 9.13 410 Note: Scores ranged from 32 (lowest) to 90 (highest). Prior knowledge In this study, prior knowledge of the issue was measur ed using four researcher developed items, which were compiled using factual information gathered from the
139 Table 43. Need for cognition scale inter item consistency statistics. Mean SD Corrected Item Total Correlation Alpha if Item Deleted NFC.1 2.93 1 .01 .43 .85 NFC.2 3.44 .89 .62 .84 NFC.3* 3.56 .90 .61 .84 NFC.4* 3.57 .89 .56 .85 NCF.5* 3.72 .87 .52 .85 NFC.6 2.96 1.00 .40 .85 NFC.7* 3.31 1.00 .45 .85 NFC.8* 3.03 1.04 .37 .85 NFC.9* 2.93 1.01 .39 .85 NFC.10 3.67 .81 .48 .85 NFC.11 3.76 .85 .58 .84 NFC.12* 3.73 .87 .50 .85 NFC.13 3.19 .92 .51 .85 NFC.14 3.40 .99 .50 .85 NFC.15 3.51 .87 .47 .85 NFC.16* 3.08 1.08 .29 .86 NFC.17* 3.59 .94 .46 .85 NFC.18 3.44 .93 .27 .85 Note: Scores based on Likert scale with 1 = strongly disagree and 5 = strongly agree. *Item was reverse coded. Valid n = 410. National Cattlemens Beef Association and the USDAs National Organic Program standards about beef production. Correct answers on each of the items were scored as +1, and the total correct answers on the four items were combined to create an index score of prior knowledge (Table 44). The information on this measure was Table 44. Means table for prior knowledge score. Treatment Group Mean SD n Benefit of organic on a high interactive site 2.73 .80 55 Benefit of organic on a low interactive site 2.87 .89 54 Benefit of conventional on a high interactive site 2.82 .97 56 Benefit of conventional on a low interactive site 2.78 .89 50 Risk of conventional on a high interactive site 3.00 .89 49 Ri sk of conventional on a low interactive site 3.02 .75 46 Risk of organic on a high interactive site 2.96 .72 50 Risk of organic on a low interactive site 3.06 .82 50 Total 2.91 .85 410 Note: Scores ranged from 1 (lowest) to 5 (highest).
140 incorporated into each of the experimental treatments, and was used to create the post experimental information recall measure. Prior attitude To measure subjects prior attitude toward the topic, Woods (2006) six item attitude scale was employed. The scale used bipol ar adjectives and asked subjects to respond on a 5point Likert scale, with scores ranging from a possible 6 at the lowest to 30 at the highest (Table 45). The total scale had standard deviations ranging from .92 to 1.03, and the overall alpha reliability 6). The scale would not have been significantly improved by eliminating any items. The overall mean score for prior attitude was 12.32, indicating a moderately positive attitude toward eating beef prior to the experimental treatment. Table 45. Means table for prior attitude score. Treatment Group Mean SD n Benefit of organic on a high interactive site 12.58 3.73 55 Benefit of organic on a low interactive site 13.68 5.59 53 Benefit of conventional on a high interactive site 11.21 3.43 56 Benefit of conventional on a low interactive site 12.30 3.82 50 Risk of conventional on a high interactive site 11.33 3.17 49 Risk of conventional on a low interactive site 13.36 4.18 45 Risk of organic on a high interactive site 11.71 3.92 51 Risk of organic on a low interactive site 12.50 4.88 50 Total 12.32 4.21 409 Note: Scores ranged from 6 (low) to 26 (high). Table 46. Prior attitude toward organic beef. Mean SD Corrected Item Total Correlation Alpha if Item Deleted Safe/Unsafe 2.0 4 1.03 .74 .91 Foolish/Wise 2.40 .92 .76 .90 Healthy/Unhealthy 2.45 1.03 .79 .90 Agreeable/Disagreeable 2.10 .99 .74 .91 Nutritious/Non nutritious 2.14 .95 .77 .90 A risk to me/A benefit to me 2.39 1.03 .79 .90 Note: Scores based on semantic differential scale with 1 = safe and 5 = unsafe. Valid n = 409.
141 Descriptive Analysis of Variables of Interest Attitude Change Subjects attitude change was measured by utilizing the attitude scale on two separate occasions and calculating the difference in scores (Kelman, 1953; Woloschuk, Harasym, & Temple. 2004). The lowest possible score on each of the attitude scales was 6, and the highest was 30. Table 47 depicts average attitude change scores by treatment group. The mean attitude score following exposure to the experimental treatment was 17.06, indicating a slightly negative attitude toward eating beef. A Cronbachs alpha test of reliability was computed on the post experimental measure scale measuring attitude after exposure to the experimental treatments had standard deviations ranging from .94 to 1.11. The scale would not have been significantly improved by removing items. Change in attitude was calculated by using the post experiment al treatment measure of attitude and subtracting the prior attitude score, yielding a change in attitude score. Scores on attitude change ranged from 7 to 12, with a grand mean of 3.84 (using absolute values). Table 47. Means table for attitude change by treatment group. Treatment Group Mean Change SD n Benefit of organic on high interactive site 3.24 3.24 55 Benefit of organic on a low interactive site 3.06 4.09 53 Benefit of conventional on a high interactive site 4.14 2.91 56 Benefit of convent ional on a low interactive site 3.32 2.92 50 Risk of conventional on a high interactive site 4.87 2.76 48 Risk of conventional on a low interactive site 3.27 3.71 44 Risk of organic on a high interactive site 4.24 2.95 50 Risk of organic on a low inter active site 4.60 3.05 50 Total 3.84 3.27 406 Note: Scores ranged from 0 (lowest no change) to 12 (highest greatest change).
142 Information Recall To measure information recall, the researcher employed a method used by Davis (2003) and created a multiplechoice examination based on information included in the message treatments. The researcher developed measure included four items, which were summed to create an index of subjects information recall. Scores ranged between 0 and 4, with an overall average score of 4 across all eight treatment conditions. Table 4 6 shows average information recall scores by treatment group. Table 48. Means table for information recall score by treatment group. Treatment Group Mean SD n Benefit of organic on high interacti ve site 3.56 .63 55 Benefit of organic on a low interactive site 3.59 .71 54 Benefit of conventional on a high interactive site 3.68 .58 56 Benefit of conventional on a low interactive site 3.44 .84 50 Risk of conventional on a high interactive site 3. 65 .78 48 Risk of conventional on a low interactive site 3.44 .83 45 Risk of organic on a high interactive site 3.45 8.93 51 Risk of organic on a low interactive site 3.54 .79 50 Total 3.55 3.17 409 Note: Scores ranged from 0 (low) to 4 (high). Manip ulation Checks To evaluate the message frame and website interactivity treatments, manipulation checks were conducted. A total of four message treatments were designed: benefit of organic ground beef, benefit of conventional ground beef, risk of conventional ground beef, and risk of organic ground beef. These message frames were then incorporated into two different websites, one with high interactivity and one with low interactivity. This yielded a total of eight treatment conditions: 1) benefit of organic ground beef on a high interactive site, 2) benefit of organic ground beef on a low interactive site, 3) benefit of conventional ground beef on a high interactive site, 4) benefit of conventional ground beef on a low interactive site, 5) risk of convention al ground beef on a high interactive
143 site, 6) risk of conventional beef on a low interactive site, 7) risk of organic beef on a high interactive site, 8) risk of organic beef on a low interactive site. Immediately following the experimental treatment, subjects were asked to describe the main theme of the article on the previous screen, using their own words for the first frame manipulation check. Responses that correctly identified the entire message frame or a portion of the message frame (either the risk /benefit contrast or the conventional/organic ground beef contrast) were coded as +1 and incorrect responses were coded as 0. Overall, 377 of the 410 subjects correctly identified the frame using their own words, while 20 incorrectly identified the frame, and 11 subjects either entered no response or provided other comments unusable in determining whether or not they correctly identified the frame (Table 49). Table 49. Openended message frame manipulation check. Assigned Frame (Treatment Condition) n Co rrect Identification Incorrect Identification No Usable Response Benefit of organic on high interactive site 53 49 0 4 Benefit of organic on a low interactive site 54 52 0 2 Benefit of conventional on a high interactive site 56 53 3 0 Benefit of conven tional on a high interactive site 49 43 6 1 Risk of conventional on a high interactive site 49 47 1 1 Risk of conventional on a low interactive site 45 39 5 1 Risk of organic on a high interactive site 51 50 1 0 Risk of organic on a low interactive sit e 50 44 4 2 Total 408 377 20 11 Note: Valid N = 408.
144 For the second frame manipulation check, a closeended item was employed. Subjects were asked to indicate whether the main theme of the story was a health risk or a health benefit (Table 410). Table 4 10. Closeended message frame manipulation check. Assigned Frame (Treatment Condition) n Correct Identification Incorrect Identification No Usable Response Benefit of organic on high interactive site 56 48 8 0 Benefit of organic on low interactive si te 54 48 6 0 Benefit of conventional on high interactive site 56 54 2 0 Benefit of conventional on low interactive site 50 41 9 0 Risk of conventional on high interactive site 48 44 4 1 Risk of conventional on low interactive site 45 40 5 0 Risk of or ganic on high interactive site 51 49 2 0 Risk of organic on low interactive site 50 47 3 0 Total 410 371 39 1 Note: Valid N = 410. For the website interactivity manipulation check, subjects were asked to indicate which statement on a closeended item best described the website on which the message was displayed (Table 411). Table 411. Website interactivity manipulation check. Assigned Frame (Treatment Condition) n Correct Identification Incorrect Identification Benefit of organic on high interact ive site 55 47 8 Benefit of organic on a low interactive site 54 39 15 Benefit of conventional on a high interactive site 56 49 7 Benefit of conventional on a high interactive site 50 35 15 Risk of conventional on a high interactive site 49 46 3 Risk of conventional on a low interactive site 45 30 15 Risk of organic on a high interactive site 51 45 6 Risk of organic on a low interactive site 50 35 15 Total 410 326 84 Note: Valid N = 410.
145 Hypothesis Testing and Research Objective The following secti on describes the statistical analysis of the hypothesis testing and analysis of the single research objective. Several hypotheses were developed based on the intervening effects of message frame and website interactivity on attitude change and information recall. Hypotheses 1 and 2 address main effects, followed by Hypotheses 3 and 4 which address interaction effects, and then Objective 1, which pertains to regression modeling, is addressed last. For this study, hypothesis testing focused on the two independent variables (message frame and website interactivity level) and attitude change and information recall. An overall means table provides insight into the average attitude change of subjects according to treatment group (Table 412). Table 412. Means ta ble for attitude change by treatment group. Treatment Group Mean Change SD n Benefit of organic on high interactive site 3.24 3.24 55 Benefit of organic on a low interactive site 3.06 4.09 53 Benefit of conventional on a high interactive site 4.14 2.91 56 Benefit of conventional on a low interactive site 3.32 2.92 50 Risk of conventional on a high interactive site 4.87 2.76 48 Risk of conventional on a low interactive site 3.27 3.71 44 Risk of organic on a high interactive site 4.24 2.95 50 Risk of organic on a low interactive site 4.60 3.05 50 Total 3.84 3.27 406 Note: Scores ranged from 7 to 12; absolute values were used in calculating statistics. An overall means table also provides insight into the average information recall of subjects, again according to treatment group (Table 413). H1: Subjects exposed to risk framed messages will have significantly different attitude change and information recall than subjects exposed to benefit framed messages. A one way betweensubjects ANOVA was condu cted to compare attitude change scores for subjects exposed to risk framed messages and those exposed to benefit
146 Table 413. Means table for information recall score by treatment group. Treatment Group Mean SD n Benefit of organic on high interactive si te 3.56 .63 55 Benefit of organic on a low interactive site 3.59 .71 54 Benefit of conventional on a high interactive site 3.68 .58 56 Benefit of conventional on a low interactive site 3.44 .84 50 Risk of conventional on a high interactive site 3.65 .7 8 48 Risk of conventional on a low interactive site 3.44 .83 45 Risk of organic on a high interactive site 3.45 8.93 51 Risk of organic on a low interactive site 3.54 .79 50 Total 3.55 3.17 409 Note: Scores ranged from 0 (low) to 4 (high). framed mes sages. Subjects were divided by treatment groups, those receiving risk framed messages and those receiving benefit framed messages. A significant difference in attitude change scores existed between subjects receiving risk framed messages ( M = 4.57, SD = 2 .69) and those receiving benefit framed messages, M = 3.60, SD = 3.16; F (1, 406) = 11.05, p = .001 (Table 4.14). The effect size, calculated using partial eta squared, was .03, illustrating what Cohen (1988) classified as a small effect, accounting for 3% of the total variance (Table 414). Table 414. One way betweensubjects ANOVA for attitude change differences between subjects receiving risk framed messages and benefit framed messages. n M SD F df p Risk framed messages 192 4.57 2.69 11.05 1 .001 B enefit framed messages 214 3.60 3.16 A one way betweensubjects ANOVA was conducted to compare information recall scores for subjects exposed to the two message frame conditions, risk and benefit. No significant difference in information recall scores existed between subjects receiving risk framed messages ( M = 3.52, SD = .763) and those receiving benefit framed messages. M = 3.57, SD = .693; F (1, 409) = .511, p = .001 ( Table 4.15). The
147 effect size, calculated using partial eta squared, was .001, ill ustrating a negligible effec t ( Table 4 1 5 ). Table 415. One way betweensubjects ANOVA for information recall differences between subjects receiving risk framed messages and benefit framed messages. n M SD F df p Risk framed messages 194 3.52 .763 .511 1 .001 Benefit framed messages 215 3. 57 .693 H2: Subjects exposed to messages on a high interactive website will have significantly different attitude change and information recall than subjects exposed to messages on a low interactive website. To t est the second hypothesis, a oneway betweensubjects ANOVA was conducted to determine whether a significant difference in attitude change score was present between subjects receiving the experimental message on a high interactive site and those receiving it on a low interactive site. The results indicated that no significant differences in attitude change scores existed between subjects receiving the experimental treatment on the high interactive website ( M = 4.10, SD = 3.01) and the low interactive websit e M = 3.56, SD = 3.50; F (1, 409) = 2.71, p = .10 (Table 416). The effect size, calculated using eta squared, was .007, illustrating what Cohen (1988) classified as a small effect. Table 416. One way betweensubjects ANOVA for attitude change differenc es between subjects receiving messages on high interactive websites and low interactive websites. n M SD F df p High interactive site 209 4.10 3.01 2.71 1 .10 Low interactive site 197 3.56 3.50 A one way betweensubjects ANOVA was conducted to determine whether a significant difference in information recall existed between subjects receiving the
148 experimental messages on a highinteractive site and those receiving them on a low interactive site. The results indicated that no significant difference in information recall scores existed between subjects receiving the experimental treatment on the high interactive website ( M = 3.59, SD = .660) and the low interactive website M = 3.51, SD = .791; F (1, 409) = 1.18, p = .28 (Table 417). The effect size, calculated using eta squared, was .007, illustrating what Cohen (1988) classified as a small effect. Table 417. One way betweensubjects ANOVA for information recall differences between subjects receiving messages on high interactive websites and low inter active websites. n M SD F df p High interactive site 210 3.59 .660 1.18 1 .28 Low interactive site 199 3.51 .791 H 3 : Subjects receiving risk framed messages on a high interactive site will have significantly less information recall than subjects receiving the benefit framed message on a low interactive site. To test the third hypothesis in the study, oneway betweensubjects ANOVA was conducted to determine if significant differences in information recall score existed between subjects receiving the risk framed message on a high interactive site and those receiving the benefit framed messages on the low interactive site. No significant differences existed between these groups of subjects. Subjects exposed to risk framed messages on a high interactiv e site ( M = 3.55, SD = .718) and those receiving benefit framed messages on a low interactive site, M = 3.52, SD = .78; F (1, 203) = .06, p = .80 ( Table 4 1 8 ). Effect size was calculated using partial eta squared, resulting in .00, indicating no effect. H 4 : Subjects receiving the risk framed message on a high interactive website will have significantly greater attitude change than subjects receiving the benefit framed message on a low interactive website.
149 Table 418. Independent samples t test for inform ation recall differences between subjects receiving risk framed messages on a high interactive website and subjects receiving benefit framed messages on low interactive websites. n M SD F df p Risk messages on high interactive site 99 3.55 .72 .06 1 .80 Benefit messages on low interactive site 104 3.52 .78 To test H2, oneway betweensubjects ANOVA was conducted to determine if significant differences in attitude change between subjects receiving risk framed messages on a high interactive website and those receiving benefit framed messages on a low interactive website. Results illustrated significant differences in attitude change scores did exist between subjects receiving risk framed messages on a high interactive site ( M = 4.55, SD = 2.86) and su bjects receiving benefit framed messages on a low interactive site M = 3.18, SD = 3.56; F (1, 201) = 8.97, p = .00 (Table 419). The effect size was calculated using partial eta squared and was .04, accounting for 4% of the variance. Table 4 19. Independe nt samples t test for attitude change differences between subjects receiving risk framed messages on a high interactive website and subjects receiving benefit framed messages on low interactive websites. n M SD F df p Risk messages on high interactive s ite 98 4.55 2.86 8.97 1 .00 Benefit messages on low interactive site 103 3.18 3.56 Objective One: Explain changes in attitude and information recall based on the individual attribute variables of issue involvement, need for cognition, prior attitude and prior knowledge of the issue.
150 To address the objective, backward stepwise multiple linear regression was conducted using individual attribute variables issue involvement, need for cognition and prior knowledge, and dependent variables attitude change and information recall. Table 4 20 displays the results of the regression model analysis and illustrates that 69% (adjusted R2) of the variance in attitude change. Of the individual predictors, only one was significant (prior attitude, p = .00) and the bet .83) indicates a negative relationship between this variable and attitude change. Table 4 20. Backwards multiple linear regression analysis for variables predicting attitude change. Attitude change Adjusted R 2 Explana tory Variable B SE B p value .690 (Constant) 12.78 .833 .020 .000 Issue Involvement .009 .012 .020 .473 Need for Cognition .005 .010 .014 .626 Prior Attitude .645 .022 .830 .000 Prior Knowledge .156 .112 .041 .165 Table 421 displays t he results of the regression model analysis and illustrates that 11% (adjusted R2) of the variance in information recall. Of the individual predictors, two were significant (need for cognition, p value = .025; prior knowledge, p value = .000) and the beta weights for need for c indicate a positive relationship between these variables and information recall. Table 4 21. Backwards multiple linear regression analysis for variables predicting information recall. Information recall Adjusted R 2 Explanatory Variable B SE B p value .110 (Constant) 2.10 .295 .000 Issue Involvement .008 .004 .085 .074 Need for Cognition .009 .004 .107 .025 Prior Attitude .005 .008 .028 .553 Prior Knowledge .261 .040 .305 .000
151 CHAPTER 5 CONCLUSION Overview This study explored the effects of combining message framing and website interactivity levels on individuals attitude change and recall of information using an experimental design. The purpose of the study was to determine if frami ng messages about ground beef using risk/benefit framing and delivering the messages on a website with either high or low levels of interactivity would affect subjects attitude regarding the issue and the amount of information recalled from the message tr eatment. Two types of messages were created for use in this study, risk framed messages and benefit framed messages. The issue context was ground beef and the messages were constructed to promote the risks or benefits of conventional or organic ground bee f. Messages were then incorporated into two different types of websites, one high in interactivity and one with a low level of interactivity, to test the combined effects of these concepts. A convenience sample of college students was used and subjects wer e randomly assigned to receive one of eight different treatments: benefit of organic beef on a high interactive site, benefit of organic beef on a low interactive site, benefit of conventional beef on a high interactive site, benefit of conventional beef on a low interactive site, risk of conventional beef on a high interactive site, risk of conventional beef on a low interactive site, risk of organic beef on a high interactive site, risk of organic beef on a low interactive site. To determine the effects o f the experimental treatments, attitude change and information recall, an attitude scale and information recall measure was employed.
152 Chapter 4 discussed data analyses and results from 410 subjects. The mean age of subjects was 20 years old and the sample was primarily composed of upper division students enrolled in the College of Agricultural and Life Sciences. The subjects reported spending between three and four hours on the Internet on an average day, and spent the least amount of their online time reading news and performing school related tasks. The largest number of subjects reported spending between 30 minutes and an hour looking for information and browsing the Internet for fun. Overall, subjects indicated they were frequent and experienced users o f the Internet. Key Findings Several individual attribute variables were measured in this study. Subjects overall had a moderate amount of personal involvement with the issue of organic beef; the sample had an average score of 28.34 out of 49 on the scale. Subjects generally had an above average need for cognition level; 60 out of a possible 90. Prior knowledge of the subject was relatively high across treatment conditions, with an overall average of 2.91 out of 5. Prior attitudes about the topic were moder ately positive; with an overall average score of 12.32 out of 30 (scores closer to 1 indicated a positive attitude and those closer to 30 indicated negative attitudes). A total of four hypotheses and one research objective were addressed in this study. Th e first hypothesis tested for differences in attitude change between subjects receiving risk framed message treatments and those receiving benefit framed message treatments, and revealed that indeed significant differences between the groups existed; subjects exposed to risk framed messages had a greater attitude change. Hypothesis one also tested differences in information recall between these two groups of subjects. No significant differences existed between these groups with respect to the amount of
153 info rmation recalled. Subjects in both groups had relatively high information recall scores. This may be partially accounted for by the relatively high prior knowledge scores, which in turn may have been impacted by the brevity of the measure used. The second hypothesis addressed differences in attitude change and information recall between subjects receiving the message treatments on high and low interactive websites. Results of this hypothesis indicated that no significant differences existed between the two groups in either attitude change or information recall. Hypothesis three predicted that subjects receiving risk framed messages on high interactive sites will have significantly lower levels of information recall than those receiving benefit framed messages on low interactive sites. Results indicated that no significant differences existed between these two groups. The fourth hypothesis stated that significant differences in attitude change would exist between subjects receiving risk framed messages on high interactive sites and those receiving benefit framed messages on low interactive sites. The researcher failed to reject this hypothesis and discovered that subjects receiving risk framed messages on the high interactive site had significantly greater atti tude change than those receiving the benefit framed messages on a low interactive website. The single research objective used multiple regression to explain changes in attitude and information recall based on issue involvement, need for cognition, prior at titude, and prior knowledge of the issue. Results of regression modeling indicated that the combination of issue involvement, need for cognition, prior attitude and prior knowledge explained 69% of the variance in the data for attitude change among subject s. However, only one statistically significant predictor emerged when evaluating
154 effect on attitude change prior attitude. When evaluating information recall, the model combining issue involvement, need for cognition, prior attitude, and prior knowledge explained just 11% of the variance in scores. Two individual predictors emerged as significant, need for cognition and prior knowledge. Conclusions and Implications The results of this study offer interesting implications for future research. The theoretic al and researchbased implications are presented first and are followed by a brief explanation of practical implications based on the effects of message framing and website interactivity described above. Theoretical Implications Previous research has illus trated the impact of message framing (Kahneman & Tversky, 1979; Meyerowitz & Chaiken, 1987; Zillmann, et al., 2004) and of website interactivity (Eveland & Dunwoody, 2002; Tremayne & Dunwoody, 2001) in affecting attitudes and recall of information. This st udy explored the combination of message frame and website interactivity and showed that, with respect to attitude change and information recall, only attitude change was significantly impacted by the experimental treatment. The first four objectives of the study aimed at determining differences between treatment groups based on message frame and website interactivity on the two dependent variables of attitude change and information recall. Results indicated that message framing impacted only attitude chang e, not information recall, and that website interactivity level did not produce an effect on either outcome variable. There are several possible contributing factors to these outcomes.
155 With respect to the message framing results, it is possible that the f ear element included in the risk framed message frame was effective in persuading subjects that organic ground beef is the safe option to avoid potential risk. Risk messages in this study discussed the possibility of the presence of E. coli O157:H7 in one or the other types of ground beef (organic or conventional) highlighted in the message. Given what we know about the impact of fear on message processing (Janis & Feshbach, 1954; LaTour & Rotfeld, 1997; Nabi, 1999), it is possible that subjects were capt ivated by this information. Research about consumer purchasing of organic products indicates that one of the primary reasons Americans purchase organic foods is because of a perceived food safety value (Hughner, McDonagh, Prothero, Schultz, & Stanton, 2007; Lockie, 2006; Onyango, Hallman, Bellows, 2007). It is possible that the combination of evidence of a risk and a way to alleviate that risk may have affected a change in attitude. The possibility that one might get sick from a food product often leads con sumers to take steps to eliminate that risk, such as purchasing organic products which they view as safer. Interestingly, no differences in attitude change emerged between groups receiving message treatments on the high interactive site versus the low interactive site. In this study, the interactivity manipulation was not effective in contributing to attitude change. A potential contributing factor to these results may have been the operationalization of interactivity employed in the study; subjects were no t asked to engage with the material by clicking on any of the hyperlinks or entering comments after having read the message treatments. It seems that the presence of interactive elements, rather than the ability of subjects to actually use the interactive elements, was a hindrance in this study.
156 Neither the message frame manipulation nor the website interactivity manipulation produced any effect on information recall in this study. There are several possible reasons for this outcome. First, the recall meas ure employed was fairly short; only four items. The overall instrument in this study contained upwards of 60 items for subjects to respond to, and thus an effort was made to keep the number of items on each measure as brief as possible. Ideally, the inform ation recall measure would have included more items and may have then been able to illustrate more effectively whether or not differences existed. Also, the instrument used only aided information recall items. It is possible that the use of unaided inform ation recall measures could have helped possible differences emerge between treatment groups. Individuals participating in this study had a relatively high level of prior knowledge regarding the subject, creating a ceiling effect; the overall prior knowledge score average was 2.91 (out of a possible 5). As the post experimental measure of information recall only included 4 items, this high level of prior knowledge may have precluded an effect on information recall, regardless of treatment. Ary, et al. (2006) state that in experimental research, effects of treatments can be underestimated because the test does not accurately measure subjects achievement at the upper end of the range. In this study, the ceiling effect is evident. Discuss prior attitude scores a significant difference among the treatment groups (conducted post hoc) existed on prior attitude; randomization of experimental design was intended to address this, but given the predominance of CALS majors, it may be that these students have prior experiences and attitudes more than those students in other colleges
157 A final possible contributing factor to the lack of effect of the interactivity manipulation is the operational definition of interactivity employed in this study. While efforts were made when designing the website image containing the treatment to make the interactive site as interactive as possible by including hyperlinked text items, images of links to share the material on the site, as well as images supporting the text of the news arti cle, in the end the treatment was delivered as a noninteractive, static image rather than on a live website where subjects could truly interact with the material. The decision to utilize an image of a website rather than a live website was made to prevent a loss of focus among subjects during the already long experimental experience. Subjects spent an average of 15 minutes completing the entire study, and a live website may have prolonged subjects time spent on the endeavor, or have contributed to distrac tion by allowing them to follow actual hyperlinks or be impacted by other material not created specifically for this study. Of the two hypotheses tested in this study, only the second was significant. Subjects having received the treatment message framed as a risk on a high interactive website had significantly different attitude change than those who received the benefit framed messages on a low interactive website. Due to the lack of impact of the website interactivity manipulation in objective three, it is possible that the message frame manipulation was responsible for the majority of the effect on attitude change. Once again, the operationalization of interactivity may have contributed to this outcome. The combination of issue involvement, need for cog nition, prior attitude and prior knowledge of the subject together explained a significant portion of the variance in the data for attitude change among the subjects in this study. However, when examining the
158 individual predictors, prior attitude emerged a s the only significant individual predictor of attitude change. This is not unexpected; as previous research illustrates, we are primed by our own experiences (Petty & Jarvis, 1996) and by the media (Roskos Ewoldsen, RoskosEwoldsen, & Carpentier, 2008). P etty (1981) noted the importance of personal relevance of an issue as the primary determining factor as to whether one is persuaded by a message and whether or not ones attitude is changed as a result. However, Petty also noted that when an issue is perso nally relevant to us, we are more likely to attend to the issue, and so it is interesting to note that issue involvement did not emerge as a significant predictor of attitude change. In evaluating the combination of the individual attribute variables and their impact on information recall, two individual predictors emerged need for cognition and prior knowledge. Although together the individual attribute variables (issue involvement, need for cognition, prior attitude and prior recall) only accounted for a small portion of the variance in the data, it is not surprising that need for cognition and prior knowledge emerged as significant individual predictors of information recall. Previous researchers have investigated need for cognition and determined that individuals with higher need for cognition scores are more likely to recall information than those with lower scores and that often, differences in individuals with high and low need for cognition scores are impacted by individuals past experiences (Caci oppo, et al., 1996). Practical Implications Subjects in this study reported spending quite a small amount of time reading news on the Internet, and thus the presentation of the experimental treatment as a news story may have been less appropriate than if i t were presented as a blog entry, opinion piece or other information package. Practitioners may benefit from investigating what
159 their specific audiences look for when exploring the Internet for information or reading news. Being able to tailor not only the message to an audience, but also the method and presentation of the message delivery may be important. Past research has indicated that website interactivity can have an impact on persuasion (Sicilia, Ruiz, & Munuera, 2005; Sundar & Kim, 2010; Sundar, K alyanaraman, & Brown, 2003). Perhaps in this study, the operational definition of interactivity was insufficient to allow true interaction to occur since the message material was delivered on a static image of a website, rather than on a live dynamic sit e, as noted above. However, those delivering persuasive messages via the Internet may take away from this research study the fact that significant differences in information recall did not exist between those receiving message treatments on high versus low interactive sites. While some previous research may have indicated that online delivery of messages may inhibit information recall, interactive websites as conceptualized in this study did not negatively impact what information subjects remembered from the messages they read. Practitioners designing messages for delivery via the Internet or online mediums may take this as an indication that elements of interactivity can be used without fear of distracting readers to the point they do not recall important m essage elements. Recommendations For Future Research Several recommendations for future research can be made following the execution of this study. First, several researchers have carried out studies which evaluate subjects need for cognition levels and s plit them into two groups, high and low, and then explore differences in message effects such as attitude strength and source
160 credibility. An extension of the present study could be conducted using high and low need for cognition subjects to explore issue involvement and attitude change. Are subjects who have high levels of need for cognition more or less likely to change their attitude, and does level of issue involvement have an impact? In future studies exploring information recall, a strong recommendat ion is to considerably lengthen the information recall measure as well as create a more rigorous instrument, possibly in combination with a ramping up of the operational definition of interactivity and to use a live, dynamic website to explore information recall. A study such as this might employ website tracking software to determine whether or not subjects moved beyond the treatment pages and material and to what sites. Conclusions might also be drawn about information seeking behavior in such a study. T his study did not ask subjects to consciously reflect on their prior experience with the issue. As we know, attitude is comprised of several elements including experience and prior knowledge, and it may be instructive, especially when including an element of fear or risk in the treatment message, to ask subjects directly about their previous experiences and knowledge. A final suggestion for future research is to explore the level of media use reported by subjects and possible interactivity effects. Again, this requires a more dynamic definition of interactivity be employed, however it is possible that as our society moves increasingly toward online environments for not only information, but for many of our daily tasks, we may be less affected by the potenti al distractions of interactivity. Especially among the digital native population in this study, it is possible that the human brain adapts and ignores interactivity as a distraction.
161 For Practice From the results of this study, it seems as if those who c reate messages intended to persuade can feel free to employ interactive functions as they wish without fearing negative consequences. However, it is important again to realize that definitions and perceptions of the concept of interactivity vary widely. F or educators, it continues to be important to teach and foster critical thinking skills as they relate to need for cognition among todays college students. Mass media will likely continue to employ the problem frame and elements of risk and fear when co vering news events, which have been shown to be persuasive. This segment of the population, especially due to their status as future leaders, must learn how to evaluate messages critically. It may also be constructive for communicators at all levels to g ain a better understanding of how students in this Web 2.0 world decide what is and is not news in their minds. While college students and young people in general are not large consumers of traditional news information, there is no doubt that when online, they are exposed to information, and they do make judgments about that information, deciding whether or not to incorporate it into their schemas (albeit probably subconsciously) and deciding whether or not it is important to them. A final recommendatio n, specifically for the field of higher education, is to design a curriculum that helps students recognize persuasive messages in the online world. While it is not possible to fully and critically elaborate on each and every message we are exposed to on a daily basis, it is important for students especially to be able to deconstruct messages and learn to think critically about the many persuasive messages they are exposed to each day.
162 Limitations While this study offers insight into message framing and webs ite interactivity for both researchers and practitioners, it includes some limitations that need to be considered. First, this study employed a one time only exposure to the experimental treatment. Research indicates that attitudes, especially those that are strongly held at either end of the continuum, are more difficult to change (Eagly & Chaiken, 1993; Krosnick & Petty, 1995; Perloff, 2003) and therefore the results of this study may only indicate a temporary change in attitude. Attitudes are presumed to be predictive of behavior, however, several other factors play in to attitude strength and attitude change; this is one focused study. Another shortcoming of the design of this study is that the use of a pretest post test design provides no way to evaluate the effect that the pretest might have on the results of subjects scores on the post test. Ary, et a;. (2006) specifically pointed out that pretesting subjects in an experiment is a challenge in studies regarding attitude change, where the pretest m ay cause subjects to think about the subject and motivate them to change their opinions about the topic merely by thinking about it, regardless of the treatment itself. Secondly, while the use of college students is often considered a study limitation, in this situation it was not the use of students that was seen as a limitation due to the specific research interest in this age group. The use of a convenience sample of subjects, however, is a limitation. Ary et al. (2006) noted that convenience sampling is the weakest of all available procedures. In this study, subjects were recruited from identified courses in one college of the university. However, efforts were made to recruit a large sample of students enrolled in a variety of majors. Courses identified and used for recruitment in this study were purposefully selected for their size (each course
163 enrolled a minimum of 100 students, one course had an enrollment of 400) and for their status as courses that met general education requirements, which tends to y ield students enrolled in a variety of majors. Once subjects elected to participate in the study, they were randomly assigned to treatment groups in an effort to help address this limitation. Another limitation involved in the execution of this study pert ains to its exploratory nature. No previous studies were found to have combined the concept of message framing and website interactivity; therefore, this study was modeled after elements of several different successful previous studies. As a result, the co mprehensive instrument (which included four preexperimental measures, the experimental treatment itself, two post experimental measures and a number of demographic items) was quite long and included a total of 83 items for which subjects were asked to provide a response. As acknowledged above, the information recall measure employed in this study was quite short, especially when compared with the length of the entire instrument. It seems that the brevity of this measure may have contributed in part to the findings pertaining to information recall. Also as acknowledged previously, the concept of interactivity used in this study did not allow for presentation of the message treatments on a live, dynamic website. It is possible that since subjects were not able to employ or explore the elements of interactivity that were present (for example, they were not able to click on the hyperlinked text or view the podcast which accompanied the article), this impacted the results of the interactivity manipulation.
164 Concl usion This study centered on combining the theory of message framing, specifically messages framed as risks and benefits, and the concept of website interactivity, which is increasingly important as we move ever further into our Web 2.0 world, and evaluati ng their impacts on attitude change and information recall. The results suggest that while message framing can be effective in persuading individuals to change their reported attitudes, the combination of website interactivity does not appear to create an effect. Information recall as a result of the experimental treatments (both the message frame and the website interactivity level) did not appear to differ significantly, and thus presumably the experimental treatments did not have an impact. Despite these findings, the study did confirm that issue involvement, prior knowledge, prior attitude and need for cognition play a part in whether or not an individual changes his or her attitude.
165 APPENDIX A VERBAL RECRUITMENT S LIDE
166 APPENDIX B FIRST EMAIL CONTACT Dear First : I am conducting a study about beef products and would like to gather your opinions about them through an online survey. Your instructor has agreed to offer you extra credit in class for participati ng in this study. The survey will take approximately 15 minutes of your time. To participate in the survey, you will need to have your unique participant ID number handy, which is ID Please take extra care to type in the correct ID number so that you get your extra credit. Do NOT delete this email if you think you might want to take the survey at a later time this email contains your unique participant ID number. The reminder emails you receive may not contain this information requi red to participate. This is an online survey. Click or copy and paste the following link into your Internet browser: Link. The link will be active through 11:59 p.m., Monday, October 11. If you have questions or problems accessing t he survey, please email me at firstname.lastname@example.org or call 352392 0502, ext. 238. Follow this link to the survey: Link Or, copy and paste the URL below into your Internet browser: Link Thank you, Karen Cannon Graduate Assistant University of Florida
167 APPENDIX C SECOND EMAIL CONTACT REMINDER Dear First : I am conducting a study about beef products and would like to gather your opinions about them through an online survey. Your instruc tor has agreed to offer you extra credit in class for participating in this study. The survey will take approximately 15 minutes of your time. To participate in the survey, you will need to have your unique participant ID number han dy, which is ID Please take extra care to type in the correct ID number so that you get your extra credit. Do NOT delete this email if you think you might want to take the survey at a later time this email contains your unique parti cipant ID number. The reminder emails you receive may not contain this information required to participate. This is an online survey. Click or copy and paste the following link into your Internet browser: Link. The link will be activ e through 11:59 p.m., Monday, October 11. If you have questions or problems accessing the survey, please email me at email@example.com or call 352392 0502, ext. 238. Follow this link to the survey: Link Or, copy and paste the U RL below into your Internet browser: Link Thank you, Karen Cannon Graduate Assistant University of Florida
168 APPENDIX D FINAL EMAIL CONTACT THANK YOU Dear Students: The extra credit survey about beef products for AEC 3030, AEC 3033, and AEC 3414 is now closed. Thank you to all of you who participated I greatly appreciate your time and effort! In the next week, I will be submitting the list of participants to your instructors to ensure that you receive extra credit for partici pating. I ask for your understanding and patience with this process as it is quite a big task! Your instructors will have the list no later than Monday, November 8. Many thanks once again to those who chose to participate! Cheers, Karen Cannon Graduate Assistant University of Florida
169 APPENDIX E INSTRUMENT
181 APPENDIX F MESSAGE TREATMENTS Treatment 1 BENEFIT of organic Headline: Study finds organic cattle produce healthier ground beef Gain esville, Fla. Researchers from the University of Florida discovered that ground beef from certified organic cattle is healthier than from cattle raised using traditional production methods (where cattle can be fed food produced with pesticides, genetical ly modified organisms, animal byproducts, as well as given supplemental antibiotics and growth hormones). The study, conducted by meat science professor Dr. John Klaussen, found that ground beef from certified organic cattle, which are fed only organic f eed (produced without the use of pesticides or genetically modified organisms and cannot contain animal byproducts) and given no supplemental antibiotics or growth hormones, had significantly higher levels of conjugated linoleic acid (CLA) when tested in a laboratory. Beef from cattle raised using conventional production methods had significantly lower levels of this beneficial acid. CLA is a naturally occurring trans fat found in food products from ruminant animals, including beef and lamb, as well as in cheese and milk. CLA is beneficial to human health, has been shown to reduce the growth of cancer cells in laboratory tests, and is considered a powerful anti carcinogen. Previous research indicates that consuming an adequate amount of CLA early in life m ay have long lasting beneficial effects on lowering an individuals risk of cancer. The average American consumer eats just over 60 pounds of beef each year. Dr. Klaussen believes this research is a breakthrough in support of the idea that organic ground beef is healthier and more beneficial for consumers. Finally we have scientific proof that organically raised cattle yield a product that is healthier for us, said Dr. Klaussen. This means that customers can feel confident that eating organic ground bee f is not only a good choice, but a healthy choice. Comment 1 More objective evidence illustrating that organic products really are better for us. Even the scientists are saying it now. Comment 2 Glad to know this research is being done! More support for living the organic lifestyle and all its benefits. Comment 3 I wonder if this is only good news for ground beef, like the article says, or for all beef. Comment 4 Proof that food produced without pesticides, GMOs, antibiotics and growth hormones are good for you! Eat organic!
182 Treatment 2 BENEFIT of conventional Headline: Study finds traditionally raised cattle produce healthier ground beef Gainesville, Fla. Researchers from the University of Florida discovered that ground beef from cattl e raised using traditional production methods (where cattle can be fed food produced using pesticides, genetically modified organisms, and animal byproducts, as well as given supplemental antibiotics and growth hormones) is actually healthier than from ce rtified organically raised cattle. The study, conducted by meat science professor Dr. John Klaussen, found that ground beef from traditionally raised cattle had significantly higher levels of conjugated linoleic acid (CLA) when tested in a laboratory. B eef from cattle raised using organic production methods had significantly lower levels of this beneficial/healthy acid. CLA is a naturally occurring trans fat found in food products from ruminant animals, including beef and lamb, as well as in cheese and milk. CLA is beneficial to human health, has been shown to reduce the growth of cancer cells in laboratory tests, and is considered a powerful anti carcinogen. Previous research indicates that consuming an adequate amount of CLA early in life may have long lasting beneficial effects on lowering an individuals risk of cancer. The average American consumer eats just over 60 pounds of beef each year. Dr. Klaussen believes this research is a breakthrough in support of the idea that ground beef from traditionally raised cattle is healthier and more beneficial for consumers. Finally we have scientific proof that traditionally raised cattle yield a product that is healthier for us, said Dr. Klaussen. This means that customers can feel confident that eating traditionally raised ground beef is not only a good choice, but a healthy choice. Comment 1 More objective evidence illustrating that traditionally raised products are good for us. You dont have to eat organic to be healthy. Comment 2 Glad to know thi s research is being done! More support for consumers who dont want to pay the higher prices of organic products. Comment 3 I wonder if this is only good news for ground beef, like the article says, or for all beef. Comment 4 I didnt know about this CLA stuff. Sounds like I need to eat more beef.
183 Treatment 3 RISK of conventional Headline: Study finds risk of E. coli contamination higher in traditionally produced ground beef Gainesville, Fla. Researchers from the University of Florida discovered that ground beef from cattle raised using traditional production methods (where cattle can be fed food produced with pesticides, genetically modified organisms and animal byproducts, as well as given supplemental antibiotics and growth hormones) have hig her levels of the bacteria E. coli O157:H7 than organically raised beef. The study, conducted by meat science professor Dr. John Klaussen, found that ground beef from cattle raised using traditional production methods had higher levels of this deadly strain of E. coli when tested in a laboratory. Beef from cattle raised using certified organic production methods had significantly lower levels of the bacteria. E. coli bacteria are found in the gut of several animals, including humans. This specific strain of the bacteria, E. coli O157:H7, found often in the gut of ruminant animals like cattle, is harmful to humans and causes vomiting, abdominal pain, bloody diarrhea. In severe cases, this strain of bacteria can cause kidney disease and even death. The average American consumer eats just over 60 pounds of beef each year. Dr. Klaussen believes that this research is a breakthrough in proving the risk of E. coli bacteria in cattle raised using traditional methods. Finally we have scientific proof that traditi onally raised cattle produce ground beef that is less safe than certified organically raised cattle, said Dr. Klaussen. This means that the risks associated with eating traditionally raised beef have been documented with scientific evidence. Comment 1 Good objective evidence illustrating that organic products really are safer to eat. Even the scientists are saying it now. Comment 2 Glad to know this research is being done! Nice to see scientists admitting that nonorganic products are riskier to eat. Comment 3 I wonder if this is only good news for ground beef, like the article says, or for all beef. Comment 4 Proof that food produced without pesticides, GMOs, antibiotics and growth hormones pose less of a risk! Eat organic!
184 Treatment 4 R ISK of organic Headline: Study finds risk of E. coli contamination higher in organic ground beef Gainesville, Fla. Researchers from the University of Florida discovered that ground beef from cattle raised using certified organic production methods (wher e cattle cannot be fed food produced with pesticides, genetically modified organisms or animal byproducts, nor given supplemental antibiotics and growth hormones) have higher levels of the bacteria E. coli O157:H7 than traditionally raised beef. The study conducted by meat science professor Dr. John Klaussen, found that ground beef from cattle raised using certified organic production methods had higher levels of this deadly strain of E. coli when tested in a laboratory. Beef from cattle raised using tra ditional production methods had significantly lower levels of the bacteria. E. coli bacteria are found in the gut of several animals, including humans. A specific strain of the bacteria, E. coli O157:H7, found often in the gut of ruminant animals such as cattle, is harmful to humans and causes vomiting, abdominal pain, bloody diarrhea. In severe cases, this strain of bacteria can cause kidney disease and even death. The average American consumer eats just over 60 pounds of beef each year. Dr. Klaussen bel ieves that this research is a breakthrough in proving the risk of E. coli bacteria in cattle raised using certified organic methods. Finally we have scientific proof that organically raised cattle produce ground beef that is less safe than traditionally r aised cattle, said Dr. Klaussen. This means that the risks associated with eating organically raised beef have been documented with scientific evidence. Comment 1 Interesting study. Guess it shows us that traditional products really are ok to eat r ather than paying more for organic. Comment 2 Glad to know this research is being done! More support for consumers who dont want to pay the higher prices of organic products. Comment 3 I wonder if this is only good news for ground beef, like the art icle says, or for all beef. Comment 4 I didnt know E. coli could kill a person, just thought it gave you a bad stomach ache.
185 APPENDIX G HIGH INTERACTIVE WEB SITE TREATMENT EXAMP LE
186 APPENDIX H LOW INTERACTIVE WEBS ITE TREATMENT EXAMPL E
187 LIST OF REFE RENCES Adams, W. (1986). Whose lives count? TV Coverage of natural disasters. Communication, 36(2), 113122. Adelson, A. (1993). Jack in the Box Franchisees Cite Parent Concern in Suit. The New York Times 3. Al Obaidi, J. A., Lamb Williams, C., & Mordas, V. (2004). The king of all mediums: A field study of college students use of mediums for news. International Journal of Instructional Media, 31 (3), 239256. Allport, G. W. (1935). Attitudes. In C. Murchison (Ed.), Handbook of Social Psychology (pp. 79888 4). Worcester, MA: Clark University Press. Altheide, D. (1997). The news media, the problem frame, and the production of fear. The Sociological Quarterly 38(4), 647668. American Meat Institute & Food Marketing Institute. (2008). The power of meat: An indepth look at meat through the shoppers eyes Paper presented at the Annual Meat Conference, Nashville, TN. Retrieved from http://www.meatconference.com/ht/a/GetDocumentAction /id/38142 An, S. K., & Gower, K. K. (2009). How do the news media frame crises? A content analysis of crisis news coverage. Public Relations Review 35(2), 107 112. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S0363811109000113 Anderson, C. A. (1997) Effects of violent movies and trait hostility on hostile feelings and aggressive thoughts. Aggressive Behavior, 23(3), 161 178. doi: 10.1002/(SICI)10982337(1997)23:3<161::AID AB2>3.0.CO;2 P Anderson, W. A. (2000). The future relationship between the media, the food industry and the consumer. British Medical Bulletin, 56 (1), 254268. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/10885119 Ankolekar, A., Krtzsch, M., Tran, T., & Vrandecic, D. (2007). Mashing up Web 2.0 and the s emantic w eb Scenario (p. 825834). International World Wide Web Conference Committee. Ariely, D. (2000). Controlling the flow: Eff ects on consumers' decision making and preferences. The Journal of Consumer Research, 27(2),233 248. Ary, D., Jacobs, L. C., Razavieh, A., & Sorensen, C. (2006). Introduction to research in education, 7th ed. Belmont, CA: Thompson Higher Education. Ashlock, M. A., Cartmell II, D. D., & Kelemen, D. B. (2006). The Cow That Stole Christmas: Framing the First U.S. Mad Cow Crisis. Journal of Applied Communications, 90(2), 29 46.
188 Baran, S. J., & Davis, D. K.(Eds.). (2009). Mass communication theory: Foundations, ferment, and future. Boston, MA: Wadsworth Cengage Learning. Bargh, J. A. (2001). The psychology of the mere. In J. A. Bargh & D. K. Apsley (Eds.), Unraveling the complexities of social life: A festschrift in honor of Robert B. Zajonc Washington, DC: Amer ican Psychological Association. Barnhurst, K. G., & Wartella, E. (1998). Young citizens, American TV newscasts and the collective memory. Critical Studies in Media Communication, 15(3), 279 305. doi: 10.1080/15295039809367049 Barnhurst, K. G., & Wartella, E. (1998). Young citizens, American TV newscasts and the collective memory. Critical Studies in Media Communication, 15(3), 279 305. doi: 10.1080/15295039809367049 Bateson, G. (1972). Steps to an ecology of mind. New York: Ballantine Books. Bernard, J. C ., Zhang, C., & Gifford, K. (2006). An Experimental Investigation of Consumer Willingness to Pay for NonGM Foods When an Organic Option Is Present. Agricultural and Resource Economics Review, 35(2), 374 385. Birdsall, W. F. (2007). Web 2.0 as a social movement. Webology, 4 (2)Article 40. Retrieved from http://www.webology.ir/2007/v4n2/a40.html Bogart, L. (2000). The death of print. Editor & Publisher, 133(48), 38. Boone, K., Meisenbach, T., & Tuck er, M. (2000). Agricultural communications: Changes and challenges. Ames: Iowa State University Press. Booth, A. (1970). The recall of news items. The Public Opinion Quarterly, 34(4), 604610. doi: 10.1086/268871 Bourn, D., & Prescott, J. (2002). A compari son of the nutritional value, sensory qualities, and food safety of organically and conventionally produced foods. Critical Reviews in Food Science and Nutrition, 42(1), 1 34. Taylor \ & Francis. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11833635 Brock, T. C. (1967). Communication discrepancy and intent to persuade as determinants of counterargument production. Journal of Experimental Social Psychology, 3 (3), 296 309. doi: 10.1016/0022 1031(67)900315 Burgoon, M., Cohen, M., Miller, M. D., & Montgomery, C. L. (1978) An empirical test of a model of resistance to persuasion. Health Communications Research, 5(1), 27 39. doi: 10.1111/j.14682958.1978.tb00620.x Burnkrant, R. E., & Unnava, H. R. ( 1989). Self referencing: A strategy for increasing processing of message content. Personality and Social Psychology Bulletin, 15(4), 628 638.
189 Burros, M. (2006, December 27). You are what you eat: 2006 and the politics of food. The New York Times pp. F2. C acioppo, J. T., & Petty, R. E. (1981). Social psychological procedures for cognitive response assessment: The thought listing technique. In T. Merluzzi, C. Glass, & M. Genest (Eds.), Cognitive assessment (pp. 309341). New York: Gilford Press. Cacioppo, J. T., & Petty, R. E. (1982). The need for cognition. Journal of Personality and Social Psychology 42(1), 116 131. American Psychological Association. Retrieved from http://content.apa.org/journal s/psp/42/1/116 Cacioppo, J., Petty, R., Feinstein, J. A., & Jarvis, W. B. G. (1996). Dispositional differences in cognitive motivation: The life and times of individuals varying in need for cognition. Psychological Bulletin, 119(2), 197 253. Cacioppo, J. T., Petty, R. E., & Kao, C. F. (1984). The efficient assessment of need for cognition. Journal of Personality Assessment, 48(3), 306 307. doi: 10.1207/s15327752jpa4803_13 Cacioppo, J. T., Petty, R. E., Kao, C. F., & Rodriguez, R. (1986). Central and peripheral routes to persuasion: An individual difference perspective. Journal of Personality and Social Psychology, 51(5), 10321043. American Psychological Association. Retrieved from http://doi.apa.org/getdoi.cfm?doi=10.1037/00223518.104.22.1682 Cacioppo, J. T., Petty, R. E., & Morris, K. J. (1983). Effects of need for cognition on message evaluation, recall, and persuasion. Journal of Personality and Social Psychology, 45 (4), 805 818. Retrieved from http://content.apa.org/journals/psp/45/4/805 California Institute for Rural Studies. (2005). Regulating organic: Impacts of the national organic standards on consumer awareness and organic consumption patterns Retrieved from http://www.cirsinc.org/Documents/Pub1205.2.PDF Campbell, D. T., & Stanley, J. C. (1963). Experimental and quasi experimental designs for research Boston, MA: Houghton Mifflin. Campbell, M. (1994). Beyond the Terms of the Contract: Mothers and Farmers. Journal of Agricultural and Environmental Ethics, 7(2) (1994), 205 220. Cannon, K. J. & Irani, T. A. (in press). Framing foot and mouth disease: coverage of the 2001 and 2007 outbreaks of FMD in the United Kingdom. Journal of Applied Communications Cannon, K. J., & Lamm, A. J. (2010, May). Understanding media consumption habits of college students: A comparison of use between agricultural and non agricultural majors. Proceedings of the Annual Conference of the American Association for Agricultural Education, 37, 455 467.
190 Caplan, P. (2000). Eating British beef with confidence: A consideration of consumers. In P. Caplan (Ed.). Risk revisited (pp. 1 14125). London: Pluto Press. Carlson, M. (2007). Blogs and Journalistic Authority. Journalism Studies, 8(2), 264279. Routledge. Retrieved from http://www.informaworld.com/openurl?genre=article& doi=10.1080/14616700601148861&magic=crossref Christenson, L. B. (2001). Experimental methodology 8th ed. Needham Heights, MA: Allyn & Bacon. Chryssochoidis, G. (2000). Repercussions of consumer confusion for late introduced differentiated products. European Journal of Marketing, 34(5/6), 705 722. Chung, D. S., & Yoo, C. Y. (2008). Audience Motivations for Using Interactive Features : Distinguishing Use of Different Types of Interactivity on an Online New spaper. Mass Communication and Society, 11(4), 375397. Chung, H., & Zhao, X. (2004). Effects of perceived i nteractivity on w eb site p reference and m emory: Role of p ersonal m otivation. Journal of Computer Mediated Communication, 10(1), 1 33. Blackwell Publ ishing Ltd. Retrieved from http://jcmc.indiana.edu/vol10/issue1/chung.html Cobb, M. D. (2005). Framing Effects on Public Opinion about Nanotechnology. Science Communication, 27(2), 221239. R etrieved from http://scx.sagepub.com/cgi/doi /10.1177/1075547005281473 Cohen, A. R., Stotland, E., & Wolfe, D. M. (1955). An experimental investigation of need for cognition. Journal of Abnormal and Social Psychology, 51(2), 291 294. Cohen, J. W. (1988). Statistical power analysis for the behavioral sciences 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates. Combs, B., & Slovic, P. (1979). Newspaper coverage of causes of death. Jour nalism Quarterly, 56(4), 83743. Coutinho, S. A. (2006). The relationship between the need for cognition, metacognition, and intellectual task performance. Educational Research and Reviews, 1(5), 162164. Crano, W. D., & Pristlin, R. (2006). Attitudes and persuasion. Annual Review of Psychology, 57 345374. doi: 10.1146/annurev.psych.57.102904.190034 Crawley, C. E. (2007). Localized Debates of Agricultural Biotechnology in Community Newspapers: A Quantitative Content Analysis of Media Frames and Sources. Science Communication, 28(3), 314346. Retrieved from http://scx.sagepub.com/ cgi/content/abstract/28/3/314
191 Curtain, P. A., Dougall, E., & Mersey, R. D. (2007). Study Compares Yahoo! News Story Preferences. Newspaper Research Journal 28(4), 2235. dHaenens, L., Jankowski, N., & Huevelman, A. (2004). News in online and print newspapers: Differences in reader consumption and recall. New Media & Society, 6 (3), 363 382. doi: 10.1177/1461444804042520 Dangour, A. D., Dodhia, S. K., Hayter, A., Allen, E., Lock, K., & Uauy, R. (2009). Nutritional quality of organic foods: a systematic review. American Journal of Clinical Nutrition 90 (3), 680 685. Am Soc Nutrition. Retrieved from http://www.ajcn.org/cgi/content/abstract/90/3/680 Das, E. H. H. J., DeWit, J. B. F., & Stroebe, W. (2003). Fear appeals motivate acceptance of action recommendations: Evidence for a positive bias in the processing of persuasive messages. Personality and Social Psychology Bulletin, 29(5), 650 664. doi: 10.1177/0146167203251527 David, P. (1998). News concreteness and visual verbal association. Human Communication Research, 25(2), 180 201. Davis, C. (2003). Cognitive and affective responses by West Texas Hispanics/Latinos to agricultural news: A comparison of four English and Spanish presentation media (Unpublished doctoral dissertation). Texas Tech University, Lubbock, TX. Davis, C. S., Akers, C., Cepica, M., Doerfert, D., Fraze, S., & Lawver, D. (2005, February). Cognitive responses by West Texas Hispanic/Latinos to agricultural news: A comparison of four English and Spanish presentation media. Paper presented at the annual meeting of the Southern Associat ion of Agricultural Scientists, Little Rock, AK. DeFleur, M. L., Davenport, L., Cronin, M., DeFleur, M. (1992). Audience recall of news stories presented by newspaper, computer, television and radio. Journalism Quarterly, 69(4), 10101022. doi: 10.1037/031 229 DeLind, L. B., & Howard, P. H. (2008). Safe at any scale? Food scares, food regulation, and scaled alternatives. Agriculture and Human Values 25(3), 301317. Retrieved from ht tp://www.springerlink.com/index/10.1007/s10460007 9112y DeNoon, D. J. (2009). Peanut butter Salmonella outbreak rages on. Medscape.com Retrieved from http://www.medscape.com/viewarticle/587195 DeWaal, E., & Schoenbach, K. (2008). Presentation style and beyond: How print newspapers and online news expand awareness of public affairs issues. Mass Communication and Society, 11(2), 161176. doi: 10.1080/15205430701668113 Dexter, L. A. (1964). Introdu ction. In L. A. Lewis & D. W. White (Ed.), People, society, and mass communications (pp. 3 25). London: The Free Press of GlencoeMacmillan Ltd.
192 Diddi, A., & LaRose, R. (2006). Getting hooked on news: Uses and gratifications and the formation of news habit s among college students in an internet environment. Journal of Broadcasting & Electronic Media, 50(2), 193 210. doi: 10.1207/s15506878jobem5002_2 Dillard, J. P., & Pfau, M. (2002). Introduction. In J. P. Dillard & M. Pfau (Eds.), The persuasion handbook: Developments in theory and practice (pp. ix xx). Thousand Oaks, CA: Sage Publications. Dillman, D. A. (2000). Mail and internet surveys: The tailored design method. New York: John Wiley & Sons, Inc Dillman, D. A., Smyth, J. D., & Christian, L. M. (2009). I nternet, mail, and mixed mode surveys: The tailored design method, 3rd ed. Hoboken, N. J.: John Wiley & Sons, Inc. Doerfert, D. (2003). Agricultural literacy: An assessment of research studies published within the agricultural education profession. Proceedings of the 22nd Annual Western Region Agricultural Education Research Conference, Portland, Oregon. Dunwoody, S. (1992). Mass media coverage of technological and environmental risks: A survey of research in the United States and Germany. Public Understanding of Science, 1(2), 199230. DuPuis, E. M. (2000). Not in my body: rBGH and the rise of organic milk. Agriculture and Human Values, 17 285 295. Durfee, J. L. (2006). "Social Change" and "Status Quo" framing effects on risk perception: An exploratory experiment. Science Communication, 27 (4), 459495. Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. New York: Wadsworth Cengage Learning. Edelman, M. J. (1993). Contestable categories and public opinion. Political Communication, 10(3), 231242 Engel, E. P. (2003). Connectivity and continuity: Influences of the digital realm on the visual information structures of print. In S. D. Reese, O. G. Gandy, & A. E. Grant (Eds.), Framing public life: Perspectives on media and our understanding of the so cial world (35 66). Mahwah, N.J.: Lawrence Erlbaum Associates. Entman, R. M. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of Communication, 43(4), 5158. Oxford University Press. Retrieved from http://www.blackwell synergy.com/doi/abs/10.1111/j.14602466.1993.tb01304.x
193 Eveland, W. P., Cortese, J., Park, H., & Dunwoody, S. (2004). How web site organization influences free recall, factual kno wledge, and knowledge structure density. Human Communication Research, 30 (2), 208233. doi: 10.1111/j.14682958.2004.tb00731.x Eveland, W. P., & Dunwoody, S. (2002). An investigation of elaboration and selective scanning as mediators of learning from the w eb versus print. Journal of Broadcasting & Electronic Media, 46(1), 34 53. Festinger, L. (1957). A theory of cognitive dissonance. Evanston, IL: Row, Peterson and Company. Field, A. (2009). Discovering statistics using SPSS London, England: Sage Publicati ons Ltd. Fischer, G. W., Morgan, M. G., Fischoff, B., Nair, I., and Lave, L. B. (1991). What risks are people concerned about? Risk Analysis, 11(2), 303 314. Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to th eory and research Reading, MA: AddisonWesley Publishing Co. Fisher, A., & Chen, Y. C. (1996). Customer perceptions of agency risk communication. Risk Analysis, 16 177 184. Fiske, S. T., & Linville, P. W. (1980). What does the schema concept buy us? Pers onality and Social Psychology Bulletin, 6(4), 543 557. doi: 10.1177/014616728064006 Fiske, S. T., & Taylor, S. E. (1984). Social cognition. Reading, MA: AddisonWesley. Fredin, E. S. (2003). Frame breaking and creativity: A frame database for hypermedia news. In S. D. Reese, O. G. Gandy, & A. E. Grant (Eds.), Framing public life: Perspectives on media and our understanding of the social world (35 66). Mahwah, N.J.: Lawrence Erlbaum Associates. Frewer, L. J. (1999). Risk perception, social trust, and public participation in strategic decision making: Implications for emerging technologies. Ambio 28(6), 569574. Frewer, L. J., Howard, C., Hedderly, D. & Shepherd, R. (1997) Public concerns about general and specific applications of genetic engineering: Risk, benefit and ethics. Science, Technology and Human Values, 22(1), 98 124. Frewer, L. J., Howard, C., Hedderly, D. & Shepherd, R. (1998) The influence of initial attitudes on responses to communication about genetic engineering in food production, Agriculture and Human Values 15 (1),15 30. doi: 10.1023/A:1007465730039
194 Frewer, L. J., Miles, S., & Marsh, R. (2002). The media and genetically modified foods: evidence in support of social amplification of risk. Risk analysis an official publication of the Society for Risk Analysis 22 (4), 701 711. Blackwell Science Ltd., Osney Mead Oxford OX 2 0 EL UK,. Retrieved from http://www.ncbi.nlm. nih.gov/pubmed/12224744 Gamson, W. A., & Modigliani, A. (1989). Medi a discourse and public opinion on nuclear power: A constructionist approach. American Journal of Sociology, 95(1), 1 37. Gamson, W. A., Croteau, D., Hoynes, W., & Sasson, T. (1992). Media Images and the Social Construction of Reality. Annual Review of Soci ology 18(1), 373393. Retrieved from http://arjournals.annualreviews.org/doi/abs/10.1146/annurev.so. 18.080192.002105 Garson, G. D. (2009). Univariate GLM, ANOVA, and ANCOVA, from Statnotes: Topics in Multivariate Analysis Retrieved from http://faculty.chass.ncsu.edu/garson/pa765/statnote.htm Gifford, K., & Bernard, J. C. (2004). The impact of message framing on organic food purchase likelihood. Journal of Food Distribution Research, 35(5), 19 28. Gifford, K., & Bernard, J. C. (2006). Influencing consumer purchase likelihood of organic food. International Journal of Consumer Studies, 30(2), 155 163. doi: 10.1111/j.14706431.2005.00472.x Gitlin, T. (1980). The whole world is watching. Mass media in the making & unmaking of the new left. Berkeley, CA: University of California Press. Green, J. M., Draper, A. K., & Dowler, E. A. (2003). Short cuts to safety: risk and rules of thumb in accounts of food choice. Health Risk and Society, 5(1), 33 52. Greene, A. (2009, April). 5 foods every woman should eat organic. Glamour 107(4). Retrieved from http://www.glamour.com/ Greene, C., Dimitri, C., Lin, B., McBride, W., Oberholtzer, L., & Smith, T. (2009 ). Emerging issues in the U. S. organic industry (Economic Information Bulletin No. 55). Retrieved from http://www.usda.gov/ers/ Goffman, E. (1974). Frame analysis: An essay on the organization of experience. Cambr idge, MA: Harvard University Press. Gonzalez, C., Dana, J., Koshino, H., & Just, M. (2005). The framing effect and risky decisions: Examining cognitive functions with fMRI. Journal of Economic Psychology, 26 (1), 1 20. Elsevier. Retrieved from http://linkinghub.elsevier. com/retrieve/pii/S016748700400073X
195 Goodman, J. R., & Goodman, B. P. (2006). Beneficial or biohazard? How the media frame biosolids. Public Understanding of Science 15(3), 359375. Retrieved from http://pus.sagepub.com/cgi/doi/10.1177/0963662506062468 Gosling, S. D., & Johnson, J. A. (2010). Advanced methods for conducting online behavioral r esearch Washington, DC: American Psychological Association. Gregory, N. G. (2000). Consumer concerns about food. Outlook on Agriculture, 29(4), 251257. Hackett, R. A. (1984). Decline of a paradigm? Bias and objectivity in news media studies. Critical Stu dies in Mass Communication, 1, 229 259. Hallahan, K. (1999). Seven Models of Framing: Implications for Public Relations. Journal of Public Relations Research, 11(3), 205 242. Retrieved from http://www.informaworld.com/openurl?genre=article&doi=10.1207/s1532754xjprr1 103_02&magic=crossref Hansen, J., Holm, L., Frewer, L., Robinson, P., & Sande, P. (2003). Beyond the knowledge deficit: recent res earch into lay and expert attitudes to food risks. Appetite 41(2), 111 121. Retrieved from http://linkinghub.elsevier.com/retrieve /pii/S0195666303000795 Harper, G. C., & Maka touni, A. (2002). Consumer perception of organic food production and farm animal welfare. British Food Journal, 104(3 5), 287299. Harris poll results show who is buying organic foods, how frequently. (2007, October). Nutrition Business Journal, 12(10), 21. Haugtvedt, C. P., Petty, R. E., & Cacioppo, J. T. (1992). Need for cognition and advertising: Understanding the role of personality variables in consumer behavior. Cognition 7 (3), 239 260. Retrieved from http://linkinghub.elsevier.com/retrieve/ pii/S1057740808800381 Haygood, J., Hagins, S., Akers, C., & Kieth, L. (2002). Associated Press wire service coverage of agricultural issues. Proceedings of the 29th Annual National Agr icultural Education Research Conference, Las Vegas Nevada. Heesacker, M., Petty, R. E., & Cacioppo, J. T. (1983). Field dependence and attitude change: Source credibility can alter persuasion by affecting messagerelevant thinking. Journal of Personality 51(4), 653666. John Wiley & Sons. Heeter, C. (2000) Interactivity in the Context of Designed Experience. Journal of Interactive Advertising, 1 (1). Retrieved from http://www.jiad.org/vol1/no1/heeter/ Heller, L. (2006). Consumer confidence in food safety on the rise? Food Production Daily March.
196 Henke, L. (1985). Perceptions and use of news media by college students. Journal of Broadcasting & Electronic Media, 29(4), 431 436. Herrmann, R., Warland, R., & Sterngold, A. (1997). Who reacts to food safety scares? Examining the Alar case. Agribusiness 5, 511 520. Hertog, J. K., & McLeod, D. M. (2003). A multiperspective approach to framing analysis: A field guide. In S.D. Reece, O.H. Gandy, & A.E. Grant (Eds.), Framing Public Life 139161. Mahwah, NJ: Erlbaum. Hill, H., & Lynchehaun, F. (2002). Organic milk: Attitudes and consumption patterns. British Food Journal, 104(7), 526 542. Howe, N., & Strauss, W. (2000). Millennials rising: The next great generation New York: Vintage Books. Howe, N. & Strauss, W. (2007). Millennials go to college: Strategies for a new generation on campus Great Falls, VA: LifeCourse Associates. Hughner, R., McDonagh, P., Prothero, A., Shultz, C. J., & Stanton, J. (2007). Who are o rganic food consumers? A compilation and review of why people purchase organic food. Journal of Consumer Behaviour 6 (2 3), 94 110. Retrieved from http://www3.interscience.wiley.com/journal/114265522/abstract Hurt, R. D. (2002). American agriculture: A brief history West Lafayette, IN: Purdue University Press. Hwang, Y. J., Roe, B., & Teisl, M. F. (2005). An Empirical Analysis of United States Consumers Concerns About Eight Food Production and Processing Technologies. AgBioforum 8 (1), 40 49. Irlbeck, E. (2009). The summer of salmonella in salsa: A framing analysis of the 2008 salmonella outreak in tomatoes and jalapenos. American Association for Agricultural Education Research Conference Proceedings (p. 61 75). Iyengar, S. (1991). Is anyone responsible? How television frames political issues Chicago: University of Chicago Press. Iyengar, S., & Kinder, D. R. (1987). News that matters Chicago: University of Chicago Press. Iyeng ar, S., Peters, M. D., & Kinder, D. R. (1982). Experimental demonstrations of the not so minimal consequences of television news programs. The American Political Science Review, 76 (4), 848858. Janis, I. L., & Feshbach, S. (1954). Personality differences associated with responsiveness to fear arousing communications. Journal of Personality, 23(2), 154166. doi: 10.1111/j.14676494.1954.tb01145.x
197 Jin, H. J., Skripnitchenko, A., & Koo, W. W. (2004). The effects of the BSE outbreak in the United States on th e beef and cattle industry Special Report 034, Center for Agricultural Policy and Trade Studies, Department of Agribusiness and Applied Economics, North Dakota State University, Fargo, ND. Jolly, D. A. (1991). Determinants of organic horticultural produc ts based on a sample of California consumers. Acta Horticulture, 295, 41148. Jones, S., JohnsonYale, C., Millermaier, S., & Perez, F. S. (2009). Everyday life, online: U.S. college students use of the Internet. First Monday, 14(10). Retrieved from http://firstmonday. org/ htbin/cgiwrap/bin/ojs/index.php/fm/article /viewArticle/2649/2301 Josephson, W. L. (1987). Television violence and childrens ag gression: Testing the priming, social script, and disinhibition predictions. Journal of Personality and Social Psychology, 53(5), 882 890. Judd, C. M., & Kulik, J. A. (1980). Schematic effects of social attitudes on information processing and recall. Jour nal of Personality and Social Psychology, 38(4), 569 578. doi: 10.1037/0022 3522.214.171.1249 Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263 292. Kahneman, D., & Tversky, A. (1984). Choices, v alues, and frames. (D. Kahneman & A. Tversky, Eds.) American Psychologist 39 (4), 341350. Cambridge University Press. Retrieved from http://content.apa.org/journals/amp/39/4/341 Kasperson, R. E ., Renn, O., Slovic, P., Brown, H. S., Emel, J., Globe, R., Kasperson J. X., & Ratick, S. (1988). The social amplification of risk: A conceptual framework. Risk Analysis, 8 177 187. Katz, D. (1960). The functional approach to the study of attitudes. Publi c Opinion Quarterly, 24(2), 163204. doi: 10.1086/266945 Katz, J. (1999). The future is the net Media Studies Journal, 13(2), 14 15. Keller, P. A., & Block, L. G. (1996). Increasing the persuasiveness of fear appeals: The effect of arousal and elaboration. The Journal of Consumer Research, 22(4), 448 459. Kelly, K., & Edwards, R. (1992). Observations: Does discussion of advertising transaction its effects? Yes, sometimes. Journal of Advertising Research, 32(2), 6569. Kelman, H. C. (1953). Attitude change as a function of response restriction. Human Relations, 6(3), 185 214. doi: 10.1177/001872675300600301
198 Kiousis, S. (2002). Interactivity: A concept explication. New Media Society, 4(3), 355 383.DOI: 10.1177/146144480200400303 Kitzinger, J. (1999). A sociology of media power: Key issues in audience reception research. In G. Philo (Ed.), Message received: Glasgow media group research 19931998, (pp. 3 20). Harlow: Longman. Kniazeva, M., & Venkatesh, A. (2007). Food for thought : A study of food consumptio n in postmodern US culture. Journal of Consumer Behaviour 435 419 435. Krosnick, J. A., & Kinder, D. R. (1990). Altering the foundations of support for the president through priming. The American Political Science Review, 84(2), 497 512. Krosnick, J. A., & Petty, R. E. (1995). Attitude strength: An overview. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences (pp. 124). Hillsdale, NJ: Lawrence Erlbaum Associates. Khberger, A. (1998). The Influence of Framing on Risky Decisions: A Metaanalysis. Organizational Behavior and Human Decision Processes 75(1), 23 55. Elsevier. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9719656 Latour, M. S., & Rotfeld, H. J. (1997). There are Threats and ( Maybe ) Fear Caused Arousal : Theory and Confusions of Appeals to Fear and Fear Arousal Itself. Journal of Advertising 26(3), 45 59. JSTOR. Retrieved from http://www.jsto r.org/stable/4189041 Lawrence, R. G. (2004). Framing Obesity: The Evolution of News Discourse on a Public Health Issue. The Harvard International Journal of PressPolitics 9 (3), 56 75. Retrieved from http://hij.sagepub.com/cgi/doi/10.1177/1081180X04266581 Lee, A. Y. (2001). The mere exposure effect: An uncertainty reduction explanation revisited. Personality and Social Psychology Bulletin, 27(10), 12551266. Leiss, W., & Powell, D. (2 004). Mad cows and mothers milk 2nd ed. Montreal: McGill Queens University Press. Leventhal, H., & Cameron, L. (1994). Persuasion and health attitudes. In S. Shavitt & T.C. Brock (Eds.) Persuasion: Psychological insights and perspectives (pp. 1542). Ne edham Heights, MA: Allyn and Bacon. Lister, M., Dovey, J., Giddings, S., Grant, I., & Kelly, K. (2003). New media: A critical introduction. Now York: Routledge. Lockie, S. (2006). Capturing the Sustainability Agenda: Organic Foods and Media Discourses on Food Scares, Environment, Genetic Engineering, and Health. Agriculture and Human Values 23 (3), 313 323. Retrieved from http://www.springerlink.com/index/10.1007/s104600069007 3
199 Lofstedt, R. E. (2006). How can we make food risk communication better: Where are we and where are we going?. Journal of Risk Research, 9 (8), 869 890. Retrieved from http://www.informaworld.com/openurl?genre=article&doi=10.1080/1366987 0601065585&magic=crossref Lundgren R.E. 1994. Risk Communication: A Handbook for Communicating Environmental, Safety, and Health Risks Columbus/Ohio: Battelle Pr ess. Lundy, L. (2006). Effect of framing on cognitive processing in public relations. Public Relations Review 32 (3), 295 301. Retrieved from http://linkinghub.elsevier.com/retri eve/pii/S0363811106000786 Lundy, L. K., & Irani, T. A. (2004). Framing biotechnology: A comparison of U. S. and British national newspapers. Journal of Applied Communication, 88 (2), 37 49. Macfarlane, R. (2002). Integrating the consumer interest in food s afety: The role of science and other factors. Food Policy, 27(1), 6580. Madden, M., & Fox, S. (2006). Riding the waves of Web 2.0: More than a buzzword, but still not easily defined. Report from the Pew Internet and American Life Project, Pew Research C enter. Retrieved from http://www.pewinternet.org/Reports/2006/ Riding the Waves of Web20.aspx Marette, S., Roosen, J., Blanchemanche, S., & Verger, P. (2006). Health information and the choice of fish species: An experiment measuring the impact of risk and benefit information (Working paper 06WP 421). Retrieved from www.card.iastate.edu Marks, L., Kalaitzandonakes, N., Wilkins, L., & Zakharova, L. (2007). Mass media framing of biotechnology news. Public Understanding of Science, 16(2), 183 203. Martin, A. (2008, May 30). Food report criticizes biofuel policies. The New York Times pp. C1. Massey, B. L., & Levy, M. R. (1999). Interactivity, online journalism, and Engli sh language web newspapers in Asia. Journalism Mass Communication Quarterly 76(1), 138 151. Association for Education in Journalism & Mass Communication. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&db=ufh&AN= 1988080&site=ehost live Maugh, T. H., & Engel, M. (2009, February 7). Peanut company lied on salmonella testing, FDA finds. Los Angeles Times Retrieved from http://articles.latimes.com/ 2009/feb/07/science/sci peanut fda7 Mccarthy, M., Brennan, M., Kelly, A., Ritson, C., Deboer, M., & Thompson, N. (2007). Who is at risk and what do they know? Segmenting a population on their food safety knowledge. Food Quality and Preference, 18 (2), 205 217. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S0950329305001576
200 McGi nn, D. (2005, March). The green machine; Whole Foods has come a long way since its hippie roots. Newsweek, 36. Retrieved from http://www.newsweek.com McQuail, D. (1994). Mass communication theory: An introduction L ondon, England: Sage Publications. McQuail, D. (2005). McQuails mass communication theory London, England: Sage Publications. Mendelsohn, H. (1964). Listening to radio. In L. A. Dexter & D. M. White (Eds.), People, society, and mass communication (pp. 23 9 249). New York: Free Press. Meyerowitz, B. E., & Chaiken, S. (1987). The effect of message framing on breast self examination attitudes, intentions, and behavior. Journal of Personality and Social Psychology, 52 (3), 500 510. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/ 3572721 Meyers, C. A. (2008). The agriculture angle: Effect of framing agricultural biotechnology messages on attitudes and intent to publish within the elaboration likelihood model (Unpublished doctoral dissertation). University of Florida, Gainesville. Meyers, C. A., & Abrams, K. (in press). Feeding the debate: A qualitative framing analysis of organic food news media coverage. Journal of Applied Communications Retrieved f rom the authors. Miller, J. M., & Peterson, D. A. M. (2004). Theoretical and empirical implications of attitude strength. The Journal of Politics, 66(3), 847867. doi: 10.1111/j.14682508.2004.00279.x Morris, M., & Ogan, C. (1996). The Internet as mass medium. Journal of Communication, 46(1), 39 50. Nabi, R. L. (1999). A cognitivefunctional model for the effects of discrete negative emotions on information processing, attitude change, and recall. Communication Theory, 9(3), 292 320. doi: 10.1111/j.1468288 5.1999.tb00172.x Nabi, R. L. (2002). Discrete emotions and persuasion. In J. P. Dillard & M. Pfau (Eds.), The persuasion handbook (289 308). Thousand Oaks, CA: Sage Publications. Nardi, B.A., Schiano, D.J., Gumbrecht, M., & Swartz, L. (2004). Why we blog. Communications of the ACM, 47(12), 41 46. National Cattlemens Beef Association. (2007). Conjugated linoleic acid and dietary beef (Report funded by The Beef Checkoff). Centennial, CO: National Cattlemens Beef Association. Retrieved from http://www.beefusa.org Nelkin, D. (1987, 1995). Selling science New York: W. H. Freeman & Company.
201 Nerlich, B., C. Hamilton, and V. Rowe. 2002. Conceptualising foot and mouth disease: The socio cultural role of metaphors, frames and narratives. Metaphorik.de 2. Retrieved from http://www.metaphorik.de/02/nerlich.htm Neuman, W. R., Just, M. R., & Crigler, A. N. (1992). Common knowledge: News and the construction of political m eaning. Chicago: University of Chicago Press. OKeefe, D. J. (1990). Persuasion: Theory and research. Newbury Park, CA: Sage Publications. OReilly, T. (2007). What i s We b 2.0: Design patterns and business models for the next generation of s oftware. Communications Strategies 1 (65), 1737. Retrieved from http://oreilly.com/web2/archive/what is web 20.html Onyango, B. M., Hallman, W. K., & Bellows, A. C. (2007). Purchasing organic food in U S food systems: A study of attitudes and practice. British Food Journal 109 (5), 399411. Retrieved from http://www.emeraldinsight.com/ 10.1108/00070700710746803 Outing, S. (1998). W hat exactly is interactivity?. Editor Publisher 131 (52), 7. Pan, Z., & Kosicki, G. M. (2003). Framing as a strategic action in public deliberation. In S. D. Reese, O. G. Gandy, & A. E. Grant (Eds.), Framing public life: Perspectives on media and our under standing of the social world (35 66). Mahwah, N.J.: Lawrence Erlbaum Associates. Pavlik, J. V. (2003). New framing and new media: Digital tools to reengage an alienated citizenry. In S. D. Reese, O. G. Gandy, & A. E. Grant (Eds.), Framing public life: Per spectives on media and our understanding of the social world (35 66). Mahwah, N.J.: Lawrence Erlbaum Associates. Perloff, R. M. (1993). The dynamics of persuasion. Hillsdale, NJ: Lawrence Erlbaum Associates. Perloff, R. M. (2003). The dynamics of persuasion: Communication and attitudes in the 21st century, 2nd ed. Mahwah, N. J.: Lawrence Erlbaum Associates. Peterson, R. (2001). On the use of college students in social science research: Insights from a secondorder meta analysis. Journal of Personality, 28(3 ), 450 461. doi: 10.1086/323732 Petty, R. E., & Cacioppo, J. T. (1977). Forewarning, cognitive responding, and resistance to persuasion. Journal of Personality and Social Psychology, 35, 645 655.
202 Petty, R. E., & Cacioppo, J. T. (1979). Issue involvement can increase or decrease persuasion by enhancing messagerelevant cognitive responses. Journal of Personality and Social Psychology 37 (10), 19151926. Retrieved from http://content.apa.org/journals/psp/37/10/1915 Petty, R. E., & Cacioppo, J. T. (1979). Issue involvement can increase or decrease persuasion by enhancing messagerelevant cognitive responses. Journal of Personality and Social Psychology, 37(10), 19151926. doi: 10.1037/00223514.37. 10.1915 Petty, R. E., & Cacioppo, J. T. (1981). Attitudes and persuasion: Classic and contemporary approaches Dubuque, IA: William C. Brown Co. Petty, R. E., & Cacioppo, J. T. (1986). Communication and Persuasion: Central and Peripheral Routes to Attitude Change New York: Springer Verlag. Petty, R. E., Cacioppo, J. T., & Goldman, R. (1981). Personal involvement as a determinant of argument based persuasion. Journal of Personality and Social Psychology, 41 847855. Petty, R. E., & Jarvis, B. G. (1996). An individual difference perspective on assessing cognitive processes. In N. Schwarz & S. Sudman (Eds.), Answering questions: Methodology for determining cognitive and communicative processes in survey research (pp. 221 257). San Francisco: Jossey Bass Publi shers. Petty, R. E., Priester, J. R., & Briol, P. (2002). Mass media attitude change: Implications of the Elaboration Likelihood Model of persuasion. In J. Bryant & D. Zillman (Eds.), Media effects: Advances in theory and research, 2nd Ed. (pp. 155198). New York: Routledge. Petty, R. E., & Wegener, D. T. (1998). Matching Versus Mismatching Attitude Functions: Implications for Scrutiny of Persuasive Messages. Personality and Social Psychology Bulletin 24 (3), 227 240. Sage Publications. Retrieved from http://psp.sagepub.com/cgi/doi/10.1177/0146167298243001 Petty, R. E., & Wegener, D. T. (1998). Attitude change: Multiple roles for persuasion variables. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (Vol. 1, pp. 323 390). New York: McGraw Hill. Petty, R. E., & Wegener, D. T. (1999). The Elaboration Likelihood Model: Current status and controversies. In S. Chaiken & Y. Trope (Eds.), Dual process theor ies in social psychology 41 72. New York: Guilford Press. Petty, R. E., DeMarree, K. G., Brinol, P., Horcajo, J., & Strathman, A. J. (2008). Need for cognition can magnify or attenuate priming effects in social judgment. Personality and Social Psychology Bulletin, 34 (7), 900 912. doi: 10.1177/0146167208316692
203 Pew Research Center Project for Excellence in Journalism. (2008). The changing newsroom Retrieved from http://www.journalism.org/node/11961 Pidge on, N.F. & Beattie, J. (1998). The psychology of risk and uncertainty. In: Handbook of Environmental Risk Assessment and Management (edited by P. Calow), p p. 289 318. Oxford: Blackwell Science Ltd. Pisano, C., & Woods, M. D. (2002). Communicating food safety across cultures: Issues of trust and credibility within diverse populations Paper presented at the international meeting of Agricultural Communicators in Education, Savannah, GA. Plough, A., & Krimsky, S. (1987). The Emergence of Risk Communication St udies : Social and Political Context. Science Technology And Human Values 12(3/4), 4 10. John Wiley and Sons. Powell, M., Dunwoody, S., Griffin, R., & Neuwirth, K. (2007). Exploring lay uncertainty about an environmental health risk. Public Understanding of Science 16(3), 323 343. Retrieved from http://pus.sagepub.com/cgi/doi/10.1177/0963662507074491 Price, V., Huang, L., & Tewksbury, D. (1997). Thirdperson effects of news coverage: Orientations toward media. Journalism and Mass Communication Quarterly, 74(3):525540. Price, V., Tewksbury, D., & Powers, E. (1997). Switching Trains of Thought: The Impact of News Frames on Readers Cognitive Responses. Communication Research, 24(5), 48 1 506. Retrieved from http://crx.sagepub.com/cgi/doi/10.1177/ 009365097024005002 Priest, S. H., & Eyck, T. (2003). News coverage of biotechnology debates. Society 40(6), 29 34. Retrieved from http://www.springerlink.com/index/ 10.1007/BF02712649 Purcell, K., Rainie, L., Mitchell, A., Rosenstiel, T., & Olmstead, K. (2010). Understanding the participatory news consumer: How internet and cell phone users have turned news into a social experience. Pew Research Center Retrieved from http://www.pewinternet.org/Reports/2010/OnlineNews.aspx Putna m, R. D. (2000). Bowling alone: The collapse and revival of American community New York: Simon & Schuster. Rainie, L., & Horrigan, J. (2005). Trends 2005 Internet: The mainstreaming of online life. Pew Research Center Retrieved from http://www.pewinternet.org/Reports/ 2005/How the internet -has woven itself into American life.aspx
204 Ramirez, J. A., & Burgoon, J. (2004). The effect of interactivity on initial interactions: the influence of information valence and modality and information richness on computer mediated interaction. Communication Monographs 71(4), 422447. Retrieved from http://www.informaworld.com/openurl?genre=article&doi= 10.1080 /0363452042000307461&magic=crossref Rasmussen, W.D. (1989). Taking the university to the people: Seventy five years of Cooperative Extension Ames, Iowa: Iowa State University Press. Reese, S. D. (2003). PrologueFraming public life: A bridging model for media research. In S. D. Reese, O. H. Gandy, & A. E. Grant (Eds.), Framing public life (pp. 732). Mahwah, NJ: Lawrence Earlbaum Associates. Reese, S. D., Rutigliano, L., Hyun, K., & Jeong, J. (2007). Mapping the blogosphere: Professional and citizenbased media in the global news arena. Journalism, 8(3), 235261. doi:10.1177/1464884907076459 Reips, UD. (2000). The web experiment method: Advantages, disadvantages, and solutions. In M. H. Birnbaum (Ed.), Psychological experiments on the Internet (pp. 89117 ). Reips, UD. (2010). Design and formatting in Internet based research. In S. D. Gosling & J. A. Johnson (Eds.), Advanced methods for conducting online behavioral research (pp. 2944). Washington, DCD: American Psychological Association. Reips, UD., & Kr antz, J. H. (2010). Conducting true experiments on the web. In S. D. Gosling & J. A. Johnson (Eds.), Advanced methods for conducting online behavioral research (pp. 193 216). Washington, DC : American Psychological Association. Rice, R. E., & Williams, F. ( 1984). Theories old and new: The study of new media. In R. E. Rice & Associates (Eds.), The new media: Communication, research and technology (pp. 55 80). Beverly Hills, CA: Sage. Robinson, S. (2006). The mission of the j blog: Recapturing journalistic aut hority online. Journalism 7 (1), 65 83. Retrieved from http://jou.sagepub.com/cgi/doi/ 10.1177/1464884906059428 RoskosEwoldsen, D. R., Roskos Ewoldsen, B., & Carpentier, F. D. (2009). In J. Bryant & M. Oliver (Eds.), Media effects: Advances in theory and research (74 93). New York: Routledge. RoskosEwoldsen, D. R., Yu, J. H., & Rhoades, N. (2004). Fear appeal messages affect accessibility of attitudes toward the threat and adaptive behaviors. Communication Monographs, 71(1), 49 69. doi: 10.1080/0363452042000228559 Rubin, A. M., & Rubin, R. C. (1985). Interface of personal and mediated communication: A research agenda. Critical Studies in Mass Communication, 3(2), 184200.
205 Ruggiero, T (2000). Uses and Gratifications Theory in the 21st Century. Mass Communication and Society 3 (1), 3 37. Routledge. Retrieved from http://www.informaworld.com/openurl?genre=article&doi=10.1207/S15327825MC S0301_02&magic=crossref Ruth, A. M., Eubanks, E. E., & Telg, R. (2005). Framing of mad cow media coverage. Journal of Applied Communications 89(4), 39 54. Schlagheck, C. (1998). Newspaper reading choices by college students. Newspaper Research Journal, 19(2) 74 87. Schudson, M. (2003). The sociology of news New York: W.W. Norton. Scheufele, D.A. (1999) Framing as a Theory of Media Effects. Journal of Communication 49(1), 103 22. Scheufele, D. A. (2000). Agendasetting, priming, and framing revisited: Another look at cognitive effects of political communication. Mass Communication & Society, 3(2), 297316. doi: 10.1207/S15327825MCS0323_07 Scheufele, D. A., & Lewenstein, B. V. (2005). The Publi c and Nanotechnology: How Citizens Make Sense of Emerging Technologies. Journal of Nanoparticle Research, 7 (6), 659 667. Retrieved from http://www.springerlink.com/index/ 10.1007/ s11051 00575262 Schultz, B., & Sheffer, M. (2009). Newspaper managers report positive attitudes about blogs. Newspaper Research Journal, 30(2), 30 43. Seevers, B., Grahman, D., Gamon, J., & Conklin, N. (1997). Education through Cooperative Extension Al bany, NY: Delmar Press. Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi experimental designs for generalized causal inference. Boston: Houghton Mifflin Company. Shah, D. V., Domke, D., & Wackman, D. B. (1996). To thine own s elf be true: Values, framing, and voter decisionmaking strategies. Communication Research, 23, 509 560. Shaw, A. (2004). Discourses of risk in lay accounts of microbiological safety and BSE: a qualitative interview study. Health Risk Society 6 (2), 1511 71. Retrieved from http://www.informaworld.com/openurl?genre=article&doi=10.1080/1369857042000 219779&magic=crossref Shih, T. J., Wijaya, R., & Brossard, D. (2008). Media Coverage of Public Health Epidemics : Linking Framing and Issue Attention Cycle Toward an Integrated Theory of Print News Coverage of Epidemics. Mass Communication Society 141 160.
206 Shoemaker, P. J., and S. D. Reese. 1996. Media ting the message: Theories of influence on mass media content New York: Longman. Sicilia, M., Ruiz, S., & Munuera, J. L. (2005). Effects of interactivity in a website: The moderating effect of need for cognition. Journal of Advertising, 34(3), 31 45. Sin ger, E., & Endreny, P. M. (1994). Reporting on risk: How the mass media portray accidents, diseases, disasters and other hazards. Risk: Health, Safety, and Environment, 5(3), 261 270. Sobal, J., & Maurer, D. (1995). Food, eating, and nutrition as social problems. In Eating agendas: Food and nutrition as social problems (edited by D. Maurer & J. Sobal), p. 3 7. New York: Aldine De Gruyter. Storck, A. (2008). Organic meat and poultry will keep growing in 2009: trade group. Meat Place Industry News Retrieved from http://www.meatingplace.com/ MembersOnly/webNews/details.aspx?item=10801 StromerGalley, J. (2000). Online interaction and why candidates avoid it. Journal of C ommunication, 50(4), 111132. Stull, C. L., Payne, M. A., Berry, S. L., & Reynolds, J. P. (2007). A review of the causes, prevention, and welfare of nonambulatory cattle. Journal of the American Veterinary Medical Association, 231(2), 227 234. Sundar, S. S ., Xu, Q., & Bellur, S. (2010). Designing interactivity in media interfaces. Proceedings of the 28th international conference on Human factors in computing systems CHI 10 2247. ACM Press. Retrieved from http://portal.acm.org/citation. cfm?doid=1753326.1753666 Tankard, J. W. (2003). The empirical approach to the study of media framing. In S. D. Reese, O. G. Gandy, & A. E. Grant (Eds.), Framing public life: Perspectives on media and our understanding of the social world (95 106). Mahwah, N.J.: Lawrence Erlbaum Associates. Taylor, S. E. (1981). The interface of cognitive and social psychology. In J. Harvey (Ed.). Cognition, social behavior and the environment. Hillsdale, N. J.: Erlbaum. Ten Eyck, T. (2000). The marginalization of food safety issues: An interpretative approach to mass media coverage. Journal of Applied Communications, 84(2), 29 47. Tewksbury, D., & Althaus, S. L. (2000). Differences in knowledge acquisition among readers o f the paper and online versions of a national newspaper. Journalism and Mass Communication Quarterly, 77(3), 457479.
207 Tourangeau, R., Rasinski, K. A., Bradburn, N., & DAndrade, R. (1989). Belief accessibility and context effects in attitude measurement. Journal of Experimental Social Psychology, 25(5), 401 421. doi:10.1016/00221031(89)900309 Touri, M. (2009). News blogs: strengthening democracy through conflict prevention. ASLIB Proceedings 61 (2), 170 184. Retrieved from http://www.emeraldinsight .com/10.1108/00012530910946910 Traub, R. E. (1994). Reliability for the social sciences: Theory and applications Thousand Oaks, CA: Sage Publications. Treise, D., & Weigold, M. F. (200 2). Advancing science communication: A survey of science communicators. Science Communication, 23(3), 310322. Tremayne, M. (2008). Manipulating interactivity with thematically hyperlinked news texts: a media learning experiment. New Media & Society 10(5), 703 727. Retrieved from http://nms.sagepub.com/cgi/doi/10.1177/1461444808094353 Tremayne, M., & Dunwoody, S. (2001). Interactivity, information processing, and learning on the World Wide Web. Science Communication, 23(2), 111 134. doi: 10.1177/1075547001023002003 Tuchman, G. (1978). Making News New York: Free Press. Tucker, M., Whaley, S. R., & Cano, J. (2003). Agricultural education and agricultural communications: Striking a proper balance in the academy. Journal of Agricultural Education, 44(1), 22 30. Tucker, M., Whaley, S. R., & Sharp, J. S. (2006). Consumer perceptions of foodrelated risks. International Journal of Food Science and Technology, 41(2), 135146. doi: 10.1111/j.13 652621.2005.01010.x Tuten, T. L. (2010). Conducting online surveys. In S. D. Gosling & J. A. Johnson (Eds.), Advanced methods for conducting online behavioral research (pp. 179192). Washington, DCD: American Psychological Association. Tversky, A., & Kahn eman, D. (1981). The framing of decisions and the psychology of choice. Science, 211(4481), 435458. Ullrich, C., Borau, K., Luo, H., Tan, X., Sh en, L., & Shen, R. (2008). Why W eb 2.0 is good for learning and for research: principles and prototypes. WWW 0 8 Proceeding of the 17th international conference on World Wide Web (p. 705714). ACM. Retrieved from http://portal.acm.org/citation.cfm?id=1367593 Ulmer, R. R., & Sellnow, T. L. (2000). Consis tent questions of ambiguity in organizational crisis communication: Jack in the box as a case study. Journal of Business Ethics, 25 (2), 143155.
208 Verbeke, W. (2005). Consumer acceptance of functional foods: sociodemographic, cognitive and attitudinal deter minants. Food Quality and Preference, 16(1), 45 57. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S0950329304000102 Vincent, R. C., & Basil, M. D. (1997). College s tudents news gratifications, media use, and current events knowledge. Journal of Broadcasting & Electronic Media, 41(3), 380392. doi: 10.1111/j.13652621.2005.01010.x Wal l, M. (2004). Blogs as Black Market Journalism: A New Paradigm for News Berglu nd Center for Internet Studies. Retrieved from http://bcis.pacificu.edu /journal/2004/02/wall.php Wardrip Fruin, N., & Montfort, N. (2003). New media reader Cambridge, MA: MIT Press. Whaley, S R., & Tucker, M. The influence of perceived food risk and source trust on media system dependency. Journal of Applied Communications, 88 (1), 9 27. Wicks, R. H., & Drew, D. G. (1991). Learning from news: Effects of message consistency and medium on recall and inference making. Journalism Quarterly, 68(1 2), 155164. Williams, C. M. Nutritional quality of organic food: Shades of grey or shades of green? Proceedings of the Nutrition Society, 61, 19 24. doi: 10.1079/PNS2001126 Williams, P. R. D., & Hammitt, J K. (2001). Perceived risks of conventional and organic produce: Pesticides, pathogens, and natural toxins. Risk Analysis, 21 (2), 319 330. Witte, K. (1998). Fear as motivator, fear as inhibitor: Using the extended parallel process model to explain fear ap peal successes and failures. In P. A. Anderson, & L. K. Guerro (Eds.), Handbook of Communication and Emotion: Research, Theory, Applications, and Contexts (pp. 423450). San Diego, CA: Academic Publishers. Witte, K. (1992). Putting the fear back into fear appeals: The extended parallel process model. Communication Monographs, 59(4), 329349. doi: 10.1080/03637759209376276 Witte, K., & Allen, M. (2000). A metaanalysis of fear appeals: Implications for effective public health campaigns. Health Education & Behavior, 27(5), 591615. doi: 10.1177/109019810002700506 Wittenberger, K., & Dohlman, E. (2010). Peanut outlook: Impacts of the 200809 foodborne illness outbreak linked to salmonella in peanuts U.S. Department of Agriculture, Economic Research Service report OCS 10a01. Retrieved from http://www.usda.gov Woloschuk, W., Harasym, P. H., & Temple, W. (2004). Attitude change during medical school: A cohort study. Medical Education, 38 522 534.
209 Wood, M. L. M. (2006). Expert ise and prior attitude: Exploring new moderators in the context of inoculation research. Unpublished doctoral dissertation, University of Wisconsin, Madison. Wood, W. (1982). Retrieval of attituderelevant information from memory: Effects on susceptibilit y to persuasion and to intrinsic motivation. Journal of Personality and Social Psychology, 42(5), 798 810. doi: 10.1037/002235126.96.36.1998 Yiridoe, E. K., Bonti Ankomah, S., & Martin, R. C. (2005). Comparison of consumer perceptions and preference toward organic versus conventionally produced foods: A review and update of the literature. Renewable Agriculture and Food Systems 20(4), 193 205. Retrieved from http://journals.cambridge.org/abstract _S1742170505000220 Zaichkowsky, J. L. (1994). The Personal Involvement Inventory: Reduction, revision, and application to advertising. Journal of Advertising, 23(4), 59 70. Zillmann, D., Chen, L., Knobloch, S., & Callison, C. (2004). Effects of Lead Framing on Selective Exposure to Internet News Reports. Communication Research, 31(1), 5881. Retrieved from http://crx.sagepub.com/cgi/content/abstract/31/1/58 Zumalt, J. R. (200 4). A primer on agricultural communications for student, librarians and researchers, Journal of Agricultural & Food Information, 5 (1), 25 33. DOI: 10.1300/J108v05n01_05 Zumalt, J. R. (2004). Identifying the core periodical literature of the Agricultural Com munications Documentation Center. Journal of Agricultural & Food Information, 8(3), 43 63.
210 BIOGRAPHICAL SKETCH Karen Cannon was born and raised in Burlingame, California, a suburb of San Francisco, and is the daughter of two professional educators. She attended Colorado State University and earned a bachelors degree in agricultural and natural resources journalism, as well as a masters degree in agricultural extension education with a focus in communication. Following her masters program, Karen went to work for the U.S. Department of Agricultures National Agricultural Statistics Service in Washington, DC, as a public affairs specialist and gained valuable experience in translating agricultural and statistical information into lay terms. Returning to Colorado after her stint in the nations capital, Karen worked as a communications management trainee at Swift & Company, the thirdlargest beef and pork processing company in the U.S. In this role, she worked closely with national and international media outlets to explain U.S. beef trade export rules related to concerns about bovine spongiform encephalopathy (BSE, or mad cow disease), and to promote the companys new research and development facility as well as its philanthropic efforts. Continuing in agric ultural public relations, she then took a position working for the National Cattlemens Beef Association as a beef safety public relations manager. Following her professional experience, Karen returned to academic life as a doctoral student in the Department of Agricultural Education and Communication at the University of Florida. In December 2010, Karen successfully defended her dissertation and will graduate with her doctoral degree in agricultural communication in May 2011. Upon graduation, Karen hopes t o begin her career as a professor of agricultural communications at a landgrant university where she is able to combine her love of teaching, agriculture, public relations, and research.